The Journal of Family Practice

SSRIs raise bleeding risk; concurrent NSAIDs raise it more

A 2014 systematic review and meta-analysis of 19 case-control and cohort studies with a total of 446,949 patients investigated the risk of UGI bleeding in patients using SSRIs and NSAIDs.¹ The studies, which included both inpatients and outpatients, were done in Europe and North America. Patients were at least 16 years old, but pooled demographics were not reported. Investigators compared SSRI use with or without concurrent NSAID use to placebo or no treatment.

SSRI use was associated with an increased risk of UGI bleeding in 15 case-control studies (393,268 patients; odds ratio [OR]=1.7; 95% confidence interval [CI], 1.4-1.9) and 4 cohort studies (53,681 patients; OR=1.7; 95% CI, 1.1-2.5). The simultaneous use of SSRIs and NSAIDs compared to nonuse of both medications was associated with a larger increase in bleeding risk (10 case-control studies, 223,336 patients; OR=4.3; 95% CI, 2.8-6.4).

The meta-analysis is limited by statistically significant heterogeneity in all of the pooled results and high risk of bias in 9 of the case-control studies and all of the cohort studies. There was no evidence of publication bias, however.

Bleeding risk rises 7 to 28 days after SSRI exposure

A 2014 case-crossover study of 5377 inpatients in Taiwan with a psychiatric diagnosis evaluated the risk of UGI bleeding within the first 28 days after SSRI exposure (SSRI-mediated inhibition of platelets occurs within the first 7 to 14 days).² The average age of the patients was 58 years and 75% of the study population was male. Each patient served as his or her own control.

ORs were calculated to compare patients who were exposed to SSRIs only during 7-, 14-, and 21-day windows immediately before a UGI bleed to controls exposed to SSRIs only during the control periods before the 7-, 14-, and 21-day windows. The ORs were adjusted through multivariate analysis to account for 7 potential confounding factors.

SSRI use was associated with an increased risk of UGI bleeding in 7-, 14-, and 21-day windows before the index event (TABLE²). An increased bleeding risk in the 14 days after SSRI initiation was observed in men (OR=2.4; 95% CI, 1.8-3.4) but not women (OR=1.0; 95% CI, 0.6-1.6). Increased bleeding risk in the 14 days after SSRI initiation was also observed in patients younger than 55 years (OR=2.1; 95% CI, 1.5-3.1), patients with a history of upper GI disease (OR=3.1; 95% CI, 1.7-6.0), and patients with no previous exposure to SSRIs (OR=2.6; 95% CI, 1.6-4.2).

This study didn’t account for SSRI indication as a potential confounder, and the study’s inclusion of inpatients, whose illnesses are typically more severe, may limit generalizability.

SSRIs raise bleeding risk; concurrent NSAIDs raise it more

A 2014 systematic review and meta-analysis of 19 case-control and cohort studies with a total of 446,949 patients investigated the risk of UGI bleeding in patients using SSRIs and NSAIDs.¹ The studies, which included both inpatients and outpatients, were done in Europe and North America. Patients were at least 16 years old, but pooled demographics were not reported. Investigators compared SSRI use with or without concurrent NSAID use to placebo or no treatment.

SSRI use was associated with an increased risk of UGI bleeding in 15 case-control studies (393,268 patients; odds ratio [OR]=1.7; 95% confidence interval [CI], 1.4-1.9) and 4 cohort studies (53,681 patients; OR=1.7; 95% CI, 1.1-2.5). The simultaneous use of SSRIs and NSAIDs compared to nonuse of both medications was associated with a larger increase in bleeding risk (10 case-control studies, 223,336 patients; OR=4.3; 95% CI, 2.8-6.4).

The meta-analysis is limited by statistically significant heterogeneity in all of the pooled results and high risk of bias in 9 of the case-control studies and all of the cohort studies. There was no evidence of publication bias, however.

Bleeding risk rises 7 to 28 days after SSRI exposure

A 2014 case-crossover study of 5377 inpatients in Taiwan with a psychiatric diagnosis evaluated the risk of UGI bleeding within the first 28 days after SSRI exposure (SSRI-mediated inhibition of platelets occurs within the first 7 to 14 days).² The average age of the patients was 58 years and 75% of the study population was male. Each patient served as his or her own control.

ORs were calculated to compare patients who were exposed to SSRIs only during 7-, 14-, and 21-day windows immediately before a UGI bleed to controls exposed to SSRIs only during the control periods before the 7-, 14-, and 21-day windows. The ORs were adjusted through multivariate analysis to account for 7 potential confounding factors.

SSRI use was associated with an increased risk of UGI bleeding in 7-, 14-, and 21-day windows before the index event (TABLE²). An increased bleeding risk in the 14 days after SSRI initiation was observed in men (OR=2.4; 95% CI, 1.8-3.4) but not women (OR=1.0; 95% CI, 0.6-1.6). Increased bleeding risk in the 14 days after SSRI initiation was also observed in patients younger than 55 years (OR=2.1; 95% CI, 1.5-3.1), patients with a history of upper GI disease (OR=3.1; 95% CI, 1.7-6.0), and patients with no previous exposure to SSRIs (OR=2.6; 95% CI, 1.6-4.2).

This study didn’t account for SSRI indication as a potential confounder, and the study’s inclusion of inpatients, whose illnesses are typically more severe, may limit generalizability.

SSRIs raise bleeding risk; concurrent NSAIDs raise it more

A 2014 systematic review and meta-analysis of 19 case-control and cohort studies with a total of 446,949 patients investigated the risk of UGI bleeding in patients using SSRIs and NSAIDs.¹ The studies, which included both inpatients and outpatients, were done in Europe and North America. Patients were at least 16 years old, but pooled demographics were not reported. Investigators compared SSRI use with or without concurrent NSAID use to placebo or no treatment.

SSRI use was associated with an increased risk of UGI bleeding in 15 case-control studies (393,268 patients; odds ratio [OR]=1.7; 95% confidence interval [CI], 1.4-1.9) and 4 cohort studies (53,681 patients; OR=1.7; 95% CI, 1.1-2.5). The simultaneous use of SSRIs and NSAIDs compared to nonuse of both medications was associated with a larger increase in bleeding risk (10 case-control studies, 223,336 patients; OR=4.3; 95% CI, 2.8-6.4).

The meta-analysis is limited by statistically significant heterogeneity in all of the pooled results and high risk of bias in 9 of the case-control studies and all of the cohort studies. There was no evidence of publication bias, however.

Bleeding risk rises 7 to 28 days after SSRI exposure

A 2014 case-crossover study of 5377 inpatients in Taiwan with a psychiatric diagnosis evaluated the risk of UGI bleeding within the first 28 days after SSRI exposure (SSRI-mediated inhibition of platelets occurs within the first 7 to 14 days).² The average age of the patients was 58 years and 75% of the study population was male. Each patient served as his or her own control.

ORs were calculated to compare patients who were exposed to SSRIs only during 7-, 14-, and 21-day windows immediately before a UGI bleed to controls exposed to SSRIs only during the control periods before the 7-, 14-, and 21-day windows. The ORs were adjusted through multivariate analysis to account for 7 potential confounding factors.

SSRI use was associated with an increased risk of UGI bleeding in 7-, 14-, and 21-day windows before the index event (TABLE²). An increased bleeding risk in the 14 days after SSRI initiation was observed in men (OR=2.4; 95% CI, 1.8-3.4) but not women (OR=1.0; 95% CI, 0.6-1.6). Increased bleeding risk in the 14 days after SSRI initiation was also observed in patients younger than 55 years (OR=2.1; 95% CI, 1.5-3.1), patients with a history of upper GI disease (OR=3.1; 95% CI, 1.7-6.0), and patients with no previous exposure to SSRIs (OR=2.6; 95% CI, 1.6-4.2).

This study didn’t account for SSRI indication as a potential confounder, and the study’s inclusion of inpatients, whose illnesses are typically more severe, may limit generalizability.

CASE 1 › George D is a 67-year-old patient who has never smoked and who has no history of malignancy. An x-ray of his ribs performed after a fall shows a 13-mm solitary nodule in his right upper lung.

CASE 2 › Cathy B is a healthy 80-year-old with no history of smoking. During a trip to the emergency department for chest pain, she had a computed tomography (CT) scan of her chest. While the chest pain was subsequently attributed to gastroesophageal reflux, the CT revealed a 9-mm part solid nodule that was 75% solid.

How should the physicians caring for each of these patients proceed with their care?

The widespread use of sensitive imaging techniques often leads to the incidental discovery of unrelated—but possibly significant—pulmonary findings. Pulmonary nodules are incidentally discovered on an estimated 0.09% to 0.2% of all chest x-rays, 13% of all chest CT angiograms,¹ 31% of all cardiac CTs performed for coronary calcium scoring,² and up to 50% of thin-section chest CT scans.¹

The widespread implementation of the US Preventive Services Task Force recommendations on lung cancer screening has further expanded the number of patients in whom asymptomatic pulmonary nodules will be detected. As a result, family physicians (FPs) will frequently encounter this challenging clinical dilemma and will need to:

• assess the patient’s risk profile
• address the patient’s concerns about malignancy while eliciting his or her preference for management
• minimize the risks of surveillance testing
• minimize patient distress while ensuring compliance with a follow-up that may extend up to 4 years
• determine when it’s appropriate to refer the patient to a pulmonologist and/or pulmonary nodule clinic or registry.

