Applied Evidence

CASE 1

Tyler M, age 40, otherwise healthy, and with a body mass index (BMI) of 30, presents to your office for his annual physical examination. He does not have a history of alcohol or tobacco use.

Mr. M’s obesity raises concern about metabolic syndrome, which warrants evaluation for hypertriglyceridemia (HTG). You offer him lipid testing to estimate his risk of atherosclerotic cardiovascular disease (ASCVD).

The only abnormal value on the lipid panel is a triglyceride (TG) level of 264 mg/dL (normal, < 175 mg/dL). Mr. M’s 10-yr ASCVD risk is determined to be < 5%.

What, if any, intervention would be triggered by the finding of moderate HTG?

© ALICIA BUELOW

CASE 2

Alicia F, age 30, with a BMI of 28 and ASCVD risk < 7.5%, comes to the clinic for evaluation of anxiety and insomnia. She reports eating a high-carbohydrate diet and drinking 3 to 5 alcoholic beverages nightly to help her sleep.

Ms. F’s daily alcohol use prompts evaluation for HTG. Results show a TG level of 1300 mg/dL and a high-density lipoprotein (HDL) level of 25 mg/dL (healthy HDL levels: adult females, ≥ 50 mg/dL; adult males, ≥ 40 mg/dL). Other test results are normal, except for elevated transaminase levels (just under twice normal).

What, if any, action would be prompted by the patient’s severe HTG and below-normal HDL level?

Continue to: How HTG is defined

HTG is most commonly caused by obesity and a sedentary lifestyle; certain associated comorbid medical conditions can also be a precipitant (Table 1^1,2). Because the condition is a result of polygenic phenotypic expression, even a genetically low-risk patient can present with HTG when exposed to certain medical conditions and environmental causes.

Primary HTG (genetic or familial) is rare. Genetic testing may be considered for patients with TG > 1000 mg/dL (severely elevated TG = 500 to 1999 mg/dL, measured in fasting state*) or a family history of early ASCVD (TABLE 1^1,2).^2,3

Even a genetically low-risk patient can present with hypertriglyceridemia when exposed to certain medical conditions and environmental causes.

Typically, HTG is asymptomatic. Xanthelasmas, xanthomas, and lipemia retinalis are found in hereditary disorders of elevated TGs. Occasionally, HTG manifests as chylomicronemia syndrome, characterized by recurrent abdominal pain, nausea, vomiting, and, in severe HTG, pancreatitis.³

Fine points of TG measurement

Triglycerides are a component of a complete lipid profile, which also includes total cholesterol, calculated low-density lipoprotein (LDL-C), and HDL.⁴ As in both case vignettes, detection of HTG is often incidental, when a lipid profile is ordered to evaluate the risk of ASCVD. (Of note, for people older than 20 years, the US Preventive Services Task Force no longer addresses the question, “Which population should be screened for dyslipidemia?” Instead, current recommendations answer the question, “For which population should statin therapy be prescribed?”⁵)

Effect on ASCVD risk assessment. TG levels are known to vary, depending on fasting or nonfasting status, with lower levels reported when fasting. An elevated TG level can lead to inaccurate calculation of LDL when using the Friedewald formula⁶:

LDL = total cholesterol – (triglycerides/5) – HDL

Continue to: The purpose of testing...

The purpose of testing lipids in a fasting state (> 9 hours) is to minimize the effects of an elevated TG level on the calculated LDL. In severe HTG, beta-quantitation by ultracentrifugation and electrophoresis can be performed to determine the LDL level directly.

Advantage of nonfasting measurement. When LDL-C is not a concern, there is, in fact, value in measuring TGs in the nonfasting state. Why? Because a nonfasting TG level is a better indicator of a patient’s average TG status: Studies have found a higher ASCVD risk in the setting of an elevated postprandial TG level accompanied by a low HDL level.⁷

Studies have found a higher ASCVD risk in the setting of an elevated postprandial triglyceride level accompanied by a low HDL level.

The Copenhagen City Heart Study identified postprandial HTG as an independent risk factor for atherogenicity, even in the setting of a normal fasting TG level.⁸ American Association of Clinical Endocrinologists and American College of Endocrinology guidelines endorse testing the nonfasting TG level when the fasting TG level is elevated in a lipid profile; if the nonfasting TG level is > 500 mg/dL, evaluation for secondary causes is warranted^9,10 (Table 1^1,2).

In a practical sense, therefore, offering patients nonfasting lipid testing allows more people to obtain access to timely care.

Pancreatitis. Acute pancreatitis commonly prompts an evaluation for HTG. The risk of acute pancreatitis in the general population is 0.04%, but that risk increases to 8% to 31% for a person with HTG.¹¹ Incidence when the TG level is > 500 mg/dL is thought to be increased because chylomicrons, acting as a TG carrier in the bloodstream, are responsible for pancreatitis.³ Treating HTG can reduce both the risk and recurrence of pancreatitis^12,13; given that the postprandial TG level can change rapidly from severe to very severe (> 2000 mg/dL), multiple guidelines recommend pharmacotherapy to a TG goal of < 500-1000 mg/dL.^1,9,13,14

Continue to: An ASCVD risk-HTG connection?

An ASCVD risk–HTG connection? In the population already at higher risk of ASCVD (> 7.5%), HTG is recognized as a risk-enhancing factor because of its atherogenic potential (Table 2²); however, there is insufficient evidence that TGs have a role as an independent risk factor for ASCVD. In a population-based study of 58,000 people, 40 to 65 years of age, conducted at Copenhagen [Denmark] University Hospital, investigators found that those who did not meet criteria for statin treatment and who had a TG level > 264 mg/dL had a 10-year risk of a major adverse cardiovascular event similar to that of people who did meet criteria for statin therapy.¹⁵

The FIELD (Fenofibrate Intervention and Event Lowering in Diabetes) and AIM-HIGH (Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides and Impact on Global Health Outcomes) studies, among others, have failed to show a significant reduction in coronary events by treating HTG.¹⁰

That said, it’s worth considering the findings of other trials:

• In the PROVE IT-TIMI 22 (Pravastatin or Atorvastatin Evaluation and Infection Therapy–Thrombolysis in Myocardial Infarction 22) trial, an overall 28% reduction in endpoint events (myocardial infarction, acute coronary syndrome) was seen with high-intensity statin therapy, compared to moderate-intensity therapy.¹⁰ However, there was a sizeable residual risk identified that was theorized by investigators to be associated with high non-HDL lipoproteins, including TGs.
• A 2016 study in Israel, in which 22 years of data on 15,355 patients with established ASCVD were studied, revealed that elevated TGs are associated with an increased long-term mortality risk that is independent of the HDL level.¹⁶
• A cross-sectional study, nested in the prospective Copenhagen City Heart Study, demonstrated that HTG is associated with an increase in ischemic stroke events.¹⁷

Treatment

Therapeutic lifestyle changes

Changes in lifestyle are the foundation of management of, and recommended first-line treatment for, all patients with HTG. Patients with a moderately elevated TG level (175-499 mg/dL, measured in a fasting or nonfasting state) can be treated with therapeutic lifestyle changes alone^1,2; a trial of 3 to 6 months (see specific interventions below) is recommended before considering adding medications.¹⁰

Weight loss. There is a strong association between BMI > 30 and HTG. Visceral adiposity is a much more significant risk than subcutaneous adipose tissue. Although weight loss to an ideal range is recommended, even a 10% to 15% reduction in an obese patient can reduce the TG level by 20%. A combination of moderate-intensity exercise and healthy eating habits appears to assist best with this intervention.¹⁸

Continue to: Exercise

Exercise. Thirty minutes a day of moderate-intensity exercise is associated with a significant drop in postprandial TG. This benefit can last as long as 3 days, suggesting a goal of at least 3 days a week of an active lifestyle. Such a program can include intermittent aerobics or mild resistance exercise.¹⁹

Healthy eating habits. The difference between a low-fat, high-carbohydrate diet and a high-fat, low-carbohydrate diet is less important than the overall benefit of weight loss from either of these diets. Complex carbohydrates are recommended over simple carbohydrates. A low-carbohydrate diet in a patient with diabetes has been demonstrated to improve the TG level, irrespective of weight change.²⁰

The risk of acute pancreatitis in the general population is 0.04%, but that risk increases to 8% to 31% for a person with hypertriglyceridemia

A Mediterranean diet can reduce the TG level by 10% to 15%, and is recommended over a low-fat diet.¹⁴ (This diet generally includes a high intake of extra virgin olive oil; leafy green vegetables, fruits, cereals, nuts, and legumes; moderate intake of fish and other meat, dairy products, and red wine; and low intake of eggs and sugars.) The American Heart Association recommends 2 servings of fatty fish a week for its omega-3 oil benefit of reducing ASCVD risk. Working with a registered dietician to assist with lipid lowering can produce better results than physician-only instruction on healthy eating.⁹

Alcohol consumption. Complete cessation or moderation of alcohol consumption (1 drink a day/women and 2 drinks a day/men*) is recommended to improve HTG. Among secondary factors, alcohol is commonly the cause of an unusually high elevation of the TG level.¹⁴

Smoking cessation. Smoking increases the postprandial TG level.¹⁰ Complete cessation for just 1 year can reduce a person’s ASCVD risk by approximately 50%. However, in a clinical trial,²² smoking cessation did not significantly decrease the TG level—possibly because of the counterbalancing effect of weight gain following cessation.

Continue to: Medical therapy

In addition to lifestyle modification, medications are recommended to reduce atherogenic potential in patients with moderate or severe HTG and an ASCVD risk > 7.5% (Table 3^4,13,18,23 and Table 4^2,9). Before initiating medical therapy, we recommend that you engage in shared decision-making with patients to (1) delineate treatment goals and (2) describe the risks and benefits of medications for HTG.²

Statins. These agents are recommended first-line therapy for reducing ASCVD risk.² If the TG level remains elevated (> 500 mg/dL) after statin therapy is maximized, an additional agent can be added—ie, a fibrate or fish oil (see below).

Fibrates. If a fibrate is used as an add-on to a statin, fenofibrate is preferred over gemfibrozil because it presents less risk of the severe myopathy that can develop when taken with a statin.¹³ Despite the effectiveness of fibrates in reducing the TG level, these drugs have not been shown to reduce overall mortality.²⁴ The evidence on improved cardiovascular outcomes is subgroup-specific (ie, prevention of a second myocardial infarction in the setting of optimal statin use and elevated non-HDL lipoproteins).¹² A study demonstrated that gemfibrozil reduced the incidence of transient ischemic attack and stroke in a subgroup of male US veterans who had coronary artery disease and a low HDL level.²⁵

Fish oil. The omega-3 ethyl esters eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), available as EPA alone or in combination with DHA, do not interact with statins and are tolerated well. They reduce hypertriglyceridemia by 20% to 50%.¹³

Eicosapentaenoic acid, EPA plus DHA, and icosapent ethyl, an ethyl ester product containing EPA without DHA, are approved by the US Food and Drug Administration for HTG > 500 mg/dL, at a dosage of 2000 mg twice daily. In the REDUCE-IT trial, adding icosapent ethyl, 2 g twice daily, to a statin in patients with HTG was associated with fewer ischemic events, compared to placebo.^23,26

Continue to: Fish oil formulations...

Fish oil formulations can inhibit platelet aggregation and increase bleeding time in otherwise healthy people; however, such episodes are minor and nonfatal. Patients on anticoagulation or an antiplatelet medication should be monitored periodically for bleeding events, although recommendations on how to monitor aren’t specified in a recent advisory by the American Heart Association.²³

There is a strong association between a BMI > 30 and hypertriglyceridemia.

DHA was thought to increase the LDL-C levels and, by doing so, potentially counterbalance benefit,^23,27 but most studies have failed to reproduce this effect.²⁸ Instead, studies have shown minimal elevation of LDL-C when DHA is used to treat HTG.^23,27

Niacin. At a dosage of 500-2000 mg/dL, niacin lowers the TG level by 10% to 30%. It also increases HDL by 10% to 40% and lowers LDL by 5% to 20%.¹³

Considerations in pancreatitis. For management of recurrent pancreatitis in patients with HTG, lifestyle modification remains the mainstay of treatment. When medication is considered for persistent severe HTG, fibrates have evidence of primary and secondary prevention of pancreatitis.¹¹ When a patient is intolerant of fibrates, consider a different option to reduce the TG level—eg, fish oil supplementation.

CASE 1

Recommendation for Mr. M: Therapeutic lifestyle changes to address moderate HTG.

Continue to: Because Mr. M's...

Because Mr. M’s 10-yr ASCVD risk is < 5%, statin therapy is not indicated for risk reduction. With a fasting TG value < 500 mg/dL, he is not considered at increased risk of pancreatitis.

CASE 2

Recommendations for Ms. F:

• Therapeutic lifestyle changes to address severe HTG. Ms. F agrees to wean off alcohol; add relaxation exercises before bedtime; do aerobic exercise 30 minutes a day, 3 times a week; decrease dietary carbohydrates daily by cutting portion size in half; and increase intake of fresh vegetables and lean protein.
• Treatment with fenofibrate to reduce the risk of pancreatitis. Ms. F begins a trial. Six months into treatment, she has reduced her BMI to 24 and the TG level has fallen to < 500 mg/dL. Ms. F also reports that she is sleeping well, believes that she is able to manage her infrequent anxiety, and is now in a routine that feels sustainable.

You congratulate Ms. F on her success and support her decision to undertake a trial of discontinuing fenofibrate, after shared decision-making about the risks and potential benefits of doing so.

Summing up: Management of HTG

Keep these treatment strategy highlights in mind:

• Lifestyle modification with a low-fat, low-carbohydrate diet, avoidance of alcohol, and moderate-intensity exercise is the mainstay of HTG management.
• The latest evidence supports that (1) HTG is a risk-enhancing factor for ­ASCVD and (2) statin therapy is recommended for patients who have HTG and an ASCVD risk > 7.5%.
• When the TG level remains elevated despite statin therapy and lifestyle changes, an omega-3 ethyl ester can be used as an adjunct for additional atherogenic risk reduction.
• For severe HTG, a regimen of therapeutic lifestyle changes plus a fibrate is recommended to reduce the risk and recurrence of pancreatitis.^1,24

* In comparison, a normal level of triglycerides is < 175 mg/dL; a moderately elevated level, measured in a fasting or nonfasting state, 175-499 mg/dL; and a very severely elevated level, ≥ 2000 mg/dL.²

CORRESPONDENCE
Ashwini Kamath Mulki, MD, Family Health Center, 1730 Chew Street, Allentown, PA 18104; Ashwini.KamathMulki@lvhn.org.

CASE 1

Tyler M, age 40, otherwise healthy, and with a body mass index (BMI) of 30, presents to your office for his annual physical examination. He does not have a history of alcohol or tobacco use.

Mr. M’s obesity raises concern about metabolic syndrome, which warrants evaluation for hypertriglyceridemia (HTG). You offer him lipid testing to estimate his risk of atherosclerotic cardiovascular disease (ASCVD).

The only abnormal value on the lipid panel is a triglyceride (TG) level of 264 mg/dL (normal, < 175 mg/dL). Mr. M’s 10-yr ASCVD risk is determined to be < 5%.

What, if any, intervention would be triggered by the finding of moderate HTG?

© ALICIA BUELOW

CASE 2

Alicia F, age 30, with a BMI of 28 and ASCVD risk < 7.5%, comes to the clinic for evaluation of anxiety and insomnia. She reports eating a high-carbohydrate diet and drinking 3 to 5 alcoholic beverages nightly to help her sleep.

Ms. F’s daily alcohol use prompts evaluation for HTG. Results show a TG level of 1300 mg/dL and a high-density lipoprotein (HDL) level of 25 mg/dL (healthy HDL levels: adult females, ≥ 50 mg/dL; adult males, ≥ 40 mg/dL). Other test results are normal, except for elevated transaminase levels (just under twice normal).

What, if any, action would be prompted by the patient’s severe HTG and below-normal HDL level?

Continue to: How HTG is defined

HTG is most commonly caused by obesity and a sedentary lifestyle; certain associated comorbid medical conditions can also be a precipitant (Table 1^1,2). Because the condition is a result of polygenic phenotypic expression, even a genetically low-risk patient can present with HTG when exposed to certain medical conditions and environmental causes.

Primary HTG (genetic or familial) is rare. Genetic testing may be considered for patients with TG > 1000 mg/dL (severely elevated TG = 500 to 1999 mg/dL, measured in fasting state*) or a family history of early ASCVD (TABLE 1^1,2).^2,3

Even a genetically low-risk patient can present with hypertriglyceridemia when exposed to certain medical conditions and environmental causes.

Typically, HTG is asymptomatic. Xanthelasmas, xanthomas, and lipemia retinalis are found in hereditary disorders of elevated TGs. Occasionally, HTG manifests as chylomicronemia syndrome, characterized by recurrent abdominal pain, nausea, vomiting, and, in severe HTG, pancreatitis.³

Fine points of TG measurement

Triglycerides are a component of a complete lipid profile, which also includes total cholesterol, calculated low-density lipoprotein (LDL-C), and HDL.⁴ As in both case vignettes, detection of HTG is often incidental, when a lipid profile is ordered to evaluate the risk of ASCVD. (Of note, for people older than 20 years, the US Preventive Services Task Force no longer addresses the question, “Which population should be screened for dyslipidemia?” Instead, current recommendations answer the question, “For which population should statin therapy be prescribed?”⁵)

Effect on ASCVD risk assessment. TG levels are known to vary, depending on fasting or nonfasting status, with lower levels reported when fasting. An elevated TG level can lead to inaccurate calculation of LDL when using the Friedewald formula⁶:

LDL = total cholesterol – (triglycerides/5) – HDL

Continue to: The purpose of testing...

The purpose of testing lipids in a fasting state (> 9 hours) is to minimize the effects of an elevated TG level on the calculated LDL. In severe HTG, beta-quantitation by ultracentrifugation and electrophoresis can be performed to determine the LDL level directly.

Advantage of nonfasting measurement. When LDL-C is not a concern, there is, in fact, value in measuring TGs in the nonfasting state. Why? Because a nonfasting TG level is a better indicator of a patient’s average TG status: Studies have found a higher ASCVD risk in the setting of an elevated postprandial TG level accompanied by a low HDL level.⁷

Studies have found a higher ASCVD risk in the setting of an elevated postprandial triglyceride level accompanied by a low HDL level.

The Copenhagen City Heart Study identified postprandial HTG as an independent risk factor for atherogenicity, even in the setting of a normal fasting TG level.⁸ American Association of Clinical Endocrinologists and American College of Endocrinology guidelines endorse testing the nonfasting TG level when the fasting TG level is elevated in a lipid profile; if the nonfasting TG level is > 500 mg/dL, evaluation for secondary causes is warranted^9,10 (Table 1^1,2).

In a practical sense, therefore, offering patients nonfasting lipid testing allows more people to obtain access to timely care.

Pancreatitis. Acute pancreatitis commonly prompts an evaluation for HTG. The risk of acute pancreatitis in the general population is 0.04%, but that risk increases to 8% to 31% for a person with HTG.¹¹ Incidence when the TG level is > 500 mg/dL is thought to be increased because chylomicrons, acting as a TG carrier in the bloodstream, are responsible for pancreatitis.³ Treating HTG can reduce both the risk and recurrence of pancreatitis^12,13; given that the postprandial TG level can change rapidly from severe to very severe (> 2000 mg/dL), multiple guidelines recommend pharmacotherapy to a TG goal of < 500-1000 mg/dL.^1,9,13,14

Continue to: An ASCVD risk-HTG connection?

An ASCVD risk–HTG connection? In the population already at higher risk of ASCVD (> 7.5%), HTG is recognized as a risk-enhancing factor because of its atherogenic potential (Table 2²); however, there is insufficient evidence that TGs have a role as an independent risk factor for ASCVD. In a population-based study of 58,000 people, 40 to 65 years of age, conducted at Copenhagen [Denmark] University Hospital, investigators found that those who did not meet criteria for statin treatment and who had a TG level > 264 mg/dL had a 10-year risk of a major adverse cardiovascular event similar to that of people who did meet criteria for statin therapy.¹⁵

The FIELD (Fenofibrate Intervention and Event Lowering in Diabetes) and AIM-HIGH (Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides and Impact on Global Health Outcomes) studies, among others, have failed to show a significant reduction in coronary events by treating HTG.¹⁰

That said, it’s worth considering the findings of other trials:

• In the PROVE IT-TIMI 22 (Pravastatin or Atorvastatin Evaluation and Infection Therapy–Thrombolysis in Myocardial Infarction 22) trial, an overall 28% reduction in endpoint events (myocardial infarction, acute coronary syndrome) was seen with high-intensity statin therapy, compared to moderate-intensity therapy.¹⁰ However, there was a sizeable residual risk identified that was theorized by investigators to be associated with high non-HDL lipoproteins, including TGs.
• A 2016 study in Israel, in which 22 years of data on 15,355 patients with established ASCVD were studied, revealed that elevated TGs are associated with an increased long-term mortality risk that is independent of the HDL level.¹⁶
• A cross-sectional study, nested in the prospective Copenhagen City Heart Study, demonstrated that HTG is associated with an increase in ischemic stroke events.¹⁷

Treatment

Therapeutic lifestyle changes

Changes in lifestyle are the foundation of management of, and recommended first-line treatment for, all patients with HTG. Patients with a moderately elevated TG level (175-499 mg/dL, measured in a fasting or nonfasting state) can be treated with therapeutic lifestyle changes alone^1,2; a trial of 3 to 6 months (see specific interventions below) is recommended before considering adding medications.¹⁰

Weight loss. There is a strong association between BMI > 30 and HTG. Visceral adiposity is a much more significant risk than subcutaneous adipose tissue. Although weight loss to an ideal range is recommended, even a 10% to 15% reduction in an obese patient can reduce the TG level by 20%. A combination of moderate-intensity exercise and healthy eating habits appears to assist best with this intervention.¹⁸

Continue to: Exercise

Exercise. Thirty minutes a day of moderate-intensity exercise is associated with a significant drop in postprandial TG. This benefit can last as long as 3 days, suggesting a goal of at least 3 days a week of an active lifestyle. Such a program can include intermittent aerobics or mild resistance exercise.¹⁹

Healthy eating habits. The difference between a low-fat, high-carbohydrate diet and a high-fat, low-carbohydrate diet is less important than the overall benefit of weight loss from either of these diets. Complex carbohydrates are recommended over simple carbohydrates. A low-carbohydrate diet in a patient with diabetes has been demonstrated to improve the TG level, irrespective of weight change.²⁰

The risk of acute pancreatitis in the general population is 0.04%, but that risk increases to 8% to 31% for a person with hypertriglyceridemia

A Mediterranean diet can reduce the TG level by 10% to 15%, and is recommended over a low-fat diet.¹⁴ (This diet generally includes a high intake of extra virgin olive oil; leafy green vegetables, fruits, cereals, nuts, and legumes; moderate intake of fish and other meat, dairy products, and red wine; and low intake of eggs and sugars.) The American Heart Association recommends 2 servings of fatty fish a week for its omega-3 oil benefit of reducing ASCVD risk. Working with a registered dietician to assist with lipid lowering can produce better results than physician-only instruction on healthy eating.⁹

Alcohol consumption. Complete cessation or moderation of alcohol consumption (1 drink a day/women and 2 drinks a day/men*) is recommended to improve HTG. Among secondary factors, alcohol is commonly the cause of an unusually high elevation of the TG level.¹⁴

Smoking cessation. Smoking increases the postprandial TG level.¹⁰ Complete cessation for just 1 year can reduce a person’s ASCVD risk by approximately 50%. However, in a clinical trial,²² smoking cessation did not significantly decrease the TG level—possibly because of the counterbalancing effect of weight gain following cessation.

Continue to: Medical therapy

In addition to lifestyle modification, medications are recommended to reduce atherogenic potential in patients with moderate or severe HTG and an ASCVD risk > 7.5% (Table 3^4,13,18,23 and Table 4^2,9). Before initiating medical therapy, we recommend that you engage in shared decision-making with patients to (1) delineate treatment goals and (2) describe the risks and benefits of medications for HTG.²

Statins. These agents are recommended first-line therapy for reducing ASCVD risk.² If the TG level remains elevated (> 500 mg/dL) after statin therapy is maximized, an additional agent can be added—ie, a fibrate or fish oil (see below).

Fibrates. If a fibrate is used as an add-on to a statin, fenofibrate is preferred over gemfibrozil because it presents less risk of the severe myopathy that can develop when taken with a statin.¹³ Despite the effectiveness of fibrates in reducing the TG level, these drugs have not been shown to reduce overall mortality.²⁴ The evidence on improved cardiovascular outcomes is subgroup-specific (ie, prevention of a second myocardial infarction in the setting of optimal statin use and elevated non-HDL lipoproteins).¹² A study demonstrated that gemfibrozil reduced the incidence of transient ischemic attack and stroke in a subgroup of male US veterans who had coronary artery disease and a low HDL level.²⁵

Fish oil. The omega-3 ethyl esters eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), available as EPA alone or in combination with DHA, do not interact with statins and are tolerated well. They reduce hypertriglyceridemia by 20% to 50%.¹³

Eicosapentaenoic acid, EPA plus DHA, and icosapent ethyl, an ethyl ester product containing EPA without DHA, are approved by the US Food and Drug Administration for HTG > 500 mg/dL, at a dosage of 2000 mg twice daily. In the REDUCE-IT trial, adding icosapent ethyl, 2 g twice daily, to a statin in patients with HTG was associated with fewer ischemic events, compared to placebo.^23,26

Continue to: Fish oil formulations...

Fish oil formulations can inhibit platelet aggregation and increase bleeding time in otherwise healthy people; however, such episodes are minor and nonfatal. Patients on anticoagulation or an antiplatelet medication should be monitored periodically for bleeding events, although recommendations on how to monitor aren’t specified in a recent advisory by the American Heart Association.²³

There is a strong association between a BMI > 30 and hypertriglyceridemia.

DHA was thought to increase the LDL-C levels and, by doing so, potentially counterbalance benefit,^23,27 but most studies have failed to reproduce this effect.²⁸ Instead, studies have shown minimal elevation of LDL-C when DHA is used to treat HTG.^23,27

Niacin. At a dosage of 500-2000 mg/dL, niacin lowers the TG level by 10% to 30%. It also increases HDL by 10% to 40% and lowers LDL by 5% to 20%.¹³

Considerations in pancreatitis. For management of recurrent pancreatitis in patients with HTG, lifestyle modification remains the mainstay of treatment. When medication is considered for persistent severe HTG, fibrates have evidence of primary and secondary prevention of pancreatitis.¹¹ When a patient is intolerant of fibrates, consider a different option to reduce the TG level—eg, fish oil supplementation.

CASE 1

Recommendation for Mr. M: Therapeutic lifestyle changes to address moderate HTG.

Continue to: Because Mr. M's...

Because Mr. M’s 10-yr ASCVD risk is < 5%, statin therapy is not indicated for risk reduction. With a fasting TG value < 500 mg/dL, he is not considered at increased risk of pancreatitis.

CASE 2

Recommendations for Ms. F:

• Therapeutic lifestyle changes to address severe HTG. Ms. F agrees to wean off alcohol; add relaxation exercises before bedtime; do aerobic exercise 30 minutes a day, 3 times a week; decrease dietary carbohydrates daily by cutting portion size in half; and increase intake of fresh vegetables and lean protein.
• Treatment with fenofibrate to reduce the risk of pancreatitis. Ms. F begins a trial. Six months into treatment, she has reduced her BMI to 24 and the TG level has fallen to < 500 mg/dL. Ms. F also reports that she is sleeping well, believes that she is able to manage her infrequent anxiety, and is now in a routine that feels sustainable.

You congratulate Ms. F on her success and support her decision to undertake a trial of discontinuing fenofibrate, after shared decision-making about the risks and potential benefits of doing so.

Summing up: Management of HTG

Keep these treatment strategy highlights in mind:

• Lifestyle modification with a low-fat, low-carbohydrate diet, avoidance of alcohol, and moderate-intensity exercise is the mainstay of HTG management.
• The latest evidence supports that (1) HTG is a risk-enhancing factor for ­ASCVD and (2) statin therapy is recommended for patients who have HTG and an ASCVD risk > 7.5%.
• When the TG level remains elevated despite statin therapy and lifestyle changes, an omega-3 ethyl ester can be used as an adjunct for additional atherogenic risk reduction.
• For severe HTG, a regimen of therapeutic lifestyle changes plus a fibrate is recommended to reduce the risk and recurrence of pancreatitis.^1,24

* In comparison, a normal level of triglycerides is < 175 mg/dL; a moderately elevated level, measured in a fasting or nonfasting state, 175-499 mg/dL; and a very severely elevated level, ≥ 2000 mg/dL.²

CORRESPONDENCE
Ashwini Kamath Mulki, MD, Family Health Center, 1730 Chew Street, Allentown, PA 18104; Ashwini.KamathMulki@lvhn.org.

CASE 1

Tyler M, age 40, otherwise healthy, and with a body mass index (BMI) of 30, presents to your office for his annual physical examination. He does not have a history of alcohol or tobacco use.

Mr. M’s obesity raises concern about metabolic syndrome, which warrants evaluation for hypertriglyceridemia (HTG). You offer him lipid testing to estimate his risk of atherosclerotic cardiovascular disease (ASCVD).

The only abnormal value on the lipid panel is a triglyceride (TG) level of 264 mg/dL (normal, < 175 mg/dL). Mr. M’s 10-yr ASCVD risk is determined to be < 5%.

What, if any, intervention would be triggered by the finding of moderate HTG?

© ALICIA BUELOW

CASE 2

Alicia F, age 30, with a BMI of 28 and ASCVD risk < 7.5%, comes to the clinic for evaluation of anxiety and insomnia. She reports eating a high-carbohydrate diet and drinking 3 to 5 alcoholic beverages nightly to help her sleep.

Ms. F’s daily alcohol use prompts evaluation for HTG. Results show a TG level of 1300 mg/dL and a high-density lipoprotein (HDL) level of 25 mg/dL (healthy HDL levels: adult females, ≥ 50 mg/dL; adult males, ≥ 40 mg/dL). Other test results are normal, except for elevated transaminase levels (just under twice normal).

What, if any, action would be prompted by the patient’s severe HTG and below-normal HDL level?

Continue to: How HTG is defined

HTG is most commonly caused by obesity and a sedentary lifestyle; certain associated comorbid medical conditions can also be a precipitant (Table 1^1,2). Because the condition is a result of polygenic phenotypic expression, even a genetically low-risk patient can present with HTG when exposed to certain medical conditions and environmental causes.

Primary HTG (genetic or familial) is rare. Genetic testing may be considered for patients with TG > 1000 mg/dL (severely elevated TG = 500 to 1999 mg/dL, measured in fasting state*) or a family history of early ASCVD (TABLE 1^1,2).^2,3

Even a genetically low-risk patient can present with hypertriglyceridemia when exposed to certain medical conditions and environmental causes.

Typically, HTG is asymptomatic. Xanthelasmas, xanthomas, and lipemia retinalis are found in hereditary disorders of elevated TGs. Occasionally, HTG manifests as chylomicronemia syndrome, characterized by recurrent abdominal pain, nausea, vomiting, and, in severe HTG, pancreatitis.³

Fine points of TG measurement

Triglycerides are a component of a complete lipid profile, which also includes total cholesterol, calculated low-density lipoprotein (LDL-C), and HDL.⁴ As in both case vignettes, detection of HTG is often incidental, when a lipid profile is ordered to evaluate the risk of ASCVD. (Of note, for people older than 20 years, the US Preventive Services Task Force no longer addresses the question, “Which population should be screened for dyslipidemia?” Instead, current recommendations answer the question, “For which population should statin therapy be prescribed?”⁵)

Effect on ASCVD risk assessment. TG levels are known to vary, depending on fasting or nonfasting status, with lower levels reported when fasting. An elevated TG level can lead to inaccurate calculation of LDL when using the Friedewald formula⁶:

LDL = total cholesterol – (triglycerides/5) – HDL

Continue to: The purpose of testing...

The purpose of testing lipids in a fasting state (> 9 hours) is to minimize the effects of an elevated TG level on the calculated LDL. In severe HTG, beta-quantitation by ultracentrifugation and electrophoresis can be performed to determine the LDL level directly.

Advantage of nonfasting measurement. When LDL-C is not a concern, there is, in fact, value in measuring TGs in the nonfasting state. Why? Because a nonfasting TG level is a better indicator of a patient’s average TG status: Studies have found a higher ASCVD risk in the setting of an elevated postprandial TG level accompanied by a low HDL level.⁷

Studies have found a higher ASCVD risk in the setting of an elevated postprandial triglyceride level accompanied by a low HDL level.

The Copenhagen City Heart Study identified postprandial HTG as an independent risk factor for atherogenicity, even in the setting of a normal fasting TG level.⁸ American Association of Clinical Endocrinologists and American College of Endocrinology guidelines endorse testing the nonfasting TG level when the fasting TG level is elevated in a lipid profile; if the nonfasting TG level is > 500 mg/dL, evaluation for secondary causes is warranted^9,10 (Table 1^1,2).

In a practical sense, therefore, offering patients nonfasting lipid testing allows more people to obtain access to timely care.

Pancreatitis. Acute pancreatitis commonly prompts an evaluation for HTG. The risk of acute pancreatitis in the general population is 0.04%, but that risk increases to 8% to 31% for a person with HTG.¹¹ Incidence when the TG level is > 500 mg/dL is thought to be increased because chylomicrons, acting as a TG carrier in the bloodstream, are responsible for pancreatitis.³ Treating HTG can reduce both the risk and recurrence of pancreatitis^12,13; given that the postprandial TG level can change rapidly from severe to very severe (> 2000 mg/dL), multiple guidelines recommend pharmacotherapy to a TG goal of < 500-1000 mg/dL.^1,9,13,14

Continue to: An ASCVD risk-HTG connection?

