Gastroesophageal reflux disease (GERD) is a chronic and common medical problem, with up to 40% of the population experiencing its symptoms at least once per month.¹ The condition develops when the reflux of stomach contents causes troublesome symptoms or complications.²

GERD symptoms can range from heartburn and regurgitation to cough and hoarseness. While many patients’ symptoms respond to medical treatment, the diagnosis and treatment in those whose symptoms do not respond to a proton pump inhibitor (PPI) may be challenging.

This article reviews the diagnosis and treatment options for GERD.

SYMPTOMS: TYPICAL, ATYPICAL, AND ALARM

Symptoms of GERD (Table 1) can be classified as typical (heartburn and regurgitation) or atypical (cough, asthma, hoarseness, chronic laryngitis, throat-clearing, chest pain, dyspepsia, and nausea). Atypical symptoms are more likely to be due to GERD if patients also have typical symptoms and if the symptoms respond to a trial of a PPI.³

Alarm symptoms. Keep in mind that extraesophageal presentations may be multifactorial, and it may be difficult to establish that reflux, even if present, is actually the cause. While chest pain may be due to GERD, it is important to rule out cardiac chest pain before considering GERD as a cause. Similarly, dysphagia along with typical or atypical symptoms warrants investigation for potential complications such as underlying motility disorder, esophageal stricture, esophageal ring, or malignancy.⁴ Other alarm symptoms include odynophagia, bleeding, weight loss, and anemia.

DIAGNOSING GERD: RESPONSE TO A PPI IS DIAGNOSTIC

Patients with typical symptoms that respond to PPI therapy need no further evaluation for a diagnosis of GERD to be made.⁵ On the other hand, further testing should be undertaken in patients with typical symptoms that do not respond to PPI therapy, in patients presenting with atypical symptoms, and in patients in whom antireflux surgery is being considered. Figure 1 shows our proposed algorithm.

Try a PPI for 6–8 weeks

Relief of heartburn and regurgitation after a 6- to 8-week course of a PPI strongly suggests GERD.⁶ However, a negative trial of a PPI does not rule out GERD, as this approach has been found to have a sensitivity of 78% and specificity of 54%.⁶

Despite this limitation, a trial of PPI therapy should be offered to patients presenting with typical symptoms and no alarm features. This approach has been found to be more cost-effective than proceeding directly to diagnostic testing.⁷

Endoscopy

Endoscopic findings in GERD can include erosive esophagitis, peptic stricture, and Barrett esophagus. Esophageal erosions are a highly specific sign of GERD; the Los Angeles classification system, a standardized scale for grading the severity of erosive esophagitis (from A to D, with D the most severe) provides an objective way to assess severity.⁸ However, most patients with heartburn and regurgitation do not have erosive disease, thus limiting the sensitivity of upper endoscopy as an initial diagnostic test in patients with suspected GERD.⁹

We recommend endoscopy for patients who present with alarm symptoms, patients with noncardiac chest pain, PPI nonresponders, and patients with chronic GERD symptoms and multiple risk factors for Barrett esophagus besides GERD, such as older age, male sex, white race, overweight, and smoking.¹⁰

Ambulatory pH and impedance monitoring

Ambulatory pH monitoring is the gold standard test for pathologic acid exposure in the esophagus. pH testing is indicated in PPI nonresponders, patients presenting with atypical symptoms, and before antireflux surgery.

In general, pH testing should be performed after the patient has been off PPI therapy for at least 7 days, as the test is highly unlikely to be abnormal while a patient is on a PPI.¹¹ It is done either with a transnasal catheter for 24 hours, or with a wireless capsule (Bravo pH System, Given Imaging Ltd, Duluth, GA), which collects 48 to 96 hours of data. Studies of the wireless system have shown that its sensitivity increases 12% to 25% when it is performed for 48 hours compared with 24 hours.^12,13

The pH test can be combined with impedance testing to evaluate for nonacid reflux.¹⁴ However, the clinical significance of nonacid reflux remains controversial, and for this reason the Esophageal Diagnostic Advisory Panel recommends that the decision to perform antireflux surgery should not be based on abnormal impedance testing.¹⁵

Ambulatory pH monitoring is the gold-standard test

During pH and impedance testing, special software can calculate how closely the patient’s symptoms correlate with esophageal acid exposure. The symptom index (SI) and symptom association probability (SAP) are the symptom measurements most commonly used in practice. The SI measures the overall strength of the relationship, and an SI greater than 50% is considered positive.¹⁶ The SAP determines whether this relationship could have occurred by chance, and an SAP greater than 95% is statistically significant.¹⁷ In patients with normal levels of esophageal acid exposure, an elevated SI or SAP may indicate a component of esophageal hypersensitivity in symptom generation.

At our institution, we generally perform pH-only transnasal or wireless testing off PPI therapy to establish that the patient has pathologic acid exposure in the distal esophagus. Combined pH-impedance testing is typically reserved for patients with atypical symptoms unresponsive to PPI therapy and abnormal results on previous pH testing, which allows for correlation of nonacid reflux and symptoms.