Taking these steps, however, can be challenging. In interviews, 15 primary care clinicians who care for patients with pulmonary nodules expressed concerns about limitations in time, knowledge, and resources, as well as a fear about such patients “falling through the cracks.”³ Familiarity with current evidence-based guidelines such as those from the American College of Chest Physicians (ACCP) and knowledge of emerging data on the management of various types of nodules are imperative.

To that end, this review will fill in the information gaps and provide guidance on how best to communicate what is known about a particular type of nodule with the patient who has one. (See “What to say to improve joint decision-making.”^4-7) But first, a word about terminology.

Solid vs subsolid pulmonary nodules

Solid pulmonary nodules. Traditionally, the term “solitary pulmonary nodule” has been used to describe a single, well-circumscribed, radiographic opacity that measures up to 3 cm in diameter and is completely surrounded by aerated lung.^1,8 The term “solitary” is now less useful because increasingly sensitive imaging techniques often reveal more than one nodule. In the absence of evidence of features that strongly suggest a benign etiology, these are now commonly referred to as indeterminate solid nodules.

If a solid nodule shows evidence of malignant growth on serial imaging, nonsurgical biopsy or surgical resection is recommended.

Subsolid nodules are pulmonary nodules that have unique characteristics and require separate guidelines for management. Subsolid nodules include pure ground glass nodules (GGNs) and part solid nodules. GGNs are focal nodular areas of increased lung attenuation through which normal parenchymal structures such as airways, vessels, and interlobular septa can be visualized.^1,8 Part solid nodules have a solid component. They are usually, but not necessarily, >50% ground glass in appearance.

Lung masses. Focal pulmonary lesions >3 cm in diameter are called lung masses and are presumed to be malignant (bronchogenic carcinoma) unless proven otherwise.^1,8

The specific approach to evaluating and monitoring a pulmonary nodule varies depending on whether the nodule is solid or subsolid and other factors, including the nodule’s size.

Monitoring solid nodules

Monitoring of an indeterminate solid nodule is largely determined by the patient’s risk profile and the characteristics of the identified nodule.¹ Independent patient predictors of malignancy include older age, smoking status, and history of prior malignancy (>5 years ago). Less established predictors are the presence of moderate or severe obstructive lung disease and exposure to particulate or sulfur oxide-related pollution.⁹

Patients who have an indeterminate nodule identified on chest x-ray should undergo a chest CT scan, preferably with thin sections through the nodule to help characterize it.¹ Nodule characteristics that can help predict a patient’s risk of malignancy include the size (>8 mm confers higher risk), malignant rate of growth, edge characteristics (spiculation or irregular edges), thickness of the wall of a cavitary pulmonary nodule (≥16 mm has a likelihood ratio [LR] 37.97 of malignancy), and the location of the nodule (upper or middle lobe [LR=1.2 to 1.6]). ¹⁰ A lack of growth over 2 years and a benign pattern of calcification eliminate the need for further evaluation.

Validated tools to help guide decision-making. Although many physicians estimate pretest probability of malignancy intuitively, validated tools are readily available and can help in clinical decision-making.¹¹ One such tool is the Mayo model, which is available at http://reference.medscape.com/calculator/solitary-pulmonary-nodule-risk. This model takes into account the patient’s age, smoking status, history of cancer, and characteristics of the nodule.

Solid nodules >8 mm to 3 cm

For a patient with a solid nodule >8 mm to 3 cm, ACCP guidelines suggest that physicians estimate the pretest probability of malignancy qualitatively using their clinical judgment and/or quantitatively by using a validated model, such as the Mayo model described above.

Based on the patient’s probability of malignancy, management options include continued CT surveillance, positron emission tomography (PET) imaging, CT-guided needle lung biopsy, bronchoscopy with biopsy, or surgical wedge resection (ALGORITHM 1).¹

CT surveillance is recommended for individuals:

• with very low (<5%) probability of malignancy
• with low to moderate (5% to 30%) or moderate to high (31% to 65%) probability of malignancy with negative functional imaging (PET)
• with high probability of malignancy (>65%) when needle biopsy is nondiagnostic and the lesion is not hypermetabolic on PET scan.

Surveillance is also recommended when a fully informed patient prefers nonaggressive management. The intervals for serial CT in this population are at 3 to 6 months, 9 to 12 months, and 18 to 24 months.

In an individual with a solid indeterminate nodule with a high probability of malignancy (>65%), functional imaging should not be performed to characterize the nodule. It may, however, be performed for staging.

Time for biopsy or resection? If a nodule shows evidence of malignant growth on serial imaging, nonsurgical biopsy (CT scan-guided transthoracic needle biopsy, bronchoscopy guided by fluoroscopy, endobronchial ultrasound, electromagnetic navigation bronchoscopy, or virtual bronchoscopy navigation) or surgical resection is recommended.

Nonsurgical biopsy is also recommended when the patient’s pretest probability and imaging test results are discordant, when a benign diagnosis requires specific medical treatment, or if a fully informed patient desires proof of diagnosis prior to surgery.

Thoracoscopy with wedge resection is the gold standard for diagnosis of a malignant nodule. It is recommended:

• when the clinical probability of malignancy is high (>65%)
• when the nodule is intensely hypermetabolic by PET scan or positive by other functional imaging tests
• when nonsurgical biopsy is suggestive of malignancy
• when a fully informed patient prefers a definitive diagnostic procedure.

CASE 1 › The FP contacts Mr. D and advises that he get a chest CT to better characterize his pulmonary nodule. A thin-slice CT of the lung reveals that the 13-mm solid nodule in the right upper lobe has spiculated margins. According to the Mayo risk calculator, Mr. D is at moderate risk of malignancy (32.5%). Mr. D and his physician discuss the findings and possible management options, and Mr. D opts to have a PET scan. The FP gives Mr. D literature on pulmonary nodules and contact information for the provider team. A PET scan shows negative uptake. Mr. D and his physician discuss CT surveillance and nonsurgical biopsy. He opts for CT surveillance. The next CT is scheduled for 3 months.

Solid nodules ≤8 mm

Management of these lesions generally follows the consensus-based guidelines that were first published by the Fleischner Society and subsequently endorsed by the ACCP.¹ The 2 main determinants that guide management of nodules ≤8 mm are the patient’s risk factors for cancer and nodule size (ALGORITHM 2).¹ The Fleischner guidelines pertain only to patients older than age 35 with no current extra pulmonary malignancy or unexplained fevers. The ACCP guidelines, although similar, do not include these limitations. Patient risk factors include history of smoking, older age, and a history of malignancy.¹

Patients with no risk factors for malignancy. The frequency of surveillance CT is determined by the size of the nodule. Nodules ≤4 mm do not need to be followed. For nodules >4 mm to 6 mm, a repeat CT in 12 months is recommended with no follow-up if stable. For nodules >6 to <8 mm, repeat CT is recommended at 6 to 12 months, and again between 18 and 24 months if unchanged.¹Patients with one or more risk factors for malignancy. Nodules ≤4 mm should be reevaluated at 12 months in patients with one or more risk factors; no additional follow-up is needed if unchanged. For nodules >4 mm to 6 mm, CT should be repeated between 6 and 12 months and again between 18 and 24 months. Nodules >6 mm to <8 mm should be followed initially between 3 to 6 months, then between 9 and 12 months and again at 24 months if unchanged.¹

Subsolid nodules have a high prevalence of premalignant and malignant disease.

Subsolid nodules require a different approach

Subsolid nodules have a high prevalence of premalignant and malignant disease (adenocarcinoma in situ, minimally invasive adenocarcinoma, and adenocarcinoma). Studies have reported subsolid nodule malignancy rates ranging from 20% to 75%.^11-15 This wide range may be a function of different nodule sizes or rates of biopsy. The prevalence increases even further in nodules with a part solid component.

These factors, plus challenges in measuring serial growth on CT and the uncertain prognosis of untreated premalignant disease, make it necessary to have separate guidelines for managing subsolid nodules. The Fleischner Society, National Comprehensive Cancer Network, and the American College of Radiology (LungRads) each have differing recommendations on the frequency of follow-up for different-sized subsolid nodules. Newer studies favor a more conservative approach.¹⁶ Here we describe the current ACCP guidelines for managing subsolid nodules (ALGORITHM 3).¹

GGNs. In an individual with a pure GGN ≤5 mm in diameter, no further evaluation is recommended. In an individual with a pure GGN >5 mm in diameter, annual surveillance with chest CT for at least 3 years is recommended.¹

Part solid nodules. In an individual with a part solid nodule ≤8 mm, conduct CT surveillance at 3, 12, and 24 months and then annually for an additional one to 3 years. In a patient with a part solid nodule >8 mm to 15 mm, repeat chest CT at 3 months followed by a PET scan, nonsurgical biopsy, and/or surgical resection if the nodule persists. A patient with a part solid nodule >15 mm should undergo a PET scan, nonsurgical biopsy, and/or surgical resection.

CASE 2 › Ms. G is seen in the office by her FP, and they discuss management options. A repeat CT is done in 3 months and shows a persistent, unchanged nodule. Ms. G opts for a transthoracic biopsy, which reveals adenocarcinoma. Following a PET scan, which shows no evidence of metastasis, curative surgical wedge resection is done.

Multiple subsolid nodules. In a patient who has a dominant nodule and one or more additional nodules, each nodule should be evaluated individually, according to recommendations from the Fleischner Society (the ACCP currently does not have guidelines for managing multiple subsolid nodules). An individual with multiple GGNs that all measure ≤5 mm should receive CT exams at 2 and 4 years.¹³ A patient with multiple GGNs that include at least one nodule >5 mm but no dominant nodule should undergo follow-up CT at 3 months and annual CT surveillance for at least 3 years.¹³

CORRESPONDENCE
Samina Yunus, MD, MPH, Cleveland Clinic, Family Medicine, 551 East Washington Street, Chagrin Falls, OH 44022; yunuss@ccf.org.