An ASCVD risk–HTG connection? In the population already at higher risk of ASCVD (> 7.5%), HTG is recognized as a risk-enhancing factor because of its atherogenic potential (Table 2²); however, there is insufficient evidence that TGs have a role as an independent risk factor for ASCVD. In a population-based study of 58,000 people, 40 to 65 years of age, conducted at Copenhagen [Denmark] University Hospital, investigators found that those who did not meet criteria for statin treatment and who had a TG level > 264 mg/dL had a 10-year risk of a major adverse cardiovascular event similar to that of people who did meet criteria for statin therapy.¹⁵

The FIELD (Fenofibrate Intervention and Event Lowering in Diabetes) and AIM-HIGH (Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides and Impact on Global Health Outcomes) studies, among others, have failed to show a significant reduction in coronary events by treating HTG.¹⁰

That said, it’s worth considering the findings of other trials:

• In the PROVE IT-TIMI 22 (Pravastatin or Atorvastatin Evaluation and Infection Therapy–Thrombolysis in Myocardial Infarction 22) trial, an overall 28% reduction in endpoint events (myocardial infarction, acute coronary syndrome) was seen with high-intensity statin therapy, compared to moderate-intensity therapy.¹⁰ However, there was a sizeable residual risk identified that was theorized by investigators to be associated with high non-HDL lipoproteins, including TGs.
• A 2016 study in Israel, in which 22 years of data on 15,355 patients with established ASCVD were studied, revealed that elevated TGs are associated with an increased long-term mortality risk that is independent of the HDL level.¹⁶
• A cross-sectional study, nested in the prospective Copenhagen City Heart Study, demonstrated that HTG is associated with an increase in ischemic stroke events.¹⁷

Treatment

Therapeutic lifestyle changes

Changes in lifestyle are the foundation of management of, and recommended first-line treatment for, all patients with HTG. Patients with a moderately elevated TG level (175-499 mg/dL, measured in a fasting or nonfasting state) can be treated with therapeutic lifestyle changes alone^1,2; a trial of 3 to 6 months (see specific interventions below) is recommended before considering adding medications.¹⁰

Weight loss. There is a strong association between BMI > 30 and HTG. Visceral adiposity is a much more significant risk than subcutaneous adipose tissue. Although weight loss to an ideal range is recommended, even a 10% to 15% reduction in an obese patient can reduce the TG level by 20%. A combination of moderate-intensity exercise and healthy eating habits appears to assist best with this intervention.¹⁸

Continue to: Exercise

Exercise. Thirty minutes a day of moderate-intensity exercise is associated with a significant drop in postprandial TG. This benefit can last as long as 3 days, suggesting a goal of at least 3 days a week of an active lifestyle. Such a program can include intermittent aerobics or mild resistance exercise.¹⁹

Healthy eating habits. The difference between a low-fat, high-carbohydrate diet and a high-fat, low-carbohydrate diet is less important than the overall benefit of weight loss from either of these diets. Complex carbohydrates are recommended over simple carbohydrates. A low-carbohydrate diet in a patient with diabetes has been demonstrated to improve the TG level, irrespective of weight change.²⁰

The risk of acute pancreatitis in the general population is 0.04%, but that risk increases to 8% to 31% for a person with hypertriglyceridemia

A Mediterranean diet can reduce the TG level by 10% to 15%, and is recommended over a low-fat diet.¹⁴ (This diet generally includes a high intake of extra virgin olive oil; leafy green vegetables, fruits, cereals, nuts, and legumes; moderate intake of fish and other meat, dairy products, and red wine; and low intake of eggs and sugars.) The American Heart Association recommends 2 servings of fatty fish a week for its omega-3 oil benefit of reducing ASCVD risk. Working with a registered dietician to assist with lipid lowering can produce better results than physician-only instruction on healthy eating.⁹

Alcohol consumption. Complete cessation or moderation of alcohol consumption (1 drink a day/women and 2 drinks a day/men*) is recommended to improve HTG. Among secondary factors, alcohol is commonly the cause of an unusually high elevation of the TG level.¹⁴

Smoking cessation. Smoking increases the postprandial TG level.¹⁰ Complete cessation for just 1 year can reduce a person’s ASCVD risk by approximately 50%. However, in a clinical trial,²² smoking cessation did not significantly decrease the TG level—possibly because of the counterbalancing effect of weight gain following cessation.

Continue to: Medical therapy

In addition to lifestyle modification, medications are recommended to reduce atherogenic potential in patients with moderate or severe HTG and an ASCVD risk > 7.5% (Table 3^4,13,18,23 and Table 4^2,9). Before initiating medical therapy, we recommend that you engage in shared decision-making with patients to (1) delineate treatment goals and (2) describe the risks and benefits of medications for HTG.²

Statins. These agents are recommended first-line therapy for reducing ASCVD risk.² If the TG level remains elevated (> 500 mg/dL) after statin therapy is maximized, an additional agent can be added—ie, a fibrate or fish oil (see below).

Fibrates. If a fibrate is used as an add-on to a statin, fenofibrate is preferred over gemfibrozil because it presents less risk of the severe myopathy that can develop when taken with a statin.¹³ Despite the effectiveness of fibrates in reducing the TG level, these drugs have not been shown to reduce overall mortality.²⁴ The evidence on improved cardiovascular outcomes is subgroup-specific (ie, prevention of a second myocardial infarction in the setting of optimal statin use and elevated non-HDL lipoproteins).¹² A study demonstrated that gemfibrozil reduced the incidence of transient ischemic attack and stroke in a subgroup of male US veterans who had coronary artery disease and a low HDL level.²⁵

Fish oil. The omega-3 ethyl esters eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), available as EPA alone or in combination with DHA, do not interact with statins and are tolerated well. They reduce hypertriglyceridemia by 20% to 50%.¹³

Eicosapentaenoic acid, EPA plus DHA, and icosapent ethyl, an ethyl ester product containing EPA without DHA, are approved by the US Food and Drug Administration for HTG > 500 mg/dL, at a dosage of 2000 mg twice daily. In the REDUCE-IT trial, adding icosapent ethyl, 2 g twice daily, to a statin in patients with HTG was associated with fewer ischemic events, compared to placebo.^23,26

Continue to: Fish oil formulations...

Fish oil formulations can inhibit platelet aggregation and increase bleeding time in otherwise healthy people; however, such episodes are minor and nonfatal. Patients on anticoagulation or an antiplatelet medication should be monitored periodically for bleeding events, although recommendations on how to monitor aren’t specified in a recent advisory by the American Heart Association.²³

There is a strong association between a BMI > 30 and hypertriglyceridemia.

DHA was thought to increase the LDL-C levels and, by doing so, potentially counterbalance benefit,^23,27 but most studies have failed to reproduce this effect.²⁸ Instead, studies have shown minimal elevation of LDL-C when DHA is used to treat HTG.^23,27

Niacin. At a dosage of 500-2000 mg/dL, niacin lowers the TG level by 10% to 30%. It also increases HDL by 10% to 40% and lowers LDL by 5% to 20%.¹³

Considerations in pancreatitis. For management of recurrent pancreatitis in patients with HTG, lifestyle modification remains the mainstay of treatment. When medication is considered for persistent severe HTG, fibrates have evidence of primary and secondary prevention of pancreatitis.¹¹ When a patient is intolerant of fibrates, consider a different option to reduce the TG level—eg, fish oil supplementation.

CASE 1

Recommendation for Mr. M: Therapeutic lifestyle changes to address moderate HTG.

Continue to: Because Mr. M's...

Because Mr. M’s 10-yr ASCVD risk is < 5%, statin therapy is not indicated for risk reduction. With a fasting TG value < 500 mg/dL, he is not considered at increased risk of pancreatitis.

CASE 2

Recommendations for Ms. F:

• Therapeutic lifestyle changes to address severe HTG. Ms. F agrees to wean off alcohol; add relaxation exercises before bedtime; do aerobic exercise 30 minutes a day, 3 times a week; decrease dietary carbohydrates daily by cutting portion size in half; and increase intake of fresh vegetables and lean protein.
• Treatment with fenofibrate to reduce the risk of pancreatitis. Ms. F begins a trial. Six months into treatment, she has reduced her BMI to 24 and the TG level has fallen to < 500 mg/dL. Ms. F also reports that she is sleeping well, believes that she is able to manage her infrequent anxiety, and is now in a routine that feels sustainable.

You congratulate Ms. F on her success and support her decision to undertake a trial of discontinuing fenofibrate, after shared decision-making about the risks and potential benefits of doing so.

Summing up: Management of HTG

Keep these treatment strategy highlights in mind:

• Lifestyle modification with a low-fat, low-carbohydrate diet, avoidance of alcohol, and moderate-intensity exercise is the mainstay of HTG management.
• The latest evidence supports that (1) HTG is a risk-enhancing factor for ­ASCVD and (2) statin therapy is recommended for patients who have HTG and an ASCVD risk > 7.5%.
• When the TG level remains elevated despite statin therapy and lifestyle changes, an omega-3 ethyl ester can be used as an adjunct for additional atherogenic risk reduction.
• For severe HTG, a regimen of therapeutic lifestyle changes plus a fibrate is recommended to reduce the risk and recurrence of pancreatitis.^1,24

* In comparison, a normal level of triglycerides is < 175 mg/dL; a moderately elevated level, measured in a fasting or nonfasting state, 175-499 mg/dL; and a very severely elevated level, ≥ 2000 mg/dL.²

CORRESPONDENCE
Ashwini Kamath Mulki, MD, Family Health Center, 1730 Chew Street, Allentown, PA 18104; Ashwini.KamathMulki@lvhn.org.

We are in the age of the microbiome. Both lay and scientific press proliferate messages about the importance of the microbiome to our health even while they often remain unclear on how to correct microbiota patterns associated with different diseases or suboptimal health states. Probiotics are defined as “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host.”¹

Certain probiotics have been shown to prevent and treat specific diseases or conditions, inside or outside the gut. But the level and quality of evidence varies greatly. In addition, the health claims allowed by government regulators depend on making discrete distinctions (food vs drug, maintaining health vs treating disease, and emerging evidence vs significant scientific agreement) along dimensions that are increasingly recognized as continuous and complex.² This leads to confusion among doctors and patients about whether to trust claims on product labels and what to make of the absence of such claims.

Find out which probiotic is effective for a patient’s condition. Simply recommending that a patient “take probiotics” is not particularly helpful when the individual wants a product that will aid a specific condition. While probiotics, to date, have not been marketed as drugs in the United States, clinicians can still approach recommending them in an evidence-based manner.

In this article, we review diseases/conditions for which probiotic products have good efficacy data. We discuss probiotic efficacy and safety, offer relevant information on regulatory categories of probiotics, and give direction for proper usage based on the current evidence base. Although this review is meant to be an easy-to-use resource for clinicians, it is not a comprehensive or detailed review of the numerous probiotic products and studies currently available.

Regulatory and commercial variances with probiotics

In the United States, probiotics have been marketed as dietary supplements, medical foods, or conventional foods, all of which require different levels of evidence and types of oversight than drugs. The efficacy of some probiotics in treating or preventing certain diseases and conditions is similar to, if not better than, effects observed with traditional drug interventions (TABLE 1^3-32). However, unlike drugs, which are subject to premarket oversight, the probiotic marketplace contains products with uneven levels of evidence, from well substantiated to greatly limited. Currently, no probiotics are sold in the United States as over-the-counter or prescription drugs, although probiotic drugs will likely enter the US market eventually.

What to consider when recommending a product. When considering probiotics, remember that strain, dosage, and indication are all important. Just as we know that not all antibiotics are equally effective for all infections, so, too, effectiveness among probiotics can—and often does—vary for any given condition. Effectiveness also may vary from patient to patient. Most recommendations made in this review are tied to specific probiotic strains and doses. In some cases, more than one probiotic may be efficacious, likely due to the same or similar underlying mechanism of action. For example, most probiotics produce short-chain fatty acids in the colon, providing a common mechanism supporting digestive health.^33-35

Some probiotics are as effective as, if not better than, drugs traditionally used to treat or prevent certain diseases and conditions.

Contrary to the blanket recommendation preferring higher dosages or a greater number of strains,³⁶ our recommendations are based on levels shown to be effective in clinical trials, which in some contexts can be as low as 100 million colony-forming units (CFU) per day.^37,38 Indeed, a survey we conducted previously of retail dietary supplement products indicated that products with lower CFUs or fewer strains could more readily be linked to evidence of efficacy than multistrain, high-CFU products.³⁹

Continue to: Understanding probiotic product labels is a good start

Understanding probiotic product labels is a good start. Information shown on the label of a probiotic dietary supplement in the United States should include the genus, species, and strains contained in the product, the dose delivered in CFU (the most common measure of the number of live microbes in a probiotic product) through the end of shelf life, and expected benefits. (For help in deciphering these labels, see the label schematic developed by the International Scientific Association for Probiotics and Prebiotics⁴⁰ at https://isappscience.org/infographics/probiotic-labelling/.)

Per guidelines from the Food and Agricultural Organization of the United Nations and the World Health Organization, all probiotic products should have this type of information clearly displayed on the product packaging.⁴¹ However, some probiotic foods display less information; for example, they may not specify the product’s strains or recommended dosage levels. Product Web sites may or may not disclose details missing from the food label. The absence of such information makes it impossible to make evidence-based recommendations about those products.

Probiotics are generally safe, with caveats

The overall safety of typical probiotics (Lactobacillus species, Bifidobacterium species, and Saccharomyces cerevisiae var. boulardii) has been well documented.^42,43 Many probiotic strains have been granted Generally Recognized as Safe status for use in foods in the United States.^44,45 Many traditional probiotic species have been evaluated by the European Food Safety Authority (similar to FDA, except jurisdiction is only over foods, not drugs) and are considered safe for use in food in the European Union.

Be aware that probiotics delivered in dietary supplements and foods are intended for the general population and not for patient populations. Manufacturers therefore are not required to assure safety in vulnerable populations. Nevertheless, probiotics are often stocked in hospital formularies.^46,47 Probiotic usage in vulnerable patient groups has been considered by an expert working group from the standpoint of quality assurance for microbiologic products used to treat and prevent disease, with the experts recommending that health care professionals (including pharmacists and physicians) seek quality information from manufacturers and that manufacturers participate in programs providing third-party (eg, United States Pharmacopeia [USP] or Underwriters Laboratories [UL]) verification of probiotic products to assure products meet applicable purity standards.^48,49

Published case studies have reported that probiotics may be a rare cause of sepsis.⁴³ Recently, Lactobacillus rhamnosus GG was linked to bacteremia in 6 critically ill patients, but all cases resolved without complications.⁵⁰ Further, the death of a premature infant was linked to administration of a probiotic contaminated with an opportunistic pathogenic mold.⁵¹ A randomized controlled trial (RCT) of a multispecies probiotic product in critically ill pancreatitis patients showed higher mortality in the group given the multispecies probiotic.⁵² However, additional examination of the data suggests that the observed higher mortality was due to problems with randomization for disease severity and other concerns, and not to the probiotic.⁵³ Much more frequently, probiotics have been administered orally in at-risk patient groups, including premature infants, cancer patients, and critically ill patients, with no significant increases in adverse events.^54-56

Continue to: Taken together...

Taken together, clinical trials have reported more adverse events in the placebo than probiotic group.⁴² Infection data collected in these trials have been used in subsequent analyses to demonstrate that in some settings, certain probiotics actually reduce the risk of infections. One notable example was a meta-analysis of 37 RCTs that showed that probiotics reduce the incidence of late-onset neonatal sepsis in premature infants.⁵⁷

At the present time, risk of probiotic use is low but still demands awareness, especially in unusual circumstances such as use in particularly vulnerable patients not yet studied or use of a product with limited available safety data. Any recommended product should be manufactured in compliance with applicable regulatory standards and preferably assured through voluntary quality audits.⁴⁹

Evidence of effectiveness is strong for many conditions

Probiotics have been studied for clinical benefit in numerous conditions (FIGURE^{3,8,11,15,19,23,54,58-65}), and systematic reviews of the clinical trials have found the overall results to be sufficiently strong to warrant recommendations, even though some individual trials were of low quality.⁶⁶ Some evidence may require confirmatory studies to clarify which specific product should be recommended.

Admittedly some of the indications are for diseases that most family physicians do not typically manage. For example, the evidence for probiotics for preventing necrotizing enterocolitis in premature infants was reviewed in a Cochrane analysis, which gave an estimated number needed to treat (NNT) of 41 and concluded, “our updated review of available evidence strongly supports a change in practice.”⁵⁴ A recent study of > 4500 infants in India found a probiotic/prebiotic supplement resulted in a 40% reduction in clinical sepsis compared with placebo.⁶⁷ Another common use of probiotics is as adjunctive therapy for mild to moderately active ulcerative colitis, where the current estimated NNT is 4.⁶³ Probiotics may also address gut and non-gut conditions and serve different functions throughout the lifespan.

Probiotic applications most relevant to primary care

We summarize in TABLE 1^3-32 probiotic uses supported by good evidence for indications of general interest in primary care medicine. This table includes endpoints with actionable evidence (including many strength of recommendation taxonomy [SORT] Level 1 studies) that allow us to make strong recommendations. Not all evidence is SORT Grade A, but we agree with the expert groups that deem evidence to be sufficient to warrant recommendations.

Continue to: The granular data...

The granular data we provide can help shape recommendations of a product for a specific indication. Numerous probiotics have been tested on suboptimal gastrointestinal health, including managing functional bowel symptoms ranging from occasional gas, bloating, or constipation through diagnosed irritable bowel syndrome (IBS). Supplements such as Bifidobacterium infantis subsp. longum 35624 (the probiotic in Align), Lactobacillus plantarum 299V (the probiotic in NatureMade Digestive Probiotic Daily Balance), and foods such as Activia yogurt, Yakult cultured milk, or Good Belly juice can be recommended for digestive symptoms.

Base your probiotic dosages on levels shown to be effective in clinical trials, which can be as low as 100 million CFU/d.

For patients experiencing gut symptoms unrelated to diagnosed disease, it may be reasonable for them to try a well-documented strain for 3 to 4 weeks. Currently it is difficult to predict success a priori; this may change as we learn more about how an individual’s microbiome, diet, and genetics affect response to specific probiotics. TABLE 2^68-71 presents sample recommendations from international expert panels for select contexts.

The popular press today commonly recommends consuming more fermented foods. Although we agree in general with this recommendation, physicians should be clear that fermented foods may be a source of live cultures, but not all fermented foods retain live microbes. Further, many fermented foods lack evidence documenting health effects, and therefore are not a source of probiotics. If the patient’s goal is to support regular diet with live microbes, any number of probiotic products or fermented foods that retain viable cultures may suffice. However, when patients request probiotics for specific needs, recommendations should be based on available evidence for specific studied products. (See also, “Questions patients frequently ask about probiotics.”)

Continue to: What to look for in the future

Basic research, human trials, and market development in the field of probiotics are progressing rapidly. Probiotics at this time are primarily from the genera Lactobacillus, Bifidobacterium, and Saccharomyces. But the potential of probiotics has spurred research into previously untapped microbial members of the healthy human microbiota. Microbes such as Akkermansia, Faecalibacterium, and Rosburia may comprise “next-generation probiotics” that will likely be developed as drugs.⁷²

Since the risks of using traditional probiotics are low, trial and error may be warranted at times.

Active areas of research holding some promise involve microbiome-driven components of intractable problems such as metabolic syndrome (obesity,⁷³ diabetes, and lipid dysregulation) and brain dysfunction⁷⁴ (depression, anxiety, cognition, autism). A guide to the clinical use of probiotic products available in the United States, updated yearly, may be a useful reference (but the reader may want to examine the referenced studies as their level of evidence is different than the SORT method).⁷⁵ Science-based videos, infographics, and other resources are available from the International Scientific Association for Probiotics and Prebiotics, (mentioned earlier; www.isappscience.org/).

It appears that probiotics will continue to be widely used and hopefully in a more evidence-based manner. As we learn more about individual microbiome variations, recommendations will likely be more patient specific. Probiotics that have robust evidence represent the strongest recommendations. Even so, since the risks of using traditional probiotics (such as Lactobacillus, Bifidobacterium and Saccharomyces strains) are low, trial and error may be warranted at times.

CORRESPONDENCE
Daniel J. Merenstein, MD, 4000 Reservoir Road NW, Building D 240, Washington, DC 20007; djm23@georgetown.edu.

ACKNOWLEDGMENT
We thank Alexandra Mannerings, PhD, for preparing the FIGURE.

We are in the age of the microbiome. Both lay and scientific press proliferate messages about the importance of the microbiome to our health even while they often remain unclear on how to correct microbiota patterns associated with different diseases or suboptimal health states. Probiotics are defined as “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host.”¹

Certain probiotics have been shown to prevent and treat specific diseases or conditions, inside or outside the gut. But the level and quality of evidence varies greatly. In addition, the health claims allowed by government regulators depend on making discrete distinctions (food vs drug, maintaining health vs treating disease, and emerging evidence vs significant scientific agreement) along dimensions that are increasingly recognized as continuous and complex.² This leads to confusion among doctors and patients about whether to trust claims on product labels and what to make of the absence of such claims.

Find out which probiotic is effective for a patient’s condition. Simply recommending that a patient “take probiotics” is not particularly helpful when the individual wants a product that will aid a specific condition. While probiotics, to date, have not been marketed as drugs in the United States, clinicians can still approach recommending them in an evidence-based manner.

In this article, we review diseases/conditions for which probiotic products have good efficacy data. We discuss probiotic efficacy and safety, offer relevant information on regulatory categories of probiotics, and give direction for proper usage based on the current evidence base. Although this review is meant to be an easy-to-use resource for clinicians, it is not a comprehensive or detailed review of the numerous probiotic products and studies currently available.

Regulatory and commercial variances with probiotics

In the United States, probiotics have been marketed as dietary supplements, medical foods, or conventional foods, all of which require different levels of evidence and types of oversight than drugs. The efficacy of some probiotics in treating or preventing certain diseases and conditions is similar to, if not better than, effects observed with traditional drug interventions (TABLE 1^3-32). However, unlike drugs, which are subject to premarket oversight, the probiotic marketplace contains products with uneven levels of evidence, from well substantiated to greatly limited. Currently, no probiotics are sold in the United States as over-the-counter or prescription drugs, although probiotic drugs will likely enter the US market eventually.

What to consider when recommending a product. When considering probiotics, remember that strain, dosage, and indication are all important. Just as we know that not all antibiotics are equally effective for all infections, so, too, effectiveness among probiotics can—and often does—vary for any given condition. Effectiveness also may vary from patient to patient. Most recommendations made in this review are tied to specific probiotic strains and doses. In some cases, more than one probiotic may be efficacious, likely due to the same or similar underlying mechanism of action. For example, most probiotics produce short-chain fatty acids in the colon, providing a common mechanism supporting digestive health.^33-35

Some probiotics are as effective as, if not better than, drugs traditionally used to treat or prevent certain diseases and conditions.

Contrary to the blanket recommendation preferring higher dosages or a greater number of strains,³⁶ our recommendations are based on levels shown to be effective in clinical trials, which in some contexts can be as low as 100 million colony-forming units (CFU) per day.^37,38 Indeed, a survey we conducted previously of retail dietary supplement products indicated that products with lower CFUs or fewer strains could more readily be linked to evidence of efficacy than multistrain, high-CFU products.³⁹

Continue to: Understanding probiotic product labels is a good start

Understanding probiotic product labels is a good start. Information shown on the label of a probiotic dietary supplement in the United States should include the genus, species, and strains contained in the product, the dose delivered in CFU (the most common measure of the number of live microbes in a probiotic product) through the end of shelf life, and expected benefits. (For help in deciphering these labels, see the label schematic developed by the International Scientific Association for Probiotics and Prebiotics⁴⁰ at https://isappscience.org/infographics/probiotic-labelling/.)

Per guidelines from the Food and Agricultural Organization of the United Nations and the World Health Organization, all probiotic products should have this type of information clearly displayed on the product packaging.⁴¹ However, some probiotic foods display less information; for example, they may not specify the product’s strains or recommended dosage levels. Product Web sites may or may not disclose details missing from the food label. The absence of such information makes it impossible to make evidence-based recommendations about those products.

Probiotics are generally safe, with caveats

The overall safety of typical probiotics (Lactobacillus species, Bifidobacterium species, and Saccharomyces cerevisiae var. boulardii) has been well documented.^42,43 Many probiotic strains have been granted Generally Recognized as Safe status for use in foods in the United States.^44,45 Many traditional probiotic species have been evaluated by the European Food Safety Authority (similar to FDA, except jurisdiction is only over foods, not drugs) and are considered safe for use in food in the European Union.

Be aware that probiotics delivered in dietary supplements and foods are intended for the general population and not for patient populations. Manufacturers therefore are not required to assure safety in vulnerable populations. Nevertheless, probiotics are often stocked in hospital formularies.^46,47 Probiotic usage in vulnerable patient groups has been considered by an expert working group from the standpoint of quality assurance for microbiologic products used to treat and prevent disease, with the experts recommending that health care professionals (including pharmacists and physicians) seek quality information from manufacturers and that manufacturers participate in programs providing third-party (eg, United States Pharmacopeia [USP] or Underwriters Laboratories [UL]) verification of probiotic products to assure products meet applicable purity standards.^48,49

Published case studies have reported that probiotics may be a rare cause of sepsis.⁴³ Recently, Lactobacillus rhamnosus GG was linked to bacteremia in 6 critically ill patients, but all cases resolved without complications.⁵⁰ Further, the death of a premature infant was linked to administration of a probiotic contaminated with an opportunistic pathogenic mold.⁵¹ A randomized controlled trial (RCT) of a multispecies probiotic product in critically ill pancreatitis patients showed higher mortality in the group given the multispecies probiotic.⁵² However, additional examination of the data suggests that the observed higher mortality was due to problems with randomization for disease severity and other concerns, and not to the probiotic.⁵³ Much more frequently, probiotics have been administered orally in at-risk patient groups, including premature infants, cancer patients, and critically ill patients, with no significant increases in adverse events.^54-56

Continue to: Taken together...

Taken together, clinical trials have reported more adverse events in the placebo than probiotic group.⁴² Infection data collected in these trials have been used in subsequent analyses to demonstrate that in some settings, certain probiotics actually reduce the risk of infections. One notable example was a meta-analysis of 37 RCTs that showed that probiotics reduce the incidence of late-onset neonatal sepsis in premature infants.⁵⁷

At the present time, risk of probiotic use is low but still demands awareness, especially in unusual circumstances such as use in particularly vulnerable patients not yet studied or use of a product with limited available safety data. Any recommended product should be manufactured in compliance with applicable regulatory standards and preferably assured through voluntary quality audits.⁴⁹

Evidence of effectiveness is strong for many conditions

Probiotics have been studied for clinical benefit in numerous conditions (FIGURE^{3,8,11,15,19,23,54,58-65}), and systematic reviews of the clinical trials have found the overall results to be sufficiently strong to warrant recommendations, even though some individual trials were of low quality.⁶⁶ Some evidence may require confirmatory studies to clarify which specific product should be recommended.

Admittedly some of the indications are for diseases that most family physicians do not typically manage. For example, the evidence for probiotics for preventing necrotizing enterocolitis in premature infants was reviewed in a Cochrane analysis, which gave an estimated number needed to treat (NNT) of 41 and concluded, “our updated review of available evidence strongly supports a change in practice.”⁵⁴ A recent study of > 4500 infants in India found a probiotic/prebiotic supplement resulted in a 40% reduction in clinical sepsis compared with placebo.⁶⁷ Another common use of probiotics is as adjunctive therapy for mild to moderately active ulcerative colitis, where the current estimated NNT is 4.⁶³ Probiotics may also address gut and non-gut conditions and serve different functions throughout the lifespan.

Probiotic applications most relevant to primary care

We summarize in TABLE 1^3-32 probiotic uses supported by good evidence for indications of general interest in primary care medicine. This table includes endpoints with actionable evidence (including many strength of recommendation taxonomy [SORT] Level 1 studies) that allow us to make strong recommendations. Not all evidence is SORT Grade A, but we agree with the expert groups that deem evidence to be sufficient to warrant recommendations.

Continue to: The granular data...

The granular data we provide can help shape recommendations of a product for a specific indication. Numerous probiotics have been tested on suboptimal gastrointestinal health, including managing functional bowel symptoms ranging from occasional gas, bloating, or constipation through diagnosed irritable bowel syndrome (IBS). Supplements such as Bifidobacterium infantis subsp. longum 35624 (the probiotic in Align), Lactobacillus plantarum 299V (the probiotic in NatureMade Digestive Probiotic Daily Balance), and foods such as Activia yogurt, Yakult cultured milk, or Good Belly juice can be recommended for digestive symptoms.

Base your probiotic dosages on levels shown to be effective in clinical trials, which can be as low as 100 million CFU/d.

For patients experiencing gut symptoms unrelated to diagnosed disease, it may be reasonable for them to try a well-documented strain for 3 to 4 weeks. Currently it is difficult to predict success a priori; this may change as we learn more about how an individual’s microbiome, diet, and genetics affect response to specific probiotics. TABLE 2^68-71 presents sample recommendations from international expert panels for select contexts.

The popular press today commonly recommends consuming more fermented foods. Although we agree in general with this recommendation, physicians should be clear that fermented foods may be a source of live cultures, but not all fermented foods retain live microbes. Further, many fermented foods lack evidence documenting health effects, and therefore are not a source of probiotics. If the patient’s goal is to support regular diet with live microbes, any number of probiotic products or fermented foods that retain viable cultures may suffice. However, when patients request probiotics for specific needs, recommendations should be based on available evidence for specific studied products. (See also, “Questions patients frequently ask about probiotics.”)

Continue to: What to look for in the future

Basic research, human trials, and market development in the field of probiotics are progressing rapidly. Probiotics at this time are primarily from the genera Lactobacillus, Bifidobacterium, and Saccharomyces. But the potential of probiotics has spurred research into previously untapped microbial members of the healthy human microbiota. Microbes such as Akkermansia, Faecalibacterium, and Rosburia may comprise “next-generation probiotics” that will likely be developed as drugs.⁷²

Since the risks of using traditional probiotics are low, trial and error may be warranted at times.

Active areas of research holding some promise involve microbiome-driven components of intractable problems such as metabolic syndrome (obesity,⁷³ diabetes, and lipid dysregulation) and brain dysfunction⁷⁴ (depression, anxiety, cognition, autism). A guide to the clinical use of probiotic products available in the United States, updated yearly, may be a useful reference (but the reader may want to examine the referenced studies as their level of evidence is different than the SORT method).⁷⁵ Science-based videos, infographics, and other resources are available from the International Scientific Association for Probiotics and Prebiotics, (mentioned earlier; www.isappscience.org/).

It appears that probiotics will continue to be widely used and hopefully in a more evidence-based manner. As we learn more about individual microbiome variations, recommendations will likely be more patient specific. Probiotics that have robust evidence represent the strongest recommendations. Even so, since the risks of using traditional probiotics (such as Lactobacillus, Bifidobacterium and Saccharomyces strains) are low, trial and error may be warranted at times.

CORRESPONDENCE
Daniel J. Merenstein, MD, 4000 Reservoir Road NW, Building D 240, Washington, DC 20007; djm23@georgetown.edu.

ACKNOWLEDGMENT
We thank Alexandra Mannerings, PhD, for preparing the FIGURE.

We are in the age of the microbiome. Both lay and scientific press proliferate messages about the importance of the microbiome to our health even while they often remain unclear on how to correct microbiota patterns associated with different diseases or suboptimal health states. Probiotics are defined as “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host.”¹

Certain probiotics have been shown to prevent and treat specific diseases or conditions, inside or outside the gut. But the level and quality of evidence varies greatly. In addition, the health claims allowed by government regulators depend on making discrete distinctions (food vs drug, maintaining health vs treating disease, and emerging evidence vs significant scientific agreement) along dimensions that are increasingly recognized as continuous and complex.² This leads to confusion among doctors and patients about whether to trust claims on product labels and what to make of the absence of such claims.

Find out which probiotic is effective for a patient’s condition. Simply recommending that a patient “take probiotics” is not particularly helpful when the individual wants a product that will aid a specific condition. While probiotics, to date, have not been marketed as drugs in the United States, clinicians can still approach recommending them in an evidence-based manner.

In this article, we review diseases/conditions for which probiotic products have good efficacy data. We discuss probiotic efficacy and safety, offer relevant information on regulatory categories of probiotics, and give direction for proper usage based on the current evidence base. Although this review is meant to be an easy-to-use resource for clinicians, it is not a comprehensive or detailed review of the numerous probiotic products and studies currently available.

Regulatory and commercial variances with probiotics

In the United States, probiotics have been marketed as dietary supplements, medical foods, or conventional foods, all of which require different levels of evidence and types of oversight than drugs. The efficacy of some probiotics in treating or preventing certain diseases and conditions is similar to, if not better than, effects observed with traditional drug interventions (TABLE 1^3-32). However, unlike drugs, which are subject to premarket oversight, the probiotic marketplace contains products with uneven levels of evidence, from well substantiated to greatly limited. Currently, no probiotics are sold in the United States as over-the-counter or prescription drugs, although probiotic drugs will likely enter the US market eventually.

What to consider when recommending a product. When considering probiotics, remember that strain, dosage, and indication are all important. Just as we know that not all antibiotics are equally effective for all infections, so, too, effectiveness among probiotics can—and often does—vary for any given condition. Effectiveness also may vary from patient to patient. Most recommendations made in this review are tied to specific probiotic strains and doses. In some cases, more than one probiotic may be efficacious, likely due to the same or similar underlying mechanism of action. For example, most probiotics produce short-chain fatty acids in the colon, providing a common mechanism supporting digestive health.^33-35

Some probiotics are as effective as, if not better than, drugs traditionally used to treat or prevent certain diseases and conditions.

Contrary to the blanket recommendation preferring higher dosages or a greater number of strains,³⁶ our recommendations are based on levels shown to be effective in clinical trials, which in some contexts can be as low as 100 million colony-forming units (CFU) per day.^37,38 Indeed, a survey we conducted previously of retail dietary supplement products indicated that products with lower CFUs or fewer strains could more readily be linked to evidence of efficacy than multistrain, high-CFU products.³⁹

Continue to: Understanding probiotic product labels is a good start

Understanding probiotic product labels is a good start. Information shown on the label of a probiotic dietary supplement in the United States should include the genus, species, and strains contained in the product, the dose delivered in CFU (the most common measure of the number of live microbes in a probiotic product) through the end of shelf life, and expected benefits. (For help in deciphering these labels, see the label schematic developed by the International Scientific Association for Probiotics and Prebiotics⁴⁰ at https://isappscience.org/infographics/probiotic-labelling/.)

Per guidelines from the Food and Agricultural Organization of the United Nations and the World Health Organization, all probiotic products should have this type of information clearly displayed on the product packaging.⁴¹ However, some probiotic foods display less information; for example, they may not specify the product’s strains or recommended dosage levels. Product Web sites may or may not disclose details missing from the food label. The absence of such information makes it impossible to make evidence-based recommendations about those products.

Probiotics are generally safe, with caveats

The overall safety of typical probiotics (Lactobacillus species, Bifidobacterium species, and Saccharomyces cerevisiae var. boulardii) has been well documented.^42,43 Many probiotic strains have been granted Generally Recognized as Safe status for use in foods in the United States.^44,45 Many traditional probiotic species have been evaluated by the European Food Safety Authority (similar to FDA, except jurisdiction is only over foods, not drugs) and are considered safe for use in food in the European Union.

Be aware that probiotics delivered in dietary supplements and foods are intended for the general population and not for patient populations. Manufacturers therefore are not required to assure safety in vulnerable populations. Nevertheless, probiotics are often stocked in hospital formularies.^46,47 Probiotic usage in vulnerable patient groups has been considered by an expert working group from the standpoint of quality assurance for microbiologic products used to treat and prevent disease, with the experts recommending that health care professionals (including pharmacists and physicians) seek quality information from manufacturers and that manufacturers participate in programs providing third-party (eg, United States Pharmacopeia [USP] or Underwriters Laboratories [UL]) verification of probiotic products to assure products meet applicable purity standards.^48,49

Published case studies have reported that probiotics may be a rare cause of sepsis.⁴³ Recently, Lactobacillus rhamnosus GG was linked to bacteremia in 6 critically ill patients, but all cases resolved without complications.⁵⁰ Further, the death of a premature infant was linked to administration of a probiotic contaminated with an opportunistic pathogenic mold.⁵¹ A randomized controlled trial (RCT) of a multispecies probiotic product in critically ill pancreatitis patients showed higher mortality in the group given the multispecies probiotic.⁵² However, additional examination of the data suggests that the observed higher mortality was due to problems with randomization for disease severity and other concerns, and not to the probiotic.⁵³ Much more frequently, probiotics have been administered orally in at-risk patient groups, including premature infants, cancer patients, and critically ill patients, with no significant increases in adverse events.^54-56

Continue to: Taken together...