Other tests

Esophageal manometry and barium esophagography have limited value in the primary diagnosis of GERD. However, they should be considered to rule out achalasia and other esophageal motility disorders in patients whose symptoms do not respond to PPIs. For this reason, esophageal manometry should be performed before considering antireflux surgery.

MANAGING GERD

Table 2 summarizes the various treatments for GERD.

Lifestyle modifications

Lifestyle modifications are the first-line therapy for GERD. Modifications that have been studied include weight loss, head-of-bed elevation, and avoidance of tobacco, alcohol, and late-night meals. Another modification that has been suggested is avoiding foods that can aggravate reflux symptoms—eg, caffeine, coffee, chocolate, spicy foods, highly acidic foods (oranges, tomatoes), and fatty foods. Of these, only weight loss and head-of-bed elevation have been proven effective.¹⁸

Three randomized controlled trials demonstrated that GERD symptoms and esophageal pH values improved with head-of-bed elevation using blocks or incline foam wedges.^19–21 Several cohort studies demonstrated reduction in GERD symptoms with weight loss.^22,23 Recently, a prospective cohort study also found that smoking cessation significantly improved GERD symptoms in patients with normal body mass index and severe symptoms.²⁴

Antacids

Several antacids (eg, sodium bicarbonate, calcium carbonate, magnesium hydroxide, aluminum hydroxide) are available over the counter.

Antacids were thought to control heartburn symptoms by increasing the pH of gastric contents that might subsequently reflux into the esophagus. However, well-controlled studies have shown that they relieve heartburn by neutralizing acid in the esophagus, with no significant effect on gastric pH.^25,26

Antacids provide rapid but short-lived relief from an existing episode of heartburn. Because they do not significantly raise the gastric pH, they do not prevent subsequent reflux episodes from repeatedly exposing the esophagus to gastric acid and causing heartburn. Additionally, antacids have not been shown to contribute to healing of erosive esophagitis.²⁷ Hence, they may not be optimal for treating frequent reflux heartburn.

Sodium alginate

Gastric acid pockets are unbuffered pools of acid that float on top of ingested food.²⁸ They develop as a result of poor mixing of newly secreted acid and food in the proximal stomach, which remains relatively quiescent after a meal compared with the distal stomach.²⁹ In GERD, proximal extension of the acid pocket above the diaphragm increases the risk of acid reflux.³⁰ The acid pocket is therefore an important source of postprandial acid in GERD and, as such, represents a unique therapeutic target.

Emerging evidence suggests that alginates may act directly on the acid pocket. Alginates are natural polysaccharide polymers that, on contact with gastric acid, precipitate within minutes into a low-density viscous gel of near-neutral pH. The change in pH triggers the sodium bicarbonate in the formulation to release carbon dioxide that becomes trapped in the alginate gel, causing it to float to the top of the gastric contents like a raft.³¹

A randomized controlled trial demonstrated that sodium alginate was as effective as omeprazole in relieving symptoms in patients with nonerosive reflux disease.³² Alginate has also been shown to provide more postprandial reflux relief than antacids.³³

Histamine-2 receptor antagonists

Histamine-2 receptor antagonists act more swiftly and increase postprandial gastric pH more rapidly than PPIs, thus making them a good option for prophylaxis against postprandial GERD.³⁴ Taking these drugs at bedtime may help in patients with objective nighttime reflux despite optimal PPI use. However, tachyphylaxis may occur as early as 1 week after starting combination therapy.³⁵

Proton pump inhibitors

There are currently seven available PPIs, including four that can be obtained over the counter (omeprazole, lansoprazole, esomeprazole, and omeprazole-sodium bicarbonate) and three available only by prescription (rabeprazole, pantoprazole, and dexlansoprazole). Studies have shown than a standard 6- to 8-week course of a PPI provides complete symptom relief in 70% to 80% of patients with erosive reflux disease and in 60% of patients with nonerosive reflux disease.^36,37 Clinically, PPIs all appear to be similar in their symptom relief.³⁸

Most PPIs should be taken 30 to 60 minutes before meals. Exceptions are omeprazole-sodium bicarbonate and dexlansoprazole, which can be taken without regard to meals. At our institution, we usually start PPIs in a once-daily standard dose for 6 to 8 weeks and consider increasing to twice-daily dosing if symptoms do not respond completely. Patients with mild intermittent GERD symptoms may benefit from “on-demand” use of PPIs. This approach is best suited for patients with nonerosive reflux disease without evidence of Barrett esophagus on endoscopy.

Safety and adverse effects of PPIs

PPIs are generally safe but can cause adverse effects (Table 3).