CASE 1 › George D is a 67-year-old patient who has never smoked and who has no history of malignancy. An x-ray of his ribs performed after a fall shows a 13-mm solitary nodule in his right upper lung.

CASE 2 › Cathy B is a healthy 80-year-old with no history of smoking. During a trip to the emergency department for chest pain, she had a computed tomography (CT) scan of her chest. While the chest pain was subsequently attributed to gastroesophageal reflux, the CT revealed a 9-mm part solid nodule that was 75% solid.

How should the physicians caring for each of these patients proceed with their care?

The widespread use of sensitive imaging techniques often leads to the incidental discovery of unrelated—but possibly significant—pulmonary findings. Pulmonary nodules are incidentally discovered on an estimated 0.09% to 0.2% of all chest x-rays, 13% of all chest CT angiograms,¹ 31% of all cardiac CTs performed for coronary calcium scoring,² and up to 50% of thin-section chest CT scans.¹

The widespread implementation of the US Preventive Services Task Force recommendations on lung cancer screening has further expanded the number of patients in whom asymptomatic pulmonary nodules will be detected. As a result, family physicians (FPs) will frequently encounter this challenging clinical dilemma and will need to:

• assess the patient’s risk profile
• address the patient’s concerns about malignancy while eliciting his or her preference for management
• minimize the risks of surveillance testing
• minimize patient distress while ensuring compliance with a follow-up that may extend up to 4 years
• determine when it’s appropriate to refer the patient to a pulmonologist and/or pulmonary nodule clinic or registry.

Taking these steps, however, can be challenging. In interviews, 15 primary care clinicians who care for patients with pulmonary nodules expressed concerns about limitations in time, knowledge, and resources, as well as a fear about such patients “falling through the cracks.”³ Familiarity with current evidence-based guidelines such as those from the American College of Chest Physicians (ACCP) and knowledge of emerging data on the management of various types of nodules are imperative.

To that end, this review will fill in the information gaps and provide guidance on how best to communicate what is known about a particular type of nodule with the patient who has one. (See “What to say to improve joint decision-making.”^4-7) But first, a word about terminology.

Solid vs subsolid pulmonary nodules

Solid pulmonary nodules. Traditionally, the term “solitary pulmonary nodule” has been used to describe a single, well-circumscribed, radiographic opacity that measures up to 3 cm in diameter and is completely surrounded by aerated lung.^1,8 The term “solitary” is now less useful because increasingly sensitive imaging techniques often reveal more than one nodule. In the absence of evidence of features that strongly suggest a benign etiology, these are now commonly referred to as indeterminate solid nodules.

If a solid nodule shows evidence of malignant growth on serial imaging, nonsurgical biopsy or surgical resection is recommended.

Subsolid nodules are pulmonary nodules that have unique characteristics and require separate guidelines for management. Subsolid nodules include pure ground glass nodules (GGNs) and part solid nodules. GGNs are focal nodular areas of increased lung attenuation through which normal parenchymal structures such as airways, vessels, and interlobular septa can be visualized.^1,8 Part solid nodules have a solid component. They are usually, but not necessarily, >50% ground glass in appearance.

Lung masses. Focal pulmonary lesions >3 cm in diameter are called lung masses and are presumed to be malignant (bronchogenic carcinoma) unless proven otherwise.^1,8

The specific approach to evaluating and monitoring a pulmonary nodule varies depending on whether the nodule is solid or subsolid and other factors, including the nodule’s size.

Monitoring solid nodules

Monitoring of an indeterminate solid nodule is largely determined by the patient’s risk profile and the characteristics of the identified nodule.¹ Independent patient predictors of malignancy include older age, smoking status, and history of prior malignancy (>5 years ago). Less established predictors are the presence of moderate or severe obstructive lung disease and exposure to particulate or sulfur oxide-related pollution.⁹

Patients who have an indeterminate nodule identified on chest x-ray should undergo a chest CT scan, preferably with thin sections through the nodule to help characterize it.¹ Nodule characteristics that can help predict a patient’s risk of malignancy include the size (>8 mm confers higher risk), malignant rate of growth, edge characteristics (spiculation or irregular edges), thickness of the wall of a cavitary pulmonary nodule (≥16 mm has a likelihood ratio [LR] 37.97 of malignancy), and the location of the nodule (upper or middle lobe [LR=1.2 to 1.6]). ¹⁰ A lack of growth over 2 years and a benign pattern of calcification eliminate the need for further evaluation.

Validated tools to help guide decision-making. Although many physicians estimate pretest probability of malignancy intuitively, validated tools are readily available and can help in clinical decision-making.¹¹ One such tool is the Mayo model, which is available at http://reference.medscape.com/calculator/solitary-pulmonary-nodule-risk. This model takes into account the patient’s age, smoking status, history of cancer, and characteristics of the nodule.

Solid nodules >8 mm to 3 cm

For a patient with a solid nodule >8 mm to 3 cm, ACCP guidelines suggest that physicians estimate the pretest probability of malignancy qualitatively using their clinical judgment and/or quantitatively by using a validated model, such as the Mayo model described above.

Based on the patient’s probability of malignancy, management options include continued CT surveillance, positron emission tomography (PET) imaging, CT-guided needle lung biopsy, bronchoscopy with biopsy, or surgical wedge resection (ALGORITHM 1).¹

CT surveillance is recommended for individuals:

• with very low (<5%) probability of malignancy
• with low to moderate (5% to 30%) or moderate to high (31% to 65%) probability of malignancy with negative functional imaging (PET)
• with high probability of malignancy (>65%) when needle biopsy is nondiagnostic and the lesion is not hypermetabolic on PET scan.

Surveillance is also recommended when a fully informed patient prefers nonaggressive management. The intervals for serial CT in this population are at 3 to 6 months, 9 to 12 months, and 18 to 24 months.

In an individual with a solid indeterminate nodule with a high probability of malignancy (>65%), functional imaging should not be performed to characterize the nodule. It may, however, be performed for staging.

Time for biopsy or resection? If a nodule shows evidence of malignant growth on serial imaging, nonsurgical biopsy (CT scan-guided transthoracic needle biopsy, bronchoscopy guided by fluoroscopy, endobronchial ultrasound, electromagnetic navigation bronchoscopy, or virtual bronchoscopy navigation) or surgical resection is recommended.

Nonsurgical biopsy is also recommended when the patient’s pretest probability and imaging test results are discordant, when a benign diagnosis requires specific medical treatment, or if a fully informed patient desires proof of diagnosis prior to surgery.

Thoracoscopy with wedge resection is the gold standard for diagnosis of a malignant nodule. It is recommended:

• when the clinical probability of malignancy is high (>65%)
• when the nodule is intensely hypermetabolic by PET scan or positive by other functional imaging tests
• when nonsurgical biopsy is suggestive of malignancy
• when a fully informed patient prefers a definitive diagnostic procedure.

CASE 1 › The FP contacts Mr. D and advises that he get a chest CT to better characterize his pulmonary nodule. A thin-slice CT of the lung reveals that the 13-mm solid nodule in the right upper lobe has spiculated margins. According to the Mayo risk calculator, Mr. D is at moderate risk of malignancy (32.5%). Mr. D and his physician discuss the findings and possible management options, and Mr. D opts to have a PET scan. The FP gives Mr. D literature on pulmonary nodules and contact information for the provider team. A PET scan shows negative uptake. Mr. D and his physician discuss CT surveillance and nonsurgical biopsy. He opts for CT surveillance. The next CT is scheduled for 3 months.

Solid nodules ≤8 mm

Management of these lesions generally follows the consensus-based guidelines that were first published by the Fleischner Society and subsequently endorsed by the ACCP.¹ The 2 main determinants that guide management of nodules ≤8 mm are the patient’s risk factors for cancer and nodule size (ALGORITHM 2).¹ The Fleischner guidelines pertain only to patients older than age 35 with no current extra pulmonary malignancy or unexplained fevers. The ACCP guidelines, although similar, do not include these limitations. Patient risk factors include history of smoking, older age, and a history of malignancy.¹

Patients with no risk factors for malignancy. The frequency of surveillance CT is determined by the size of the nodule. Nodules ≤4 mm do not need to be followed. For nodules >4 mm to 6 mm, a repeat CT in 12 months is recommended with no follow-up if stable. For nodules >6 to <8 mm, repeat CT is recommended at 6 to 12 months, and again between 18 and 24 months if unchanged.¹Patients with one or more risk factors for malignancy. Nodules ≤4 mm should be reevaluated at 12 months in patients with one or more risk factors; no additional follow-up is needed if unchanged. For nodules >4 mm to 6 mm, CT should be repeated between 6 and 12 months and again between 18 and 24 months. Nodules >6 mm to <8 mm should be followed initially between 3 to 6 months, then between 9 and 12 months and again at 24 months if unchanged.¹

Subsolid nodules have a high prevalence of premalignant and malignant disease.

Subsolid nodules require a different approach

Subsolid nodules have a high prevalence of premalignant and malignant disease (adenocarcinoma in situ, minimally invasive adenocarcinoma, and adenocarcinoma). Studies have reported subsolid nodule malignancy rates ranging from 20% to 75%.^11-15 This wide range may be a function of different nodule sizes or rates of biopsy. The prevalence increases even further in nodules with a part solid component.