Taken together, clinical trials have reported more adverse events in the placebo than probiotic group.⁴² Infection data collected in these trials have been used in subsequent analyses to demonstrate that in some settings, certain probiotics actually reduce the risk of infections. One notable example was a meta-analysis of 37 RCTs that showed that probiotics reduce the incidence of late-onset neonatal sepsis in premature infants.⁵⁷

At the present time, risk of probiotic use is low but still demands awareness, especially in unusual circumstances such as use in particularly vulnerable patients not yet studied or use of a product with limited available safety data. Any recommended product should be manufactured in compliance with applicable regulatory standards and preferably assured through voluntary quality audits.⁴⁹

Evidence of effectiveness is strong for many conditions

Probiotics have been studied for clinical benefit in numerous conditions (FIGURE^{3,8,11,15,19,23,54,58-65}), and systematic reviews of the clinical trials have found the overall results to be sufficiently strong to warrant recommendations, even though some individual trials were of low quality.⁶⁶ Some evidence may require confirmatory studies to clarify which specific product should be recommended.

Admittedly some of the indications are for diseases that most family physicians do not typically manage. For example, the evidence for probiotics for preventing necrotizing enterocolitis in premature infants was reviewed in a Cochrane analysis, which gave an estimated number needed to treat (NNT) of 41 and concluded, “our updated review of available evidence strongly supports a change in practice.”⁵⁴ A recent study of > 4500 infants in India found a probiotic/prebiotic supplement resulted in a 40% reduction in clinical sepsis compared with placebo.⁶⁷ Another common use of probiotics is as adjunctive therapy for mild to moderately active ulcerative colitis, where the current estimated NNT is 4.⁶³ Probiotics may also address gut and non-gut conditions and serve different functions throughout the lifespan.

Probiotic applications most relevant to primary care

We summarize in TABLE 1^3-32 probiotic uses supported by good evidence for indications of general interest in primary care medicine. This table includes endpoints with actionable evidence (including many strength of recommendation taxonomy [SORT] Level 1 studies) that allow us to make strong recommendations. Not all evidence is SORT Grade A, but we agree with the expert groups that deem evidence to be sufficient to warrant recommendations.

Continue to: The granular data...

The granular data we provide can help shape recommendations of a product for a specific indication. Numerous probiotics have been tested on suboptimal gastrointestinal health, including managing functional bowel symptoms ranging from occasional gas, bloating, or constipation through diagnosed irritable bowel syndrome (IBS). Supplements such as Bifidobacterium infantis subsp. longum 35624 (the probiotic in Align), Lactobacillus plantarum 299V (the probiotic in NatureMade Digestive Probiotic Daily Balance), and foods such as Activia yogurt, Yakult cultured milk, or Good Belly juice can be recommended for digestive symptoms.

Base your probiotic dosages on levels shown to be effective in clinical trials, which can be as low as 100 million CFU/d.

For patients experiencing gut symptoms unrelated to diagnosed disease, it may be reasonable for them to try a well-documented strain for 3 to 4 weeks. Currently it is difficult to predict success a priori; this may change as we learn more about how an individual’s microbiome, diet, and genetics affect response to specific probiotics. TABLE 2^68-71 presents sample recommendations from international expert panels for select contexts.

The popular press today commonly recommends consuming more fermented foods. Although we agree in general with this recommendation, physicians should be clear that fermented foods may be a source of live cultures, but not all fermented foods retain live microbes. Further, many fermented foods lack evidence documenting health effects, and therefore are not a source of probiotics. If the patient’s goal is to support regular diet with live microbes, any number of probiotic products or fermented foods that retain viable cultures may suffice. However, when patients request probiotics for specific needs, recommendations should be based on available evidence for specific studied products. (See also, “Questions patients frequently ask about probiotics.”)

Continue to: What to look for in the future

Basic research, human trials, and market development in the field of probiotics are progressing rapidly. Probiotics at this time are primarily from the genera Lactobacillus, Bifidobacterium, and Saccharomyces. But the potential of probiotics has spurred research into previously untapped microbial members of the healthy human microbiota. Microbes such as Akkermansia, Faecalibacterium, and Rosburia may comprise “next-generation probiotics” that will likely be developed as drugs.⁷²

Since the risks of using traditional probiotics are low, trial and error may be warranted at times.

Active areas of research holding some promise involve microbiome-driven components of intractable problems such as metabolic syndrome (obesity,⁷³ diabetes, and lipid dysregulation) and brain dysfunction⁷⁴ (depression, anxiety, cognition, autism). A guide to the clinical use of probiotic products available in the United States, updated yearly, may be a useful reference (but the reader may want to examine the referenced studies as their level of evidence is different than the SORT method).⁷⁵ Science-based videos, infographics, and other resources are available from the International Scientific Association for Probiotics and Prebiotics, (mentioned earlier; www.isappscience.org/).

It appears that probiotics will continue to be widely used and hopefully in a more evidence-based manner. As we learn more about individual microbiome variations, recommendations will likely be more patient specific. Probiotics that have robust evidence represent the strongest recommendations. Even so, since the risks of using traditional probiotics (such as Lactobacillus, Bifidobacterium and Saccharomyces strains) are low, trial and error may be warranted at times.

CORRESPONDENCE
Daniel J. Merenstein, MD, 4000 Reservoir Road NW, Building D 240, Washington, DC 20007; djm23@georgetown.edu.

ACKNOWLEDGMENT
We thank Alexandra Mannerings, PhD, for preparing the FIGURE.

Recent advances in our understanding of asthma pathophysiology have led to the development of new treatment approaches to this chronic respiratory condition, which affects 25 million Americans or nearly 8% of the population.¹ As a result, asthma treatment options have expanded from just simple inhalers and corticosteroids to include biological therapies, immunotherapy, bronchial thermoplasty, and anti-inflammatory agents. This review will focus on the newer therapeutic options and provide guidance on when to seek expert pulmonary consultation (TABLE 1). But before we begin, it’s important to briefly review the pathophysiology of asthma and several key diagnostic considerations.

The pathophysiology of asthma provides key targets for therapy

There are 2 basic phenotypes of asthma—neutrophilic predominant and eosinophilic predominant—and 3 key components to its pathophysiology²:

Airway inflammation. Asthma is mediated through either a type 1 T-helper (Th-1) cell or a type 2 T-helper (Th-2) cell response, the pathways of which have a fair amount of overlap (FIGURE). In the neutrophilic-predominant phenotype, irritants, pollutants, and viruses trigger an innate Th-1 cell–mediated pathway that leads to subsequent neutrophil release. This asthma phenotype responds poorly to standard asthma therapy.^2-4

In the eosinophilic-predominant phenotype, environmental allergic antigens induce a Th-2 cell–mediated response in the airways of patients with asthma.^5-7 This creates a downstream effect on the release of interleukins (IL) including IL-4, IL-5, and IL-13. IL-4 triggers immunoglobulin (Ig) E release, which subsequently induces mast cells to release inflammatory cytokines, while IL-5 and IL-13 are responsible for eosinophilic response. These cytokines and eosinophils induce airway hyperresponsiveness, remodeling, and mucus production. Through repeated exposure, chronic inflammation develops and subsequently causes structural changes related to increased smooth muscle mass, goblet cell hyperplasia, and thickening of lamina reticularis.^8,9 Understanding of this pathobiological pathway has led to the development of anti-IgE and anti-IL-5 drugs (to be discussed shortly).

It is important to determine whether patients with asthma-COPD overlap syndrome are asthma predominant or COPD predominant, because appropriate first-line treatment will differ.

Airway obstruction. Early asthmatic response is due to acute bronchoconstriction secondary to IgE; this is followed by airway edema occurring 6 to 24 hours after an acute event (called late asthmatic response). The obstruction is worsened by an overproduction of mucus, which may take weeks to resolve.¹⁰ Longstanding inflammation can lead to structural changes and reduced airflow reversibility.

Bronchial hyperresponsiveness is induced by various forms of allergens, pollutants, or viral upper respiratory infections. Sympathetic control in the airway is mediated via beta-2 adrenoceptors expressed on airway smooth muscle, which are responsible for the effect of bronchodilation in response to albuterol.^11,12 Cholinergic pathways may further contribute to bronchial hyperresponsiveness and form the basis for the efficacy of anticholinergic therapy.^12,13

What we’ve learned about asthma can inform treatment decisions

Presentation may vary, as asthma has many forms including cough-variant asthma and exercise-induced asthma. Airflow limitation is typically identified through spirometry and characterized by reduced (< 70% in adults) forced expiratory volume in 1 second (FEV₁)/forced vital capacity (FVC) or bronchodilator response positivity (an increase in post-bronchodilator FEV₁ > 12% or FVC > 200 mL from baseline).² If spirometry is not diagnostic but suspicion for asthma remains, bronchial provocation testing or exercise challenge testing may be needed.

Continue to: Additional diagnostic considerations...

Additional diagnostic considerations may impact the treatment plan for patients with asthma:

Asthma and COPD. A history of smoking is a key factor in the diagnosis of chronic obstructive pulmonary disease (COPD)—but many patients with asthma are also smokers. This subgroup may have asthma-COPD overlap syndrome (ACOS). It is important to determine whether these patients are asthma predominant or COPD predominant, because appropriate first-line treatment will differ. Patients who are COPD predominant demonstrate reduced diffusion capacity (DLCO) and abnormal PaCO2 on arterial blood gas. They also may show more structural damage on chest computed tomography (CT) than patients with asthma do. Asthma-predominant patients are more likely to have eosinophilia.¹⁴

Patients with severe persistent asthma or frequent exacerbations, or those receiving step-up therapy, may require additional serologic testing. Specialized testing for IgE and eosinophil count, as well as a sensitized allergy panel, may help clinicians in selecting specific biological therapies for treatment of severe asthma (further discussion to follow). We recommend using a serum allergy panel, as it is a quick and easy way to identify patients with extrinsic allergies, whereas skin-based testing is often time consuming and may require referral to a specialist.^2,5,15

Aspergillus. An additional consideration is testing for Aspergillus antibodies. Aspergillus is a ubiquitous fungus found in the airways of humans. In patients with asthma, however, it can trigger an intense inflammatory response known as allergic bronchopulmonary aspergillosis. ABPA is not an infection. It should be considered in patients who have lived in a damp, old housing environment with possible mold exposure. Treatment of ABPA involves oral corticosteroids; there are varying reports of efficacy with voriconazole or itraconazole as suppressive therapy or steroid-sparing treatment.^16-18

Getting a handle on an ever-expanding asthma Tx arsenal

The goals of asthma treatment are symptom control and risk minimization. Treatment choices are dictated in part by disease severity (mild, moderate, severe) and classification (intermittent, persistent). Asthma therapy is traditionally described as step-up and step-down; TABLE 2 summarizes available pharmacotherapy for asthma and provides a framework for add-on therapy as the disease advances.

Continue to: Over the past decade...

Over the past decade, a number of therapeutic options have been introduced or added to the pantheon of asthma treatment.

Inhaled medications

This category includes inhaled corticosteroids (ICS), which are recommended for use alone or in combination with long-acting beta-agonists (LABA) or with long-acting muscarinic agonists (LAMA).

ICS is the first choice for long-term control of persistent asthma.² Its molecular effects include activating anti-inflammatory genes, switching off inflammatory genes, and inhibiting inflammatory cells, combined with enhancement of beta-2-adrenergic receptor expression. The cumulative effect is reduction in airway responsiveness in asthma patients.^19-22

LABAs are next in line in the step-up, step-down model of symptom management. LABAs should not be prescribed as stand-alone therapy in patients with asthma, as they have received a black box warning from the US Food and Drug Administration (FDA) for an increase in asthma-related death²³—a concern that has not been demonstrated with the combination of ICS-LABA.

LABAs should not be prescribed as stand-alone therapy in patients with asthma, as they have received an FDA black box warning.

LABAs cause smooth muscle relaxation in the lungs.²⁴ There are 3 combination products currently available: once-daily fluticasone furoate/vilanterol (Breo), twice-daily fluticasone propionate/salmeterol (Advair), and twice-daily budesonide/formoterol (Symbicort).

Continue to: Once-daily fluticasone furoate/vilanterol...

Once-daily fluticasone furoate/vilanterol has been shown to improve mean FEV₁.²⁵ In a 24-week, open-label, multicenter randomized controlled trial to evaluate the efficacy and safety of all 3 combination ICS-LABAs, preliminary results indicated that—at least in a tightly controlled setting—once-daily fluticasone furoate/vilanterol provides asthma control similar to the twice-daily combinations and is well tolerated.²⁶

Two ultra-long-acting (24-hour) LABAs, olodaterol (Striverdi Respimat) and indacaterol (Arcapta Neohaler), are being studied for possible use in asthma treatment. In a phase 2 trial investigating therapy for moderate-to-severe persistent asthma, 24-hour FEV₁ improved with olodeaterol when compared to placebo.²⁷

Another ongoing clinical trial is studying the effects of ultra-long-acting bronchodilator therapy (olodaterol vs combination olodaterol/tiotropium) in asthma patients who smoke and who are already using ICS (ClinicalTrials.gov NCT02682862). Indacaterol has been shown to be effective in the treatment of moderate-to-severe asthma in a once-a-day dosing regimen.²⁸ However, when compared to mometasone alone, a combination of indacaterol and mometasone demonstrated no statistically significant reduction in time to serious exacerbation.²⁹

The LAMA tiotropium is recommended as add-on therapy for patients whose asthma is uncontrolled despite use of low-dose ICS-LABA or as an alternative to high-dose ICS-LABA, per Global Initiative for Asthma (GINA) 2019 guidelines.¹⁵

Ideal candidates for biological therapy are patients who have exhausted other forms of severe asthma treatment.

Tiotropium induces bronchodilation by selectively inhibiting the action of acetylcholine at muscarinic (M) receptors in bronchial smooth muscles; it has a longer duration of action because of its slower dissociation from receptor types M₁ and M₃.³⁰ Tiotropium respimat (Spiriva, Tiova) has been approved for COPD for many years; in 2013, it was shown to prevent worsening of symptomatic asthma and increase time to first severe exacerbation.¹³ The FDA subsequently approved tiotropium as an add-on treatment for patients with uncontrolled asthma despite use of ICS-LABA.

Continue to: Glycopyrronium bromide...

Glycopyrronium bromide (glycopyrrolate, multiple brand names) and umeclidinium (Incruse Ellipta) are LAMAs that are approved for COPD treatment but have not yet been approved for patients who have asthma only.³¹

Biological therapies

In the past few years, improved understanding of asthma’s pathophysiology has led to the development of biological therapy for severe asthma. This therapy is directed at Th-2 inflammatory pathways (FIGURE) and targets various inflammatory markers, such as IgE, IL-5, and eosinophils.

Biologicals are not the first-line therapy for the management of severe asthma. Ideal candidates for this therapy are patients who have exhausted other forms of severe asthma treatment, including ICS-LABA, LAMA, leukotriene receptor antagonists, and mucus-clearing agents. Patients with frequent exacerbations who need continuous steroids or need steroids at least twice a year should be considered for biologicals.³²

All biological therapies must be administered in a clinical setting, as they carry risk for anaphylaxis. TABLE 3^15,33-47 summarizes all approved biologicals for the management of severe asthma.

Anti-IgE therapy. Omalizumab (Xolair) was the first approved biological therapy for severe asthma (in 2003). It is a recombinant humanized IgG1 monoclonal antibody that binds to free IgE and down regulates the inflammatory cascade. It is therefore best suited for patients with early-onset allergic asthma with a high IgE count. The dose and frequency (once or twice per month) of omalizumab are based on IgE levels and patient weight. Omalizumab reduces asthma exacerbation (up to 45%) and hospitalization (up to 85%).³⁴ Omalizumab also reduces the need for high-dose ICS-LABA therapy and improves quality of life (QoL).^33,34

Continue to: Its efficacy and safety...

Mepolizumab has shown a trend of greater benefit in patients with a very high eosinophil count.

Its efficacy and safety have been proven outside the clinical trial setting. Treatment response should be assessed over a 3- to 4-month period, using fractional exhalation of nitric oxide (FeNO); serial measurement of IgE levels is not recommended for this purpose. Once started, treatment should be considered long term, as discontinuation of treatment has been shown to lead to recurrence of symptoms and exacerbation.^35,36 Of note, the GINA guidelines recommend omalizumab over prednisone as add-on therapy for severe persistent asthma.¹⁵

Anti-IL-5 therapy. IL-5 is the main cytokine for growth, differentiation, and activation of eosinophils in the Th-2-mediated inflammatory cascade. Mepolizumab, reslizumab, and benralizumab are 3 FDA-approved anti-IL-5 monoclonal antibody therapies for severe eosinophilic asthma. Mepolizumab has been the most commonly studied anti-IL-5 therapy, while benralizumab, the latest of the 3, has a unique property of inducing eosinophilic apoptosis. There has been no direct comparison of the different anti-IL-5 therapies.

Mepolizumab (Nucala) is a mouse anti-human monoclonal antibody that binds to IL-5 and prevents it from binding to IL-5 receptors on the eosinophil surface. Mepolizumab should be considered in patients with a peripheral eosinophil count > 150 cells/mcL; it has shown a trend of greater benefit in patients with a very high eosinophil count (75% reduction in exacerbation with blood eosinophil count > 500 cells/mcL compared to 56% exacerbation reduction with blood eosinophil count > 150 cells/mcL).³⁷

Mepolizumab has consistently been shown to reduce asthma exacerbation (by about 50%) and emergency department (ED) visits and hospitalization (60%), when compared with placebo in clinical trials.^37,38 It also reduces the need for oral corticosteroids, an effect sustained for up to 52 weeks.^39,40 The Mepolizumab adjUnctive therapy in subjects with Severe eosinophiliC Asthma (MUSCA) study showed that mepolizumab was associated with significant improvement of health-related QoL, lung function, and asthma symptoms in patients with severe eosinophilic asthma.³⁸

GINA guidelines recommend mepolizumab as an add-on therapy for severe asthma. Mepolizumab is given as a fixed dose of 100 mg every 4 weeks. A 300-mg dose has also been approved for eosinophilic granulomatosis with polyangiitis. Monitoring with serial eosinophils might be of value in determining the efficacy of the drug. Mepolizumab is currently in clinical trials for a broad spectrum of diseases, including COPD, hyper-eosinophilic syndrome, and ABPA.

Continue to: Reslizumab (Cinqair)...

Reslizumab (Cinqair) is a rat anti-human monoclonal antibody of the IgG4κ subtype that binds to a small region of IL-5 and subsequently blocks IL-5 from binding to the IL-5 receptor complex on the cell surface of eosinophils. It is currently approved for use as a 3-mg/kg IV infusion every 4 weeks. In large clinical trials,^41-43 reslizumab decreased asthma exacerbation and improved QoL, asthma control, and lung function. Most of the study populations had an eosinophil count > 400 cells/mcL. A small study also suggested patients with severe eosinophilic asthma with prednisone dependency (10 mg/d) had better sputum eosinophilia suppression and asthma control with reslizumab when compared with mepolizumab.⁴⁴

Benralizumab (Fasenra) is a humanized IgG1 anti-IL-5 receptor α monoclonal antibody derived from mice. It induces apoptosis of eosinophils and, to a lesser extent, of basophils.⁴⁵ In clinical trials, it demonstrated a reduction in asthma exacerbation rate and improvement in prebronchodilator FEV₁ and asthma symptoms.^46,47 It does not need reconstitution, as the drug is dispensed as prefilled syringes with fixed non-weight-based dosing. Another potential advantage to benralizumab is that after the loading dose, subsequent doses are given every 8 weeks.

Bronchial thermoplasty

Bronchial thermoplasty (BT) is a novel nonpharmacologic intervention that entails the delivery of controlled radiofrequency-­generated heat via a catheter inserted into the bronchial tree of the lungs through a flexible bronchoscope. The potential mechanism of action is reduction in airway smooth muscle mass and inflammatory markers.

Evidence for BT started with the Asthma Intervention Research (AIR) and Research in Severe Asthma (RISA) trials.^48,49 In the AIR study, BT was shown to reduce the rate of mild exacerbations and improve morning peak expiratory flow and asthma scores at 12 months.⁴⁸ In the RISA trial, BT resulted in improvements in Asthma Quality of Life Questionnaire (AQLQ) score and need for rescue medication at 52 weeks, as well as a trend toward decrease in steroid use.⁴⁹

Bronchial thermoplasty results in clinically worthwhile reductions in severe exacerbations and ED visits in the year post treatment.

However, these studies were criticized for not having a placebo group—an issue addressed in the AIR2 trial, which compared bronchial thermoplasty with a sham procedure. AIR2 demonstrated improvements in AQLQ score and a 32% reduction in severe exacerbations and 84% fewer ED visits in the post-treatment period (up to 1 year post treatment).⁵⁰

Continue to: Both treatment groups...

Both treatment groups experienced an increase in respiratory adverse events: during the treatment period (up to 6 weeks post procedure), 16 subjects (8.4%) in the BT group required 19 hospitalizations for respiratory symptoms and 2 subjects (2%) in the sham group required 2 hospitalizations. A follow-up observational study involving a cohort of AIR2 patients demonstrated long-lasting effects of BT in asthma exacerbation frequency, ED visits, and stabilization of FEV₁ for up to 5 years.⁵¹

The Post-market Post-FDA Approval Clinical Trial Evaluating Bronchial Thermoplasty in Severe Persistent Asthma (PAS2) showed similar beneficial effects of BT on asthma control despite enrolling subjects who may have had poorer asthma control in the “real world” setting.⁵²

In summary, BT results in modest improvements in AQLQ scores and clinically worthwhile reductions in severe exacerbations and ED visits in the year post treatment, which may persist for up to 5 years. BT causes short-term increases in asthma-related morbidity, including hospital admissions. While there is encouraging data and the scope is increasing, BT remains limited to carefully selected (by a specialist) patients with severe asthma that is poorly controlled despite maximal inhaled therapy.

Immunotherapy

Immunotherapy for allergic disease is aimed at inducing immune tolerance to an allergen and alleviating allergic symptoms. This is done by administration of the allergen to which the patient is sensitive. There are 2 approaches: subcutaneous immunotherapy (SCIT) and sublingual immunotherapy (SLIT; a dissolvable tablet under the tongue or an aqueous or liquid extract).

Immunotherapy is generally reserved for patients who have allergic symptoms with exposure to a trigger and evidence (through skin or serum testing) of specific IgE to that trigger, especially if there is poor response to pharmacotherapy and allergen avoidance. Overall, evidence in this field is limited: Most studies have included patients with mild asthma, and few studies have compared immunotherapy with pharmacologic therapy or used standardized outcomes, such as exacerbations.

Continue to: SCIT

SCIT. A 2010 Cochrane review concluded that SCIT reduces asthma symptoms and use of asthma medications and improves bronchial hyperreactivity. Adverse effects include uncommon anaphylactic reactions, which may be life-threatening.⁵³

SLIT has advantages over SCIT as it can be administered by patients or caregivers, does not require injections, and carries a much lower risk for anaphylaxis. Modest benefits have been seen in adults and children, but there is concern about the design of many early studies.

The potential benefits of allergen immunotherapy must be weighed against the risk for adverse effects, including anaphylaxis.

A 2015 Cochrane review of SLIT in asthma recommended further research using validated scales and important outcomes for patients and decision makers so that SLIT can be properly assessed as a clinical treatment for asthma.⁵⁴ A subsequently published study of SLIT for house dust mites (HDM) in patients with asthma and HDM allergic rhinitis demonstrated a modest reduction in use of ICS with high-dose SLIT.⁵⁵

In another recent study, among adults with HDM allergy-related asthma not well controlled by ICS, the addition of HDM SLIT to maintenance medications improved time to first moderate-or-severe asthma exacerbation during ICS reduction.⁵⁶ Additional studies are needed to assess long-term efficacy and safety. However, for patients who experience exacerbations despite use of a low-dose or medium-dose ICS-LABA combination, SLIT can now be considered as an add-on therapy.

Per the GINA guidelines, the potential benefits of allergen immunotherapy must be weighed against the risk for adverse effects, including anaphylaxis, and the inconvenience and cost of the prolonged course of therapy.¹⁵

Continue to: Azithromycin

Azithromycin

Macrolides have immunomodulatory and anti-inflammatory effects in addition to their antibacterial effects. Maintenance treatment with macrolides such as azithromycin has been proven to be effective in chronic neutrophilic airway diseases (FIGURE). There have been attempts to assess whether this therapy can be useful in asthma management, as well. Some randomized controlled trials and meta-analyses have shown conflicting results, and early studies were limited by lack of data, heterogeneous results, and inadequate study designs.

The AZithromycin Against pLacebo in Exacerbations of Asthma (AZALEA) study was a randomized, multicenter, double-blind, placebo-controlled clinical trial in the United Kingdom among patients requiring emergency care for acute asthma exacerbations. Azithromycin added to standard care for asthma attacks did not result in clinical benefit.⁵⁷ While azithromycin in acute exacerbation is not currently recommended, recent trials in outpatient settings have shown promise.

The AZIthromycin in Severe ASThma study (AZISAST) was a randomized, double-blind, placebo-controlled trial in subjects with exacerbation-prone severe asthma in Belgium. Low-dose azithromycin (250 mg 3 times a week) as an add-on treatment to combination ICS-LABA therapy for 6 months did not reduce the rate of severe asthma exacerbations or lower respiratory tract infection (LRTI). However, subjects with a non-eosinophilic variant (neutrophilic phenotype) experienced significant reduction in the rate of exacerbation and LRTI.⁵⁸

Chronic azithromycin might prove to be a useful agent in the long-term management of asthma patients whose disease is not well controlled on inhaled therapy

The recently published Asthma and Macrolides: the AZithromycin Efficacy and Safety Study (AMAZES) shows promise for chronic azithromycin therapy as an add-on to medium-to-high-dose inhaled steroids and a long-acting bronchodilator in adults with uncontrolled persistent asthma. This was a large multicenter, randomized, double-blind, placebo-controlled, parallel group trial in New Zealand and Australia. Patients were excluded if they had hearing impairment or abnormally prolonged QTc. Azithromycin at a dose of 500 mg 3 times a week for 48 months reduced asthma exacerbations and improved QoL compared to placebo. The effect was sustained between subgroups based on phenotypes (eosinophilic vs noneosinophilic; frequent exacerbators vs nonfrequent exacerbators) and even among those with symptom differences at baseline (eg, cough or sputum positivity). The rate of antibiotic courses for respiratory infectious episodes was significantly reduced in the azithromycin-treated group.⁵⁹

The take-away: Chronic azithromycin might prove to be a useful agent in the long-term management of asthma patients whose disease is not well controlled on inhaled therapy. Further studies on mechanism and effects of prolonged antibiotic use will shed more light. For more information, see When guideline treatment of asthma fails, consider a macrolide antibiotic; http://bit.ly/2vDAWc6.

Continue to: A new era

We have entered an exciting era of asthma management, with the introduction of several novel modalities, such as biological therapy and bronchial thermoplasty, as well as use of known drugs such as macrolides, immunotherapy, and LAMA. This was made possible through a better understanding of the biological pathways of asthma. Asthma management has moved toward more personalized, targeted therapy based on asthma phenotypes.

It’s important to remember, however, that pharmacological and nonpharmacological aspects of management—including inhaler techniques, adherence to inhaler therapy, vaccinations, control of asthma triggers, and smoking cessation—remain the foundation of optimal asthma management and need to be aggressively addressed before embarking on advanced treatment options. Patients whose asthma is not well controlled with inhaled medications or who have frequent exacerbations (requiring use of steroids) should be comanaged by an expert asthma specialist to explore all possible therapies.

CORRESPONDENCE
Mayur Rali, MD, 995 Newbridge Road, Bellmore, NY 11710; mrali@northwell.edu

Recent advances in our understanding of asthma pathophysiology have led to the development of new treatment approaches to this chronic respiratory condition, which affects 25 million Americans or nearly 8% of the population.¹ As a result, asthma treatment options have expanded from just simple inhalers and corticosteroids to include biological therapies, immunotherapy, bronchial thermoplasty, and anti-inflammatory agents. This review will focus on the newer therapeutic options and provide guidance on when to seek expert pulmonary consultation (TABLE 1). But before we begin, it’s important to briefly review the pathophysiology of asthma and several key diagnostic considerations.

The pathophysiology of asthma provides key targets for therapy

There are 2 basic phenotypes of asthma—neutrophilic predominant and eosinophilic predominant—and 3 key components to its pathophysiology²:

Airway inflammation. Asthma is mediated through either a type 1 T-helper (Th-1) cell or a type 2 T-helper (Th-2) cell response, the pathways of which have a fair amount of overlap (FIGURE). In the neutrophilic-predominant phenotype, irritants, pollutants, and viruses trigger an innate Th-1 cell–mediated pathway that leads to subsequent neutrophil release. This asthma phenotype responds poorly to standard asthma therapy.^2-4

In the eosinophilic-predominant phenotype, environmental allergic antigens induce a Th-2 cell–mediated response in the airways of patients with asthma.^5-7 This creates a downstream effect on the release of interleukins (IL) including IL-4, IL-5, and IL-13. IL-4 triggers immunoglobulin (Ig) E release, which subsequently induces mast cells to release inflammatory cytokines, while IL-5 and IL-13 are responsible for eosinophilic response. These cytokines and eosinophils induce airway hyperresponsiveness, remodeling, and mucus production. Through repeated exposure, chronic inflammation develops and subsequently causes structural changes related to increased smooth muscle mass, goblet cell hyperplasia, and thickening of lamina reticularis.^8,9 Understanding of this pathobiological pathway has led to the development of anti-IgE and anti-IL-5 drugs (to be discussed shortly).

It is important to determine whether patients with asthma-COPD overlap syndrome are asthma predominant or COPD predominant, because appropriate first-line treatment will differ.

Airway obstruction. Early asthmatic response is due to acute bronchoconstriction secondary to IgE; this is followed by airway edema occurring 6 to 24 hours after an acute event (called late asthmatic response). The obstruction is worsened by an overproduction of mucus, which may take weeks to resolve.¹⁰ Longstanding inflammation can lead to structural changes and reduced airflow reversibility.

Bronchial hyperresponsiveness is induced by various forms of allergens, pollutants, or viral upper respiratory infections. Sympathetic control in the airway is mediated via beta-2 adrenoceptors expressed on airway smooth muscle, which are responsible for the effect of bronchodilation in response to albuterol.^11,12 Cholinergic pathways may further contribute to bronchial hyperresponsiveness and form the basis for the efficacy of anticholinergic therapy.^12,13

What we’ve learned about asthma can inform treatment decisions

Presentation may vary, as asthma has many forms including cough-variant asthma and exercise-induced asthma. Airflow limitation is typically identified through spirometry and characterized by reduced (< 70% in adults) forced expiratory volume in 1 second (FEV₁)/forced vital capacity (FVC) or bronchodilator response positivity (an increase in post-bronchodilator FEV₁ > 12% or FVC > 200 mL from baseline).² If spirometry is not diagnostic but suspicion for asthma remains, bronchial provocation testing or exercise challenge testing may be needed.

Continue to: Additional diagnostic considerations...

Additional diagnostic considerations may impact the treatment plan for patients with asthma:

Asthma and COPD. A history of smoking is a key factor in the diagnosis of chronic obstructive pulmonary disease (COPD)—but many patients with asthma are also smokers. This subgroup may have asthma-COPD overlap syndrome (ACOS). It is important to determine whether these patients are asthma predominant or COPD predominant, because appropriate first-line treatment will differ. Patients who are COPD predominant demonstrate reduced diffusion capacity (DLCO) and abnormal PaCO2 on arterial blood gas. They also may show more structural damage on chest computed tomography (CT) than patients with asthma do. Asthma-predominant patients are more likely to have eosinophilia.¹⁴

Patients with severe persistent asthma or frequent exacerbations, or those receiving step-up therapy, may require additional serologic testing. Specialized testing for IgE and eosinophil count, as well as a sensitized allergy panel, may help clinicians in selecting specific biological therapies for treatment of severe asthma (further discussion to follow). We recommend using a serum allergy panel, as it is a quick and easy way to identify patients with extrinsic allergies, whereas skin-based testing is often time consuming and may require referral to a specialist.^2,5,15

Aspergillus. An additional consideration is testing for Aspergillus antibodies. Aspergillus is a ubiquitous fungus found in the airways of humans. In patients with asthma, however, it can trigger an intense inflammatory response known as allergic bronchopulmonary aspergillosis. ABPA is not an infection. It should be considered in patients who have lived in a damp, old housing environment with possible mold exposure. Treatment of ABPA involves oral corticosteroids; there are varying reports of efficacy with voriconazole or itraconazole as suppressive therapy or steroid-sparing treatment.^16-18

Getting a handle on an ever-expanding asthma Tx arsenal

The goals of asthma treatment are symptom control and risk minimization. Treatment choices are dictated in part by disease severity (mild, moderate, severe) and classification (intermittent, persistent). Asthma therapy is traditionally described as step-up and step-down; TABLE 2 summarizes available pharmacotherapy for asthma and provides a framework for add-on therapy as the disease advances.

Continue to: Over the past decade...

Over the past decade, a number of therapeutic options have been introduced or added to the pantheon of asthma treatment.

Inhaled medications

This category includes inhaled corticosteroids (ICS), which are recommended for use alone or in combination with long-acting beta-agonists (LABA) or with long-acting muscarinic agonists (LAMA).

ICS is the first choice for long-term control of persistent asthma.² Its molecular effects include activating anti-inflammatory genes, switching off inflammatory genes, and inhibiting inflammatory cells, combined with enhancement of beta-2-adrenergic receptor expression. The cumulative effect is reduction in airway responsiveness in asthma patients.^19-22

LABAs are next in line in the step-up, step-down model of symptom management. LABAs should not be prescribed as stand-alone therapy in patients with asthma, as they have received a black box warning from the US Food and Drug Administration (FDA) for an increase in asthma-related death²³—a concern that has not been demonstrated with the combination of ICS-LABA.

LABAs should not be prescribed as stand-alone therapy in patients with asthma, as they have received an FDA black box warning.

LABAs cause smooth muscle relaxation in the lungs.²⁴ There are 3 combination products currently available: once-daily fluticasone furoate/vilanterol (Breo), twice-daily fluticasone propionate/salmeterol (Advair), and twice-daily budesonide/formoterol (Symbicort).

Continue to: Once-daily fluticasone furoate/vilanterol...

Once-daily fluticasone furoate/vilanterol has been shown to improve mean FEV₁.²⁵ In a 24-week, open-label, multicenter randomized controlled trial to evaluate the efficacy and safety of all 3 combination ICS-LABAs, preliminary results indicated that—at least in a tightly controlled setting—once-daily fluticasone furoate/vilanterol provides asthma control similar to the twice-daily combinations and is well tolerated.²⁶

Two ultra-long-acting (24-hour) LABAs, olodaterol (Striverdi Respimat) and indacaterol (Arcapta Neohaler), are being studied for possible use in asthma treatment. In a phase 2 trial investigating therapy for moderate-to-severe persistent asthma, 24-hour FEV₁ improved with olodeaterol when compared to placebo.²⁷

Another ongoing clinical trial is studying the effects of ultra-long-acting bronchodilator therapy (olodaterol vs combination olodaterol/tiotropium) in asthma patients who smoke and who are already using ICS (ClinicalTrials.gov NCT02682862). Indacaterol has been shown to be effective in the treatment of moderate-to-severe asthma in a once-a-day dosing regimen.²⁸ However, when compared to mometasone alone, a combination of indacaterol and mometasone demonstrated no statistically significant reduction in time to serious exacerbation.²⁹

The LAMA tiotropium is recommended as add-on therapy for patients whose asthma is uncontrolled despite use of low-dose ICS-LABA or as an alternative to high-dose ICS-LABA, per Global Initiative for Asthma (GINA) 2019 guidelines.¹⁵

Ideal candidates for biological therapy are patients who have exhausted other forms of severe asthma treatment.