These factors, plus challenges in measuring serial growth on CT and the uncertain prognosis of untreated premalignant disease, make it necessary to have separate guidelines for managing subsolid nodules. The Fleischner Society, National Comprehensive Cancer Network, and the American College of Radiology (LungRads) each have differing recommendations on the frequency of follow-up for different-sized subsolid nodules. Newer studies favor a more conservative approach.¹⁶ Here we describe the current ACCP guidelines for managing subsolid nodules (ALGORITHM 3).¹

GGNs. In an individual with a pure GGN ≤5 mm in diameter, no further evaluation is recommended. In an individual with a pure GGN >5 mm in diameter, annual surveillance with chest CT for at least 3 years is recommended.¹

Part solid nodules. In an individual with a part solid nodule ≤8 mm, conduct CT surveillance at 3, 12, and 24 months and then annually for an additional one to 3 years. In a patient with a part solid nodule >8 mm to 15 mm, repeat chest CT at 3 months followed by a PET scan, nonsurgical biopsy, and/or surgical resection if the nodule persists. A patient with a part solid nodule >15 mm should undergo a PET scan, nonsurgical biopsy, and/or surgical resection.

CASE 2 › Ms. G is seen in the office by her FP, and they discuss management options. A repeat CT is done in 3 months and shows a persistent, unchanged nodule. Ms. G opts for a transthoracic biopsy, which reveals adenocarcinoma. Following a PET scan, which shows no evidence of metastasis, curative surgical wedge resection is done.

Multiple subsolid nodules. In a patient who has a dominant nodule and one or more additional nodules, each nodule should be evaluated individually, according to recommendations from the Fleischner Society (the ACCP currently does not have guidelines for managing multiple subsolid nodules). An individual with multiple GGNs that all measure ≤5 mm should receive CT exams at 2 and 4 years.¹³ A patient with multiple GGNs that include at least one nodule >5 mm but no dominant nodule should undergo follow-up CT at 3 months and annual CT surveillance for at least 3 years.¹³

CORRESPONDENCE
Samina Yunus, MD, MPH, Cleveland Clinic, Family Medicine, 551 East Washington Street, Chagrin Falls, OH 44022; yunuss@ccf.org.

CASE 1 › George D is a 67-year-old patient who has never smoked and who has no history of malignancy. An x-ray of his ribs performed after a fall shows a 13-mm solitary nodule in his right upper lung.

CASE 2 › Cathy B is a healthy 80-year-old with no history of smoking. During a trip to the emergency department for chest pain, she had a computed tomography (CT) scan of her chest. While the chest pain was subsequently attributed to gastroesophageal reflux, the CT revealed a 9-mm part solid nodule that was 75% solid.

How should the physicians caring for each of these patients proceed with their care?

The widespread use of sensitive imaging techniques often leads to the incidental discovery of unrelated—but possibly significant—pulmonary findings. Pulmonary nodules are incidentally discovered on an estimated 0.09% to 0.2% of all chest x-rays, 13% of all chest CT angiograms,¹ 31% of all cardiac CTs performed for coronary calcium scoring,² and up to 50% of thin-section chest CT scans.¹

The widespread implementation of the US Preventive Services Task Force recommendations on lung cancer screening has further expanded the number of patients in whom asymptomatic pulmonary nodules will be detected. As a result, family physicians (FPs) will frequently encounter this challenging clinical dilemma and will need to:

• assess the patient’s risk profile
• address the patient’s concerns about malignancy while eliciting his or her preference for management
• minimize the risks of surveillance testing
• minimize patient distress while ensuring compliance with a follow-up that may extend up to 4 years
• determine when it’s appropriate to refer the patient to a pulmonologist and/or pulmonary nodule clinic or registry.

Taking these steps, however, can be challenging. In interviews, 15 primary care clinicians who care for patients with pulmonary nodules expressed concerns about limitations in time, knowledge, and resources, as well as a fear about such patients “falling through the cracks.”³ Familiarity with current evidence-based guidelines such as those from the American College of Chest Physicians (ACCP) and knowledge of emerging data on the management of various types of nodules are imperative.

To that end, this review will fill in the information gaps and provide guidance on how best to communicate what is known about a particular type of nodule with the patient who has one. (See “What to say to improve joint decision-making.”^4-7) But first, a word about terminology.

Solid vs subsolid pulmonary nodules

Solid pulmonary nodules. Traditionally, the term “solitary pulmonary nodule” has been used to describe a single, well-circumscribed, radiographic opacity that measures up to 3 cm in diameter and is completely surrounded by aerated lung.^1,8 The term “solitary” is now less useful because increasingly sensitive imaging techniques often reveal more than one nodule. In the absence of evidence of features that strongly suggest a benign etiology, these are now commonly referred to as indeterminate solid nodules.

If a solid nodule shows evidence of malignant growth on serial imaging, nonsurgical biopsy or surgical resection is recommended.

Subsolid nodules are pulmonary nodules that have unique characteristics and require separate guidelines for management. Subsolid nodules include pure ground glass nodules (GGNs) and part solid nodules. GGNs are focal nodular areas of increased lung attenuation through which normal parenchymal structures such as airways, vessels, and interlobular septa can be visualized.^1,8 Part solid nodules have a solid component. They are usually, but not necessarily, >50% ground glass in appearance.

Lung masses. Focal pulmonary lesions >3 cm in diameter are called lung masses and are presumed to be malignant (bronchogenic carcinoma) unless proven otherwise.^1,8

The specific approach to evaluating and monitoring a pulmonary nodule varies depending on whether the nodule is solid or subsolid and other factors, including the nodule’s size.

Monitoring solid nodules

Monitoring of an indeterminate solid nodule is largely determined by the patient’s risk profile and the characteristics of the identified nodule.¹ Independent patient predictors of malignancy include older age, smoking status, and history of prior malignancy (>5 years ago). Less established predictors are the presence of moderate or severe obstructive lung disease and exposure to particulate or sulfur oxide-related pollution.⁹

Patients who have an indeterminate nodule identified on chest x-ray should undergo a chest CT scan, preferably with thin sections through the nodule to help characterize it.¹ Nodule characteristics that can help predict a patient’s risk of malignancy include the size (>8 mm confers higher risk), malignant rate of growth, edge characteristics (spiculation or irregular edges), thickness of the wall of a cavitary pulmonary nodule (≥16 mm has a likelihood ratio [LR] 37.97 of malignancy), and the location of the nodule (upper or middle lobe [LR=1.2 to 1.6]). ¹⁰ A lack of growth over 2 years and a benign pattern of calcification eliminate the need for further evaluation.

Validated tools to help guide decision-making. Although many physicians estimate pretest probability of malignancy intuitively, validated tools are readily available and can help in clinical decision-making.¹¹ One such tool is the Mayo model, which is available at http://reference.medscape.com/calculator/solitary-pulmonary-nodule-risk. This model takes into account the patient’s age, smoking status, history of cancer, and characteristics of the nodule.

Solid nodules >8 mm to 3 cm

For a patient with a solid nodule >8 mm to 3 cm, ACCP guidelines suggest that physicians estimate the pretest probability of malignancy qualitatively using their clinical judgment and/or quantitatively by using a validated model, such as the Mayo model described above.

Based on the patient’s probability of malignancy, management options include continued CT surveillance, positron emission tomography (PET) imaging, CT-guided needle lung biopsy, bronchoscopy with biopsy, or surgical wedge resection (ALGORITHM 1).¹

CT surveillance is recommended for individuals:

• with very low (<5%) probability of malignancy
• with low to moderate (5% to 30%) or moderate to high (31% to 65%) probability of malignancy with negative functional imaging (PET)
• with high probability of malignancy (>65%) when needle biopsy is nondiagnostic and the lesion is not hypermetabolic on PET scan.

Surveillance is also recommended when a fully informed patient prefers nonaggressive management. The intervals for serial CT in this population are at 3 to 6 months, 9 to 12 months, and 18 to 24 months.

In an individual with a solid indeterminate nodule with a high probability of malignancy (>65%), functional imaging should not be performed to characterize the nodule. It may, however, be performed for staging.

Time for biopsy or resection? If a nodule shows evidence of malignant growth on serial imaging, nonsurgical biopsy (CT scan-guided transthoracic needle biopsy, bronchoscopy guided by fluoroscopy, endobronchial ultrasound, electromagnetic navigation bronchoscopy, or virtual bronchoscopy navigation) or surgical resection is recommended.

Nonsurgical biopsy is also recommended when the patient’s pretest probability and imaging test results are discordant, when a benign diagnosis requires specific medical treatment, or if a fully informed patient desires proof of diagnosis prior to surgery.

Thoracoscopy with wedge resection is the gold standard for diagnosis of a malignant nodule. It is recommended:

• when the clinical probability of malignancy is high (>65%)
• when the nodule is intensely hypermetabolic by PET scan or positive by other functional imaging tests
• when nonsurgical biopsy is suggestive of malignancy
• when a fully informed patient prefers a definitive diagnostic procedure.

CASE 1 › The FP contacts Mr. D and advises that he get a chest CT to better characterize his pulmonary nodule. A thin-slice CT of the lung reveals that the 13-mm solid nodule in the right upper lobe has spiculated margins. According to the Mayo risk calculator, Mr. D is at moderate risk of malignancy (32.5%). Mr. D and his physician discuss the findings and possible management options, and Mr. D opts to have a PET scan. The FP gives Mr. D literature on pulmonary nodules and contact information for the provider team. A PET scan shows negative uptake. Mr. D and his physician discuss CT surveillance and nonsurgical biopsy. He opts for CT surveillance. The next CT is scheduled for 3 months.

Solid nodules ≤8 mm

Management of these lesions generally follows the consensus-based guidelines that were first published by the Fleischner Society and subsequently endorsed by the ACCP.¹ The 2 main determinants that guide management of nodules ≤8 mm are the patient’s risk factors for cancer and nodule size (ALGORITHM 2).¹ The Fleischner guidelines pertain only to patients older than age 35 with no current extra pulmonary malignancy or unexplained fevers. The ACCP guidelines, although similar, do not include these limitations. Patient risk factors include history of smoking, older age, and a history of malignancy.¹

Patients with no risk factors for malignancy. The frequency of surveillance CT is determined by the size of the nodule. Nodules ≤4 mm do not need to be followed. For nodules >4 mm to 6 mm, a repeat CT in 12 months is recommended with no follow-up if stable. For nodules >6 to <8 mm, repeat CT is recommended at 6 to 12 months, and again between 18 and 24 months if unchanged.¹Patients with one or more risk factors for malignancy. Nodules ≤4 mm should be reevaluated at 12 months in patients with one or more risk factors; no additional follow-up is needed if unchanged. For nodules >4 mm to 6 mm, CT should be repeated between 6 and 12 months and again between 18 and 24 months. Nodules >6 mm to <8 mm should be followed initially between 3 to 6 months, then between 9 and 12 months and again at 24 months if unchanged.¹

Subsolid nodules have a high prevalence of premalignant and malignant disease.

Subsolid nodules require a different approach

Subsolid nodules have a high prevalence of premalignant and malignant disease (adenocarcinoma in situ, minimally invasive adenocarcinoma, and adenocarcinoma). Studies have reported subsolid nodule malignancy rates ranging from 20% to 75%.^11-15 This wide range may be a function of different nodule sizes or rates of biopsy. The prevalence increases even further in nodules with a part solid component.

These factors, plus challenges in measuring serial growth on CT and the uncertain prognosis of untreated premalignant disease, make it necessary to have separate guidelines for managing subsolid nodules. The Fleischner Society, National Comprehensive Cancer Network, and the American College of Radiology (LungRads) each have differing recommendations on the frequency of follow-up for different-sized subsolid nodules. Newer studies favor a more conservative approach.¹⁶ Here we describe the current ACCP guidelines for managing subsolid nodules (ALGORITHM 3).¹

GGNs. In an individual with a pure GGN ≤5 mm in diameter, no further evaluation is recommended. In an individual with a pure GGN >5 mm in diameter, annual surveillance with chest CT for at least 3 years is recommended.¹

Part solid nodules. In an individual with a part solid nodule ≤8 mm, conduct CT surveillance at 3, 12, and 24 months and then annually for an additional one to 3 years. In a patient with a part solid nodule >8 mm to 15 mm, repeat chest CT at 3 months followed by a PET scan, nonsurgical biopsy, and/or surgical resection if the nodule persists. A patient with a part solid nodule >15 mm should undergo a PET scan, nonsurgical biopsy, and/or surgical resection.

CASE 2 › Ms. G is seen in the office by her FP, and they discuss management options. A repeat CT is done in 3 months and shows a persistent, unchanged nodule. Ms. G opts for a transthoracic biopsy, which reveals adenocarcinoma. Following a PET scan, which shows no evidence of metastasis, curative surgical wedge resection is done.

Multiple subsolid nodules. In a patient who has a dominant nodule and one or more additional nodules, each nodule should be evaluated individually, according to recommendations from the Fleischner Society (the ACCP currently does not have guidelines for managing multiple subsolid nodules). An individual with multiple GGNs that all measure ≤5 mm should receive CT exams at 2 and 4 years.¹³ A patient with multiple GGNs that include at least one nodule >5 mm but no dominant nodule should undergo follow-up CT at 3 months and annual CT surveillance for at least 3 years.¹³

CORRESPONDENCE
Samina Yunus, MD, MPH, Cleveland Clinic, Family Medicine, 551 East Washington Street, Chagrin Falls, OH 44022; yunuss@ccf.org.

CASE 1 › Sheila S, age 27, has irritable bowel syndrome (IBS) and comes to your office for a follow-up visit. Over the past 6 months she has started taking a fiber supplement, drinking more water, and looking for links between stress and her symptoms. She has read about probiotics and wonders if you would consider recommending them in her situation.

CASE 2 › Mark M, age 45, has type 2 diabetes and is overweight. He is motivated to change his diet and has started to exercise more. He is taking metformin 2000 mg/d but his hemoglobin A1c remains slightly elevated at 7.2%. He heard on television that probiotics might help to keep him from needing to add another medication.

Most of the living organisms that comprise the human microbiome—all of the microbes that live on or in humans—are found in the gastrointestinal (GI) tract. The gut flora contribute 99% of the genetic material in the human body. The composition of the gut flora is remarkably diverse across the population; each individual has a unique microbial footprint. Within this microbial diversity, there appears to be a stable number of genes that are responsible for the major functions of the gut flora.¹ These microbes:

• supply essential nutrients by breaking down complex carbohydrates;
• generate secondary bile acids that assist in digesting fats;²
• synthesize vitamins such as K, B12, folate, and biotin;³
• contribute to the defensive barrier in the colon by keeping pathogenic bacteria from crossing the colonic mucosa; and
• interact with our systemic immune system in a way that maintains a level of homeostasis, allowing for appropriate activation in the face of pathogens without developing autoimmunity.⁴

The gut flora also play a role in the communication between the central nervous system and the enteric nervous system by modulating the hormonal and neural pathways that have been labeled the “gut-brain axis.” The gut-brain axis has been associated with numerous disease states, including irritable bowel syndrome and certain psychiatric disorders.⁵

Researchers are investigating interventions that target the microbiome to increase microbial diversity and the presence of certain species to prevent or treat various diseases. The use of probiotics and dietary changes to increase intake of soluble fiber have been the most studied of these interventions. The thought is that these interventions can correct an imbalance, or dysbiosis, of the gut flora.⁶ Studies have shown that decreased microbial diversity is associated with elevations of certain disease markers (eg, adiposity, insulin, triglycerides, C–reactive protein)⁷ and that increases in soluble fiber lead to the greatest long-term improvement in microbial diversity.⁸ Fecal transplant—the transfer of a processed mixture of stool that contains “healthy” bacteria from a donor into the intestines of a patient—is being explored as a method of replacing colonic gut flora, but evidence is limited.

The following review takes a closer look at these options and identifies those that are most likely to benefit patients in the treatment—and prevention—of several diseases (TABLE 1).^9-16

Evidence is best for using probiotics for digestive diseases

Dietary interventions for digestive diseases have long been studied, but are getting renewed attention for their potential impact on the microbiome.¹⁷ Beyond dietary modification, other similar treatment options include probiotics (live microorganisms thought to confer a beneficial effect on the host), prebiotics (non-digestible food ingredients, including oligosaccharides and inulin, thought to promote the growth of “helpful” gut flora), and synbiotics (combinations of the 2).¹⁸

Irritable bowel syndrome (IBS) is a heterogeneous disorder characterized by altered intestinal transit, low-grade colonic inflammation, and/or alterations in the gutbrain axis. Research has increasingly focused on recently discovered increases in intestinal immune activation, intestinal permeability, and alterations in the colonic microbiome (decreased diversity and increased pathogenic bacteria) associated with IBS.¹⁹

Meta-analyses have found that combination probiotics benefit patients with ulcerative colitis, but not those with Crohn’s disease.

A meta-analysis of 43 randomized control trials (RCTs) found probiotics ranging from Lactobacillus to Saccharomyces can significantly decrease global IBS symptoms, abdominal pain, bloating, and flatulence.⁹ For a patient such as Ms. S, the evidence suggests a probiotic that contains a mixture of Lactobacillus and Bifidobacterium might help relieve her symptoms.⁹ In terms of dietary modifications, soluble fiber, which is already known to help treat IBS,²⁰ has profound effects on improving microbiota diversity and in shifting the composition toward less pathogenic strains.²¹ The Institute of Medicine's daily recommended intake of soluble fiber is about 15 g/d.²²

Inflammatory bowel disease (IBD) is caused by inflammation of the GI lining due to an overactive immune response. Evidence shows that patients with IBD have an altered microbial composition—specifically, an increase in bacteria that produce pro-inflammatory molecules and a decrease in bacteria that have a dampening effect on immune activation.²³

Most studies evaluating probiotics as a treatment for IBD have been small and have used a wide variety of bacterial mixtures, which makes comparisons difficult. Recent meta-analyses found combination probiotics can both induce and maintain remission in patients with ulcerative colitis, but have no beneficial effects in Crohn’s disease.¹⁰ In a review of 9 case series of patients with IBD, fecal transplant reduced IBD symptoms, and patients were able to decrease medication use.²⁴

Diarrheal illness. The human intestine is protected from diarrheal illness by healthy bacteria that block the actions of pathogenic bacteria. This mechanism is called colonization resistance. Moderate levels of evidence support the use of probiotics to prevent or treat several types of diarrheal illness.¹⁴

Antibiotic-associated diarrhea (AAD) is caused when antibiotic use alters the microbial balance. Recent meta-analyses have shown probiotics can prevent AAD and Clostridium difficile-associated diarrhea.^11,12 Several case series and one RCT have found that fecal transplants are safe and efficacious for treating recurrent Clostridium difficile infection.²⁵ Using probiotics to treat symptoms of AAD has been less studied.