Tiotropium induces bronchodilation by selectively inhibiting the action of acetylcholine at muscarinic (M) receptors in bronchial smooth muscles; it has a longer duration of action because of its slower dissociation from receptor types M₁ and M₃.³⁰ Tiotropium respimat (Spiriva, Tiova) has been approved for COPD for many years; in 2013, it was shown to prevent worsening of symptomatic asthma and increase time to first severe exacerbation.¹³ The FDA subsequently approved tiotropium as an add-on treatment for patients with uncontrolled asthma despite use of ICS-LABA.

Continue to: Glycopyrronium bromide...

Glycopyrronium bromide (glycopyrrolate, multiple brand names) and umeclidinium (Incruse Ellipta) are LAMAs that are approved for COPD treatment but have not yet been approved for patients who have asthma only.³¹

Biological therapies

In the past few years, improved understanding of asthma’s pathophysiology has led to the development of biological therapy for severe asthma. This therapy is directed at Th-2 inflammatory pathways (FIGURE) and targets various inflammatory markers, such as IgE, IL-5, and eosinophils.

Biologicals are not the first-line therapy for the management of severe asthma. Ideal candidates for this therapy are patients who have exhausted other forms of severe asthma treatment, including ICS-LABA, LAMA, leukotriene receptor antagonists, and mucus-clearing agents. Patients with frequent exacerbations who need continuous steroids or need steroids at least twice a year should be considered for biologicals.³²

All biological therapies must be administered in a clinical setting, as they carry risk for anaphylaxis. TABLE 3^15,33-47 summarizes all approved biologicals for the management of severe asthma.

Anti-IgE therapy. Omalizumab (Xolair) was the first approved biological therapy for severe asthma (in 2003). It is a recombinant humanized IgG1 monoclonal antibody that binds to free IgE and down regulates the inflammatory cascade. It is therefore best suited for patients with early-onset allergic asthma with a high IgE count. The dose and frequency (once or twice per month) of omalizumab are based on IgE levels and patient weight. Omalizumab reduces asthma exacerbation (up to 45%) and hospitalization (up to 85%).³⁴ Omalizumab also reduces the need for high-dose ICS-LABA therapy and improves quality of life (QoL).^33,34

Continue to: Its efficacy and safety...

Mepolizumab has shown a trend of greater benefit in patients with a very high eosinophil count.

Its efficacy and safety have been proven outside the clinical trial setting. Treatment response should be assessed over a 3- to 4-month period, using fractional exhalation of nitric oxide (FeNO); serial measurement of IgE levels is not recommended for this purpose. Once started, treatment should be considered long term, as discontinuation of treatment has been shown to lead to recurrence of symptoms and exacerbation.^35,36 Of note, the GINA guidelines recommend omalizumab over prednisone as add-on therapy for severe persistent asthma.¹⁵

Anti-IL-5 therapy. IL-5 is the main cytokine for growth, differentiation, and activation of eosinophils in the Th-2-mediated inflammatory cascade. Mepolizumab, reslizumab, and benralizumab are 3 FDA-approved anti-IL-5 monoclonal antibody therapies for severe eosinophilic asthma. Mepolizumab has been the most commonly studied anti-IL-5 therapy, while benralizumab, the latest of the 3, has a unique property of inducing eosinophilic apoptosis. There has been no direct comparison of the different anti-IL-5 therapies.

Mepolizumab (Nucala) is a mouse anti-human monoclonal antibody that binds to IL-5 and prevents it from binding to IL-5 receptors on the eosinophil surface. Mepolizumab should be considered in patients with a peripheral eosinophil count > 150 cells/mcL; it has shown a trend of greater benefit in patients with a very high eosinophil count (75% reduction in exacerbation with blood eosinophil count > 500 cells/mcL compared to 56% exacerbation reduction with blood eosinophil count > 150 cells/mcL).³⁷

Mepolizumab has consistently been shown to reduce asthma exacerbation (by about 50%) and emergency department (ED) visits and hospitalization (60%), when compared with placebo in clinical trials.^37,38 It also reduces the need for oral corticosteroids, an effect sustained for up to 52 weeks.^39,40 The Mepolizumab adjUnctive therapy in subjects with Severe eosinophiliC Asthma (MUSCA) study showed that mepolizumab was associated with significant improvement of health-related QoL, lung function, and asthma symptoms in patients with severe eosinophilic asthma.³⁸

GINA guidelines recommend mepolizumab as an add-on therapy for severe asthma. Mepolizumab is given as a fixed dose of 100 mg every 4 weeks. A 300-mg dose has also been approved for eosinophilic granulomatosis with polyangiitis. Monitoring with serial eosinophils might be of value in determining the efficacy of the drug. Mepolizumab is currently in clinical trials for a broad spectrum of diseases, including COPD, hyper-eosinophilic syndrome, and ABPA.

Continue to: Reslizumab (Cinqair)...

Reslizumab (Cinqair) is a rat anti-human monoclonal antibody of the IgG4κ subtype that binds to a small region of IL-5 and subsequently blocks IL-5 from binding to the IL-5 receptor complex on the cell surface of eosinophils. It is currently approved for use as a 3-mg/kg IV infusion every 4 weeks. In large clinical trials,^41-43 reslizumab decreased asthma exacerbation and improved QoL, asthma control, and lung function. Most of the study populations had an eosinophil count > 400 cells/mcL. A small study also suggested patients with severe eosinophilic asthma with prednisone dependency (10 mg/d) had better sputum eosinophilia suppression and asthma control with reslizumab when compared with mepolizumab.⁴⁴

Benralizumab (Fasenra) is a humanized IgG1 anti-IL-5 receptor α monoclonal antibody derived from mice. It induces apoptosis of eosinophils and, to a lesser extent, of basophils.⁴⁵ In clinical trials, it demonstrated a reduction in asthma exacerbation rate and improvement in prebronchodilator FEV₁ and asthma symptoms.^46,47 It does not need reconstitution, as the drug is dispensed as prefilled syringes with fixed non-weight-based dosing. Another potential advantage to benralizumab is that after the loading dose, subsequent doses are given every 8 weeks.

Bronchial thermoplasty

Bronchial thermoplasty (BT) is a novel nonpharmacologic intervention that entails the delivery of controlled radiofrequency-­generated heat via a catheter inserted into the bronchial tree of the lungs through a flexible bronchoscope. The potential mechanism of action is reduction in airway smooth muscle mass and inflammatory markers.

Evidence for BT started with the Asthma Intervention Research (AIR) and Research in Severe Asthma (RISA) trials.^48,49 In the AIR study, BT was shown to reduce the rate of mild exacerbations and improve morning peak expiratory flow and asthma scores at 12 months.⁴⁸ In the RISA trial, BT resulted in improvements in Asthma Quality of Life Questionnaire (AQLQ) score and need for rescue medication at 52 weeks, as well as a trend toward decrease in steroid use.⁴⁹

Bronchial thermoplasty results in clinically worthwhile reductions in severe exacerbations and ED visits in the year post treatment.

However, these studies were criticized for not having a placebo group—an issue addressed in the AIR2 trial, which compared bronchial thermoplasty with a sham procedure. AIR2 demonstrated improvements in AQLQ score and a 32% reduction in severe exacerbations and 84% fewer ED visits in the post-treatment period (up to 1 year post treatment).⁵⁰

Continue to: Both treatment groups...

Both treatment groups experienced an increase in respiratory adverse events: during the treatment period (up to 6 weeks post procedure), 16 subjects (8.4%) in the BT group required 19 hospitalizations for respiratory symptoms and 2 subjects (2%) in the sham group required 2 hospitalizations. A follow-up observational study involving a cohort of AIR2 patients demonstrated long-lasting effects of BT in asthma exacerbation frequency, ED visits, and stabilization of FEV₁ for up to 5 years.⁵¹

The Post-market Post-FDA Approval Clinical Trial Evaluating Bronchial Thermoplasty in Severe Persistent Asthma (PAS2) showed similar beneficial effects of BT on asthma control despite enrolling subjects who may have had poorer asthma control in the “real world” setting.⁵²

In summary, BT results in modest improvements in AQLQ scores and clinically worthwhile reductions in severe exacerbations and ED visits in the year post treatment, which may persist for up to 5 years. BT causes short-term increases in asthma-related morbidity, including hospital admissions. While there is encouraging data and the scope is increasing, BT remains limited to carefully selected (by a specialist) patients with severe asthma that is poorly controlled despite maximal inhaled therapy.

Immunotherapy

Immunotherapy for allergic disease is aimed at inducing immune tolerance to an allergen and alleviating allergic symptoms. This is done by administration of the allergen to which the patient is sensitive. There are 2 approaches: subcutaneous immunotherapy (SCIT) and sublingual immunotherapy (SLIT; a dissolvable tablet under the tongue or an aqueous or liquid extract).

Immunotherapy is generally reserved for patients who have allergic symptoms with exposure to a trigger and evidence (through skin or serum testing) of specific IgE to that trigger, especially if there is poor response to pharmacotherapy and allergen avoidance. Overall, evidence in this field is limited: Most studies have included patients with mild asthma, and few studies have compared immunotherapy with pharmacologic therapy or used standardized outcomes, such as exacerbations.

Continue to: SCIT

SCIT. A 2010 Cochrane review concluded that SCIT reduces asthma symptoms and use of asthma medications and improves bronchial hyperreactivity. Adverse effects include uncommon anaphylactic reactions, which may be life-threatening.⁵³

SLIT has advantages over SCIT as it can be administered by patients or caregivers, does not require injections, and carries a much lower risk for anaphylaxis. Modest benefits have been seen in adults and children, but there is concern about the design of many early studies.

The potential benefits of allergen immunotherapy must be weighed against the risk for adverse effects, including anaphylaxis.

A 2015 Cochrane review of SLIT in asthma recommended further research using validated scales and important outcomes for patients and decision makers so that SLIT can be properly assessed as a clinical treatment for asthma.⁵⁴ A subsequently published study of SLIT for house dust mites (HDM) in patients with asthma and HDM allergic rhinitis demonstrated a modest reduction in use of ICS with high-dose SLIT.⁵⁵

In another recent study, among adults with HDM allergy-related asthma not well controlled by ICS, the addition of HDM SLIT to maintenance medications improved time to first moderate-or-severe asthma exacerbation during ICS reduction.⁵⁶ Additional studies are needed to assess long-term efficacy and safety. However, for patients who experience exacerbations despite use of a low-dose or medium-dose ICS-LABA combination, SLIT can now be considered as an add-on therapy.

Per the GINA guidelines, the potential benefits of allergen immunotherapy must be weighed against the risk for adverse effects, including anaphylaxis, and the inconvenience and cost of the prolonged course of therapy.¹⁵

Continue to: Azithromycin

Azithromycin

Macrolides have immunomodulatory and anti-inflammatory effects in addition to their antibacterial effects. Maintenance treatment with macrolides such as azithromycin has been proven to be effective in chronic neutrophilic airway diseases (FIGURE). There have been attempts to assess whether this therapy can be useful in asthma management, as well. Some randomized controlled trials and meta-analyses have shown conflicting results, and early studies were limited by lack of data, heterogeneous results, and inadequate study designs.

The AZithromycin Against pLacebo in Exacerbations of Asthma (AZALEA) study was a randomized, multicenter, double-blind, placebo-controlled clinical trial in the United Kingdom among patients requiring emergency care for acute asthma exacerbations. Azithromycin added to standard care for asthma attacks did not result in clinical benefit.⁵⁷ While azithromycin in acute exacerbation is not currently recommended, recent trials in outpatient settings have shown promise.

The AZIthromycin in Severe ASThma study (AZISAST) was a randomized, double-blind, placebo-controlled trial in subjects with exacerbation-prone severe asthma in Belgium. Low-dose azithromycin (250 mg 3 times a week) as an add-on treatment to combination ICS-LABA therapy for 6 months did not reduce the rate of severe asthma exacerbations or lower respiratory tract infection (LRTI). However, subjects with a non-eosinophilic variant (neutrophilic phenotype) experienced significant reduction in the rate of exacerbation and LRTI.⁵⁸

Chronic azithromycin might prove to be a useful agent in the long-term management of asthma patients whose disease is not well controlled on inhaled therapy

The recently published Asthma and Macrolides: the AZithromycin Efficacy and Safety Study (AMAZES) shows promise for chronic azithromycin therapy as an add-on to medium-to-high-dose inhaled steroids and a long-acting bronchodilator in adults with uncontrolled persistent asthma. This was a large multicenter, randomized, double-blind, placebo-controlled, parallel group trial in New Zealand and Australia. Patients were excluded if they had hearing impairment or abnormally prolonged QTc. Azithromycin at a dose of 500 mg 3 times a week for 48 months reduced asthma exacerbations and improved QoL compared to placebo. The effect was sustained between subgroups based on phenotypes (eosinophilic vs noneosinophilic; frequent exacerbators vs nonfrequent exacerbators) and even among those with symptom differences at baseline (eg, cough or sputum positivity). The rate of antibiotic courses for respiratory infectious episodes was significantly reduced in the azithromycin-treated group.⁵⁹

The take-away: Chronic azithromycin might prove to be a useful agent in the long-term management of asthma patients whose disease is not well controlled on inhaled therapy. Further studies on mechanism and effects of prolonged antibiotic use will shed more light. For more information, see When guideline treatment of asthma fails, consider a macrolide antibiotic; http://bit.ly/2vDAWc6.

Continue to: A new era

We have entered an exciting era of asthma management, with the introduction of several novel modalities, such as biological therapy and bronchial thermoplasty, as well as use of known drugs such as macrolides, immunotherapy, and LAMA. This was made possible through a better understanding of the biological pathways of asthma. Asthma management has moved toward more personalized, targeted therapy based on asthma phenotypes.

It’s important to remember, however, that pharmacological and nonpharmacological aspects of management—including inhaler techniques, adherence to inhaler therapy, vaccinations, control of asthma triggers, and smoking cessation—remain the foundation of optimal asthma management and need to be aggressively addressed before embarking on advanced treatment options. Patients whose asthma is not well controlled with inhaled medications or who have frequent exacerbations (requiring use of steroids) should be comanaged by an expert asthma specialist to explore all possible therapies.

CORRESPONDENCE
Mayur Rali, MD, 995 Newbridge Road, Bellmore, NY 11710; mrali@northwell.edu

Recent advances in our understanding of asthma pathophysiology have led to the development of new treatment approaches to this chronic respiratory condition, which affects 25 million Americans or nearly 8% of the population.¹ As a result, asthma treatment options have expanded from just simple inhalers and corticosteroids to include biological therapies, immunotherapy, bronchial thermoplasty, and anti-inflammatory agents. This review will focus on the newer therapeutic options and provide guidance on when to seek expert pulmonary consultation (TABLE 1). But before we begin, it’s important to briefly review the pathophysiology of asthma and several key diagnostic considerations.

The pathophysiology of asthma provides key targets for therapy

There are 2 basic phenotypes of asthma—neutrophilic predominant and eosinophilic predominant—and 3 key components to its pathophysiology²:

Airway inflammation. Asthma is mediated through either a type 1 T-helper (Th-1) cell or a type 2 T-helper (Th-2) cell response, the pathways of which have a fair amount of overlap (FIGURE). In the neutrophilic-predominant phenotype, irritants, pollutants, and viruses trigger an innate Th-1 cell–mediated pathway that leads to subsequent neutrophil release. This asthma phenotype responds poorly to standard asthma therapy.^2-4

In the eosinophilic-predominant phenotype, environmental allergic antigens induce a Th-2 cell–mediated response in the airways of patients with asthma.^5-7 This creates a downstream effect on the release of interleukins (IL) including IL-4, IL-5, and IL-13. IL-4 triggers immunoglobulin (Ig) E release, which subsequently induces mast cells to release inflammatory cytokines, while IL-5 and IL-13 are responsible for eosinophilic response. These cytokines and eosinophils induce airway hyperresponsiveness, remodeling, and mucus production. Through repeated exposure, chronic inflammation develops and subsequently causes structural changes related to increased smooth muscle mass, goblet cell hyperplasia, and thickening of lamina reticularis.^8,9 Understanding of this pathobiological pathway has led to the development of anti-IgE and anti-IL-5 drugs (to be discussed shortly).

It is important to determine whether patients with asthma-COPD overlap syndrome are asthma predominant or COPD predominant, because appropriate first-line treatment will differ.

Airway obstruction. Early asthmatic response is due to acute bronchoconstriction secondary to IgE; this is followed by airway edema occurring 6 to 24 hours after an acute event (called late asthmatic response). The obstruction is worsened by an overproduction of mucus, which may take weeks to resolve.¹⁰ Longstanding inflammation can lead to structural changes and reduced airflow reversibility.

Bronchial hyperresponsiveness is induced by various forms of allergens, pollutants, or viral upper respiratory infections. Sympathetic control in the airway is mediated via beta-2 adrenoceptors expressed on airway smooth muscle, which are responsible for the effect of bronchodilation in response to albuterol.^11,12 Cholinergic pathways may further contribute to bronchial hyperresponsiveness and form the basis for the efficacy of anticholinergic therapy.^12,13

What we’ve learned about asthma can inform treatment decisions

Presentation may vary, as asthma has many forms including cough-variant asthma and exercise-induced asthma. Airflow limitation is typically identified through spirometry and characterized by reduced (< 70% in adults) forced expiratory volume in 1 second (FEV₁)/forced vital capacity (FVC) or bronchodilator response positivity (an increase in post-bronchodilator FEV₁ > 12% or FVC > 200 mL from baseline).² If spirometry is not diagnostic but suspicion for asthma remains, bronchial provocation testing or exercise challenge testing may be needed.

Continue to: Additional diagnostic considerations...

Additional diagnostic considerations may impact the treatment plan for patients with asthma:

Asthma and COPD. A history of smoking is a key factor in the diagnosis of chronic obstructive pulmonary disease (COPD)—but many patients with asthma are also smokers. This subgroup may have asthma-COPD overlap syndrome (ACOS). It is important to determine whether these patients are asthma predominant or COPD predominant, because appropriate first-line treatment will differ. Patients who are COPD predominant demonstrate reduced diffusion capacity (DLCO) and abnormal PaCO2 on arterial blood gas. They also may show more structural damage on chest computed tomography (CT) than patients with asthma do. Asthma-predominant patients are more likely to have eosinophilia.¹⁴

Patients with severe persistent asthma or frequent exacerbations, or those receiving step-up therapy, may require additional serologic testing. Specialized testing for IgE and eosinophil count, as well as a sensitized allergy panel, may help clinicians in selecting specific biological therapies for treatment of severe asthma (further discussion to follow). We recommend using a serum allergy panel, as it is a quick and easy way to identify patients with extrinsic allergies, whereas skin-based testing is often time consuming and may require referral to a specialist.^2,5,15

Aspergillus. An additional consideration is testing for Aspergillus antibodies. Aspergillus is a ubiquitous fungus found in the airways of humans. In patients with asthma, however, it can trigger an intense inflammatory response known as allergic bronchopulmonary aspergillosis. ABPA is not an infection. It should be considered in patients who have lived in a damp, old housing environment with possible mold exposure. Treatment of ABPA involves oral corticosteroids; there are varying reports of efficacy with voriconazole or itraconazole as suppressive therapy or steroid-sparing treatment.^16-18

Getting a handle on an ever-expanding asthma Tx arsenal

The goals of asthma treatment are symptom control and risk minimization. Treatment choices are dictated in part by disease severity (mild, moderate, severe) and classification (intermittent, persistent). Asthma therapy is traditionally described as step-up and step-down; TABLE 2 summarizes available pharmacotherapy for asthma and provides a framework for add-on therapy as the disease advances.

Continue to: Over the past decade...

Over the past decade, a number of therapeutic options have been introduced or added to the pantheon of asthma treatment.

Inhaled medications

This category includes inhaled corticosteroids (ICS), which are recommended for use alone or in combination with long-acting beta-agonists (LABA) or with long-acting muscarinic agonists (LAMA).

ICS is the first choice for long-term control of persistent asthma.² Its molecular effects include activating anti-inflammatory genes, switching off inflammatory genes, and inhibiting inflammatory cells, combined with enhancement of beta-2-adrenergic receptor expression. The cumulative effect is reduction in airway responsiveness in asthma patients.^19-22

LABAs are next in line in the step-up, step-down model of symptom management. LABAs should not be prescribed as stand-alone therapy in patients with asthma, as they have received a black box warning from the US Food and Drug Administration (FDA) for an increase in asthma-related death²³—a concern that has not been demonstrated with the combination of ICS-LABA.

LABAs should not be prescribed as stand-alone therapy in patients with asthma, as they have received an FDA black box warning.

LABAs cause smooth muscle relaxation in the lungs.²⁴ There are 3 combination products currently available: once-daily fluticasone furoate/vilanterol (Breo), twice-daily fluticasone propionate/salmeterol (Advair), and twice-daily budesonide/formoterol (Symbicort).

Continue to: Once-daily fluticasone furoate/vilanterol...

Once-daily fluticasone furoate/vilanterol has been shown to improve mean FEV₁.²⁵ In a 24-week, open-label, multicenter randomized controlled trial to evaluate the efficacy and safety of all 3 combination ICS-LABAs, preliminary results indicated that—at least in a tightly controlled setting—once-daily fluticasone furoate/vilanterol provides asthma control similar to the twice-daily combinations and is well tolerated.²⁶

Two ultra-long-acting (24-hour) LABAs, olodaterol (Striverdi Respimat) and indacaterol (Arcapta Neohaler), are being studied for possible use in asthma treatment. In a phase 2 trial investigating therapy for moderate-to-severe persistent asthma, 24-hour FEV₁ improved with olodeaterol when compared to placebo.²⁷

Another ongoing clinical trial is studying the effects of ultra-long-acting bronchodilator therapy (olodaterol vs combination olodaterol/tiotropium) in asthma patients who smoke and who are already using ICS (ClinicalTrials.gov NCT02682862). Indacaterol has been shown to be effective in the treatment of moderate-to-severe asthma in a once-a-day dosing regimen.²⁸ However, when compared to mometasone alone, a combination of indacaterol and mometasone demonstrated no statistically significant reduction in time to serious exacerbation.²⁹

The LAMA tiotropium is recommended as add-on therapy for patients whose asthma is uncontrolled despite use of low-dose ICS-LABA or as an alternative to high-dose ICS-LABA, per Global Initiative for Asthma (GINA) 2019 guidelines.¹⁵

Ideal candidates for biological therapy are patients who have exhausted other forms of severe asthma treatment.

Tiotropium induces bronchodilation by selectively inhibiting the action of acetylcholine at muscarinic (M) receptors in bronchial smooth muscles; it has a longer duration of action because of its slower dissociation from receptor types M₁ and M₃.³⁰ Tiotropium respimat (Spiriva, Tiova) has been approved for COPD for many years; in 2013, it was shown to prevent worsening of symptomatic asthma and increase time to first severe exacerbation.¹³ The FDA subsequently approved tiotropium as an add-on treatment for patients with uncontrolled asthma despite use of ICS-LABA.

Continue to: Glycopyrronium bromide...

Glycopyrronium bromide (glycopyrrolate, multiple brand names) and umeclidinium (Incruse Ellipta) are LAMAs that are approved for COPD treatment but have not yet been approved for patients who have asthma only.³¹

Biological therapies

In the past few years, improved understanding of asthma’s pathophysiology has led to the development of biological therapy for severe asthma. This therapy is directed at Th-2 inflammatory pathways (FIGURE) and targets various inflammatory markers, such as IgE, IL-5, and eosinophils.

Biologicals are not the first-line therapy for the management of severe asthma. Ideal candidates for this therapy are patients who have exhausted other forms of severe asthma treatment, including ICS-LABA, LAMA, leukotriene receptor antagonists, and mucus-clearing agents. Patients with frequent exacerbations who need continuous steroids or need steroids at least twice a year should be considered for biologicals.³²

All biological therapies must be administered in a clinical setting, as they carry risk for anaphylaxis. TABLE 3^15,33-47 summarizes all approved biologicals for the management of severe asthma.

Anti-IgE therapy. Omalizumab (Xolair) was the first approved biological therapy for severe asthma (in 2003). It is a recombinant humanized IgG1 monoclonal antibody that binds to free IgE and down regulates the inflammatory cascade. It is therefore best suited for patients with early-onset allergic asthma with a high IgE count. The dose and frequency (once or twice per month) of omalizumab are based on IgE levels and patient weight. Omalizumab reduces asthma exacerbation (up to 45%) and hospitalization (up to 85%).³⁴ Omalizumab also reduces the need for high-dose ICS-LABA therapy and improves quality of life (QoL).^33,34

Continue to: Its efficacy and safety...

Mepolizumab has shown a trend of greater benefit in patients with a very high eosinophil count.

Its efficacy and safety have been proven outside the clinical trial setting. Treatment response should be assessed over a 3- to 4-month period, using fractional exhalation of nitric oxide (FeNO); serial measurement of IgE levels is not recommended for this purpose. Once started, treatment should be considered long term, as discontinuation of treatment has been shown to lead to recurrence of symptoms and exacerbation.^35,36 Of note, the GINA guidelines recommend omalizumab over prednisone as add-on therapy for severe persistent asthma.¹⁵

Anti-IL-5 therapy. IL-5 is the main cytokine for growth, differentiation, and activation of eosinophils in the Th-2-mediated inflammatory cascade. Mepolizumab, reslizumab, and benralizumab are 3 FDA-approved anti-IL-5 monoclonal antibody therapies for severe eosinophilic asthma. Mepolizumab has been the most commonly studied anti-IL-5 therapy, while benralizumab, the latest of the 3, has a unique property of inducing eosinophilic apoptosis. There has been no direct comparison of the different anti-IL-5 therapies.

Mepolizumab (Nucala) is a mouse anti-human monoclonal antibody that binds to IL-5 and prevents it from binding to IL-5 receptors on the eosinophil surface. Mepolizumab should be considered in patients with a peripheral eosinophil count > 150 cells/mcL; it has shown a trend of greater benefit in patients with a very high eosinophil count (75% reduction in exacerbation with blood eosinophil count > 500 cells/mcL compared to 56% exacerbation reduction with blood eosinophil count > 150 cells/mcL).³⁷

Mepolizumab has consistently been shown to reduce asthma exacerbation (by about 50%) and emergency department (ED) visits and hospitalization (60%), when compared with placebo in clinical trials.^37,38 It also reduces the need for oral corticosteroids, an effect sustained for up to 52 weeks.^39,40 The Mepolizumab adjUnctive therapy in subjects with Severe eosinophiliC Asthma (MUSCA) study showed that mepolizumab was associated with significant improvement of health-related QoL, lung function, and asthma symptoms in patients with severe eosinophilic asthma.³⁸

GINA guidelines recommend mepolizumab as an add-on therapy for severe asthma. Mepolizumab is given as a fixed dose of 100 mg every 4 weeks. A 300-mg dose has also been approved for eosinophilic granulomatosis with polyangiitis. Monitoring with serial eosinophils might be of value in determining the efficacy of the drug. Mepolizumab is currently in clinical trials for a broad spectrum of diseases, including COPD, hyper-eosinophilic syndrome, and ABPA.

Continue to: Reslizumab (Cinqair)...

Reslizumab (Cinqair) is a rat anti-human monoclonal antibody of the IgG4κ subtype that binds to a small region of IL-5 and subsequently blocks IL-5 from binding to the IL-5 receptor complex on the cell surface of eosinophils. It is currently approved for use as a 3-mg/kg IV infusion every 4 weeks. In large clinical trials,^41-43 reslizumab decreased asthma exacerbation and improved QoL, asthma control, and lung function. Most of the study populations had an eosinophil count > 400 cells/mcL. A small study also suggested patients with severe eosinophilic asthma with prednisone dependency (10 mg/d) had better sputum eosinophilia suppression and asthma control with reslizumab when compared with mepolizumab.⁴⁴

Benralizumab (Fasenra) is a humanized IgG1 anti-IL-5 receptor α monoclonal antibody derived from mice. It induces apoptosis of eosinophils and, to a lesser extent, of basophils.⁴⁵ In clinical trials, it demonstrated a reduction in asthma exacerbation rate and improvement in prebronchodilator FEV₁ and asthma symptoms.^46,47 It does not need reconstitution, as the drug is dispensed as prefilled syringes with fixed non-weight-based dosing. Another potential advantage to benralizumab is that after the loading dose, subsequent doses are given every 8 weeks.

Bronchial thermoplasty

Bronchial thermoplasty (BT) is a novel nonpharmacologic intervention that entails the delivery of controlled radiofrequency-­generated heat via a catheter inserted into the bronchial tree of the lungs through a flexible bronchoscope. The potential mechanism of action is reduction in airway smooth muscle mass and inflammatory markers.

Evidence for BT started with the Asthma Intervention Research (AIR) and Research in Severe Asthma (RISA) trials.^48,49 In the AIR study, BT was shown to reduce the rate of mild exacerbations and improve morning peak expiratory flow and asthma scores at 12 months.⁴⁸ In the RISA trial, BT resulted in improvements in Asthma Quality of Life Questionnaire (AQLQ) score and need for rescue medication at 52 weeks, as well as a trend toward decrease in steroid use.⁴⁹

Bronchial thermoplasty results in clinically worthwhile reductions in severe exacerbations and ED visits in the year post treatment.

However, these studies were criticized for not having a placebo group—an issue addressed in the AIR2 trial, which compared bronchial thermoplasty with a sham procedure. AIR2 demonstrated improvements in AQLQ score and a 32% reduction in severe exacerbations and 84% fewer ED visits in the post-treatment period (up to 1 year post treatment).⁵⁰

Continue to: Both treatment groups...

Both treatment groups experienced an increase in respiratory adverse events: during the treatment period (up to 6 weeks post procedure), 16 subjects (8.4%) in the BT group required 19 hospitalizations for respiratory symptoms and 2 subjects (2%) in the sham group required 2 hospitalizations. A follow-up observational study involving a cohort of AIR2 patients demonstrated long-lasting effects of BT in asthma exacerbation frequency, ED visits, and stabilization of FEV₁ for up to 5 years.⁵¹

The Post-market Post-FDA Approval Clinical Trial Evaluating Bronchial Thermoplasty in Severe Persistent Asthma (PAS2) showed similar beneficial effects of BT on asthma control despite enrolling subjects who may have had poorer asthma control in the “real world” setting.⁵²

In summary, BT results in modest improvements in AQLQ scores and clinically worthwhile reductions in severe exacerbations and ED visits in the year post treatment, which may persist for up to 5 years. BT causes short-term increases in asthma-related morbidity, including hospital admissions. While there is encouraging data and the scope is increasing, BT remains limited to carefully selected (by a specialist) patients with severe asthma that is poorly controlled despite maximal inhaled therapy.

Immunotherapy

Immunotherapy for allergic disease is aimed at inducing immune tolerance to an allergen and alleviating allergic symptoms. This is done by administration of the allergen to which the patient is sensitive. There are 2 approaches: subcutaneous immunotherapy (SCIT) and sublingual immunotherapy (SLIT; a dissolvable tablet under the tongue or an aqueous or liquid extract).

Immunotherapy is generally reserved for patients who have allergic symptoms with exposure to a trigger and evidence (through skin or serum testing) of specific IgE to that trigger, especially if there is poor response to pharmacotherapy and allergen avoidance. Overall, evidence in this field is limited: Most studies have included patients with mild asthma, and few studies have compared immunotherapy with pharmacologic therapy or used standardized outcomes, such as exacerbations.

Continue to: SCIT

SCIT. A 2010 Cochrane review concluded that SCIT reduces asthma symptoms and use of asthma medications and improves bronchial hyperreactivity. Adverse effects include uncommon anaphylactic reactions, which may be life-threatening.⁵³

SLIT has advantages over SCIT as it can be administered by patients or caregivers, does not require injections, and carries a much lower risk for anaphylaxis. Modest benefits have been seen in adults and children, but there is concern about the design of many early studies.

The potential benefits of allergen immunotherapy must be weighed against the risk for adverse effects, including anaphylaxis.

A 2015 Cochrane review of SLIT in asthma recommended further research using validated scales and important outcomes for patients and decision makers so that SLIT can be properly assessed as a clinical treatment for asthma.⁵⁴ A subsequently published study of SLIT for house dust mites (HDM) in patients with asthma and HDM allergic rhinitis demonstrated a modest reduction in use of ICS with high-dose SLIT.⁵⁵

In another recent study, among adults with HDM allergy-related asthma not well controlled by ICS, the addition of HDM SLIT to maintenance medications improved time to first moderate-or-severe asthma exacerbation during ICS reduction.⁵⁶ Additional studies are needed to assess long-term efficacy and safety. However, for patients who experience exacerbations despite use of a low-dose or medium-dose ICS-LABA combination, SLIT can now be considered as an add-on therapy.

Per the GINA guidelines, the potential benefits of allergen immunotherapy must be weighed against the risk for adverse effects, including anaphylaxis, and the inconvenience and cost of the prolonged course of therapy.¹⁵

Continue to: Azithromycin

Azithromycin

Macrolides have immunomodulatory and anti-inflammatory effects in addition to their antibacterial effects. Maintenance treatment with macrolides such as azithromycin has been proven to be effective in chronic neutrophilic airway diseases (FIGURE). There have been attempts to assess whether this therapy can be useful in asthma management, as well. Some randomized controlled trials and meta-analyses have shown conflicting results, and early studies were limited by lack of data, heterogeneous results, and inadequate study designs.

The AZithromycin Against pLacebo in Exacerbations of Asthma (AZALEA) study was a randomized, multicenter, double-blind, placebo-controlled clinical trial in the United Kingdom among patients requiring emergency care for acute asthma exacerbations. Azithromycin added to standard care for asthma attacks did not result in clinical benefit.⁵⁷ While azithromycin in acute exacerbation is not currently recommended, recent trials in outpatient settings have shown promise.

The AZIthromycin in Severe ASThma study (AZISAST) was a randomized, double-blind, placebo-controlled trial in subjects with exacerbation-prone severe asthma in Belgium. Low-dose azithromycin (250 mg 3 times a week) as an add-on treatment to combination ICS-LABA therapy for 6 months did not reduce the rate of severe asthma exacerbations or lower respiratory tract infection (LRTI). However, subjects with a non-eosinophilic variant (neutrophilic phenotype) experienced significant reduction in the rate of exacerbation and LRTI.⁵⁸

Chronic azithromycin might prove to be a useful agent in the long-term management of asthma patients whose disease is not well controlled on inhaled therapy

The recently published Asthma and Macrolides: the AZithromycin Efficacy and Safety Study (AMAZES) shows promise for chronic azithromycin therapy as an add-on to medium-to-high-dose inhaled steroids and a long-acting bronchodilator in adults with uncontrolled persistent asthma. This was a large multicenter, randomized, double-blind, placebo-controlled, parallel group trial in New Zealand and Australia. Patients were excluded if they had hearing impairment or abnormally prolonged QTc. Azithromycin at a dose of 500 mg 3 times a week for 48 months reduced asthma exacerbations and improved QoL compared to placebo. The effect was sustained between subgroups based on phenotypes (eosinophilic vs noneosinophilic; frequent exacerbators vs nonfrequent exacerbators) and even among those with symptom differences at baseline (eg, cough or sputum positivity). The rate of antibiotic courses for respiratory infectious episodes was significantly reduced in the azithromycin-treated group.⁵⁹

The take-away: Chronic azithromycin might prove to be a useful agent in the long-term management of asthma patients whose disease is not well controlled on inhaled therapy. Further studies on mechanism and effects of prolonged antibiotic use will shed more light. For more information, see When guideline treatment of asthma fails, consider a macrolide antibiotic; http://bit.ly/2vDAWc6.