Acute infectious diarrhea and traveler’s diarrhea (TD). A Cochrane review found that probiotics decreased the duration of diarrheal episodes by 25 hours, decreased the risk of an episode lasting more than 4 days by 59%, and led to one less diarrheal stool per day by the second day of the intervention.¹³ In a separate meta-analysis of 12 studies, probiotics significantly prevented 85% of cases of TD.¹⁴

Encouraging early evidence for several other illnesses

Metabolic disorders. Both animal and human studies support the theory that gut flora contribute to energy homeostasis, and in some genetically predisposed people dysbiosis may lead to obesity and diabetes. The traditional western diet⁴ and possibly decreased physical activity²⁶ are major contributors to gut flora dysbiosis. Healthy bacteria in the gut break down soluble fiber into short chain fatty acids (SCFAs). SCFAs are associated with increased satiety, decreased food intake, lower levels of inflammation, and improvement in insulin signaling in adipose tissue. In addition to decreased SFCA production, dysbiosis also leads to increased lipid deposition through higher levels of lipoprotein lipase.²⁷

Obesity. The bacteria in our gut affect energy metabolism. In patients with obesity, increased amounts of bacteria in the taxa Firmicutes and a corresponding decrease in Bacteroidetes is associated with an increased energy harvest and decreased SCFA production, which leads to a pro-inflammatory state.²⁸ Probiotics that contain Bifidobacterium and Lactobacillus are thought to help correct this dysbiosis by increasing production of SCFAs.²⁸

A recent meta-analysis of 4 RCTs found no significant difference between supplementation with probiotics and placebo on weight reduction.²⁹ However, lower-quality studies with more subjects and longer duration have shown a statistically significant improvement in weight reduction with probiotic use compared to placebo.²⁹

The evidence suggests that fecal transplants are safe and efficacious for treating recurrent Clostridium difficile infection.

Diabetes. Although dietary interventions to improve glycemic control have long been an important cornerstone of treatment, probiotic supplementation to further alter gut flora composition is also being evaluated. Studies have found probiotics have largely beneficial effects on glycemic control, especially in animals. The largest systematic review to date looked at 33 studies, including 5 human trials. The human studies each found a significant reduction in at least one of 6 parameters of glycemic control (levels of fasting plasma glucose, postprandial blood glucose, glycated hemoglobin, insulin, insulin resistance, and onset of diabetes).¹⁶ It is unclear which probiotic strains confer benefit, and if those benefits are sustainable without dietary modification and increased physical activity.

Psychiatric illnesses. The gut-brain axis is thought to impact mental health by several mechanisms, including modulating the hypothalamic-pituitary-adrenal axis, activating the immune system, producing active metabolites, and affecting the vagus nerve. It is unclear which of these pathways may be clinically relevant.^5,30 The few human studies that have looked for a potential link between gut flora and psychiatric illness have focused on depression and autism spectrum disorders (ASD).

Depression. Small studies comparing the microbiome composition of depressed patients vs healthy controls have found differences in patterns of both over- and underrepresented microbiota species in depressed patients, although the patterns across studies have been inconsistent.^31,32 One small functional magnetic resonance imaging study of healthy women showed that a fermented milk product that contained probiotics affected activity in areas of the brain that control emotion and sensation.³³ A few small studies have shown that patients who used probiotics had improved depression scores.³⁴ Further studies are needed.

ASD. Children with ASD have GI disturbances—most commonly diarrhea, constipation, and/or bloating—more often than healthy controls.^35,36 This association has led to speculation of a connection between the gut and brain. The microbial composition and diversity appears to be different in individuals with ASD; several studies have found an increase in Clostridia species.³⁷

Research on probiotics for treating ASD has been primarily in preclinical models. Human studies of probiotics for ASD are lacking.³⁸ Small studies on dietary modifications such as gluten-free and casein-free diets have had varying results; to what extent these dietary changes exert their influence via the intestinal microbiome is unknown.³⁸

Eczema. Several studies have looked at the role of prebiotics and probiotics in reducing the risk for allergic disease. A 2013 Cochrane review found strong evidence that certain prebiotics can prevent eczema in children under age 2.¹⁵ There is limited evidence that probiotics may also play a role in preventing eczema.^39,40 However, probiotics do not appear to be effective for treating eczema.⁴¹

Several studies have found a link between the use of probiotics and significant reductions in at least one of 6 parameters of glycemic control.

Rheumatoid arthritis (RA). Patients with RA have a change in the balance of function of different T helper cells subsets, and several studies have shown that changes in the gut microbiome can affect this balance.⁴² A recent small study of patients with RA found that 75% of those with new onset RA had Prevotella copri bacteria as the predominant species, and patients with chronic RA had a decrease in Bacteroides species compared to healthy counterparts.^42-44 The exact influence of gut flora dysbiosis on RA is unknown.⁴⁵ Small studies suggest dietary changes may improve RA symptoms, while data on the use of probiotics to alleviate symptoms is mixed.⁴⁶

What to tell patients about gut flora and health

There is increasing evidence that the gut microbiome and the genes contained therein have an impact on an individual’s health. (See TABLE 2 for additional resources.) The best preventive advice for patients and their families is to eat a diet rich in fruits and vegetables. This measure has well proven benefits beyond its potential effects on gut flora.

Correcting dysbiosis with diet or probiotics may play a role in treating chronic conditions; however, in many cases, further research is required to elucidate specific recommendations. In the meantime, given the safety profile of probiotics and dietary fiber, it is reasonable to consider using these interventions, particularly probiotics for treating IBS, ulcerative colitis, and acute infectious diarrhea; probiotics for preventing antibiotic-associated diarrhea and traveler’s diarrhea; and prebiotics for preventing eczema in high-risk infants.

CORRESPONDENCE
Jill Schneiderhan, MD, Family Medicine at Domino’s Farms, 24 Frank Lloyd Wright Dr., Lobby H, Suite 2300, Ann Arbor, MI 48105; jillsch@umich.edu.

CASE 1 › Sheila S, age 27, has irritable bowel syndrome (IBS) and comes to your office for a follow-up visit. Over the past 6 months she has started taking a fiber supplement, drinking more water, and looking for links between stress and her symptoms. She has read about probiotics and wonders if you would consider recommending them in her situation.

CASE 2 › Mark M, age 45, has type 2 diabetes and is overweight. He is motivated to change his diet and has started to exercise more. He is taking metformin 2000 mg/d but his hemoglobin A1c remains slightly elevated at 7.2%. He heard on television that probiotics might help to keep him from needing to add another medication.

Most of the living organisms that comprise the human microbiome—all of the microbes that live on or in humans—are found in the gastrointestinal (GI) tract. The gut flora contribute 99% of the genetic material in the human body. The composition of the gut flora is remarkably diverse across the population; each individual has a unique microbial footprint. Within this microbial diversity, there appears to be a stable number of genes that are responsible for the major functions of the gut flora.¹ These microbes:

• supply essential nutrients by breaking down complex carbohydrates;
• generate secondary bile acids that assist in digesting fats;²
• synthesize vitamins such as K, B12, folate, and biotin;³
• contribute to the defensive barrier in the colon by keeping pathogenic bacteria from crossing the colonic mucosa; and
• interact with our systemic immune system in a way that maintains a level of homeostasis, allowing for appropriate activation in the face of pathogens without developing autoimmunity.⁴

The gut flora also play a role in the communication between the central nervous system and the enteric nervous system by modulating the hormonal and neural pathways that have been labeled the “gut-brain axis.” The gut-brain axis has been associated with numerous disease states, including irritable bowel syndrome and certain psychiatric disorders.⁵

Researchers are investigating interventions that target the microbiome to increase microbial diversity and the presence of certain species to prevent or treat various diseases. The use of probiotics and dietary changes to increase intake of soluble fiber have been the most studied of these interventions. The thought is that these interventions can correct an imbalance, or dysbiosis, of the gut flora.⁶ Studies have shown that decreased microbial diversity is associated with elevations of certain disease markers (eg, adiposity, insulin, triglycerides, C–reactive protein)⁷ and that increases in soluble fiber lead to the greatest long-term improvement in microbial diversity.⁸ Fecal transplant—the transfer of a processed mixture of stool that contains “healthy” bacteria from a donor into the intestines of a patient—is being explored as a method of replacing colonic gut flora, but evidence is limited.

The following review takes a closer look at these options and identifies those that are most likely to benefit patients in the treatment—and prevention—of several diseases (TABLE 1).^9-16

Evidence is best for using probiotics for digestive diseases

Dietary interventions for digestive diseases have long been studied, but are getting renewed attention for their potential impact on the microbiome.¹⁷ Beyond dietary modification, other similar treatment options include probiotics (live microorganisms thought to confer a beneficial effect on the host), prebiotics (non-digestible food ingredients, including oligosaccharides and inulin, thought to promote the growth of “helpful” gut flora), and synbiotics (combinations of the 2).¹⁸

Irritable bowel syndrome (IBS) is a heterogeneous disorder characterized by altered intestinal transit, low-grade colonic inflammation, and/or alterations in the gutbrain axis. Research has increasingly focused on recently discovered increases in intestinal immune activation, intestinal permeability, and alterations in the colonic microbiome (decreased diversity and increased pathogenic bacteria) associated with IBS.¹⁹

Meta-analyses have found that combination probiotics benefit patients with ulcerative colitis, but not those with Crohn’s disease.