Continue to: A new era

We have entered an exciting era of asthma management, with the introduction of several novel modalities, such as biological therapy and bronchial thermoplasty, as well as use of known drugs such as macrolides, immunotherapy, and LAMA. This was made possible through a better understanding of the biological pathways of asthma. Asthma management has moved toward more personalized, targeted therapy based on asthma phenotypes.

It’s important to remember, however, that pharmacological and nonpharmacological aspects of management—including inhaler techniques, adherence to inhaler therapy, vaccinations, control of asthma triggers, and smoking cessation—remain the foundation of optimal asthma management and need to be aggressively addressed before embarking on advanced treatment options. Patients whose asthma is not well controlled with inhaled medications or who have frequent exacerbations (requiring use of steroids) should be comanaged by an expert asthma specialist to explore all possible therapies.

CORRESPONDENCE
Mayur Rali, MD, 995 Newbridge Road, Bellmore, NY 11710; mrali@northwell.edu

The vast majority of patients with tendon problems are successfully treated nonoperatively. But which treatments should you try (and when), and which are not quite ready for prime time? This review presents the evidence for the treatment options available to you. But first, it’s important to get our terminology right.

Tendinitis vs tendinosis vs paratenonitis: Words matter

The term “tendinopathy” encompasses many issues related to tendon pathology including tendinitis, tendinosis, and paratenonitis.^1,2 The clinical syndrome consists of pain, swelling, and functional impairment associated with activities of daily living or athletic performance.³ Tendinopathy may be acute or chronic, but most cases result from overuse.¹

In healthy tendons, the collagen fibers are packed tightly and organized in a linear pattern (FIGURE 1A). However, tendons that are chronically overused develop cumulative microtrauma that leads to a degenerative process within the tendon that is slow (typically measured in months) to heal. This is due to the relative lack of vasculature and the slow rate of tissue turnover in tendons.^2,4,5

Sports and manual labor are the most common causes of tendinopathy, but medical conditions including obesity, high blood pressure, diabetes, and high cholesterol are associated risk factors. Medications, particularly fluoroquinolones and statins, can cause tendon problems, and steroids, particularly those injected intratendinously, have been implicated in tendon rupture.^4,6

The term “tendinitis” has long been used for all tendon disorders although it is best reserved for acute inflammatory conditions. For most tendon conditions resulting from overuse, the term “tendinosis” is now more widely recognized and preferred.^7,8 Family physicians (FPs) should recognize that tendinitis and tendinosis differ greatly in pathophysiology and treatment.³

Tendinitis: Not as common as you think

Tendinitis is an acute inflammatory condition that accounts for only about 3% of all tendon disorders.³ Patients presenting with tendinitis usually have acute onset of pain and swelling typically either from a new activity or one to which they are unaccustomed (eg, lateral elbow pain after painting a house) or from an acute injury. Partial tearing of the affected tendon is likely, especially following injury.^2,3

Tendinosis: A degenerative condition

In contrast to the acute inflammation of tendinitis, tendinosis is a degenerative condition induced by chronic overuse. It is typically encountered in athletes and laborers.^2,5,8,9 Tendinotic tissue is generally regarded as noninflammatory, but recent research supports inflammation playing at least a small role, especially in closely associated tissues such as bursae and the paratenon tissue.¹⁰

Continue to: Histologically, tendinosis shows...

Histologically, tendinosis shows loss of the typical linear collagen fiber organization, increased mucoid ground substance, hypercellularity, and increased growth of nerves and vessels (FIGURE 1B).

Tendinosis is not always symptomatic.^5,11 When pain is present, experts have proposed that it is neurogenically derived rather than from local chemical inflammation. This is supported by evidence of increases in the excitatory neurotransmitter glutamate and its receptor N-methyl-D-aspartate in tendinotic tissue with nerve ingrowth. Tendinotic tissue also contains substance P and calcitonin gene-related peptide, neuropeptides that are associated with pain and nociceptive nerve endings.^2,3,6,10

Patients with tendinosis typically present with an insidious onset of a painful, thickened tendon.¹¹ The most common tendons affected include the Achilles, the patellar, the supraspinatus, and the common extensor tendon of the lateral elbow.² Lower extremity tendinosis is common in athletes, while upper extremity tendinopathies are more often work-related.³

Paratenonitis: Inflammation surrounding the tendon

Occasionally, tendinosis may be associated with paratenonitis, which is inflammation of the paratenon (tissue surrounding some tendons).^2,5,10 Paratendinous tissue contains a higher concentration of sensory nerves than the tendon itself and may generate significant discomfort.^10,11

The clinical presentation of paratenonitis includes a swollen and erythematous tendon.⁵ The classic example—de Quervain disease—involves the first dorsal wrist compartment, in which the abductor pollicis longus and extensor pollicis brevis tendons are encased in a synovial sheath. The term tenosynovitis is commonly used to indicate inflammation of both the paratenon and synovial sheath (TABLE 1^2,3,5,6,9-11).⁵

Continue to: Treatment demands time and patience

Treating tendon conditions is challenging for both the patient and the clinician. Improvement takes time and several different treatment strategies may be required for success. Given the large number of available treatment options and the often weak or limited supporting evidence of their efficacy, designing a treatment plan can be difficult. TABLE 2 summarizes the information detailed below about specific treatment options.

First-line treatments. The vast majority of patients with tendon problems are successfully treated nonoperatively. Reasonable first-line treatments, especially for inflammatory conditions like tendinitis, tenosynovitis, and paratenonitis, include relative rest, activity modification, cryotherapy, and bracing.^12-14

Nonsteroidal anti-inflammatory drugs (NSAIDs) for pain control are somewhat controversial. At best, they provide pain relief in the short term (7-14 days); at worst, some studies suggest potential detrimental effects to the tendon.¹⁴ If considered, NSAIDs should be used for no longer than 2 weeks. They are ideally reserved for pain control in patients with acute injuries when an inflammatory condition is likely. An alternative for pain control in inflammatory cases is a short course of oral steroids, but the adverse effects of these medications may be challenging for some patients.

Other options. If these more conservative treatments fail, or the patient is experiencing significant and debilitating pain, FPs may consider a corticosteroid injection. If this fails, or the condition is clearly past an inflammatory stage, then physical therapy should be considered. More advanced treatments, such as platelet-rich plasma injections and percutaneous needle tenotomies, are typically reserved for chronic, recalcitrant cases of tendinosis. Various other treatment options are detailed below and can be used on a case-by-case basis. Surgical management should be considered only as a last resort.

Reserve use of NSAIDs for pain control in patients with acute injuries when an inflammatory condition is likely.

Realize that certain barriers may exist to some of these treatments. With extracorporeal shockwave therapy, for example, access to a machine can be challenging, as they are typically only found in major metropolitan areas. Polidocanol, used during sclerotherapy, can be difficult to obtain in the United States. Another challenge is cost. Not all of these procedures are covered by insurance, and they can be expensive when paying out of pocket.

Continue to: Rehabilitation...

Studies show that eccentric exercises (EEs) help to decrease vascularity and nerve presence in affected tendons, modulate expression of neuronal substances, and may stimulate formation of load-tolerant fibroblasts.^2,3

Plateletrich plasma injections are typically expensive; whole blood is less expensive because there's no manipulation of the blood product.

For Achilles tendinosis, EE is a well-­established treatment supported by multiple randomized controlled trials (RCTs). Improvements in patient satisfaction and pain range from 60% to 90%; evidence suggests greater success in midsubstance vs insertional Achilles tendinosis.¹⁵ The addition of deep-friction massage (DFM), which we’ll discuss in a moment, to EE appears to improve outcomes even more than EE alone.¹⁶

EE is also a beneficial treatment for patellar tendinosis,^3,14 and it appears to benefit rotator cuff tendinosis,³ but research has shown EE for lateral epicondylosis to be no more effective than stretching alone.¹⁷

Recommend deep-friction massage for tendinosis—not inflammatory conditions.

DFM is for treating tendinosis—not inflammatory conditions. Mechanical stimulation of the tissue being massaged releases cell mediators and growth factors that activate fibroblasts. It is typically performed with plastic or metal tools.¹⁶ DFM appears to be a reliable treatment option for the lateral elbow.¹⁸

Extracorporeal shockwave therapy appears promising; evidence is limited

Research has shown that extracorporeal shockwave therapy (ESWT) promotes the production of TGF-β1 and IGF-1 in rat models,² and it is believed to be able to disintegrate calcium deposits and stimulate tissue repair.¹⁴ Research is generally supportive of its effectiveness in treating tendinosis; however, evidence is limited by great variability in studies in terms of treatment intensity, frequency, duration, timing, number of treatments, and use of a local anesthetic.¹⁴ ESWT appears to be useful in augmenting treatment with EE, particularly with regard to the rotator cuff.¹⁹

Continue to: A review of 10 RCTs...

A review of 10 RCTs demonstrated the effectiveness of ESWT for tennis elbow.² ESWT for greater trochanteric pain syndrome (GTPS, formerly known as trochanteric bursitis) appears to be more effective than corticosteroids and home exercises for outcomes at 4 months and equivalent to home exercises at 15 months.²⁰ In patellar tendinosis, ESWT has been shown to be an effective treatment, especially under ultrasound guidance.¹² Studies involving the use of ESWT for Achilles tendinosis have had mixed results for midsubstance tendinosis, and more positive results for insertional tendinosis.¹⁵ For a video on how the therapy is administered, see https://www.youtube.com/watch?v=Fq5yqiWByX4.

Glyceryl trinitrate patches: Mixed results

Basic science studies have shown that nitric oxide modulates tendon healing by enhancing fibroblast proliferation and collagen synthesis,^2,14 but that it should be used with caution in cardiac patients and in those who take PDE-5 inhibitors. Common adverse effects include rash, headache, and dizziness.

Use glyceryl trinitrate patches with caution in cardiac patients and in those who take PDE-5 inhibitors.

In clinical studies, glyceryl trinitrate (GTN) patches show mixed results. For the upper extremity, GTN appears to be helpful for pain in the short term when combined with physical therapy, but long-term positive outcomes have been absent.²¹ In one Level 1 study for patellar tendinosis comparing GTN patches with EE to a placebo patch with EE, no significant difference was noted at 24 weeks.²² Benefit for Achilles tendinosis also appears to be lacking.^3,23

Corticosteroid injections: Mechanism unknown

The mechanism for the beneficial effects of corticosteroid injections (CSIs) for tendinosis remains controversial. Proposed mechanisms include lysis of peritendinous adhesions, disruption of the nociceptors in the region of the injection, and decreased vascularization.^10,15 Given tendinosis is generally regarded as a noninflammatory condition, and the fact that these medications have demonstrated potential negative effects on tendon healing, exercise caution when considering CSIs.^2,24

Although steroids can effectively reduce pain in the short term, intermediate- and long-term studies generally show no difference or worse outcomes when they are compared to no treatment, placebo, or other treatment modalities. In fact, strong evidence exists for negative effects of steroids on lateral epicondylosis in both the intermediate (6 months) and long (1 year) term.²⁴ Particular care is required when administering a CSI for medial epicondylosis, as the ulnar nerve is immediately posterior to the medial epicondyle.²⁵

Continue to: In contrast...

In contrast, CSIs appear to be a reliable treatment option for de Quervain disease.²⁶ Landmark-guided injections for GTPS can improve pain in the short term (< 1 month), but are inferior to either home exercises or ESWT beyond a few months. Thus, CSIs are a reasonable option for pain control in GTPS, but should not be the sole treatment modality.²⁰

Studies regarding corticosteroid use for Achilles and patellar tendinosis have had mixed results. Patients can hope for mild improvement in pain at best, and the risk for relapse and tendon rupture is ever present.²⁷ This is especially concerning given the significant load-bearing of the patellar and Achilles tendons.^14,15 If you are considering a CSI for these purposes, use imaging guidance to ensure the injection is not placed intratendinously.

Platelet-rich plasma and whole blood: Inducing an anabolic healing response

Platelet-rich plasma (PRP) and whole blood injections both aim to deliver autologous growth factors (eg, VEGF, PDGF, and IGF-1) and bioactive mediators to the site of tendinosis to induce an anabolic healing response. PRP therapy differs from whole blood therapy in that it is withdrawn and then concentrated in a centrifuge before being injected. Patients are typically injected under ultrasound guidance. The great variation in PRP preparation, platelet concentration, use of adjunctive treatments, leukocyte concentration, and number and technique of injections makes it difficult to determine the optimal PRP treatment protocol.^10,28,29

In 1 prospective RCT comparing subacromial PRP injections to CSI for the shoulder, the PRP group had better outcomes at 3 months, but similar outcomes at 6 months. The suggestion was made that PRP therapy could be an alternative treatment for individuals with a contraindication to CSIs.³⁰

PRP therapy appears to be an effective treatment option for patellar tendinopathy.^28,31 A Level 1 study comparing dry needle tenotomy and EE to dry needle tenotomy with both PRP therapy and EE found faster recovery in the PRP group.³² In another patellar tendon study comparing ESWT to PRP therapy, both were found to be effective, but PRP performed better in terms of pain, function, and satisfaction at 6 and 12 months.¹² For Achilles tendons, however, the evidence is mixed; case series have had generally positive outcomes, but the only double-blind RCT found no benefit.^28,31

Continue to: In lateral epicondylosis...

In lateral epicondylosis, the use of auto-logous whole blood or PRP injections appears to help both pain and function, with several studies failing to demonstrate superiority of 1 modality over the other.^24,25,28,33 This raises the issue of whether PRP therapy is any more effective than whole blood for the treatment of other tendinopathies. Unfortunately, there is a paucity of studies comparing the effectiveness of 1 modality to the other, apart from those for lateral epicondylosis.

Prolotherapy: An option for these 3 conditions

Prolotherapy involves the injection of hypertonic dextrose and local anesthetic, which is believed to lead to an upregulation of inflammatory mediators and growth factors. This treatment usually involves several injections spaced 2 to 6 weeks apart over several months. High-quality studies are not available to clarify the optimal dextrose concentration or number of injections required. The few high-quality studies available support prolotherapy for lateral epicondylosis, rotator cuff tendinopathy, and Osgood Schlatter disease. Lesser-quality studies support its use for refractory pain of the Achilles, hip adductors, and plantar fascia.^24,34

Sclerotherapy: Not just for veins

As discussed earlier, tendinotic tissue can have neovascularization that is easily detected on Doppler ultrasound. Sensory nerves typically grow alongside the new vessels. Sclerosing agents, such as polidocanol, can be injected with ultrasound guidance into areas of neovascularization, with the intention of causing denervation and pain relief.¹⁵ Neovascularization does not always correlate with pathology, so careful patient selection is necessary.³⁵

Exercise caution when considering corticosteroid injections for tendinosis.

Studies of sclerotherapy for patellar tendinopathy are generally favorable. One comparing sclerotherapy to arthroscopic debridement showed improvement in pain from both treatments at 6 and 12 months, but the arthroscopy group had less pain, better satisfaction scores, and a faster return to sport.¹⁴ Sclerotherapy is also effective for Achilles tendinosis.¹⁵

Stem cells: Not at this time

Stem cell use for tendinosis is based on the theory that these cells possess the capability to differentiate into tenocytes to produce new, healthy tendon tissue. Additionally, stem cell injections are believed to create a local immune response, recruiting local growth factors and cytokines to aide in tendon repair. A recent systematic review failed to identify any high-quality studies (Level 4 data at best) supporting the use of stem cells in tendinopathy, and the researchers did not recommend stem cell use outside of clinical trials at this time.³⁶

Continue to: Percutaneous needle tenotomy...

Percutaneous needle tenotomy is thought to benefit tendinosis by disrupting the tendinotic tissue via needling, while simultaneously causing bleeding and the release of growth factors to aid in healing. Unlike surgical ­tenotomy, the procedure is typically performed with ultrasound guidance in the office or other ambulatory setting. After local anesthesia is administered, a needle is passed multiple times through the entire region of abnormality noted on ultrasound. Generally, around 20 to 30 needle fenestrations are performed.^37,38

Consider prolotherapy for lateral epicondylosis, rotator cuff tendinopathy, and Osgood Schlatter disease.

In one retrospective study evaluating 47 patellar tendons, 81% had excellent or good results.³⁸ In a retrospective study for lateral epicondylosis, 80% had good to excellent results.³⁹

CORRESPONDENCE
Kyle Goerl, MD, CAQSM, Lafene Health Center, 1105 Sunset Avenue, Manhattan, KS, 66502-3761; kvg3355@ksu.edu.

The vast majority of patients with tendon problems are successfully treated nonoperatively. But which treatments should you try (and when), and which are not quite ready for prime time? This review presents the evidence for the treatment options available to you. But first, it’s important to get our terminology right.

Tendinitis vs tendinosis vs paratenonitis: Words matter

The term “tendinopathy” encompasses many issues related to tendon pathology including tendinitis, tendinosis, and paratenonitis.^1,2 The clinical syndrome consists of pain, swelling, and functional impairment associated with activities of daily living or athletic performance.³ Tendinopathy may be acute or chronic, but most cases result from overuse.¹

In healthy tendons, the collagen fibers are packed tightly and organized in a linear pattern (FIGURE 1A). However, tendons that are chronically overused develop cumulative microtrauma that leads to a degenerative process within the tendon that is slow (typically measured in months) to heal. This is due to the relative lack of vasculature and the slow rate of tissue turnover in tendons.^2,4,5

Sports and manual labor are the most common causes of tendinopathy, but medical conditions including obesity, high blood pressure, diabetes, and high cholesterol are associated risk factors. Medications, particularly fluoroquinolones and statins, can cause tendon problems, and steroids, particularly those injected intratendinously, have been implicated in tendon rupture.^4,6

The term “tendinitis” has long been used for all tendon disorders although it is best reserved for acute inflammatory conditions. For most tendon conditions resulting from overuse, the term “tendinosis” is now more widely recognized and preferred.^7,8 Family physicians (FPs) should recognize that tendinitis and tendinosis differ greatly in pathophysiology and treatment.³

Tendinitis: Not as common as you think

Tendinitis is an acute inflammatory condition that accounts for only about 3% of all tendon disorders.³ Patients presenting with tendinitis usually have acute onset of pain and swelling typically either from a new activity or one to which they are unaccustomed (eg, lateral elbow pain after painting a house) or from an acute injury. Partial tearing of the affected tendon is likely, especially following injury.^2,3

Tendinosis: A degenerative condition

In contrast to the acute inflammation of tendinitis, tendinosis is a degenerative condition induced by chronic overuse. It is typically encountered in athletes and laborers.^2,5,8,9 Tendinotic tissue is generally regarded as noninflammatory, but recent research supports inflammation playing at least a small role, especially in closely associated tissues such as bursae and the paratenon tissue.¹⁰

Continue to: Histologically, tendinosis shows...

Histologically, tendinosis shows loss of the typical linear collagen fiber organization, increased mucoid ground substance, hypercellularity, and increased growth of nerves and vessels (FIGURE 1B).

Tendinosis is not always symptomatic.^5,11 When pain is present, experts have proposed that it is neurogenically derived rather than from local chemical inflammation. This is supported by evidence of increases in the excitatory neurotransmitter glutamate and its receptor N-methyl-D-aspartate in tendinotic tissue with nerve ingrowth. Tendinotic tissue also contains substance P and calcitonin gene-related peptide, neuropeptides that are associated with pain and nociceptive nerve endings.^2,3,6,10

Patients with tendinosis typically present with an insidious onset of a painful, thickened tendon.¹¹ The most common tendons affected include the Achilles, the patellar, the supraspinatus, and the common extensor tendon of the lateral elbow.² Lower extremity tendinosis is common in athletes, while upper extremity tendinopathies are more often work-related.³

Paratenonitis: Inflammation surrounding the tendon

Occasionally, tendinosis may be associated with paratenonitis, which is inflammation of the paratenon (tissue surrounding some tendons).^2,5,10 Paratendinous tissue contains a higher concentration of sensory nerves than the tendon itself and may generate significant discomfort.^10,11

The clinical presentation of paratenonitis includes a swollen and erythematous tendon.⁵ The classic example—de Quervain disease—involves the first dorsal wrist compartment, in which the abductor pollicis longus and extensor pollicis brevis tendons are encased in a synovial sheath. The term tenosynovitis is commonly used to indicate inflammation of both the paratenon and synovial sheath (TABLE 1^2,3,5,6,9-11).⁵

Continue to: Treatment demands time and patience

Treating tendon conditions is challenging for both the patient and the clinician. Improvement takes time and several different treatment strategies may be required for success. Given the large number of available treatment options and the often weak or limited supporting evidence of their efficacy, designing a treatment plan can be difficult. TABLE 2 summarizes the information detailed below about specific treatment options.

First-line treatments. The vast majority of patients with tendon problems are successfully treated nonoperatively. Reasonable first-line treatments, especially for inflammatory conditions like tendinitis, tenosynovitis, and paratenonitis, include relative rest, activity modification, cryotherapy, and bracing.^12-14

Nonsteroidal anti-inflammatory drugs (NSAIDs) for pain control are somewhat controversial. At best, they provide pain relief in the short term (7-14 days); at worst, some studies suggest potential detrimental effects to the tendon.¹⁴ If considered, NSAIDs should be used for no longer than 2 weeks. They are ideally reserved for pain control in patients with acute injuries when an inflammatory condition is likely. An alternative for pain control in inflammatory cases is a short course of oral steroids, but the adverse effects of these medications may be challenging for some patients.

Other options. If these more conservative treatments fail, or the patient is experiencing significant and debilitating pain, FPs may consider a corticosteroid injection. If this fails, or the condition is clearly past an inflammatory stage, then physical therapy should be considered. More advanced treatments, such as platelet-rich plasma injections and percutaneous needle tenotomies, are typically reserved for chronic, recalcitrant cases of tendinosis. Various other treatment options are detailed below and can be used on a case-by-case basis. Surgical management should be considered only as a last resort.

Reserve use of NSAIDs for pain control in patients with acute injuries when an inflammatory condition is likely.

Realize that certain barriers may exist to some of these treatments. With extracorporeal shockwave therapy, for example, access to a machine can be challenging, as they are typically only found in major metropolitan areas. Polidocanol, used during sclerotherapy, can be difficult to obtain in the United States. Another challenge is cost. Not all of these procedures are covered by insurance, and they can be expensive when paying out of pocket.

Continue to: Rehabilitation...

Studies show that eccentric exercises (EEs) help to decrease vascularity and nerve presence in affected tendons, modulate expression of neuronal substances, and may stimulate formation of load-tolerant fibroblasts.^2,3

Plateletrich plasma injections are typically expensive; whole blood is less expensive because there's no manipulation of the blood product.

For Achilles tendinosis, EE is a well-­established treatment supported by multiple randomized controlled trials (RCTs). Improvements in patient satisfaction and pain range from 60% to 90%; evidence suggests greater success in midsubstance vs insertional Achilles tendinosis.¹⁵ The addition of deep-friction massage (DFM), which we’ll discuss in a moment, to EE appears to improve outcomes even more than EE alone.¹⁶

EE is also a beneficial treatment for patellar tendinosis,^3,14 and it appears to benefit rotator cuff tendinosis,³ but research has shown EE for lateral epicondylosis to be no more effective than stretching alone.¹⁷

Recommend deep-friction massage for tendinosis—not inflammatory conditions.

DFM is for treating tendinosis—not inflammatory conditions. Mechanical stimulation of the tissue being massaged releases cell mediators and growth factors that activate fibroblasts. It is typically performed with plastic or metal tools.¹⁶ DFM appears to be a reliable treatment option for the lateral elbow.¹⁸

Extracorporeal shockwave therapy appears promising; evidence is limited

Research has shown that extracorporeal shockwave therapy (ESWT) promotes the production of TGF-β1 and IGF-1 in rat models,² and it is believed to be able to disintegrate calcium deposits and stimulate tissue repair.¹⁴ Research is generally supportive of its effectiveness in treating tendinosis; however, evidence is limited by great variability in studies in terms of treatment intensity, frequency, duration, timing, number of treatments, and use of a local anesthetic.¹⁴ ESWT appears to be useful in augmenting treatment with EE, particularly with regard to the rotator cuff.¹⁹

Continue to: A review of 10 RCTs...

A review of 10 RCTs demonstrated the effectiveness of ESWT for tennis elbow.² ESWT for greater trochanteric pain syndrome (GTPS, formerly known as trochanteric bursitis) appears to be more effective than corticosteroids and home exercises for outcomes at 4 months and equivalent to home exercises at 15 months.²⁰ In patellar tendinosis, ESWT has been shown to be an effective treatment, especially under ultrasound guidance.¹² Studies involving the use of ESWT for Achilles tendinosis have had mixed results for midsubstance tendinosis, and more positive results for insertional tendinosis.¹⁵ For a video on how the therapy is administered, see https://www.youtube.com/watch?v=Fq5yqiWByX4.

Glyceryl trinitrate patches: Mixed results

Basic science studies have shown that nitric oxide modulates tendon healing by enhancing fibroblast proliferation and collagen synthesis,^2,14 but that it should be used with caution in cardiac patients and in those who take PDE-5 inhibitors. Common adverse effects include rash, headache, and dizziness.

Use glyceryl trinitrate patches with caution in cardiac patients and in those who take PDE-5 inhibitors.

In clinical studies, glyceryl trinitrate (GTN) patches show mixed results. For the upper extremity, GTN appears to be helpful for pain in the short term when combined with physical therapy, but long-term positive outcomes have been absent.²¹ In one Level 1 study for patellar tendinosis comparing GTN patches with EE to a placebo patch with EE, no significant difference was noted at 24 weeks.²² Benefit for Achilles tendinosis also appears to be lacking.^3,23

Corticosteroid injections: Mechanism unknown

The mechanism for the beneficial effects of corticosteroid injections (CSIs) for tendinosis remains controversial. Proposed mechanisms include lysis of peritendinous adhesions, disruption of the nociceptors in the region of the injection, and decreased vascularization.^10,15 Given tendinosis is generally regarded as a noninflammatory condition, and the fact that these medications have demonstrated potential negative effects on tendon healing, exercise caution when considering CSIs.^2,24

Although steroids can effectively reduce pain in the short term, intermediate- and long-term studies generally show no difference or worse outcomes when they are compared to no treatment, placebo, or other treatment modalities. In fact, strong evidence exists for negative effects of steroids on lateral epicondylosis in both the intermediate (6 months) and long (1 year) term.²⁴ Particular care is required when administering a CSI for medial epicondylosis, as the ulnar nerve is immediately posterior to the medial epicondyle.²⁵

Continue to: In contrast...

In contrast, CSIs appear to be a reliable treatment option for de Quervain disease.²⁶ Landmark-guided injections for GTPS can improve pain in the short term (< 1 month), but are inferior to either home exercises or ESWT beyond a few months. Thus, CSIs are a reasonable option for pain control in GTPS, but should not be the sole treatment modality.²⁰

Studies regarding corticosteroid use for Achilles and patellar tendinosis have had mixed results. Patients can hope for mild improvement in pain at best, and the risk for relapse and tendon rupture is ever present.²⁷ This is especially concerning given the significant load-bearing of the patellar and Achilles tendons.^14,15 If you are considering a CSI for these purposes, use imaging guidance to ensure the injection is not placed intratendinously.

Platelet-rich plasma and whole blood: Inducing an anabolic healing response

Platelet-rich plasma (PRP) and whole blood injections both aim to deliver autologous growth factors (eg, VEGF, PDGF, and IGF-1) and bioactive mediators to the site of tendinosis to induce an anabolic healing response. PRP therapy differs from whole blood therapy in that it is withdrawn and then concentrated in a centrifuge before being injected. Patients are typically injected under ultrasound guidance. The great variation in PRP preparation, platelet concentration, use of adjunctive treatments, leukocyte concentration, and number and technique of injections makes it difficult to determine the optimal PRP treatment protocol.^10,28,29

In 1 prospective RCT comparing subacromial PRP injections to CSI for the shoulder, the PRP group had better outcomes at 3 months, but similar outcomes at 6 months. The suggestion was made that PRP therapy could be an alternative treatment for individuals with a contraindication to CSIs.³⁰

PRP therapy appears to be an effective treatment option for patellar tendinopathy.^28,31 A Level 1 study comparing dry needle tenotomy and EE to dry needle tenotomy with both PRP therapy and EE found faster recovery in the PRP group.³² In another patellar tendon study comparing ESWT to PRP therapy, both were found to be effective, but PRP performed better in terms of pain, function, and satisfaction at 6 and 12 months.¹² For Achilles tendons, however, the evidence is mixed; case series have had generally positive outcomes, but the only double-blind RCT found no benefit.^28,31

Continue to: In lateral epicondylosis...

In lateral epicondylosis, the use of auto-logous whole blood or PRP injections appears to help both pain and function, with several studies failing to demonstrate superiority of 1 modality over the other.^24,25,28,33 This raises the issue of whether PRP therapy is any more effective than whole blood for the treatment of other tendinopathies. Unfortunately, there is a paucity of studies comparing the effectiveness of 1 modality to the other, apart from those for lateral epicondylosis.

Prolotherapy: An option for these 3 conditions

Prolotherapy involves the injection of hypertonic dextrose and local anesthetic, which is believed to lead to an upregulation of inflammatory mediators and growth factors. This treatment usually involves several injections spaced 2 to 6 weeks apart over several months. High-quality studies are not available to clarify the optimal dextrose concentration or number of injections required. The few high-quality studies available support prolotherapy for lateral epicondylosis, rotator cuff tendinopathy, and Osgood Schlatter disease. Lesser-quality studies support its use for refractory pain of the Achilles, hip adductors, and plantar fascia.^24,34

Sclerotherapy: Not just for veins

As discussed earlier, tendinotic tissue can have neovascularization that is easily detected on Doppler ultrasound. Sensory nerves typically grow alongside the new vessels. Sclerosing agents, such as polidocanol, can be injected with ultrasound guidance into areas of neovascularization, with the intention of causing denervation and pain relief.¹⁵ Neovascularization does not always correlate with pathology, so careful patient selection is necessary.³⁵

Exercise caution when considering corticosteroid injections for tendinosis.

Studies of sclerotherapy for patellar tendinopathy are generally favorable. One comparing sclerotherapy to arthroscopic debridement showed improvement in pain from both treatments at 6 and 12 months, but the arthroscopy group had less pain, better satisfaction scores, and a faster return to sport.¹⁴ Sclerotherapy is also effective for Achilles tendinosis.¹⁵

Stem cells: Not at this time

Stem cell use for tendinosis is based on the theory that these cells possess the capability to differentiate into tenocytes to produce new, healthy tendon tissue. Additionally, stem cell injections are believed to create a local immune response, recruiting local growth factors and cytokines to aide in tendon repair. A recent systematic review failed to identify any high-quality studies (Level 4 data at best) supporting the use of stem cells in tendinopathy, and the researchers did not recommend stem cell use outside of clinical trials at this time.³⁶

Continue to: Percutaneous needle tenotomy...

Percutaneous needle tenotomy is thought to benefit tendinosis by disrupting the tendinotic tissue via needling, while simultaneously causing bleeding and the release of growth factors to aid in healing. Unlike surgical ­tenotomy, the procedure is typically performed with ultrasound guidance in the office or other ambulatory setting. After local anesthesia is administered, a needle is passed multiple times through the entire region of abnormality noted on ultrasound. Generally, around 20 to 30 needle fenestrations are performed.^37,38

Consider prolotherapy for lateral epicondylosis, rotator cuff tendinopathy, and Osgood Schlatter disease.

In one retrospective study evaluating 47 patellar tendons, 81% had excellent or good results.³⁸ In a retrospective study for lateral epicondylosis, 80% had good to excellent results.³⁹

CORRESPONDENCE
Kyle Goerl, MD, CAQSM, Lafene Health Center, 1105 Sunset Avenue, Manhattan, KS, 66502-3761; kvg3355@ksu.edu.

The vast majority of patients with tendon problems are successfully treated nonoperatively. But which treatments should you try (and when), and which are not quite ready for prime time? This review presents the evidence for the treatment options available to you. But first, it’s important to get our terminology right.

Tendinitis vs tendinosis vs paratenonitis: Words matter

The term “tendinopathy” encompasses many issues related to tendon pathology including tendinitis, tendinosis, and paratenonitis.^1,2 The clinical syndrome consists of pain, swelling, and functional impairment associated with activities of daily living or athletic performance.³ Tendinopathy may be acute or chronic, but most cases result from overuse.¹

In healthy tendons, the collagen fibers are packed tightly and organized in a linear pattern (FIGURE 1A). However, tendons that are chronically overused develop cumulative microtrauma that leads to a degenerative process within the tendon that is slow (typically measured in months) to heal. This is due to the relative lack of vasculature and the slow rate of tissue turnover in tendons.^2,4,5

Sports and manual labor are the most common causes of tendinopathy, but medical conditions including obesity, high blood pressure, diabetes, and high cholesterol are associated risk factors. Medications, particularly fluoroquinolones and statins, can cause tendon problems, and steroids, particularly those injected intratendinously, have been implicated in tendon rupture.^4,6

The term “tendinitis” has long been used for all tendon disorders although it is best reserved for acute inflammatory conditions. For most tendon conditions resulting from overuse, the term “tendinosis” is now more widely recognized and preferred.^7,8 Family physicians (FPs) should recognize that tendinitis and tendinosis differ greatly in pathophysiology and treatment.³

Tendinitis: Not as common as you think

Tendinitis is an acute inflammatory condition that accounts for only about 3% of all tendon disorders.³ Patients presenting with tendinitis usually have acute onset of pain and swelling typically either from a new activity or one to which they are unaccustomed (eg, lateral elbow pain after painting a house) or from an acute injury. Partial tearing of the affected tendon is likely, especially following injury.^2,3

Tendinosis: A degenerative condition

In contrast to the acute inflammation of tendinitis, tendinosis is a degenerative condition induced by chronic overuse. It is typically encountered in athletes and laborers.^2,5,8,9 Tendinotic tissue is generally regarded as noninflammatory, but recent research supports inflammation playing at least a small role, especially in closely associated tissues such as bursae and the paratenon tissue.¹⁰

Continue to: Histologically, tendinosis shows...

Histologically, tendinosis shows loss of the typical linear collagen fiber organization, increased mucoid ground substance, hypercellularity, and increased growth of nerves and vessels (FIGURE 1B).