A meta-analysis of 43 randomized control trials (RCTs) found probiotics ranging from Lactobacillus to Saccharomyces can significantly decrease global IBS symptoms, abdominal pain, bloating, and flatulence.⁹ For a patient such as Ms. S, the evidence suggests a probiotic that contains a mixture of Lactobacillus and Bifidobacterium might help relieve her symptoms.⁹ In terms of dietary modifications, soluble fiber, which is already known to help treat IBS,²⁰ has profound effects on improving microbiota diversity and in shifting the composition toward less pathogenic strains.²¹ The Institute of Medicine's daily recommended intake of soluble fiber is about 15 g/d.²²

Inflammatory bowel disease (IBD) is caused by inflammation of the GI lining due to an overactive immune response. Evidence shows that patients with IBD have an altered microbial composition—specifically, an increase in bacteria that produce pro-inflammatory molecules and a decrease in bacteria that have a dampening effect on immune activation.²³

Most studies evaluating probiotics as a treatment for IBD have been small and have used a wide variety of bacterial mixtures, which makes comparisons difficult. Recent meta-analyses found combination probiotics can both induce and maintain remission in patients with ulcerative colitis, but have no beneficial effects in Crohn’s disease.¹⁰ In a review of 9 case series of patients with IBD, fecal transplant reduced IBD symptoms, and patients were able to decrease medication use.²⁴

Diarrheal illness. The human intestine is protected from diarrheal illness by healthy bacteria that block the actions of pathogenic bacteria. This mechanism is called colonization resistance. Moderate levels of evidence support the use of probiotics to prevent or treat several types of diarrheal illness.¹⁴

Antibiotic-associated diarrhea (AAD) is caused when antibiotic use alters the microbial balance. Recent meta-analyses have shown probiotics can prevent AAD and Clostridium difficile-associated diarrhea.^11,12 Several case series and one RCT have found that fecal transplants are safe and efficacious for treating recurrent Clostridium difficile infection.²⁵ Using probiotics to treat symptoms of AAD has been less studied.

Acute infectious diarrhea and traveler’s diarrhea (TD). A Cochrane review found that probiotics decreased the duration of diarrheal episodes by 25 hours, decreased the risk of an episode lasting more than 4 days by 59%, and led to one less diarrheal stool per day by the second day of the intervention.¹³ In a separate meta-analysis of 12 studies, probiotics significantly prevented 85% of cases of TD.¹⁴

Encouraging early evidence for several other illnesses

Metabolic disorders. Both animal and human studies support the theory that gut flora contribute to energy homeostasis, and in some genetically predisposed people dysbiosis may lead to obesity and diabetes. The traditional western diet⁴ and possibly decreased physical activity²⁶ are major contributors to gut flora dysbiosis. Healthy bacteria in the gut break down soluble fiber into short chain fatty acids (SCFAs). SCFAs are associated with increased satiety, decreased food intake, lower levels of inflammation, and improvement in insulin signaling in adipose tissue. In addition to decreased SFCA production, dysbiosis also leads to increased lipid deposition through higher levels of lipoprotein lipase.²⁷

Obesity. The bacteria in our gut affect energy metabolism. In patients with obesity, increased amounts of bacteria in the taxa Firmicutes and a corresponding decrease in Bacteroidetes is associated with an increased energy harvest and decreased SCFA production, which leads to a pro-inflammatory state.²⁸ Probiotics that contain Bifidobacterium and Lactobacillus are thought to help correct this dysbiosis by increasing production of SCFAs.²⁸

A recent meta-analysis of 4 RCTs found no significant difference between supplementation with probiotics and placebo on weight reduction.²⁹ However, lower-quality studies with more subjects and longer duration have shown a statistically significant improvement in weight reduction with probiotic use compared to placebo.²⁹

The evidence suggests that fecal transplants are safe and efficacious for treating recurrent Clostridium difficile infection.

Diabetes. Although dietary interventions to improve glycemic control have long been an important cornerstone of treatment, probiotic supplementation to further alter gut flora composition is also being evaluated. Studies have found probiotics have largely beneficial effects on glycemic control, especially in animals. The largest systematic review to date looked at 33 studies, including 5 human trials. The human studies each found a significant reduction in at least one of 6 parameters of glycemic control (levels of fasting plasma glucose, postprandial blood glucose, glycated hemoglobin, insulin, insulin resistance, and onset of diabetes).¹⁶ It is unclear which probiotic strains confer benefit, and if those benefits are sustainable without dietary modification and increased physical activity.

Psychiatric illnesses. The gut-brain axis is thought to impact mental health by several mechanisms, including modulating the hypothalamic-pituitary-adrenal axis, activating the immune system, producing active metabolites, and affecting the vagus nerve. It is unclear which of these pathways may be clinically relevant.^5,30 The few human studies that have looked for a potential link between gut flora and psychiatric illness have focused on depression and autism spectrum disorders (ASD).

Depression. Small studies comparing the microbiome composition of depressed patients vs healthy controls have found differences in patterns of both over- and underrepresented microbiota species in depressed patients, although the patterns across studies have been inconsistent.^31,32 One small functional magnetic resonance imaging study of healthy women showed that a fermented milk product that contained probiotics affected activity in areas of the brain that control emotion and sensation.³³ A few small studies have shown that patients who used probiotics had improved depression scores.³⁴ Further studies are needed.

ASD. Children with ASD have GI disturbances—most commonly diarrhea, constipation, and/or bloating—more often than healthy controls.^35,36 This association has led to speculation of a connection between the gut and brain. The microbial composition and diversity appears to be different in individuals with ASD; several studies have found an increase in Clostridia species.³⁷

Research on probiotics for treating ASD has been primarily in preclinical models. Human studies of probiotics for ASD are lacking.³⁸ Small studies on dietary modifications such as gluten-free and casein-free diets have had varying results; to what extent these dietary changes exert their influence via the intestinal microbiome is unknown.³⁸

Eczema. Several studies have looked at the role of prebiotics and probiotics in reducing the risk for allergic disease. A 2013 Cochrane review found strong evidence that certain prebiotics can prevent eczema in children under age 2.¹⁵ There is limited evidence that probiotics may also play a role in preventing eczema.^39,40 However, probiotics do not appear to be effective for treating eczema.⁴¹

Several studies have found a link between the use of probiotics and significant reductions in at least one of 6 parameters of glycemic control.

Rheumatoid arthritis (RA). Patients with RA have a change in the balance of function of different T helper cells subsets, and several studies have shown that changes in the gut microbiome can affect this balance.⁴² A recent small study of patients with RA found that 75% of those with new onset RA had Prevotella copri bacteria as the predominant species, and patients with chronic RA had a decrease in Bacteroides species compared to healthy counterparts.^42-44 The exact influence of gut flora dysbiosis on RA is unknown.⁴⁵ Small studies suggest dietary changes may improve RA symptoms, while data on the use of probiotics to alleviate symptoms is mixed.⁴⁶

What to tell patients about gut flora and health

There is increasing evidence that the gut microbiome and the genes contained therein have an impact on an individual’s health. (See TABLE 2 for additional resources.) The best preventive advice for patients and their families is to eat a diet rich in fruits and vegetables. This measure has well proven benefits beyond its potential effects on gut flora.

Correcting dysbiosis with diet or probiotics may play a role in treating chronic conditions; however, in many cases, further research is required to elucidate specific recommendations. In the meantime, given the safety profile of probiotics and dietary fiber, it is reasonable to consider using these interventions, particularly probiotics for treating IBS, ulcerative colitis, and acute infectious diarrhea; probiotics for preventing antibiotic-associated diarrhea and traveler’s diarrhea; and prebiotics for preventing eczema in high-risk infants.

CORRESPONDENCE
Jill Schneiderhan, MD, Family Medicine at Domino’s Farms, 24 Frank Lloyd Wright Dr., Lobby H, Suite 2300, Ann Arbor, MI 48105; jillsch@umich.edu.

CASE 1 › Sheila S, age 27, has irritable bowel syndrome (IBS) and comes to your office for a follow-up visit. Over the past 6 months she has started taking a fiber supplement, drinking more water, and looking for links between stress and her symptoms. She has read about probiotics and wonders if you would consider recommending them in her situation.

CASE 2 › Mark M, age 45, has type 2 diabetes and is overweight. He is motivated to change his diet and has started to exercise more. He is taking metformin 2000 mg/d but his hemoglobin A1c remains slightly elevated at 7.2%. He heard on television that probiotics might help to keep him from needing to add another medication.

Most of the living organisms that comprise the human microbiome—all of the microbes that live on or in humans—are found in the gastrointestinal (GI) tract. The gut flora contribute 99% of the genetic material in the human body. The composition of the gut flora is remarkably diverse across the population; each individual has a unique microbial footprint. Within this microbial diversity, there appears to be a stable number of genes that are responsible for the major functions of the gut flora.¹ These microbes:

• supply essential nutrients by breaking down complex carbohydrates;
• generate secondary bile acids that assist in digesting fats;²
• synthesize vitamins such as K, B12, folate, and biotin;³
• contribute to the defensive barrier in the colon by keeping pathogenic bacteria from crossing the colonic mucosa; and
• interact with our systemic immune system in a way that maintains a level of homeostasis, allowing for appropriate activation in the face of pathogens without developing autoimmunity.⁴

The gut flora also play a role in the communication between the central nervous system and the enteric nervous system by modulating the hormonal and neural pathways that have been labeled the “gut-brain axis.” The gut-brain axis has been associated with numerous disease states, including irritable bowel syndrome and certain psychiatric disorders.⁵

Researchers are investigating interventions that target the microbiome to increase microbial diversity and the presence of certain species to prevent or treat various diseases. The use of probiotics and dietary changes to increase intake of soluble fiber have been the most studied of these interventions. The thought is that these interventions can correct an imbalance, or dysbiosis, of the gut flora.⁶ Studies have shown that decreased microbial diversity is associated with elevations of certain disease markers (eg, adiposity, insulin, triglycerides, C–reactive protein)⁷ and that increases in soluble fiber lead to the greatest long-term improvement in microbial diversity.⁸ Fecal transplant—the transfer of a processed mixture of stool that contains “healthy” bacteria from a donor into the intestines of a patient—is being explored as a method of replacing colonic gut flora, but evidence is limited.

The following review takes a closer look at these options and identifies those that are most likely to benefit patients in the treatment—and prevention—of several diseases (TABLE 1).^9-16

Evidence is best for using probiotics for digestive diseases

Dietary interventions for digestive diseases have long been studied, but are getting renewed attention for their potential impact on the microbiome.¹⁷ Beyond dietary modification, other similar treatment options include probiotics (live microorganisms thought to confer a beneficial effect on the host), prebiotics (non-digestible food ingredients, including oligosaccharides and inulin, thought to promote the growth of “helpful” gut flora), and synbiotics (combinations of the 2).¹⁸

Irritable bowel syndrome (IBS) is a heterogeneous disorder characterized by altered intestinal transit, low-grade colonic inflammation, and/or alterations in the gutbrain axis. Research has increasingly focused on recently discovered increases in intestinal immune activation, intestinal permeability, and alterations in the colonic microbiome (decreased diversity and increased pathogenic bacteria) associated with IBS.¹⁹

Meta-analyses have found that combination probiotics benefit patients with ulcerative colitis, but not those with Crohn’s disease.

A meta-analysis of 43 randomized control trials (RCTs) found probiotics ranging from Lactobacillus to Saccharomyces can significantly decrease global IBS symptoms, abdominal pain, bloating, and flatulence.⁹ For a patient such as Ms. S, the evidence suggests a probiotic that contains a mixture of Lactobacillus and Bifidobacterium might help relieve her symptoms.⁹ In terms of dietary modifications, soluble fiber, which is already known to help treat IBS,²⁰ has profound effects on improving microbiota diversity and in shifting the composition toward less pathogenic strains.²¹ The Institute of Medicine's daily recommended intake of soluble fiber is about 15 g/d.²²

Inflammatory bowel disease (IBD) is caused by inflammation of the GI lining due to an overactive immune response. Evidence shows that patients with IBD have an altered microbial composition—specifically, an increase in bacteria that produce pro-inflammatory molecules and a decrease in bacteria that have a dampening effect on immune activation.²³

Most studies evaluating probiotics as a treatment for IBD have been small and have used a wide variety of bacterial mixtures, which makes comparisons difficult. Recent meta-analyses found combination probiotics can both induce and maintain remission in patients with ulcerative colitis, but have no beneficial effects in Crohn’s disease.¹⁰ In a review of 9 case series of patients with IBD, fecal transplant reduced IBD symptoms, and patients were able to decrease medication use.²⁴

Diarrheal illness. The human intestine is protected from diarrheal illness by healthy bacteria that block the actions of pathogenic bacteria. This mechanism is called colonization resistance. Moderate levels of evidence support the use of probiotics to prevent or treat several types of diarrheal illness.¹⁴

Antibiotic-associated diarrhea (AAD) is caused when antibiotic use alters the microbial balance. Recent meta-analyses have shown probiotics can prevent AAD and Clostridium difficile-associated diarrhea.^11,12 Several case series and one RCT have found that fecal transplants are safe and efficacious for treating recurrent Clostridium difficile infection.²⁵ Using probiotics to treat symptoms of AAD has been less studied.

Acute infectious diarrhea and traveler’s diarrhea (TD). A Cochrane review found that probiotics decreased the duration of diarrheal episodes by 25 hours, decreased the risk of an episode lasting more than 4 days by 59%, and led to one less diarrheal stool per day by the second day of the intervention.¹³ In a separate meta-analysis of 12 studies, probiotics significantly prevented 85% of cases of TD.¹⁴

Encouraging early evidence for several other illnesses

Metabolic disorders. Both animal and human studies support the theory that gut flora contribute to energy homeostasis, and in some genetically predisposed people dysbiosis may lead to obesity and diabetes. The traditional western diet⁴ and possibly decreased physical activity²⁶ are major contributors to gut flora dysbiosis. Healthy bacteria in the gut break down soluble fiber into short chain fatty acids (SCFAs). SCFAs are associated with increased satiety, decreased food intake, lower levels of inflammation, and improvement in insulin signaling in adipose tissue. In addition to decreased SFCA production, dysbiosis also leads to increased lipid deposition through higher levels of lipoprotein lipase.²⁷

Obesity. The bacteria in our gut affect energy metabolism. In patients with obesity, increased amounts of bacteria in the taxa Firmicutes and a corresponding decrease in Bacteroidetes is associated with an increased energy harvest and decreased SCFA production, which leads to a pro-inflammatory state.²⁸ Probiotics that contain Bifidobacterium and Lactobacillus are thought to help correct this dysbiosis by increasing production of SCFAs.²⁸

A recent meta-analysis of 4 RCTs found no significant difference between supplementation with probiotics and placebo on weight reduction.²⁹ However, lower-quality studies with more subjects and longer duration have shown a statistically significant improvement in weight reduction with probiotic use compared to placebo.²⁹

The evidence suggests that fecal transplants are safe and efficacious for treating recurrent Clostridium difficile infection.

Diabetes. Although dietary interventions to improve glycemic control have long been an important cornerstone of treatment, probiotic supplementation to further alter gut flora composition is also being evaluated. Studies have found probiotics have largely beneficial effects on glycemic control, especially in animals. The largest systematic review to date looked at 33 studies, including 5 human trials. The human studies each found a significant reduction in at least one of 6 parameters of glycemic control (levels of fasting plasma glucose, postprandial blood glucose, glycated hemoglobin, insulin, insulin resistance, and onset of diabetes).¹⁶ It is unclear which probiotic strains confer benefit, and if those benefits are sustainable without dietary modification and increased physical activity.

Psychiatric illnesses. The gut-brain axis is thought to impact mental health by several mechanisms, including modulating the hypothalamic-pituitary-adrenal axis, activating the immune system, producing active metabolites, and affecting the vagus nerve. It is unclear which of these pathways may be clinically relevant.^5,30 The few human studies that have looked for a potential link between gut flora and psychiatric illness have focused on depression and autism spectrum disorders (ASD).

Depression. Small studies comparing the microbiome composition of depressed patients vs healthy controls have found differences in patterns of both over- and underrepresented microbiota species in depressed patients, although the patterns across studies have been inconsistent.^31,32 One small functional magnetic resonance imaging study of healthy women showed that a fermented milk product that contained probiotics affected activity in areas of the brain that control emotion and sensation.³³ A few small studies have shown that patients who used probiotics had improved depression scores.³⁴ Further studies are needed.

ASD. Children with ASD have GI disturbances—most commonly diarrhea, constipation, and/or bloating—more often than healthy controls.^35,36 This association has led to speculation of a connection between the gut and brain. The microbial composition and diversity appears to be different in individuals with ASD; several studies have found an increase in Clostridia species.³⁷

Research on probiotics for treating ASD has been primarily in preclinical models. Human studies of probiotics for ASD are lacking.³⁸ Small studies on dietary modifications such as gluten-free and casein-free diets have had varying results; to what extent these dietary changes exert their influence via the intestinal microbiome is unknown.³⁸

Eczema. Several studies have looked at the role of prebiotics and probiotics in reducing the risk for allergic disease. A 2013 Cochrane review found strong evidence that certain prebiotics can prevent eczema in children under age 2.¹⁵ There is limited evidence that probiotics may also play a role in preventing eczema.^39,40 However, probiotics do not appear to be effective for treating eczema.⁴¹

Several studies have found a link between the use of probiotics and significant reductions in at least one of 6 parameters of glycemic control.

Rheumatoid arthritis (RA). Patients with RA have a change in the balance of function of different T helper cells subsets, and several studies have shown that changes in the gut microbiome can affect this balance.⁴² A recent small study of patients with RA found that 75% of those with new onset RA had Prevotella copri bacteria as the predominant species, and patients with chronic RA had a decrease in Bacteroides species compared to healthy counterparts.^42-44 The exact influence of gut flora dysbiosis on RA is unknown.⁴⁵ Small studies suggest dietary changes may improve RA symptoms, while data on the use of probiotics to alleviate symptoms is mixed.⁴⁶

What to tell patients about gut flora and health

There is increasing evidence that the gut microbiome and the genes contained therein have an impact on an individual’s health. (See TABLE 2 for additional resources.) The best preventive advice for patients and their families is to eat a diet rich in fruits and vegetables. This measure has well proven benefits beyond its potential effects on gut flora.

Correcting dysbiosis with diet or probiotics may play a role in treating chronic conditions; however, in many cases, further research is required to elucidate specific recommendations. In the meantime, given the safety profile of probiotics and dietary fiber, it is reasonable to consider using these interventions, particularly probiotics for treating IBS, ulcerative colitis, and acute infectious diarrhea; probiotics for preventing antibiotic-associated diarrhea and traveler’s diarrhea; and prebiotics for preventing eczema in high-risk infants.

CORRESPONDENCE
Jill Schneiderhan, MD, Family Medicine at Domino’s Farms, 24 Frank Lloyd Wright Dr., Lobby H, Suite 2300, Ann Arbor, MI 48105; jillsch@umich.edu.