Tendinosis is not always symptomatic.^5,11 When pain is present, experts have proposed that it is neurogenically derived rather than from local chemical inflammation. This is supported by evidence of increases in the excitatory neurotransmitter glutamate and its receptor N-methyl-D-aspartate in tendinotic tissue with nerve ingrowth. Tendinotic tissue also contains substance P and calcitonin gene-related peptide, neuropeptides that are associated with pain and nociceptive nerve endings.^2,3,6,10

Patients with tendinosis typically present with an insidious onset of a painful, thickened tendon.¹¹ The most common tendons affected include the Achilles, the patellar, the supraspinatus, and the common extensor tendon of the lateral elbow.² Lower extremity tendinosis is common in athletes, while upper extremity tendinopathies are more often work-related.³

Paratenonitis: Inflammation surrounding the tendon

Occasionally, tendinosis may be associated with paratenonitis, which is inflammation of the paratenon (tissue surrounding some tendons).^2,5,10 Paratendinous tissue contains a higher concentration of sensory nerves than the tendon itself and may generate significant discomfort.^10,11

The clinical presentation of paratenonitis includes a swollen and erythematous tendon.⁵ The classic example—de Quervain disease—involves the first dorsal wrist compartment, in which the abductor pollicis longus and extensor pollicis brevis tendons are encased in a synovial sheath. The term tenosynovitis is commonly used to indicate inflammation of both the paratenon and synovial sheath (TABLE 1^2,3,5,6,9-11).⁵

Continue to: Treatment demands time and patience

Treating tendon conditions is challenging for both the patient and the clinician. Improvement takes time and several different treatment strategies may be required for success. Given the large number of available treatment options and the often weak or limited supporting evidence of their efficacy, designing a treatment plan can be difficult. TABLE 2 summarizes the information detailed below about specific treatment options.

First-line treatments. The vast majority of patients with tendon problems are successfully treated nonoperatively. Reasonable first-line treatments, especially for inflammatory conditions like tendinitis, tenosynovitis, and paratenonitis, include relative rest, activity modification, cryotherapy, and bracing.^12-14

Nonsteroidal anti-inflammatory drugs (NSAIDs) for pain control are somewhat controversial. At best, they provide pain relief in the short term (7-14 days); at worst, some studies suggest potential detrimental effects to the tendon.¹⁴ If considered, NSAIDs should be used for no longer than 2 weeks. They are ideally reserved for pain control in patients with acute injuries when an inflammatory condition is likely. An alternative for pain control in inflammatory cases is a short course of oral steroids, but the adverse effects of these medications may be challenging for some patients.

Other options. If these more conservative treatments fail, or the patient is experiencing significant and debilitating pain, FPs may consider a corticosteroid injection. If this fails, or the condition is clearly past an inflammatory stage, then physical therapy should be considered. More advanced treatments, such as platelet-rich plasma injections and percutaneous needle tenotomies, are typically reserved for chronic, recalcitrant cases of tendinosis. Various other treatment options are detailed below and can be used on a case-by-case basis. Surgical management should be considered only as a last resort.

Reserve use of NSAIDs for pain control in patients with acute injuries when an inflammatory condition is likely.

Realize that certain barriers may exist to some of these treatments. With extracorporeal shockwave therapy, for example, access to a machine can be challenging, as they are typically only found in major metropolitan areas. Polidocanol, used during sclerotherapy, can be difficult to obtain in the United States. Another challenge is cost. Not all of these procedures are covered by insurance, and they can be expensive when paying out of pocket.

Continue to: Rehabilitation...

Studies show that eccentric exercises (EEs) help to decrease vascularity and nerve presence in affected tendons, modulate expression of neuronal substances, and may stimulate formation of load-tolerant fibroblasts.^2,3

Plateletrich plasma injections are typically expensive; whole blood is less expensive because there's no manipulation of the blood product.

For Achilles tendinosis, EE is a well-­established treatment supported by multiple randomized controlled trials (RCTs). Improvements in patient satisfaction and pain range from 60% to 90%; evidence suggests greater success in midsubstance vs insertional Achilles tendinosis.¹⁵ The addition of deep-friction massage (DFM), which we’ll discuss in a moment, to EE appears to improve outcomes even more than EE alone.¹⁶

EE is also a beneficial treatment for patellar tendinosis,^3,14 and it appears to benefit rotator cuff tendinosis,³ but research has shown EE for lateral epicondylosis to be no more effective than stretching alone.¹⁷

Recommend deep-friction massage for tendinosis—not inflammatory conditions.

DFM is for treating tendinosis—not inflammatory conditions. Mechanical stimulation of the tissue being massaged releases cell mediators and growth factors that activate fibroblasts. It is typically performed with plastic or metal tools.¹⁶ DFM appears to be a reliable treatment option for the lateral elbow.¹⁸

Extracorporeal shockwave therapy appears promising; evidence is limited

Research has shown that extracorporeal shockwave therapy (ESWT) promotes the production of TGF-β1 and IGF-1 in rat models,² and it is believed to be able to disintegrate calcium deposits and stimulate tissue repair.¹⁴ Research is generally supportive of its effectiveness in treating tendinosis; however, evidence is limited by great variability in studies in terms of treatment intensity, frequency, duration, timing, number of treatments, and use of a local anesthetic.¹⁴ ESWT appears to be useful in augmenting treatment with EE, particularly with regard to the rotator cuff.¹⁹

Continue to: A review of 10 RCTs...

A review of 10 RCTs demonstrated the effectiveness of ESWT for tennis elbow.² ESWT for greater trochanteric pain syndrome (GTPS, formerly known as trochanteric bursitis) appears to be more effective than corticosteroids and home exercises for outcomes at 4 months and equivalent to home exercises at 15 months.²⁰ In patellar tendinosis, ESWT has been shown to be an effective treatment, especially under ultrasound guidance.¹² Studies involving the use of ESWT for Achilles tendinosis have had mixed results for midsubstance tendinosis, and more positive results for insertional tendinosis.¹⁵ For a video on how the therapy is administered, see https://www.youtube.com/watch?v=Fq5yqiWByX4.

Glyceryl trinitrate patches: Mixed results

Basic science studies have shown that nitric oxide modulates tendon healing by enhancing fibroblast proliferation and collagen synthesis,^2,14 but that it should be used with caution in cardiac patients and in those who take PDE-5 inhibitors. Common adverse effects include rash, headache, and dizziness.

Use glyceryl trinitrate patches with caution in cardiac patients and in those who take PDE-5 inhibitors.

In clinical studies, glyceryl trinitrate (GTN) patches show mixed results. For the upper extremity, GTN appears to be helpful for pain in the short term when combined with physical therapy, but long-term positive outcomes have been absent.²¹ In one Level 1 study for patellar tendinosis comparing GTN patches with EE to a placebo patch with EE, no significant difference was noted at 24 weeks.²² Benefit for Achilles tendinosis also appears to be lacking.^3,23

Corticosteroid injections: Mechanism unknown

The mechanism for the beneficial effects of corticosteroid injections (CSIs) for tendinosis remains controversial. Proposed mechanisms include lysis of peritendinous adhesions, disruption of the nociceptors in the region of the injection, and decreased vascularization.^10,15 Given tendinosis is generally regarded as a noninflammatory condition, and the fact that these medications have demonstrated potential negative effects on tendon healing, exercise caution when considering CSIs.^2,24

Although steroids can effectively reduce pain in the short term, intermediate- and long-term studies generally show no difference or worse outcomes when they are compared to no treatment, placebo, or other treatment modalities. In fact, strong evidence exists for negative effects of steroids on lateral epicondylosis in both the intermediate (6 months) and long (1 year) term.²⁴ Particular care is required when administering a CSI for medial epicondylosis, as the ulnar nerve is immediately posterior to the medial epicondyle.²⁵

Continue to: In contrast...

In contrast, CSIs appear to be a reliable treatment option for de Quervain disease.²⁶ Landmark-guided injections for GTPS can improve pain in the short term (< 1 month), but are inferior to either home exercises or ESWT beyond a few months. Thus, CSIs are a reasonable option for pain control in GTPS, but should not be the sole treatment modality.²⁰

Studies regarding corticosteroid use for Achilles and patellar tendinosis have had mixed results. Patients can hope for mild improvement in pain at best, and the risk for relapse and tendon rupture is ever present.²⁷ This is especially concerning given the significant load-bearing of the patellar and Achilles tendons.^14,15 If you are considering a CSI for these purposes, use imaging guidance to ensure the injection is not placed intratendinously.

Platelet-rich plasma and whole blood: Inducing an anabolic healing response

Platelet-rich plasma (PRP) and whole blood injections both aim to deliver autologous growth factors (eg, VEGF, PDGF, and IGF-1) and bioactive mediators to the site of tendinosis to induce an anabolic healing response. PRP therapy differs from whole blood therapy in that it is withdrawn and then concentrated in a centrifuge before being injected. Patients are typically injected under ultrasound guidance. The great variation in PRP preparation, platelet concentration, use of adjunctive treatments, leukocyte concentration, and number and technique of injections makes it difficult to determine the optimal PRP treatment protocol.^10,28,29

In 1 prospective RCT comparing subacromial PRP injections to CSI for the shoulder, the PRP group had better outcomes at 3 months, but similar outcomes at 6 months. The suggestion was made that PRP therapy could be an alternative treatment for individuals with a contraindication to CSIs.³⁰

PRP therapy appears to be an effective treatment option for patellar tendinopathy.^28,31 A Level 1 study comparing dry needle tenotomy and EE to dry needle tenotomy with both PRP therapy and EE found faster recovery in the PRP group.³² In another patellar tendon study comparing ESWT to PRP therapy, both were found to be effective, but PRP performed better in terms of pain, function, and satisfaction at 6 and 12 months.¹² For Achilles tendons, however, the evidence is mixed; case series have had generally positive outcomes, but the only double-blind RCT found no benefit.^28,31

Continue to: In lateral epicondylosis...

In lateral epicondylosis, the use of auto-logous whole blood or PRP injections appears to help both pain and function, with several studies failing to demonstrate superiority of 1 modality over the other.^24,25,28,33 This raises the issue of whether PRP therapy is any more effective than whole blood for the treatment of other tendinopathies. Unfortunately, there is a paucity of studies comparing the effectiveness of 1 modality to the other, apart from those for lateral epicondylosis.

Prolotherapy: An option for these 3 conditions

Prolotherapy involves the injection of hypertonic dextrose and local anesthetic, which is believed to lead to an upregulation of inflammatory mediators and growth factors. This treatment usually involves several injections spaced 2 to 6 weeks apart over several months. High-quality studies are not available to clarify the optimal dextrose concentration or number of injections required. The few high-quality studies available support prolotherapy for lateral epicondylosis, rotator cuff tendinopathy, and Osgood Schlatter disease. Lesser-quality studies support its use for refractory pain of the Achilles, hip adductors, and plantar fascia.^24,34

Sclerotherapy: Not just for veins

As discussed earlier, tendinotic tissue can have neovascularization that is easily detected on Doppler ultrasound. Sensory nerves typically grow alongside the new vessels. Sclerosing agents, such as polidocanol, can be injected with ultrasound guidance into areas of neovascularization, with the intention of causing denervation and pain relief.¹⁵ Neovascularization does not always correlate with pathology, so careful patient selection is necessary.³⁵

Exercise caution when considering corticosteroid injections for tendinosis.

Studies of sclerotherapy for patellar tendinopathy are generally favorable. One comparing sclerotherapy to arthroscopic debridement showed improvement in pain from both treatments at 6 and 12 months, but the arthroscopy group had less pain, better satisfaction scores, and a faster return to sport.¹⁴ Sclerotherapy is also effective for Achilles tendinosis.¹⁵

Stem cells: Not at this time

Stem cell use for tendinosis is based on the theory that these cells possess the capability to differentiate into tenocytes to produce new, healthy tendon tissue. Additionally, stem cell injections are believed to create a local immune response, recruiting local growth factors and cytokines to aide in tendon repair. A recent systematic review failed to identify any high-quality studies (Level 4 data at best) supporting the use of stem cells in tendinopathy, and the researchers did not recommend stem cell use outside of clinical trials at this time.³⁶

Continue to: Percutaneous needle tenotomy...

Percutaneous needle tenotomy is thought to benefit tendinosis by disrupting the tendinotic tissue via needling, while simultaneously causing bleeding and the release of growth factors to aid in healing. Unlike surgical ­tenotomy, the procedure is typically performed with ultrasound guidance in the office or other ambulatory setting. After local anesthesia is administered, a needle is passed multiple times through the entire region of abnormality noted on ultrasound. Generally, around 20 to 30 needle fenestrations are performed.^37,38

Consider prolotherapy for lateral epicondylosis, rotator cuff tendinopathy, and Osgood Schlatter disease.

In one retrospective study evaluating 47 patellar tendons, 81% had excellent or good results.³⁸ In a retrospective study for lateral epicondylosis, 80% had good to excellent results.³⁹

CORRESPONDENCE
Kyle Goerl, MD, CAQSM, Lafene Health Center, 1105 Sunset Avenue, Manhattan, KS, 66502-3761; kvg3355@ksu.edu.

As of 2015, an estimated 30.2 million adults in the United States—12.2% of the population— had diabetes mellitus (DM). During that year, approximately 1.5 million new cases (6.7 cases for every 1000 people) were diagnosed in adults (≥ 18 years of age).¹

As the number of people with DM increases, so will the number of cases of diabetic retinopathy, the main cause of new cases of blindness in adults in the United States² and the leading cause of blindness among US working-age (20 to 74 years) adults.³ It is estimated that 4.1 million Americans have diabetic retinopathy³; it is projected that prevalence will reach 6 million this year.⁴

Blindness related to DM costs the United States approximately $500 million each year,⁵ including health care utilization: physician office visits, diagnostic testing, medication and other treatments, and hospitalization.⁶ Impairment of vision also results in social isolation, dependence on others to perform daily functions, and a decline in physical activity.

Several professional organizations, including the American Diabetes Association and the American Academy of Ophthalmology, have developed practice guidelines for diabetic retinopathy screening. Guidelines notwithstanding, only about 55% of people with DM in the United States receive the recommended dilated eye examination at established intervals.^2,3 In addition to screening by an ophthalmologist or optometrist, adherence to clinical guidelines for risk assessment, prevention, and early referral helps reduce the incidence and severity of retinopathy.⁵

The Wisconsin Epidemiologic Study of Diabetic Retinopathy found that the prevalence of diabetic retinopathy varied from 28.8% in people who had diabetes mellitus (DM) for < 5 years to 77.8% in people who had DM for ≥ 15 years.

This article describes how to assess the risk of diabetic retinopathy in your patients, details the crucial role that you, the primary care physician, can play in prevention, and emphasizes the importance of referral to an eye specialist for screening, evaluation, treatment (when indicated), and follow-up.

Pathophysiology and classification

Diabetic retinopathy, the result of progressive blood vessel damage to the retina, has 2 major forms: nonproliferative and proliferative. Those forms are distinguished by the absence or presence of new growth of blood vessels (retinal neovascularization).^3,7 To improve communication and coordination among physicians who care for patients with DM worldwide, the International Clinical Diabetic Retinopathy Disease Severity Scale for diabetic retinopathy was developed,^8-10 comprising 5 levels of severity that are based on findings on dilated ophthalmoscopy (Table 1^8-10):

• Level 1. No apparent retinopathy. Funduscopic abnormalities are absent.
• Level 2. Mild nonproliferative diabetic retinopathy (NPDR). Only a few microaneurysms are seen.
• Level 3: Moderate NPDR. Characterized by microaneurysms and by intraretinal hemorrhage and venous beading, but less severe than what is seen in Level 4.
• Level 4. Severe NPDR. More than 20 intraretinal hemorrhages in each quadrant of the retina, definite venous beading in > 2 quadrants, intraretinal microvascular abnormalities in > 1 quadrant, or any combination of these findings.
• Level 5. Proliferative diabetic retinopathy. Characterized by neovascularization of the disc, retina, iris, or angle; vitreous hemorrhage; retinal detachment; or any combination of these findings. Further classified as “mild,” “moderate,” or “severe” if macular edema is present; severity is dependent on the distance of thickening and exudates from the center of the macula.⁹

Be attentive to risk factors

There are several risk factors for diabetic retinopathy, including duration of disease, type 1 DM, male gender, black race (non-Hispanic), elevated hemoglobin A1C(HbA1C) level, elevated systolic and diastolic blood pressure (BP), and insulin therapy. ^4,5,11,12

Continue to: Time since diagnosis

Time since diagnosis. The Wisconsin Epidemiologic Study of Diabetic Retinopathy found that the prevalence of diabetic retinopathy varied from 28.8% in people who had DM for < 5 years to 77.8% in people who had DM for ≥ 15 years. The rate of proliferative diabetic retinopathy was 2% in people who had DM for < 5 years and 15.5% in those who had DM for ≥ 15 years.¹¹

Diabetic retinopathy can deteriorate during pregnancy but generally reverts to the pre-pregnancy level; long-term progression of retinopathy is not affected.

Demographic variables. The prevalence of diabetic retinopathy is higher in men, non-Hispanic blacks (38.8%), and ­people with type 1 DM.^4,5,11-13 The Veterans ­Affairs Diabetes Trial found a higher prevalence of moderate-to-severe diabetic retinopathy in Hispanics (36%) and African Americans (29%) than in non-Hispanic whites (22%).¹⁴

Among people with DM who have ­diabetic retinopathy, systolic and diastolic BP and the HbA1C level tend to be higher. They are more likely to use insulin to control disease.^4,5,13 In a recent cross-sectional analysis, the prevalence of vision-threatening retinopathy was higher among people ≥ 65 years of age (1%; 95% confidence interval [CI], 0.7%-1.5%) than among people 40 to 64 years of age (0.4%; 95% CI, 0.3%-0.7%) (P = .009).⁵

Does pregnancy exacerbate retinopathy? Controversy surrounds the role of pregnancy in the development and progression of diabetic retinopathy. The Diabetes Control and Complications Trial found a short-term increase in the level of retinopathy during pregnancy that persisted into the first postpartum year. A 1.63-fold greater risk of any deterioration of retinopathy was observed in women who received intensive DM treatment from before to during pregnancy (P < .05); pregnant women who received conventional treatment had a 2.48-fold greater risk than nonpregnant women with DM who received conventional treatment (P < .001).

Deterioration of retinopathy during pregnancy had no long-term consequences, however, regardless of type of treatment.¹⁵ More importantly, in most cases, changes in the level of retinopathy revert to the pre-­pregnancy level after 1 year or longer, and pregnancy does not appear to affect long-term progression of retinopathy.¹⁵

Continue to: Proven primary prevention strategies

Glycemic control. Optimal glycemic control is an essential component of prevention of diabetic retinopathy. From 1983 to 1993, the Diabetes Control and Complications Trial randomized 1441 patients with type 1 DM to receive intensive therapy (median HbA1C level, 7.2%) or conventional therapy (median HbA1C level, 9.1%). During a mean of 6 years of follow-up, intensive therapy reduced the adjusted mean risk of retinopathy by 76% (95% CI, 62%-85%).^16,17 A 2007 systematic review of 44 studies of the treatment of diabetic retinopathy found that strict glycemic control was beneficial in reducing the incidence and progression of retinopathy.¹⁷

The American Diabetes Association’s Standards of Medical Care in Diabetes—2019 Abridged for Primary Care Providers recommends that most nonpregnant adults maintain an HbA1Clevel < 7%. For patients with a history of hypoglycemia, limited life expectancy, advanced microvascular or macrovascular disease, other significant comorbid conditions, or longstanding DM in which it is difficult to achieve the optimal goal, a higher HbA1clevel (< 8%) might be appropriate.¹⁸

Control of BP. Strict control of BP is a major modifier of the incidence and progression of diabetic retinopathy.^17,19 In the United Kingdom Prospective Diabetes Study, 1148 patients with type 2 DM and a mean BP of 160/94 mm Hg at the onset of the study were randomly assigned to either (1) a “tight” blood pressure group (< 150/85 mm Hg) or (2) a “less-tight” group (< 180/105 mm Hg). The primary therapy for controlling BP was captopril or atenolol. After 9 years of follow-up, the tight-control group had a 34% mean reduction in risk in the percentage of patients with deterioration of retinopathy (99% CI, 11%-50%; P = .0004) and a 47% reduction in risk (99% CI, 7%-70%; P = .004) of deterioration in visual acuity.²⁰

Most patients with DM and hypertension should be treated to maintain a BP < 140/90 mm Hg. Although there is insufficient evidence to recommend a specific antihypertensive agent for preventing diabetic retinopathy, therapy should include agents from drug classes that have a demonstrated reduction in cardiovascular events in patients with DM. These include angiotensin-­converting enzyme inhibitors, angiotensin receptor blockers, thiazide diuretics, and dihydropyridine calcium channel blockers.¹⁸

Lipid management. The benefit of targeted therapy for lowering lipids for the prevention of diabetic retinopathy is not well established.¹⁷ In the Collaborative Atorvastatin Diabetes Study, 2838 patients with type 2 DM were randomized to atorvastatin (10 mg) or placebo; microvascular endpoint analysis demonstrated that patients taking atorvastatin needed less laser therapy (P = .14); however, progression of diabetic retinopathy was not reduced.²¹ Similarly, in the Action to Control Cardiovascular Risk in Diabetes Eye Study, slowing of progression to retinopathy was observed in patients with type 2 DM who were treated with fenofibrate (ie, progression in 6.5%, compared with progression in 10.2% of untreated subjects [odds ratio = 0.60 (95% CI, 0.42-0.87); P = .0056]).²²

Continue to: Despite limited data on...

Nonpregnant adults with diabetes should maintain an HbA1C level < 7%; a higher level (< 8%) might be appropriate in patients with various complicating factors.

Despite limited data on the impact of ­lipid-lowering agents on patients with diabetic retinopathy, those with type 2 DM (especially) and those who have, or are at risk of, atherosclerotic cardiovascular disease should receive statin therapy.¹⁸

Aspirin therapy. Aspirin has not been found to be beneficial for slowing progression of diabetic retinopathy. However, aspirin did not cause further deterioration of disease, specifically in patients with vitreous hemorrhages⁴; patients with diabetic retinopathy who require aspirin therapy for other medical reasons can therefore continue to take it without increasing the risk of damage to the retina.^4,18

When should you refer patients for screening?

Screening for diabetic retinopathy is important because affected patients can be asymptomatic but have significant disease. Early detection also helps determine which patients need treatment when it is most beneficial: early in its course.⁴

Type 1 DM. Retinopathy can become apparent as early as 6 or 7 years after the onset of disease, and is rare in children prior to puberty.^4,11 As a result, patients with type 1 DM should first be screened with a comprehensive eye examination by an ophthalmologist or optometrist within 5 years of DM onset.^4,18

Type 2 DM. Because of the insidious onset of type 2 DM, patients who are given a diagnosis of DM after 30 years of age might already have high-risk features of retinopathy.⁹ In patients with type 2 DM, therefore, initial screening for diabetic retinopathy should begin at the time of diagnosis and include a comprehensive eye examination by an ophthalmologist or optometrist.^4,18,23

Continue to: Components of the exam

Components of the exam. Initial evaluation by the ophthalmologist or optometrist should include a detailed history and comprehensive eye exam with pupil dilation. Table 2⁴ lists elements of the initial physical exam, which should assess for features that often lead to visual impairment. These features include macular edema, retinal hemorrhage, venous beading, neovascularization, and vitreous hemorrhage.⁴

Frequency of follow-up. The interval between subsequent examinations should be individualized, based on the findings of the initial assessment. Consider that:

• Screening should occur every 1 or 2 years in patients without evidence of retinopathy and with adequate glycemic control.^4,18,23
• Screening every 1 or 2 years appears to be cost-effective in patients who have had 1 or more normal eye exams.
• A 3-year screening interval does not appear to present a risk in well-controlled patients with type 2 DM.²⁴
• Women with type 1 or type 2 DM who are planning pregnancy or who are pregnant should have an eye exam prior to pregnancy or early in the first trimester.^4,18,23 They should then be monitored each trimester and at the end of the first postpartum year, depending on the severity of retinopathy.¹⁸

Alternative screening modalities

Seven-field stereoscopic fundus photography is an alternative screening tool that compares favorably to ophthalmoscopy when performed by an experienced ophthalmologist, optometrist, or ophthalmologic technician.²⁵ Nonmydriatic digital stereoscopic retinal imaging has been shown to be a cost-effective method of screening patients for diabetic retinopathy.²⁶ In a study that compared digital imaging with dilated funduscopic examination, investigators reported that, of 311 eyes evaluated, there was agreement between the methods in 86% of cases. Disagreement was mostly related to the greater frequency of finding mild-to-moderate NPDR when using digital imaging.²⁷

Screening in primary care

Programs that use telemedicine-based fundus photography to screen for diabetic retinopathy during primary care visits, followed by remote interpretation by an ophthalmologist, have been shown to increase the rate of retinal screening by offering an option other than direct referral to an ophthalmologist or optometrist.²⁸ However, telemedicine-based retinal photography can be successful as a screening tool for retinopathy only if timely referral to an eye specialist is arranged when indicated by findings.¹⁸

What therapy will your referred patients receive?

Patients found to have signs of diabetic retinopathy should be referred to an ophthalmologist who is knowledgeable and experienced in the management of diabetic retinopathy. Care will be managed according to the severity of the patient’s diabetic retinopathy.

Continue to: Patients with mild-to-moderate NPDR but without macular edema

Patients with mild-to-moderate NPDR but without macular edema. Treatment is generally not recommended. Patients should be reevaluated every 6 to 12 months because they have an increased risk of progression.⁵

Patients with mild-to-moderate NPDR and clinically significant macular edema (CSME). It is important for the eye specialist to assess for edema at the center of the macula because the risk of vision loss and need for treatment is greater when the center is involved. Vascular–endothelial growth factor (VEGF) is an important mediator of neovascularization and macular edema in diabetic retinopathy. For patients with center-­involving CSME, intravitreous injection of an anti-VEGF agent provides significant benefit and is first-line treatment in these cases.^4,29

Patients with DM (especially those with type 2 disease) who have, or are at risk of, atherosclerotic cardiovascular disease should receive statin therapy.

The Early Treatment for Diabetic Retinopathy Study evaluated the efficacy of focal photocoagulation, a painless laser therapy, for CSME and demonstrated that this modality reduces the risk of moderate visual loss; increases the likelihood of improvement in vision; and decreases the frequency of persistent macular edema.³⁰ Focal photocoagulation has been found effective in both non-center-involving CSME and center-involving CSME.⁵

Patients with severe NPDR. The recommendation is to initiate full panretinal photocoagulation prior to progression to proliferative diabetic retinopathy PDR. Researchers noted a 50% reduction in vision loss and vitrectomy when patients with type 2 DM were treated with panretinal photocoagulation early, compared with those in whom treatment was deferred until PDR developed.^4,31 The role of anti-VEGF treatment of severe NPDR is under investigation.⁴

Patients with high-risk and severe PDR. Panretinal photocoagulation is the recommended treatment for patients with high-risk and severe PDR, and usually induces regression of retinal neovascularization. In patients with CSME and high-risk PDR, the combination of anti-VEGF therapy and panretinal photocoagulation should be considered. Vitrectomy should be considered for patients who have failed panretinal photocoagulation or are not amenable to photocoagulation.⁴

CORRESPONDENCE
Bryan Farford, DO, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224; Farford.Bryan@mayo.edu.

As of 2015, an estimated 30.2 million adults in the United States—12.2% of the population— had diabetes mellitus (DM). During that year, approximately 1.5 million new cases (6.7 cases for every 1000 people) were diagnosed in adults (≥ 18 years of age).¹

As the number of people with DM increases, so will the number of cases of diabetic retinopathy, the main cause of new cases of blindness in adults in the United States² and the leading cause of blindness among US working-age (20 to 74 years) adults.³ It is estimated that 4.1 million Americans have diabetic retinopathy³; it is projected that prevalence will reach 6 million this year.⁴

Blindness related to DM costs the United States approximately $500 million each year,⁵ including health care utilization: physician office visits, diagnostic testing, medication and other treatments, and hospitalization.⁶ Impairment of vision also results in social isolation, dependence on others to perform daily functions, and a decline in physical activity.

Several professional organizations, including the American Diabetes Association and the American Academy of Ophthalmology, have developed practice guidelines for diabetic retinopathy screening. Guidelines notwithstanding, only about 55% of people with DM in the United States receive the recommended dilated eye examination at established intervals.^2,3 In addition to screening by an ophthalmologist or optometrist, adherence to clinical guidelines for risk assessment, prevention, and early referral helps reduce the incidence and severity of retinopathy.⁵

The Wisconsin Epidemiologic Study of Diabetic Retinopathy found that the prevalence of diabetic retinopathy varied from 28.8% in people who had diabetes mellitus (DM) for < 5 years to 77.8% in people who had DM for ≥ 15 years.

This article describes how to assess the risk of diabetic retinopathy in your patients, details the crucial role that you, the primary care physician, can play in prevention, and emphasizes the importance of referral to an eye specialist for screening, evaluation, treatment (when indicated), and follow-up.

Pathophysiology and classification

Diabetic retinopathy, the result of progressive blood vessel damage to the retina, has 2 major forms: nonproliferative and proliferative. Those forms are distinguished by the absence or presence of new growth of blood vessels (retinal neovascularization).^3,7 To improve communication and coordination among physicians who care for patients with DM worldwide, the International Clinical Diabetic Retinopathy Disease Severity Scale for diabetic retinopathy was developed,^8-10 comprising 5 levels of severity that are based on findings on dilated ophthalmoscopy (Table 1^8-10):

• Level 1. No apparent retinopathy. Funduscopic abnormalities are absent.
• Level 2. Mild nonproliferative diabetic retinopathy (NPDR). Only a few microaneurysms are seen.
• Level 3: Moderate NPDR. Characterized by microaneurysms and by intraretinal hemorrhage and venous beading, but less severe than what is seen in Level 4.
• Level 4. Severe NPDR. More than 20 intraretinal hemorrhages in each quadrant of the retina, definite venous beading in > 2 quadrants, intraretinal microvascular abnormalities in > 1 quadrant, or any combination of these findings.
• Level 5. Proliferative diabetic retinopathy. Characterized by neovascularization of the disc, retina, iris, or angle; vitreous hemorrhage; retinal detachment; or any combination of these findings. Further classified as “mild,” “moderate,” or “severe” if macular edema is present; severity is dependent on the distance of thickening and exudates from the center of the macula.⁹

Be attentive to risk factors

There are several risk factors for diabetic retinopathy, including duration of disease, type 1 DM, male gender, black race (non-Hispanic), elevated hemoglobin A1C(HbA1C) level, elevated systolic and diastolic blood pressure (BP), and insulin therapy. ^4,5,11,12

Continue to: Time since diagnosis

Time since diagnosis. The Wisconsin Epidemiologic Study of Diabetic Retinopathy found that the prevalence of diabetic retinopathy varied from 28.8% in people who had DM for < 5 years to 77.8% in people who had DM for ≥ 15 years. The rate of proliferative diabetic retinopathy was 2% in people who had DM for < 5 years and 15.5% in those who had DM for ≥ 15 years.¹¹

Diabetic retinopathy can deteriorate during pregnancy but generally reverts to the pre-pregnancy level; long-term progression of retinopathy is not affected.

Demographic variables. The prevalence of diabetic retinopathy is higher in men, non-Hispanic blacks (38.8%), and ­people with type 1 DM.^4,5,11-13 The Veterans ­Affairs Diabetes Trial found a higher prevalence of moderate-to-severe diabetic retinopathy in Hispanics (36%) and African Americans (29%) than in non-Hispanic whites (22%).¹⁴

Among people with DM who have ­diabetic retinopathy, systolic and diastolic BP and the HbA1C level tend to be higher. They are more likely to use insulin to control disease.^4,5,13 In a recent cross-sectional analysis, the prevalence of vision-threatening retinopathy was higher among people ≥ 65 years of age (1%; 95% confidence interval [CI], 0.7%-1.5%) than among people 40 to 64 years of age (0.4%; 95% CI, 0.3%-0.7%) (P = .009).⁵

Does pregnancy exacerbate retinopathy? Controversy surrounds the role of pregnancy in the development and progression of diabetic retinopathy. The Diabetes Control and Complications Trial found a short-term increase in the level of retinopathy during pregnancy that persisted into the first postpartum year. A 1.63-fold greater risk of any deterioration of retinopathy was observed in women who received intensive DM treatment from before to during pregnancy (P < .05); pregnant women who received conventional treatment had a 2.48-fold greater risk than nonpregnant women with DM who received conventional treatment (P < .001).

Deterioration of retinopathy during pregnancy had no long-term consequences, however, regardless of type of treatment.¹⁵ More importantly, in most cases, changes in the level of retinopathy revert to the pre-­pregnancy level after 1 year or longer, and pregnancy does not appear to affect long-term progression of retinopathy.¹⁵

Continue to: Proven primary prevention strategies

Glycemic control. Optimal glycemic control is an essential component of prevention of diabetic retinopathy. From 1983 to 1993, the Diabetes Control and Complications Trial randomized 1441 patients with type 1 DM to receive intensive therapy (median HbA1C level, 7.2%) or conventional therapy (median HbA1C level, 9.1%). During a mean of 6 years of follow-up, intensive therapy reduced the adjusted mean risk of retinopathy by 76% (95% CI, 62%-85%).^16,17 A 2007 systematic review of 44 studies of the treatment of diabetic retinopathy found that strict glycemic control was beneficial in reducing the incidence and progression of retinopathy.¹⁷

The American Diabetes Association’s Standards of Medical Care in Diabetes—2019 Abridged for Primary Care Providers recommends that most nonpregnant adults maintain an HbA1Clevel < 7%. For patients with a history of hypoglycemia, limited life expectancy, advanced microvascular or macrovascular disease, other significant comorbid conditions, or longstanding DM in which it is difficult to achieve the optimal goal, a higher HbA1clevel (< 8%) might be appropriate.¹⁸

Control of BP. Strict control of BP is a major modifier of the incidence and progression of diabetic retinopathy.^17,19 In the United Kingdom Prospective Diabetes Study, 1148 patients with type 2 DM and a mean BP of 160/94 mm Hg at the onset of the study were randomly assigned to either (1) a “tight” blood pressure group (< 150/85 mm Hg) or (2) a “less-tight” group (< 180/105 mm Hg). The primary therapy for controlling BP was captopril or atenolol. After 9 years of follow-up, the tight-control group had a 34% mean reduction in risk in the percentage of patients with deterioration of retinopathy (99% CI, 11%-50%; P = .0004) and a 47% reduction in risk (99% CI, 7%-70%; P = .004) of deterioration in visual acuity.²⁰

Most patients with DM and hypertension should be treated to maintain a BP < 140/90 mm Hg. Although there is insufficient evidence to recommend a specific antihypertensive agent for preventing diabetic retinopathy, therapy should include agents from drug classes that have a demonstrated reduction in cardiovascular events in patients with DM. These include angiotensin-­converting enzyme inhibitors, angiotensin receptor blockers, thiazide diuretics, and dihydropyridine calcium channel blockers.¹⁸

Lipid management. The benefit of targeted therapy for lowering lipids for the prevention of diabetic retinopathy is not well established.¹⁷ In the Collaborative Atorvastatin Diabetes Study, 2838 patients with type 2 DM were randomized to atorvastatin (10 mg) or placebo; microvascular endpoint analysis demonstrated that patients taking atorvastatin needed less laser therapy (P = .14); however, progression of diabetic retinopathy was not reduced.²¹ Similarly, in the Action to Control Cardiovascular Risk in Diabetes Eye Study, slowing of progression to retinopathy was observed in patients with type 2 DM who were treated with fenofibrate (ie, progression in 6.5%, compared with progression in 10.2% of untreated subjects [odds ratio = 0.60 (95% CI, 0.42-0.87); P = .0056]).²²

Continue to: Despite limited data on...

Nonpregnant adults with diabetes should maintain an HbA1C level < 7%; a higher level (< 8%) might be appropriate in patients with various complicating factors.

Despite limited data on the impact of ­lipid-lowering agents on patients with diabetic retinopathy, those with type 2 DM (especially) and those who have, or are at risk of, atherosclerotic cardiovascular disease should receive statin therapy.¹⁸

Aspirin therapy. Aspirin has not been found to be beneficial for slowing progression of diabetic retinopathy. However, aspirin did not cause further deterioration of disease, specifically in patients with vitreous hemorrhages⁴; patients with diabetic retinopathy who require aspirin therapy for other medical reasons can therefore continue to take it without increasing the risk of damage to the retina.^4,18

When should you refer patients for screening?

Screening for diabetic retinopathy is important because affected patients can be asymptomatic but have significant disease. Early detection also helps determine which patients need treatment when it is most beneficial: early in its course.⁴

Type 1 DM. Retinopathy can become apparent as early as 6 or 7 years after the onset of disease, and is rare in children prior to puberty.^4,11 As a result, patients with type 1 DM should first be screened with a comprehensive eye examination by an ophthalmologist or optometrist within 5 years of DM onset.^4,18

Type 2 DM. Because of the insidious onset of type 2 DM, patients who are given a diagnosis of DM after 30 years of age might already have high-risk features of retinopathy.⁹ In patients with type 2 DM, therefore, initial screening for diabetic retinopathy should begin at the time of diagnosis and include a comprehensive eye examination by an ophthalmologist or optometrist.^4,18,23

Continue to: Components of the exam

Components of the exam. Initial evaluation by the ophthalmologist or optometrist should include a detailed history and comprehensive eye exam with pupil dilation. Table 2⁴ lists elements of the initial physical exam, which should assess for features that often lead to visual impairment. These features include macular edema, retinal hemorrhage, venous beading, neovascularization, and vitreous hemorrhage.⁴

Frequency of follow-up. The interval between subsequent examinations should be individualized, based on the findings of the initial assessment. Consider that:

• Screening should occur every 1 or 2 years in patients without evidence of retinopathy and with adequate glycemic control.^4,18,23
• Screening every 1 or 2 years appears to be cost-effective in patients who have had 1 or more normal eye exams.
• A 3-year screening interval does not appear to present a risk in well-controlled patients with type 2 DM.²⁴
• Women with type 1 or type 2 DM who are planning pregnancy or who are pregnant should have an eye exam prior to pregnancy or early in the first trimester.^4,18,23 They should then be monitored each trimester and at the end of the first postpartum year, depending on the severity of retinopathy.¹⁸

Alternative screening modalities

Seven-field stereoscopic fundus photography is an alternative screening tool that compares favorably to ophthalmoscopy when performed by an experienced ophthalmologist, optometrist, or ophthalmologic technician.²⁵ Nonmydriatic digital stereoscopic retinal imaging has been shown to be a cost-effective method of screening patients for diabetic retinopathy.²⁶ In a study that compared digital imaging with dilated funduscopic examination, investigators reported that, of 311 eyes evaluated, there was agreement between the methods in 86% of cases. Disagreement was mostly related to the greater frequency of finding mild-to-moderate NPDR when using digital imaging.²⁷

Screening in primary care

Programs that use telemedicine-based fundus photography to screen for diabetic retinopathy during primary care visits, followed by remote interpretation by an ophthalmologist, have been shown to increase the rate of retinal screening by offering an option other than direct referral to an ophthalmologist or optometrist.²⁸ However, telemedicine-based retinal photography can be successful as a screening tool for retinopathy only if timely referral to an eye specialist is arranged when indicated by findings.¹⁸

What therapy will your referred patients receive?

Patients found to have signs of diabetic retinopathy should be referred to an ophthalmologist who is knowledgeable and experienced in the management of diabetic retinopathy. Care will be managed according to the severity of the patient’s diabetic retinopathy.

Continue to: Patients with mild-to-moderate NPDR but without macular edema

Patients with mild-to-moderate NPDR but without macular edema. Treatment is generally not recommended. Patients should be reevaluated every 6 to 12 months because they have an increased risk of progression.⁵

Patients with mild-to-moderate NPDR and clinically significant macular edema (CSME). It is important for the eye specialist to assess for edema at the center of the macula because the risk of vision loss and need for treatment is greater when the center is involved. Vascular–endothelial growth factor (VEGF) is an important mediator of neovascularization and macular edema in diabetic retinopathy. For patients with center-­involving CSME, intravitreous injection of an anti-VEGF agent provides significant benefit and is first-line treatment in these cases.^4,29

Patients with DM (especially those with type 2 disease) who have, or are at risk of, atherosclerotic cardiovascular disease should receive statin therapy.

The Early Treatment for Diabetic Retinopathy Study evaluated the efficacy of focal photocoagulation, a painless laser therapy, for CSME and demonstrated that this modality reduces the risk of moderate visual loss; increases the likelihood of improvement in vision; and decreases the frequency of persistent macular edema.³⁰ Focal photocoagulation has been found effective in both non-center-involving CSME and center-involving CSME.⁵

Patients with severe NPDR. The recommendation is to initiate full panretinal photocoagulation prior to progression to proliferative diabetic retinopathy PDR. Researchers noted a 50% reduction in vision loss and vitrectomy when patients with type 2 DM were treated with panretinal photocoagulation early, compared with those in whom treatment was deferred until PDR developed.^4,31 The role of anti-VEGF treatment of severe NPDR is under investigation.⁴

Patients with high-risk and severe PDR. Panretinal photocoagulation is the recommended treatment for patients with high-risk and severe PDR, and usually induces regression of retinal neovascularization. In patients with CSME and high-risk PDR, the combination of anti-VEGF therapy and panretinal photocoagulation should be considered. Vitrectomy should be considered for patients who have failed panretinal photocoagulation or are not amenable to photocoagulation.⁴

CORRESPONDENCE
Bryan Farford, DO, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224; Farford.Bryan@mayo.edu.

As of 2015, an estimated 30.2 million adults in the United States—12.2% of the population— had diabetes mellitus (DM). During that year, approximately 1.5 million new cases (6.7 cases for every 1000 people) were diagnosed in adults (≥ 18 years of age).¹

As the number of people with DM increases, so will the number of cases of diabetic retinopathy, the main cause of new cases of blindness in adults in the United States² and the leading cause of blindness among US working-age (20 to 74 years) adults.³ It is estimated that 4.1 million Americans have diabetic retinopathy³; it is projected that prevalence will reach 6 million this year.⁴

Blindness related to DM costs the United States approximately $500 million each year,⁵ including health care utilization: physician office visits, diagnostic testing, medication and other treatments, and hospitalization.⁶ Impairment of vision also results in social isolation, dependence on others to perform daily functions, and a decline in physical activity.

Several professional organizations, including the American Diabetes Association and the American Academy of Ophthalmology, have developed practice guidelines for diabetic retinopathy screening. Guidelines notwithstanding, only about 55% of people with DM in the United States receive the recommended dilated eye examination at established intervals.^2,3 In addition to screening by an ophthalmologist or optometrist, adherence to clinical guidelines for risk assessment, prevention, and early referral helps reduce the incidence and severity of retinopathy.⁵

The Wisconsin Epidemiologic Study of Diabetic Retinopathy found that the prevalence of diabetic retinopathy varied from 28.8% in people who had diabetes mellitus (DM) for < 5 years to 77.8% in people who had DM for ≥ 15 years.

This article describes how to assess the risk of diabetic retinopathy in your patients, details the crucial role that you, the primary care physician, can play in prevention, and emphasizes the importance of referral to an eye specialist for screening, evaluation, treatment (when indicated), and follow-up.

Pathophysiology and classification

Diabetic retinopathy, the result of progressive blood vessel damage to the retina, has 2 major forms: nonproliferative and proliferative. Those forms are distinguished by the absence or presence of new growth of blood vessels (retinal neovascularization).^3,7 To improve communication and coordination among physicians who care for patients with DM worldwide, the International Clinical Diabetic Retinopathy Disease Severity Scale for diabetic retinopathy was developed,^8-10 comprising 5 levels of severity that are based on findings on dilated ophthalmoscopy (Table 1^8-10):

• Level 1. No apparent retinopathy. Funduscopic abnormalities are absent.
• Level 2. Mild nonproliferative diabetic retinopathy (NPDR). Only a few microaneurysms are seen.
• Level 3: Moderate NPDR. Characterized by microaneurysms and by intraretinal hemorrhage and venous beading, but less severe than what is seen in Level 4.
• Level 4. Severe NPDR. More than 20 intraretinal hemorrhages in each quadrant of the retina, definite venous beading in > 2 quadrants, intraretinal microvascular abnormalities in > 1 quadrant, or any combination of these findings.
• Level 5. Proliferative diabetic retinopathy. Characterized by neovascularization of the disc, retina, iris, or angle; vitreous hemorrhage; retinal detachment; or any combination of these findings. Further classified as “mild,” “moderate,” or “severe” if macular edema is present; severity is dependent on the distance of thickening and exudates from the center of the macula.⁹

Be attentive to risk factors

There are several risk factors for diabetic retinopathy, including duration of disease, type 1 DM, male gender, black race (non-Hispanic), elevated hemoglobin A1C(HbA1C) level, elevated systolic and diastolic blood pressure (BP), and insulin therapy. ^4,5,11,12

Continue to: Time since diagnosis

Time since diagnosis. The Wisconsin Epidemiologic Study of Diabetic Retinopathy found that the prevalence of diabetic retinopathy varied from 28.8% in people who had DM for < 5 years to 77.8% in people who had DM for ≥ 15 years. The rate of proliferative diabetic retinopathy was 2% in people who had DM for < 5 years and 15.5% in those who had DM for ≥ 15 years.¹¹

Diabetic retinopathy can deteriorate during pregnancy but generally reverts to the pre-pregnancy level; long-term progression of retinopathy is not affected.

Demographic variables. The prevalence of diabetic retinopathy is higher in men, non-Hispanic blacks (38.8%), and ­people with type 1 DM.^4,5,11-13 The Veterans ­Affairs Diabetes Trial found a higher prevalence of moderate-to-severe diabetic retinopathy in Hispanics (36%) and African Americans (29%) than in non-Hispanic whites (22%).¹⁴

Among people with DM who have ­diabetic retinopathy, systolic and diastolic BP and the HbA1C level tend to be higher. They are more likely to use insulin to control disease.^4,5,13 In a recent cross-sectional analysis, the prevalence of vision-threatening retinopathy was higher among people ≥ 65 years of age (1%; 95% confidence interval [CI], 0.7%-1.5%) than among people 40 to 64 years of age (0.4%; 95% CI, 0.3%-0.7%) (P = .009).⁵

Does pregnancy exacerbate retinopathy? Controversy surrounds the role of pregnancy in the development and progression of diabetic retinopathy. The Diabetes Control and Complications Trial found a short-term increase in the level of retinopathy during pregnancy that persisted into the first postpartum year. A 1.63-fold greater risk of any deterioration of retinopathy was observed in women who received intensive DM treatment from before to during pregnancy (P < .05); pregnant women who received conventional treatment had a 2.48-fold greater risk than nonpregnant women with DM who received conventional treatment (P < .001).

Deterioration of retinopathy during pregnancy had no long-term consequences, however, regardless of type of treatment.¹⁵ More importantly, in most cases, changes in the level of retinopathy revert to the pre-­pregnancy level after 1 year or longer, and pregnancy does not appear to affect long-term progression of retinopathy.¹⁵

Continue to: Proven primary prevention strategies

Glycemic control. Optimal glycemic control is an essential component of prevention of diabetic retinopathy. From 1983 to 1993, the Diabetes Control and Complications Trial randomized 1441 patients with type 1 DM to receive intensive therapy (median HbA1C level, 7.2%) or conventional therapy (median HbA1C level, 9.1%). During a mean of 6 years of follow-up, intensive therapy reduced the adjusted mean risk of retinopathy by 76% (95% CI, 62%-85%).^16,17 A 2007 systematic review of 44 studies of the treatment of diabetic retinopathy found that strict glycemic control was beneficial in reducing the incidence and progression of retinopathy.¹⁷

The American Diabetes Association’s Standards of Medical Care in Diabetes—2019 Abridged for Primary Care Providers recommends that most nonpregnant adults maintain an HbA1Clevel < 7%. For patients with a history of hypoglycemia, limited life expectancy, advanced microvascular or macrovascular disease, other significant comorbid conditions, or longstanding DM in which it is difficult to achieve the optimal goal, a higher HbA1clevel (< 8%) might be appropriate.¹⁸

Control of BP. Strict control of BP is a major modifier of the incidence and progression of diabetic retinopathy.^17,19 In the United Kingdom Prospective Diabetes Study, 1148 patients with type 2 DM and a mean BP of 160/94 mm Hg at the onset of the study were randomly assigned to either (1) a “tight” blood pressure group (< 150/85 mm Hg) or (2) a “less-tight” group (< 180/105 mm Hg). The primary therapy for controlling BP was captopril or atenolol. After 9 years of follow-up, the tight-control group had a 34% mean reduction in risk in the percentage of patients with deterioration of retinopathy (99% CI, 11%-50%; P = .0004) and a 47% reduction in risk (99% CI, 7%-70%; P = .004) of deterioration in visual acuity.²⁰

Most patients with DM and hypertension should be treated to maintain a BP < 140/90 mm Hg. Although there is insufficient evidence to recommend a specific antihypertensive agent for preventing diabetic retinopathy, therapy should include agents from drug classes that have a demonstrated reduction in cardiovascular events in patients with DM. These include angiotensin-­converting enzyme inhibitors, angiotensin receptor blockers, thiazide diuretics, and dihydropyridine calcium channel blockers.¹⁸

Lipid management. The benefit of targeted therapy for lowering lipids for the prevention of diabetic retinopathy is not well established.¹⁷ In the Collaborative Atorvastatin Diabetes Study, 2838 patients with type 2 DM were randomized to atorvastatin (10 mg) or placebo; microvascular endpoint analysis demonstrated that patients taking atorvastatin needed less laser therapy (P = .14); however, progression of diabetic retinopathy was not reduced.²¹ Similarly, in the Action to Control Cardiovascular Risk in Diabetes Eye Study, slowing of progression to retinopathy was observed in patients with type 2 DM who were treated with fenofibrate (ie, progression in 6.5%, compared with progression in 10.2% of untreated subjects [odds ratio = 0.60 (95% CI, 0.42-0.87); P = .0056]).²²

Continue to: Despite limited data on...

Nonpregnant adults with diabetes should maintain an HbA1C level < 7%; a higher level (< 8%) might be appropriate in patients with various complicating factors.

Despite limited data on the impact of ­lipid-lowering agents on patients with diabetic retinopathy, those with type 2 DM (especially) and those who have, or are at risk of, atherosclerotic cardiovascular disease should receive statin therapy.¹⁸

Aspirin therapy. Aspirin has not been found to be beneficial for slowing progression of diabetic retinopathy. However, aspirin did not cause further deterioration of disease, specifically in patients with vitreous hemorrhages⁴; patients with diabetic retinopathy who require aspirin therapy for other medical reasons can therefore continue to take it without increasing the risk of damage to the retina.^4,18

When should you refer patients for screening?

Screening for diabetic retinopathy is important because affected patients can be asymptomatic but have significant disease. Early detection also helps determine which patients need treatment when it is most beneficial: early in its course.⁴

Type 1 DM. Retinopathy can become apparent as early as 6 or 7 years after the onset of disease, and is rare in children prior to puberty.^4,11 As a result, patients with type 1 DM should first be screened with a comprehensive eye examination by an ophthalmologist or optometrist within 5 years of DM onset.^4,18

Type 2 DM. Because of the insidious onset of type 2 DM, patients who are given a diagnosis of DM after 30 years of age might already have high-risk features of retinopathy.⁹ In patients with type 2 DM, therefore, initial screening for diabetic retinopathy should begin at the time of diagnosis and include a comprehensive eye examination by an ophthalmologist or optometrist.^4,18,23

Continue to: Components of the exam

Components of the exam. Initial evaluation by the ophthalmologist or optometrist should include a detailed history and comprehensive eye exam with pupil dilation. Table 2⁴ lists elements of the initial physical exam, which should assess for features that often lead to visual impairment. These features include macular edema, retinal hemorrhage, venous beading, neovascularization, and vitreous hemorrhage.⁴

Frequency of follow-up. The interval between subsequent examinations should be individualized, based on the findings of the initial assessment. Consider that:

• Screening should occur every 1 or 2 years in patients without evidence of retinopathy and with adequate glycemic control.^4,18,23
• Screening every 1 or 2 years appears to be cost-effective in patients who have had 1 or more normal eye exams.
• A 3-year screening interval does not appear to present a risk in well-controlled patients with type 2 DM.²⁴
• Women with type 1 or type 2 DM who are planning pregnancy or who are pregnant should have an eye exam prior to pregnancy or early in the first trimester.^4,18,23 They should then be monitored each trimester and at the end of the first postpartum year, depending on the severity of retinopathy.¹⁸

Alternative screening modalities

Seven-field stereoscopic fundus photography is an alternative screening tool that compares favorably to ophthalmoscopy when performed by an experienced ophthalmologist, optometrist, or ophthalmologic technician.²⁵ Nonmydriatic digital stereoscopic retinal imaging has been shown to be a cost-effective method of screening patients for diabetic retinopathy.²⁶ In a study that compared digital imaging with dilated funduscopic examination, investigators reported that, of 311 eyes evaluated, there was agreement between the methods in 86% of cases. Disagreement was mostly related to the greater frequency of finding mild-to-moderate NPDR when using digital imaging.²⁷

Screening in primary care

Programs that use telemedicine-based fundus photography to screen for diabetic retinopathy during primary care visits, followed by remote interpretation by an ophthalmologist, have been shown to increase the rate of retinal screening by offering an option other than direct referral to an ophthalmologist or optometrist.²⁸ However, telemedicine-based retinal photography can be successful as a screening tool for retinopathy only if timely referral to an eye specialist is arranged when indicated by findings.¹⁸

What therapy will your referred patients receive?

Patients found to have signs of diabetic retinopathy should be referred to an ophthalmologist who is knowledgeable and experienced in the management of diabetic retinopathy. Care will be managed according to the severity of the patient’s diabetic retinopathy.

Continue to: Patients with mild-to-moderate NPDR but without macular edema

Patients with mild-to-moderate NPDR but without macular edema. Treatment is generally not recommended. Patients should be reevaluated every 6 to 12 months because they have an increased risk of progression.⁵

Patients with mild-to-moderate NPDR and clinically significant macular edema (CSME). It is important for the eye specialist to assess for edema at the center of the macula because the risk of vision loss and need for treatment is greater when the center is involved. Vascular–endothelial growth factor (VEGF) is an important mediator of neovascularization and macular edema in diabetic retinopathy. For patients with center-­involving CSME, intravitreous injection of an anti-VEGF agent provides significant benefit and is first-line treatment in these cases.^4,29

Patients with DM (especially those with type 2 disease) who have, or are at risk of, atherosclerotic cardiovascular disease should receive statin therapy.

The Early Treatment for Diabetic Retinopathy Study evaluated the efficacy of focal photocoagulation, a painless laser therapy, for CSME and demonstrated that this modality reduces the risk of moderate visual loss; increases the likelihood of improvement in vision; and decreases the frequency of persistent macular edema.³⁰ Focal photocoagulation has been found effective in both non-center-involving CSME and center-involving CSME.⁵

Patients with severe NPDR. The recommendation is to initiate full panretinal photocoagulation prior to progression to proliferative diabetic retinopathy PDR. Researchers noted a 50% reduction in vision loss and vitrectomy when patients with type 2 DM were treated with panretinal photocoagulation early, compared with those in whom treatment was deferred until PDR developed.^4,31 The role of anti-VEGF treatment of severe NPDR is under investigation.⁴

Patients with high-risk and severe PDR. Panretinal photocoagulation is the recommended treatment for patients with high-risk and severe PDR, and usually induces regression of retinal neovascularization. In patients with CSME and high-risk PDR, the combination of anti-VEGF therapy and panretinal photocoagulation should be considered. Vitrectomy should be considered for patients who have failed panretinal photocoagulation or are not amenable to photocoagulation.⁴

CORRESPONDENCE
Bryan Farford, DO, Mayo Clinic, 4500 San Pablo Road, Jacksonville, FL 32224; Farford.Bryan@mayo.edu.

CASE

A 38-year-old woman presents for a routine physical. Other than urgent care visits for 1 episode of influenza and 2 upper respiratory illnesses, she has not seen a physician for a physical in 5 years. She denies any significant medical history. She takes naproxen occasionally for chronic right knee pain. She does not use tobacco or alcohol. Recently, she has started using a meal replacement shake at lunchtime for weight management. She performs aerobic exercise 30 to 40 minutes per day, 5 days per week. Her family history is significant for type 2 diabetes mellitus, arthritis, heart disease, and hyperlipidemia on her mother’s side. She is single, is not currently sexually active, works as a pharmacy technician, and has no children. A high-risk human papillomavirus test was normal 4 years ago.

A review of systems is notable for a 20-pound weight gain over the past year, worsening heartburn over the past 2 weeks, and chronic knee pain, which is greater in the right knee than the left. She denies weakness, fatigue, nausea, diarrhea, constipation, or abdominal pain. Vital signs reveal a blood pressure of 146/88 mm Hg, a heart rate of 63 bpm, a temperature of 98°F (36.7°C), a respiratory rate of 16, a height of 5’7’’ (1.7 m), a weight of 217 lbs (98.4 kg), and a peripheral capillary oxygen saturation (SpO₂) of 99% on room air. The physical exam reveals a body mass index (BMI) of 34, warm dry skin, and coarse brittle hair.

Lab results reveal a thyroid-stimulating hormone (TSH) level of 11.17 mIU/L (reference range, 0.45-4.5 mIU/L) and a free thyroxine (T4) of 0.58 ng/dL (reference range, 0.8-2.8 ng/dL). A basic metabolic panel and hemoglobin A1C level are normal.

What would you recommend?

In the United States, the prevalence of overt hypothyroidism (defined as a TSH level > 4.5 mIU/L and a low free T4) among people ≥ 12 years of age was estimated at 0.3% based on National Health and Nutrition Examination Survey (NHANES) data from 1999-2002.¹ Subclinical hypothyroidism (TSH level > 4.5 mIU/L but < 10 mIU/L and a normal T4 level) is even more common, with an estimated prevalence of 3.4%.¹ Hypothyroidism is more common in females and occurs more frequently in Caucasian Americans and Mexican Americans than in African Americans.¹

The most common etiologies of hypothyroidism include autoimmune thyroiditis (eg, Hashimoto thyroiditis, atrophic autoimmune thyroiditis) and iatrogenic causes (eg, after radioactive iodine ablation or thyroidectomy) (TABLE 1).^2-4

Initiating thyroid hormone replacement

Factors to consider when starting a patient on thyroid hormone replacement include age, weight, symptom severity, TSH level, goal TSH value, adverse effects from thyroid supplements, history of cardiac disease, and, for women of child-bearing age, the desire for pregnancy vs the use of contraceptives. Most adult patients < 50 years with overt hypothyroidism can begin a weight-based dose of levothyroxine: ~1.6 mcg/kg/d (based on ideal body weight).³

Continue to: For adults with cardiac disease...

For adults with cardiac disease, the risk of over-replacement limits initial dosing to 25 to 50 mcg/d for patients < 50 years (12.5-25 mcg/d; ≥ 50 years).³ For adults with subclinical hypothyroidism, it is reasonable to begin therapy at a lower daily dose (eg, 25-75 mcg/d) depending on baseline TSH level, symptoms (the patient may be asymptomatic), and the presence of cardiac disease (TABLE 2^3,4). Consider treatment in patients with subclinical hypothyroidism particularly when patients have a goiter or dyslipidemia and in women contemplating pregnancy in the near future. Elderly patients may require a dose 20% to 25% lower than younger adults because of decreased body mass.³

Levothyroxine is considered first-line therapy for hypothyroidism because of its low cost, dose consistency, low risk of allergic reactions, and potential to cause fewer cardiac adverse effects than triiodothyronine (T3) products such as desiccated thyroid extract.⁵ Although data have not shown an absolute increase in cardiovascular adverse effects, T3 products have a higher T3 vs T4 ratio, giving them a theoretically increased risk.^5,6 Desiccated thyroid extract also has been associated with allergic reactions.⁵

Consider treatment in adults with subclinical hypothyroidism when they have a goiter or dyslipidemia and in women contemplating pregnancy in the near future.

Use of liothyronine alone or in combination with levothyroxine lacks evidence and guideline support.⁴ Furthermore, it is dosed twice daily, which makes it less convenient, and concerns still exist that there may be an increase in cardiovascular adverse effects.^4,6 See TABLE 3⁷ for a summary of available products and their equivalent doses.

Maintaining patients on therapy

The maintenance phase begins once hypothyroidism is diagnosed and treatment is initiated. This phase includes regular monitoring with laboratory studies, office visits, and as-needed adjustments in hormone replacement dosing. The frequency at which all of these occur is variable and based on a number of factors including the patient’s other medical conditions, use of other medications including over-the-counter agents, the patient’s age, weight changes, and pregnancy status.^3,4,8 In general, dosage adjustments of 12.5 to 25 mcg can be made at 6- to 8-week intervals based on repeat TSH measurements, patient symptoms, and comorbidities.³

Instruct patients to take levothyroxine either in the morning at least 30 to 60 minutes before eating or 3 to 4 hours after the last meal of the day.

Once a patient is symptomatically stable and laboratory values have normalized, the recommended frequency of laboratory evaluation and office visits is every 12 months, barring significant changes in any of the factors mentioned above. At each visit, physicians should perform medication (including supplements) reconciliation and discuss any health condition updates. Changes to the therapy plan, including frequency or timing of laboratory tests, may be necessary if patients begin taking medications that alter the absorption or function of levothyroxine (eg, steroids).

Continue to: To maximize absorption...

To maximize absorption, providers should review with patients the optimal way to take thyroid hormones. Levothyroxine is approximately 70% to 80% absorbed under ideal conditions, which means taking it in the morning at least 30 to 60 minutes before eating or 3 to 4 hours after the last meal of the day.^3,9-13 Of note, TSH levels may increase slightly in patients taking proton pump inhibitors, but this does not usually require a dose increase of thyroid hormone.¹¹ Given that some supplements, particularly iron and calcium, can interfere with absorption, it is recommended to maintain a 3- to 4-hour gap between taking those supplements and taking levothyroxine.^12-14 For those patients unable or unwilling to adhere to these recommendations, an increase in levothyroxine dose may be required in order to compensate for the decreased absorption.

Don’t adjust hormone therapy based on clinical presentation alone. While clinical symptoms are important, it is not recommended to adjust hormone therapy based solely on clinical presentation. Common hypothyroid symptoms of dry skin, edema, weight gain, and fatigue may be caused by other medical conditions. While indices including Achilles reflex time and basal metabolic rate have shown some correlation to thyroid dysfunction, there has been limited evidence to show that longitudinal index changes reflect subtle changes in thyroid hormone levels.³

The most recent guidelines from the American Thyroid Association recommend that, “Symptoms should be followed, but considered in the context of serum thyrotropin values, relevant comorbidities, and other potential causes.”³

Special populations/circumstances to keep in mind

Malabsorption conditions. When a higher than expected weight-based dose of levothyroxine is required, physicians should review administration timing, adherence, and comorbid medical conditions that can affect absorption.

Several studies, for example, have demonstrated the impact of Helicobacter pylori gastritis on levothyroxine absorption and subsequent TSH levels.^15-17 In one nonrandomized prospective study, patients with H pylori and hypothyroidism who were previously thought to be unresponsive to levothyroxine therapy had a decrease in average TSH level from 30.5 mIU/L to 4.2 mIU/L after H pylori was eradicated.¹⁵ Autoimmune atrophic gastritis and celiac disease, both of which are more common in those with other autoimmune diseases, are also associated with the need for higher than expected levothyroxine doses.^17,18

Continue to: A history of gastric bypass surgery...

A history of gastric bypass surgery alone is not considered a risk factor for poor absorption of thyroid hormone, given that the majority of levothyroxine absorption occurs in the ileum.^19,20 However, advancing age (> 70 years) and extreme obesity (BMI > 40) are independent risk factors for decreased levothyroxine absorption.^20,21

Women of reproductive age and pregnant women. Overt untreated or undertreated hypothyroidism can be associated with increased risk of maternal and fetal complications including decreased fertility, miscarriage, preterm delivery, lower birth rates, and infant cognitive deficits.^3,22 Therefore, the main focus should be optimization of thyroid hormone levels prior to and during pregnancy.^3,4,8,22 Thyroid hormone replacement needs to be increased during pregnancy in approximately 50% to 85% of women using thyroid replacement prior to pregnancy, but the dose requirements vary based on the underlying etiology of thyroid dysfunction.

One initial option for patients on a stable dose before pregnancy is to increase their daily dose by a half tablet (1.5 × daily dose) immediately after home confirmation of pregnancy, until finer dose adjustments (usually increases of 25%-60% ) can be made by a physician. Experts recommend that a TSH level be obtained every 4 weeks until mid-gestation and then at least once around 30 weeks’ gestation to ensure specific targets are being met with dose adjustments.²² Optimal thyrotropin reference ranges during conception and pregnancy can be found in the literature.²³

Iron and calcium supplements can interfere with the absorption of levothyroxine, so a 3- to 4-hour gap between taking these supplements and levothyroxine is recommended.

Patients who have positive antibodies and normal thyroid function tests. Patients who are screened for thyroid disorders may demonstrate normal thyroid function (ie, euthyroid) with TSH, free T4, and, if checked, free T3, all within normal ranges. Despite these normal lab results, patients may have additional test results that demonstrate positive thyroid autoantibodies including thyroglobulin antibodies and/or thyroid peroxidase antibodies. Thyroid ­autoimmunity itself has been associated with a range of other autoimmune conditions as well as an increased risk of thyroid cancer in those with Hashimoto thyroiditis.²⁴ Two studies showed that prophylactic treatment of euthyroid patients with levothyroxine led to a reduction in antibody levels and a lower TSH level.^25,26 However, no studies have focused on patient-oriented outcomes such as hospitalizations, quality of life, or symptoms. If the patient remains asymptomatic, we recommend no treatment, but that the patient’s TSH levels be monitored every 12 months.²⁷

Elderly patients. Population data have shown that TSH increases normally with age, with a TSH level of 7.5 mIU/L being the ­upper limit of normal for a population of healthy adults > 80 years of age.^28,29 Overall, studies have failed to show any benefit in treating elderly patients with subclinical hypothyroidism unless their TSH level exceeds 10 mIU/L.^6,21 The one exception is elderly patients with heart failure in whom untreated subclinical hypothyroidism has been shown to be associated with higher mortality.³⁰

Continue to: Elderly patients are at higher risk...

Elderly patients are at higher risk for adverse effects of thyroid over-replacement, including atrial fibrillation and osteoporosis. While there have been no randomized trials examining target TSH levels in this population, a reasonable recommendation is a goal TSH level of 4 to 6 mIU/L for elderly patients ≥ 70 years.⁴

CASE

As a result of the patient’s elevated TSH level and symptoms of hypothyroidism, you start levothyroxine 150 mcg/d by mouth, counsel her on potential adverse effects, and schedule a follow-up visit with another TSH check in 6 weeks.

Follow-up laboratory studies 6 weeks later reveal a TSH level of 5.86 mIU/L (reference range, 0.45-4.5 mIU/L) and a free T4 level of 0.74 ng/dL (reference range, 0.8-2.8 ng/dL). Based on those results, you increase the dose of levothyroxine to 175 mcg/d.

Advancing age (> 70 years) and extreme obesity (BMI > 40) are independent risk factors for decreased levothyroxine absorption.

At her follow-up visit 12 weeks after initial presentation, her TSH level is 3.85 mIU/L. She reports feeling better overall with less fatigue, and she has lost 5 pounds since her last visit. You recommend she continue levothyroxine 175 mcg/d after reviewing medication compliance with the patient and ensuring she is indeed taking it in the morning, at least 30 minutes prior to eating. With improved but not resolved symptoms, she agrees to follow-up with repeat TSH laboratory studies in 6 weeks to determine whether further dose adjustments are necessary. Given that she is of reproductive age and her TSH level is suboptimal for pregnancy, you caution her about heightened pregnancy/fetal risks with a suboptimal TSH and recommend that she use reliable contraception.

CORRESPONDENCE
Christopher Bunt, MD, FAAFP, 5 Charleston Center Drive, Suite 263, MSC 192,Charleston, SC 29425; buntc@musc.edu

CASE

A 38-year-old woman presents for a routine physical. Other than urgent care visits for 1 episode of influenza and 2 upper respiratory illnesses, she has not seen a physician for a physical in 5 years. She denies any significant medical history. She takes naproxen occasionally for chronic right knee pain. She does not use tobacco or alcohol. Recently, she has started using a meal replacement shake at lunchtime for weight management. She performs aerobic exercise 30 to 40 minutes per day, 5 days per week. Her family history is significant for type 2 diabetes mellitus, arthritis, heart disease, and hyperlipidemia on her mother’s side. She is single, is not currently sexually active, works as a pharmacy technician, and has no children. A high-risk human papillomavirus test was normal 4 years ago.

A review of systems is notable for a 20-pound weight gain over the past year, worsening heartburn over the past 2 weeks, and chronic knee pain, which is greater in the right knee than the left. She denies weakness, fatigue, nausea, diarrhea, constipation, or abdominal pain. Vital signs reveal a blood pressure of 146/88 mm Hg, a heart rate of 63 bpm, a temperature of 98°F (36.7°C), a respiratory rate of 16, a height of 5’7’’ (1.7 m), a weight of 217 lbs (98.4 kg), and a peripheral capillary oxygen saturation (SpO₂) of 99% on room air. The physical exam reveals a body mass index (BMI) of 34, warm dry skin, and coarse brittle hair.

Lab results reveal a thyroid-stimulating hormone (TSH) level of 11.17 mIU/L (reference range, 0.45-4.5 mIU/L) and a free thyroxine (T4) of 0.58 ng/dL (reference range, 0.8-2.8 ng/dL). A basic metabolic panel and hemoglobin A1C level are normal.

What would you recommend?

In the United States, the prevalence of overt hypothyroidism (defined as a TSH level > 4.5 mIU/L and a low free T4) among people ≥ 12 years of age was estimated at 0.3% based on National Health and Nutrition Examination Survey (NHANES) data from 1999-2002.¹ Subclinical hypothyroidism (TSH level > 4.5 mIU/L but < 10 mIU/L and a normal T4 level) is even more common, with an estimated prevalence of 3.4%.¹ Hypothyroidism is more common in females and occurs more frequently in Caucasian Americans and Mexican Americans than in African Americans.¹

The most common etiologies of hypothyroidism include autoimmune thyroiditis (eg, Hashimoto thyroiditis, atrophic autoimmune thyroiditis) and iatrogenic causes (eg, after radioactive iodine ablation or thyroidectomy) (TABLE 1).^2-4

Initiating thyroid hormone replacement

Factors to consider when starting a patient on thyroid hormone replacement include age, weight, symptom severity, TSH level, goal TSH value, adverse effects from thyroid supplements, history of cardiac disease, and, for women of child-bearing age, the desire for pregnancy vs the use of contraceptives. Most adult patients < 50 years with overt hypothyroidism can begin a weight-based dose of levothyroxine: ~1.6 mcg/kg/d (based on ideal body weight).³

Continue to: For adults with cardiac disease...

For adults with cardiac disease, the risk of over-replacement limits initial dosing to 25 to 50 mcg/d for patients < 50 years (12.5-25 mcg/d; ≥ 50 years).³ For adults with subclinical hypothyroidism, it is reasonable to begin therapy at a lower daily dose (eg, 25-75 mcg/d) depending on baseline TSH level, symptoms (the patient may be asymptomatic), and the presence of cardiac disease (TABLE 2^3,4). Consider treatment in patients with subclinical hypothyroidism particularly when patients have a goiter or dyslipidemia and in women contemplating pregnancy in the near future. Elderly patients may require a dose 20% to 25% lower than younger adults because of decreased body mass.³

Levothyroxine is considered first-line therapy for hypothyroidism because of its low cost, dose consistency, low risk of allergic reactions, and potential to cause fewer cardiac adverse effects than triiodothyronine (T3) products such as desiccated thyroid extract.⁵ Although data have not shown an absolute increase in cardiovascular adverse effects, T3 products have a higher T3 vs T4 ratio, giving them a theoretically increased risk.^5,6 Desiccated thyroid extract also has been associated with allergic reactions.⁵

Consider treatment in adults with subclinical hypothyroidism when they have a goiter or dyslipidemia and in women contemplating pregnancy in the near future.

Use of liothyronine alone or in combination with levothyroxine lacks evidence and guideline support.⁴ Furthermore, it is dosed twice daily, which makes it less convenient, and concerns still exist that there may be an increase in cardiovascular adverse effects.^4,6 See TABLE 3⁷ for a summary of available products and their equivalent doses.

Maintaining patients on therapy

The maintenance phase begins once hypothyroidism is diagnosed and treatment is initiated. This phase includes regular monitoring with laboratory studies, office visits, and as-needed adjustments in hormone replacement dosing. The frequency at which all of these occur is variable and based on a number of factors including the patient’s other medical conditions, use of other medications including over-the-counter agents, the patient’s age, weight changes, and pregnancy status.^3,4,8 In general, dosage adjustments of 12.5 to 25 mcg can be made at 6- to 8-week intervals based on repeat TSH measurements, patient symptoms, and comorbidities.³

Instruct patients to take levothyroxine either in the morning at least 30 to 60 minutes before eating or 3 to 4 hours after the last meal of the day.

Once a patient is symptomatically stable and laboratory values have normalized, the recommended frequency of laboratory evaluation and office visits is every 12 months, barring significant changes in any of the factors mentioned above. At each visit, physicians should perform medication (including supplements) reconciliation and discuss any health condition updates. Changes to the therapy plan, including frequency or timing of laboratory tests, may be necessary if patients begin taking medications that alter the absorption or function of levothyroxine (eg, steroids).

Continue to: To maximize absorption...

To maximize absorption, providers should review with patients the optimal way to take thyroid hormones. Levothyroxine is approximately 70% to 80% absorbed under ideal conditions, which means taking it in the morning at least 30 to 60 minutes before eating or 3 to 4 hours after the last meal of the day.^3,9-13 Of note, TSH levels may increase slightly in patients taking proton pump inhibitors, but this does not usually require a dose increase of thyroid hormone.¹¹ Given that some supplements, particularly iron and calcium, can interfere with absorption, it is recommended to maintain a 3- to 4-hour gap between taking those supplements and taking levothyroxine.^12-14 For those patients unable or unwilling to adhere to these recommendations, an increase in levothyroxine dose may be required in order to compensate for the decreased absorption.

Don’t adjust hormone therapy based on clinical presentation alone. While clinical symptoms are important, it is not recommended to adjust hormone therapy based solely on clinical presentation. Common hypothyroid symptoms of dry skin, edema, weight gain, and fatigue may be caused by other medical conditions. While indices including Achilles reflex time and basal metabolic rate have shown some correlation to thyroid dysfunction, there has been limited evidence to show that longitudinal index changes reflect subtle changes in thyroid hormone levels.³

The most recent guidelines from the American Thyroid Association recommend that, “Symptoms should be followed, but considered in the context of serum thyrotropin values, relevant comorbidities, and other potential causes.”³

Special populations/circumstances to keep in mind

Malabsorption conditions. When a higher than expected weight-based dose of levothyroxine is required, physicians should review administration timing, adherence, and comorbid medical conditions that can affect absorption.

Several studies, for example, have demonstrated the impact of Helicobacter pylori gastritis on levothyroxine absorption and subsequent TSH levels.^15-17 In one nonrandomized prospective study, patients with H pylori and hypothyroidism who were previously thought to be unresponsive to levothyroxine therapy had a decrease in average TSH level from 30.5 mIU/L to 4.2 mIU/L after H pylori was eradicated.¹⁵ Autoimmune atrophic gastritis and celiac disease, both of which are more common in those with other autoimmune diseases, are also associated with the need for higher than expected levothyroxine doses.^17,18

Continue to: A history of gastric bypass surgery...

A history of gastric bypass surgery alone is not considered a risk factor for poor absorption of thyroid hormone, given that the majority of levothyroxine absorption occurs in the ileum.^19,20 However, advancing age (> 70 years) and extreme obesity (BMI > 40) are independent risk factors for decreased levothyroxine absorption.^20,21

Women of reproductive age and pregnant women. Overt untreated or undertreated hypothyroidism can be associated with increased risk of maternal and fetal complications including decreased fertility, miscarriage, preterm delivery, lower birth rates, and infant cognitive deficits.^3,22 Therefore, the main focus should be optimization of thyroid hormone levels prior to and during pregnancy.^3,4,8,22 Thyroid hormone replacement needs to be increased during pregnancy in approximately 50% to 85% of women using thyroid replacement prior to pregnancy, but the dose requirements vary based on the underlying etiology of thyroid dysfunction.

One initial option for patients on a stable dose before pregnancy is to increase their daily dose by a half tablet (1.5 × daily dose) immediately after home confirmation of pregnancy, until finer dose adjustments (usually increases of 25%-60% ) can be made by a physician. Experts recommend that a TSH level be obtained every 4 weeks until mid-gestation and then at least once around 30 weeks’ gestation to ensure specific targets are being met with dose adjustments.²² Optimal thyrotropin reference ranges during conception and pregnancy can be found in the literature.²³

Iron and calcium supplements can interfere with the absorption of levothyroxine, so a 3- to 4-hour gap between taking these supplements and levothyroxine is recommended.

Patients who have positive antibodies and normal thyroid function tests. Patients who are screened for thyroid disorders may demonstrate normal thyroid function (ie, euthyroid) with TSH, free T4, and, if checked, free T3, all within normal ranges. Despite these normal lab results, patients may have additional test results that demonstrate positive thyroid autoantibodies including thyroglobulin antibodies and/or thyroid peroxidase antibodies. Thyroid ­autoimmunity itself has been associated with a range of other autoimmune conditions as well as an increased risk of thyroid cancer in those with Hashimoto thyroiditis.²⁴ Two studies showed that prophylactic treatment of euthyroid patients with levothyroxine led to a reduction in antibody levels and a lower TSH level.^25,26 However, no studies have focused on patient-oriented outcomes such as hospitalizations, quality of life, or symptoms. If the patient remains asymptomatic, we recommend no treatment, but that the patient’s TSH levels be monitored every 12 months.²⁷

Elderly patients. Population data have shown that TSH increases normally with age, with a TSH level of 7.5 mIU/L being the ­upper limit of normal for a population of healthy adults > 80 years of age.^28,29 Overall, studies have failed to show any benefit in treating elderly patients with subclinical hypothyroidism unless their TSH level exceeds 10 mIU/L.^6,21 The one exception is elderly patients with heart failure in whom untreated subclinical hypothyroidism has been shown to be associated with higher mortality.³⁰

Continue to: Elderly patients are at higher risk...

Elderly patients are at higher risk for adverse effects of thyroid over-replacement, including atrial fibrillation and osteoporosis. While there have been no randomized trials examining target TSH levels in this population, a reasonable recommendation is a goal TSH level of 4 to 6 mIU/L for elderly patients ≥ 70 years.⁴

CASE

As a result of the patient’s elevated TSH level and symptoms of hypothyroidism, you start levothyroxine 150 mcg/d by mouth, counsel her on potential adverse effects, and schedule a follow-up visit with another TSH check in 6 weeks.

Follow-up laboratory studies 6 weeks later reveal a TSH level of 5.86 mIU/L (reference range, 0.45-4.5 mIU/L) and a free T4 level of 0.74 ng/dL (reference range, 0.8-2.8 ng/dL). Based on those results, you increase the dose of levothyroxine to 175 mcg/d.

Advancing age (> 70 years) and extreme obesity (BMI > 40) are independent risk factors for decreased levothyroxine absorption.

At her follow-up visit 12 weeks after initial presentation, her TSH level is 3.85 mIU/L. She reports feeling better overall with less fatigue, and she has lost 5 pounds since her last visit. You recommend she continue levothyroxine 175 mcg/d after reviewing medication compliance with the patient and ensuring she is indeed taking it in the morning, at least 30 minutes prior to eating. With improved but not resolved symptoms, she agrees to follow-up with repeat TSH laboratory studies in 6 weeks to determine whether further dose adjustments are necessary. Given that she is of reproductive age and her TSH level is suboptimal for pregnancy, you caution her about heightened pregnancy/fetal risks with a suboptimal TSH and recommend that she use reliable contraception.

CORRESPONDENCE
Christopher Bunt, MD, FAAFP, 5 Charleston Center Drive, Suite 263, MSC 192,Charleston, SC 29425; buntc@musc.edu

CASE

A 38-year-old woman presents for a routine physical. Other than urgent care visits for 1 episode of influenza and 2 upper respiratory illnesses, she has not seen a physician for a physical in 5 years. She denies any significant medical history. She takes naproxen occasionally for chronic right knee pain. She does not use tobacco or alcohol. Recently, she has started using a meal replacement shake at lunchtime for weight management. She performs aerobic exercise 30 to 40 minutes per day, 5 days per week. Her family history is significant for type 2 diabetes mellitus, arthritis, heart disease, and hyperlipidemia on her mother’s side. She is single, is not currently sexually active, works as a pharmacy technician, and has no children. A high-risk human papillomavirus test was normal 4 years ago.

A review of systems is notable for a 20-pound weight gain over the past year, worsening heartburn over the past 2 weeks, and chronic knee pain, which is greater in the right knee than the left. She denies weakness, fatigue, nausea, diarrhea, constipation, or abdominal pain. Vital signs reveal a blood pressure of 146/88 mm Hg, a heart rate of 63 bpm, a temperature of 98°F (36.7°C), a respiratory rate of 16, a height of 5’7’’ (1.7 m), a weight of 217 lbs (98.4 kg), and a peripheral capillary oxygen saturation (SpO₂) of 99% on room air. The physical exam reveals a body mass index (BMI) of 34, warm dry skin, and coarse brittle hair.

Lab results reveal a thyroid-stimulating hormone (TSH) level of 11.17 mIU/L (reference range, 0.45-4.5 mIU/L) and a free thyroxine (T4) of 0.58 ng/dL (reference range, 0.8-2.8 ng/dL). A basic metabolic panel and hemoglobin A1C level are normal.

What would you recommend?

In the United States, the prevalence of overt hypothyroidism (defined as a TSH level > 4.5 mIU/L and a low free T4) among people ≥ 12 years of age was estimated at 0.3% based on National Health and Nutrition Examination Survey (NHANES) data from 1999-2002.¹ Subclinical hypothyroidism (TSH level > 4.5 mIU/L but < 10 mIU/L and a normal T4 level) is even more common, with an estimated prevalence of 3.4%.¹ Hypothyroidism is more common in females and occurs more frequently in Caucasian Americans and Mexican Americans than in African Americans.¹

The most common etiologies of hypothyroidism include autoimmune thyroiditis (eg, Hashimoto thyroiditis, atrophic autoimmune thyroiditis) and iatrogenic causes (eg, after radioactive iodine ablation or thyroidectomy) (TABLE 1).^2-4

Initiating thyroid hormone replacement

Factors to consider when starting a patient on thyroid hormone replacement include age, weight, symptom severity, TSH level, goal TSH value, adverse effects from thyroid supplements, history of cardiac disease, and, for women of child-bearing age, the desire for pregnancy vs the use of contraceptives. Most adult patients < 50 years with overt hypothyroidism can begin a weight-based dose of levothyroxine: ~1.6 mcg/kg/d (based on ideal body weight).³

Continue to: For adults with cardiac disease...

For adults with cardiac disease, the risk of over-replacement limits initial dosing to 25 to 50 mcg/d for patients < 50 years (12.5-25 mcg/d; ≥ 50 years).³ For adults with subclinical hypothyroidism, it is reasonable to begin therapy at a lower daily dose (eg, 25-75 mcg/d) depending on baseline TSH level, symptoms (the patient may be asymptomatic), and the presence of cardiac disease (TABLE 2^3,4). Consider treatment in patients with subclinical hypothyroidism particularly when patients have a goiter or dyslipidemia and in women contemplating pregnancy in the near future. Elderly patients may require a dose 20% to 25% lower than younger adults because of decreased body mass.³

Levothyroxine is considered first-line therapy for hypothyroidism because of its low cost, dose consistency, low risk of allergic reactions, and potential to cause fewer cardiac adverse effects than triiodothyronine (T3) products such as desiccated thyroid extract.⁵ Although data have not shown an absolute increase in cardiovascular adverse effects, T3 products have a higher T3 vs T4 ratio, giving them a theoretically increased risk.^5,6 Desiccated thyroid extract also has been associated with allergic reactions.⁵

Consider treatment in adults with subclinical hypothyroidism when they have a goiter or dyslipidemia and in women contemplating pregnancy in the near future.

Use of liothyronine alone or in combination with levothyroxine lacks evidence and guideline support.⁴ Furthermore, it is dosed twice daily, which makes it less convenient, and concerns still exist that there may be an increase in cardiovascular adverse effects.^4,6 See TABLE 3⁷ for a summary of available products and their equivalent doses.

Maintaining patients on therapy

The maintenance phase begins once hypothyroidism is diagnosed and treatment is initiated. This phase includes regular monitoring with laboratory studies, office visits, and as-needed adjustments in hormone replacement dosing. The frequency at which all of these occur is variable and based on a number of factors including the patient’s other medical conditions, use of other medications including over-the-counter agents, the patient’s age, weight changes, and pregnancy status.^3,4,8 In general, dosage adjustments of 12.5 to 25 mcg can be made at 6- to 8-week intervals based on repeat TSH measurements, patient symptoms, and comorbidities.³

Instruct patients to take levothyroxine either in the morning at least 30 to 60 minutes before eating or 3 to 4 hours after the last meal of the day.

Once a patient is symptomatically stable and laboratory values have normalized, the recommended frequency of laboratory evaluation and office visits is every 12 months, barring significant changes in any of the factors mentioned above. At each visit, physicians should perform medication (including supplements) reconciliation and discuss any health condition updates. Changes to the therapy plan, including frequency or timing of laboratory tests, may be necessary if patients begin taking medications that alter the absorption or function of levothyroxine (eg, steroids).

Continue to: To maximize absorption...

To maximize absorption, providers should review with patients the optimal way to take thyroid hormones. Levothyroxine is approximately 70% to 80% absorbed under ideal conditions, which means taking it in the morning at least 30 to 60 minutes before eating or 3 to 4 hours after the last meal of the day.^3,9-13 Of note, TSH levels may increase slightly in patients taking proton pump inhibitors, but this does not usually require a dose increase of thyroid hormone.¹¹ Given that some supplements, particularly iron and calcium, can interfere with absorption, it is recommended to maintain a 3- to 4-hour gap between taking those supplements and taking levothyroxine.^12-14 For those patients unable or unwilling to adhere to these recommendations, an increase in levothyroxine dose may be required in order to compensate for the decreased absorption.

Don’t adjust hormone therapy based on clinical presentation alone. While clinical symptoms are important, it is not recommended to adjust hormone therapy based solely on clinical presentation. Common hypothyroid symptoms of dry skin, edema, weight gain, and fatigue may be caused by other medical conditions. While indices including Achilles reflex time and basal metabolic rate have shown some correlation to thyroid dysfunction, there has been limited evidence to show that longitudinal index changes reflect subtle changes in thyroid hormone levels.³

The most recent guidelines from the American Thyroid Association recommend that, “Symptoms should be followed, but considered in the context of serum thyrotropin values, relevant comorbidities, and other potential causes.”³

Special populations/circumstances to keep in mind

Malabsorption conditions. When a higher than expected weight-based dose of levothyroxine is required, physicians should review administration timing, adherence, and comorbid medical conditions that can affect absorption.

Several studies, for example, have demonstrated the impact of Helicobacter pylori gastritis on levothyroxine absorption and subsequent TSH levels.^15-17 In one nonrandomized prospective study, patients with H pylori and hypothyroidism who were previously thought to be unresponsive to levothyroxine therapy had a decrease in average TSH level from 30.5 mIU/L to 4.2 mIU/L after H pylori was eradicated.¹⁵ Autoimmune atrophic gastritis and celiac disease, both of which are more common in those with other autoimmune diseases, are also associated with the need for higher than expected levothyroxine doses.^17,18

Continue to: A history of gastric bypass surgery...

A history of gastric bypass surgery alone is not considered a risk factor for poor absorption of thyroid hormone, given that the majority of levothyroxine absorption occurs in the ileum.^19,20 However, advancing age (> 70 years) and extreme obesity (BMI > 40) are independent risk factors for decreased levothyroxine absorption.^20,21

Women of reproductive age and pregnant women. Overt untreated or undertreated hypothyroidism can be associated with increased risk of maternal and fetal complications including decreased fertility, miscarriage, preterm delivery, lower birth rates, and infant cognitive deficits.^3,22 Therefore, the main focus should be optimization of thyroid hormone levels prior to and during pregnancy.^3,4,8,22 Thyroid hormone replacement needs to be increased during pregnancy in approximately 50% to 85% of women using thyroid replacement prior to pregnancy, but the dose requirements vary based on the underlying etiology of thyroid dysfunction.

One initial option for patients on a stable dose before pregnancy is to increase their daily dose by a half tablet (1.5 × daily dose) immediately after home confirmation of pregnancy, until finer dose adjustments (usually increases of 25%-60% ) can be made by a physician. Experts recommend that a TSH level be obtained every 4 weeks until mid-gestation and then at least once around 30 weeks’ gestation to ensure specific targets are being met with dose adjustments.²² Optimal thyrotropin reference ranges during conception and pregnancy can be found in the literature.²³

Iron and calcium supplements can interfere with the absorption of levothyroxine, so a 3- to 4-hour gap between taking these supplements and levothyroxine is recommended.

Patients who have positive antibodies and normal thyroid function tests. Patients who are screened for thyroid disorders may demonstrate normal thyroid function (ie, euthyroid) with TSH, free T4, and, if checked, free T3, all within normal ranges. Despite these normal lab results, patients may have additional test results that demonstrate positive thyroid autoantibodies including thyroglobulin antibodies and/or thyroid peroxidase antibodies. Thyroid ­autoimmunity itself has been associated with a range of other autoimmune conditions as well as an increased risk of thyroid cancer in those with Hashimoto thyroiditis.²⁴ Two studies showed that prophylactic treatment of euthyroid patients with levothyroxine led to a reduction in antibody levels and a lower TSH level.^25,26 However, no studies have focused on patient-oriented outcomes such as hospitalizations, quality of life, or symptoms. If the patient remains asymptomatic, we recommend no treatment, but that the patient’s TSH levels be monitored every 12 months.²⁷

Elderly patients. Population data have shown that TSH increases normally with age, with a TSH level of 7.5 mIU/L being the ­upper limit of normal for a population of healthy adults > 80 years of age.^28,29 Overall, studies have failed to show any benefit in treating elderly patients with subclinical hypothyroidism unless their TSH level exceeds 10 mIU/L.^6,21 The one exception is elderly patients with heart failure in whom untreated subclinical hypothyroidism has been shown to be associated with higher mortality.³⁰

Continue to: Elderly patients are at higher risk...

Elderly patients are at higher risk for adverse effects of thyroid over-replacement, including atrial fibrillation and osteoporosis. While there have been no randomized trials examining target TSH levels in this population, a reasonable recommendation is a goal TSH level of 4 to 6 mIU/L for elderly patients ≥ 70 years.⁴

CASE

As a result of the patient’s elevated TSH level and symptoms of hypothyroidism, you start levothyroxine 150 mcg/d by mouth, counsel her on potential adverse effects, and schedule a follow-up visit with another TSH check in 6 weeks.

Follow-up laboratory studies 6 weeks later reveal a TSH level of 5.86 mIU/L (reference range, 0.45-4.5 mIU/L) and a free T4 level of 0.74 ng/dL (reference range, 0.8-2.8 ng/dL). Based on those results, you increase the dose of levothyroxine to 175 mcg/d.

Advancing age (> 70 years) and extreme obesity (BMI > 40) are independent risk factors for decreased levothyroxine absorption.

At her follow-up visit 12 weeks after initial presentation, her TSH level is 3.85 mIU/L. She reports feeling better overall with less fatigue, and she has lost 5 pounds since her last visit. You recommend she continue levothyroxine 175 mcg/d after reviewing medication compliance with the patient and ensuring she is indeed taking it in the morning, at least 30 minutes prior to eating. With improved but not resolved symptoms, she agrees to follow-up with repeat TSH laboratory studies in 6 weeks to determine whether further dose adjustments are necessary. Given that she is of reproductive age and her TSH level is suboptimal for pregnancy, you caution her about heightened pregnancy/fetal risks with a suboptimal TSH and recommend that she use reliable contraception.

CORRESPONDENCE
Christopher Bunt, MD, FAAFP, 5 Charleston Center Drive, Suite 263, MSC 192,Charleston, SC 29425; buntc@musc.edu

Blood pressure (BP) measurement is an essential component of the physical examination. The information gleaned through this simple but vitally important assessment provides a basis for critical decisions about diagnosis, prognosis, and therapy in a variety of health care settings. In the emergency department, it helps guide resuscitation efforts; in the intensive care unit, it helps to identify the deteriorating patient and guide vasopressor drug titration; in the ambulatory office setting, it helps to identify hypertension and the need for antihypertensive therapy.

In the office setting, inaccurate BP measurement can have profound effects. An overestimation by only 5 mm Hg would result in an erroneous diagnosis and unnecessary treatment of hypertension for about 27 million patients—entailing medication costs, potential adverse effects, and psychologic issues associated with this diagnosis. Conversely, underestimation by 5 mm Hg would miss about 21 million patients who actually have hypertension.¹

Why accurate BP measurement matters so much

About 75 million adults in the United States have high BP,² which costs the nation $46 billion annually in health care services, antihypertensive medications, and missed days of work.³ Among US adults ages 20 or older, the age-­adjusted prevalence of hypertension is estimated to be 34%, equivalent to 85.7 million adults.⁴

Defining hypertension. For the general population, the Eighth Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure ­(JNC-8) defines hypertension as a BP of 140/90 mm Hg or higher in adults younger than 60 and a BP of 150/90 mm Hg or higher in adults ages 60 or older. For patients with comorbid hypertension and diabetes, JNC-8 recommends pharmacologic treatment when BP is 140/90 mm Hg or higher, regardless of age.⁵

Accurate measurement of BP provides the rational basis for the management of hypertension, which in turn may decrease the risk for stroke, congestive heart failure, and other cardiovascular diseases. Several investigators^6-8 have observed that differences in interarm systolic BP are associated with an increased risk for peripheral vascular disease, stroke, and other cardiovascular problems.

Multiple factors impact accuracy; some might surprise you

A number of factors may influence the accuracy of BP measurement in the office; these are generally classified as related to the patient, the observer, the technique or procedure, or the equipment used. A recent systematic review by Kallioinen et al⁹ empirically evaluated 29 potential sources of inaccuracy in the measurement of adult resting BP. Among them were

Patient-related: Recent meal or alcohol intake; recent caffeine or nicotine use; full bladder distention; cold exposure; white-coat effect. Given the simplicity of assessing for these influences, it is worthwhile for office staff to ask patients, prior to the recommended 3 to 5 minutes of rest before BP measurement, if they were rushing to make their appointment, need to void their bladder, or have consumed food or drink or used tobacco within the past 30 minutes.

Continue to: Observer-related...

Observer-related: Hearing deficit; terminal digit bias (ie, preference for rounding BP reading to a specific end digit, eg, 0); measurement of diastolic BP at Korotkoff phase IV rather than phase V.

Procedure-related: Patient’s body position (eg, standing vs supine; legs crossed at knee; unsupported back or arm; arm lower than heart level); incorrect size or placement of cuff; talking during measurement (the content of conversation may influence results); and reliance on a single BP measurement.

Equipment-related: Device model bias; device calibration error.

The USPSTF, the UK’s National Institute for Health and Clinical Excellence, and other groups endorse ambulatory BP monitoring as the optimal method for BP measurement.

As reported by Kallioinen et al⁹, the magnitude of these potential errors ranges from small to large in both the positive and negative direction for both systolic and diastolic BP, and several sources of error are potentially bidirectional. For example, talking during BP measurement may result in an increase in systolic BP of 4 to 19 mm Hg and in diastolic BP of 5 to 14.3 mm Hg; measurement of diastolic BP at Korotkoff phase IV rather than phase V significantly increases diastolic BP by 12.5 mm Hg; and recent alcohol intake can affect systolic BP by –23.6 to +24 mm Hg and diastolic BP by –14 to +16 mm Hg. Overall, the researchers found significant directional effects for 27 of the 29 potential sources of error, ranging from a mean –24 mm Hg to +33 mm Hg error for estimating systolic BP and a mean –14 mm Hg to +23 mm Hg for estimating diastolic BP.⁹

Careful adherence to guidelines ensures accurate BP measurement

Adequate training and standardized procedures can target and mitigate many of the identified sources of error; accordingly, all clinical staff responsible for obtaining a patient’s BP measurement should be trained not only in the correct method for accurate measurement but also in the identification of factors that may introduce errors.

Continue to: The American Heart Association...

The American Heart Association (AHA) recommends that BP be measured in both arms at the initial evaluation, with the higher measurement used for monitoring BP. The AHA also recommends obtaining at least 2 readings at least 1 minute apart and averaging them as the patient’s BP.¹⁰ Other research recommends using a fully automated sphygmomanometer to take multiple readings with the patient resting quietly alone in either the exam room or the waiting room¹¹ as an effective and efficient method for accurate BP averaging.

The 2 principal noninvasive methods of BP measurement are the manual auscultatory technique and the oscillatory technique. Because of its simplicity and relative degree of accuracy (when correctly performed), the auscultatory measurement remains common in everyday medical practice. Remarkably, it is one of only a few techniques for clinical examination of patients that has remained relatively unchanged since it was introduced by the Russian physician and scientist Nikolai Sergeevich Korotkoff in 1905.¹² However, accurate performance of the auscultatory method requires adequate training and experience.

In contrast, automated oscillometric BP measurement is easily performed and requires minimal training. However, it is important to note that any condition altering oscillation amplitude or regularity (eg, arterial wall stiffness or cardiac arrhythmia) will produce erroneous results, and the reading must be confirmed by auscultatory measurement.^{13, 14}

Auscultatory methods of BP ­measurement

The mainstay of clinical BP measurement has been auscultatory methods to detect the Korotkoff sounds, using a stethoscope and either mercury, aneroid, or “hybrid” sphygmomanometers. Traditionally, the mercury device was the “gold standard,” but the widespread ban of mercury in health care settings has now all but eliminated its use.

Aneroid gauge sphygmomanometers have a metallic spring and a metal membrane that flexes elastically to translate pressure signals from the cuff and operate a needle in the gauge. Owing to their complexity, these devices require regular recalibration, since inaccurate results may occur anytime the needle does not rest on 0 before use.

Contine to: The newer hybrid sphygmomanometers...

The newer hybrid sphygmomanometers have an electronic transducer in place of a mercury column; BP measurement is performed in the same fashion as with a mercury device, using a stethoscope and auscultation for the Korotkoff sounds.

Variations in technique for BP measurement can result in significantly different readings. In 2005, the AHA published recommendations for BP monitoring to increase the accuracy of in-clinic measurements.¹⁰ Recommendations for accurate BP measurement include:

Patient preparation. The patient should be seated in a chair with his or her back supported, legs uncrossed, and feet flat on the floor. The patient’s bare arm should be supported such that the midpoint of the upper arm is at heart level. An appropriately sized cuff (ie, bladder encircles 80% of the arm for an adult or 100% of the arm for a child younger than 13 years) should be secured around the bare upper arm and the bladder centered over the brachial artery, with the lower edge of the cuff about 2 cm above the antecubital fossa.¹⁰

Technique. The cuff is inflated while palpating the radial artery to the approximate systolic pressure (ie, the point at which the radial pulse is no longer palpated). The bell of the stethoscope is placed just proximal and medial to the antecubital fossa and the cuff is inflated another 20 to 30 mm Hg above the point at which the radial pulse is no longer felt. The cuff is deflated at a rate of about 2 mm Hg per second.¹⁰

BP recording. The systolic BP is recorded at the appearance of the Korotkoff sounds (phase I) for an auscultatory measurement. The diastolic BP is recorded at the disappearance of the Korotkoff sounds (phase V) in adults and at the muffling of sounds (phase IV) in children for an auscultatory measurement.¹⁰

Continue to: Oscillometric methods of BP measurement

Oscillometric methods of BP measurement

The auscultatory methods of BP measurement are gradually being replaced by oscillometric techniques that are better suited to automated methods of measurement. When oscillations of pressure in the gradually deflating bladder cuff are sensed and recorded, the point of maximal oscillation corresponds to the mean intra-arterial pressure.¹⁵ The oscillations sensed are vibrations in the arterial wall that are detected and transduced to an electric signal, producing a digital readout, and correspond approximately to the systolic pressure and continue below the diastolic pressure. The actual systolic and diastolic pressures are indirectly estimated according to a proprietary, empirically derived algorithm that differs from 1 manufacturer to another.

In older patients or those with diabetes who have reduced arterial wall elasticity, oscillometric BP measurements overestimate systolic pressure and underestimate diastolic pressure.

Validated oscillometric techniques have been successfully used in ambulatory BP monitors, which record pressure at regular intervals (typically 20 to 30 minutes) over a 24-hour period while patients perform normal daily activities, including sleep. The US Preventive Services Task Force¹⁶, the UK’s National Institute for Health and Clinical Excellence¹⁷, the European Society of Hypertension¹⁸, and the Canadian Hypertension Education Program¹⁹ collectively endorse ambulatory BP monitoring as the optimal method for BP measurement.

The oscillometric method has also been used for automated office BP measurement, which averages multiple BP readings recorded with a fully automated device while the patient rests alone in a quiet room in clinic. Compared with conventional auscultatory office BP measurement, this method has been promoted to provide a more standardized BP measurement by reducing observer error and the “white coat” effect.^20-22

There are some limitations to oscillometric methods. The amplitude of oscillations is influenced by factors other than BP, notably, arterial wall stiffness. Therefore, in older patients¹³ or those with diabetes¹⁴ who have reduced arterial wall elasticity, oscillometric BP measurements overestimate systolic pressure and underestimate diastolic pressure. In contrast, acutely ill patients, particularly those with hypovolemia and more compliant arterial walls, may have significant underestimation of BP by oscillometric techniques.²³ In patients with peripheral arterial disease, calcified leg vessels can affect the diagnostic accuracy of oscillometric measurement of the ankle-brachial index (ABI).²⁴ A meta-analysis reported that in patients with atrial fibrillation, oscillometric measurement accurately assesses systolic BP but not diastolic BP, and therefore it may be inappropriate for office measurement of BP in these patients.²⁵ Other studies have reported that atrial fibrillation does not significantly affect the accuracy of oscillometric BP measurement if 3 repeated measurements are performed.^26,27

If oscillometric BP measurement is performed in patients with atrial fibrillation, at least 3 repeated measurements should be done to improve accuracy.

Moreover, the algorithms used in these devices are proprietary trade secrets that can be modified by the manufacturer at any time without notice. Therefore, different devices—and even different models from the same manufacturer—may function differently. Only devices calibrated using a validated protocol should be used.^10,28 There are currently 4 unique protocols for validation of BP devices, although an international collaborative group recently published recommendations for a universal protocol for validation of BP measurement devices.²⁹

Continue to: The takeaway

Accurate office BP measurement is essential for patient evaluation and provides the basis for critical decisions about diagnosis, prognosis, and treatment of hypertensive disease. It is imperative to control for factors that may introduce error in BP determination by using a standard protocol and calibrated BP measurement equipment.

Both manual auscultatory and oscillometric methods of measurement are appropriate for office assessment, but oscillometric evaluation is inappropriate for patients with severe atherosclerotic disease, peripheral arterial disease (for ABI), or small arm circumference. If oscillometric BP measurement is performed in patients with atrial fibrillation, at least 3 repeated measurements should be done to improve accuracy. Automated oscillometric BP assessment that records multiple measurements in the quietly resting patient has been promoted to provide a more standardized BP measurement by reducing observer error and the “white coat” effect. Ambulatory oscillometric BP monitoring has been widely endorsed as the optimal method for BP measurement.

CORRESPONDENCE
Darrell R. Over, MD, MSc, FAAFP, 1601 West 40th Street, Pine Bluff, AR 71603; OverDarrellR@uams.edu

Blood pressure (BP) measurement is an essential component of the physical examination. The information gleaned through this simple but vitally important assessment provides a basis for critical decisions about diagnosis, prognosis, and therapy in a variety of health care settings. In the emergency department, it helps guide resuscitation efforts; in the intensive care unit, it helps to identify the deteriorating patient and guide vasopressor drug titration; in the ambulatory office setting, it helps to identify hypertension and the need for antihypertensive therapy.

In the office setting, inaccurate BP measurement can have profound effects. An overestimation by only 5 mm Hg would result in an erroneous diagnosis and unnecessary treatment of hypertension for about 27 million patients—entailing medication costs, potential adverse effects, and psychologic issues associated with this diagnosis. Conversely, underestimation by 5 mm Hg would miss about 21 million patients who actually have hypertension.¹

Why accurate BP measurement matters so much

About 75 million adults in the United States have high BP,² which costs the nation $46 billion annually in health care services, antihypertensive medications, and missed days of work.³ Among US adults ages 20 or older, the age-­adjusted prevalence of hypertension is estimated to be 34%, equivalent to 85.7 million adults.⁴

Defining hypertension. For the general population, the Eighth Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure ­(JNC-8) defines hypertension as a BP of 140/90 mm Hg or higher in adults younger than 60 and a BP of 150/90 mm Hg or higher in adults ages 60 or older. For patients with comorbid hypertension and diabetes, JNC-8 recommends pharmacologic treatment when BP is 140/90 mm Hg or higher, regardless of age.⁵

Accurate measurement of BP provides the rational basis for the management of hypertension, which in turn may decrease the risk for stroke, congestive heart failure, and other cardiovascular diseases. Several investigators^6-8 have observed that differences in interarm systolic BP are associated with an increased risk for peripheral vascular disease, stroke, and other cardiovascular problems.

Multiple factors impact accuracy; some might surprise you

A number of factors may influence the accuracy of BP measurement in the office; these are generally classified as related to the patient, the observer, the technique or procedure, or the equipment used. A recent systematic review by Kallioinen et al⁹ empirically evaluated 29 potential sources of inaccuracy in the measurement of adult resting BP. Among them were