Osteoporosis. In 2010, the US Food and Drug Administration issued warnings regarding the potential for wrist, hip, and spine fractures in PPI users.²⁶ Most recent evidence suggests that PPIs may be associated with a small increase in risk of hip fractures in patients already at high risk.^39,40 However, the 2013 American College of Gastroenterology (ACG) guidelines say that patients with known osteoporosis can remain on PPI therapy, and concern for hip fractures and osteoporosis should not affect the decision to use PPIs long-term except in patients with other risk factors for hip fracture.⁴¹

Community-acquired pneumonia. An increased risk of community-acquired pneumonia cannot be clearly documented in association with PPI therapy. Multiple studies, including randomized controlled trials, investigated this potential correlation. However, evidence suggests that short-term but not long-term PPI use may be associated with an overall increased risk of community-acquired pneumonia.^42,43 Current guidelines suggest that in patients who need a PPI, the drug should not be withheld on the basis of a potential risk of community-acquired pneumonia.⁴¹

Clostridium difficile infection. In theory, PPIs may increase the risk of C difficile infection by increasing the ability of the spore to convert to the vegetative form and to survive intraluminally. In fact, studies and meta-analyses have suggested that PPIs do increase the risk of development and recurrence of C difficile infection.^44,45 Therefore, PPIs should be used with care in patients who are at risk.⁴¹

Interaction with clopidogrel. The antiplatelet activity of clopidogrel requires activation by CYP2C19, the same pathway required for metabolism of some PPIs. Concern was raised about decreased antiplatelet activity of clopidogrel in the presence of PPIs. This was extensively studied, and there now appears to be no increased risk of adverse cardiovascular events in patients on PPIs, based on data from well-controlled randomized trials.^46,47 A consensus panel of the American College of Cardiology Foundation, the American Heart Association, and the ACG said that PPIs may be used for appropriate indications in patients taking clopidogrel.⁴⁷

Hypomagnesemia. By an unknown molecular mechanism, PPIs are thought to reduce intestinal magnesium absorption, leading to hypomagnesemia. A meta-analysis published in 2011 showed that PPI-induced hypomagnesemia is a drug-class effect and occurred after a median of 5.5 years of PPI use. Stopping the PPI resulted in magnesium recovery in 4 days, and rechallenge led to recurrence within 4 days.⁴⁸

Hence, to avoid putting patients on long-term PPI therapy at risk, clinicians should anticipate this problem. Our practice is to check the magnesium level before starting a patient on long-term PPI therapy, and then to repeat the measurement every 1 to 2 years.

Baclofen

Transient lower esophageal sphincter relaxation has been shown to be a cause of reflux in healthy people and in patients with GERD.⁴⁹

Baclofen, a muscle relaxant with selective gamma-aminobutyric acid receptor class B agonist properties, reduces transient lower esophageal sphincter relaxation in humans.⁵⁰ In a well-designed, double-blind, randomized controlled trial, baclofen was associated with a significant decrease in upright reflux on 24-hour pH monitoring and significant improvement in belching and overall reflux symptoms.⁵¹ However, baclofen is not approved by the US Food and Drug Administration for the treatment of GERD, and its use may be limited by side effects such as somnolence and dizziness.

Antireflux surgery

Antireflux surgery is a reasonable option for selected patients with chronic GERD. The main types of surgery are laparoscopic fundoplication and, for obese patients, gastric bypass. Reasons to consider antireflux surgery include desire to stop PPI therapy, esophagitis not healed by PPIs, symptomatic hiatal hernia, and refractory reflux documented by pH testing.⁴¹

Clinically, all PPIs appear to be similar in their symptom relief

In general, surgical therapy may be considered in patients who respond to PPIs, but patients who do not respond to PPIs are less likely to respond to antireflux surgery.¹⁵ Other patients less likely to respond are those with symptoms of dyspepsia, such as nausea, vomiting, and epigastric pain.⁴¹

Common adverse effects of antireflux surgery include gas-bloat syndrome (up to 85% of patients), dysphagia (10% to 50% of patients), diarrhea (18% to 33% of patients), and recurrent heartburn (10% to 62% of patients).⁵²

Endoscopic and minimally invasive antireflux procedures include endoscopic plication of the lower esophageal sphincter, radiofrequency augmentation of the lower esophageal sphincter, and sphincter augmentation by a string of titanium beads. While some have shown promise, they are not recommended by the most recent ACG guidelines, given lack of long-term data.⁴¹

REFRACTORY GERD

There is no consensus on the definition of refractory GERD. However, for the sake of simplicity, we can define it as persistence of suspected GERD symptoms despite treatment with a PPI. This may vary from a partial response to PPI therapy to a complete absence of response.

It is extremely important to rule out non-GERD causes of the ongoing symptoms, such as achalasia, gastroparesis, eosinophilic esophagitis, rumination, and aerophagia. PPI nonresponders are more likely to be obese, poorly compliant, and have extraesophageal symptoms.^53–56 As previously discussed, PPIs should be taken 30 to 60 minutes before meals. For patients whose symptoms fail to respond to standard-dose daily PPI therapy, switching to another PPI or doubling the dose is common, although data to support this practice are limited. Of note, omeprazole-sodium bicarbonate has been shown to provide more symptom relief in nocturnal GERD.⁵⁷ Additionally, adding a nighttime histamine-2 receptor antagonist may also help in patients with objective nighttime reflux.⁴¹

After noncompliance and suboptimal PPI dosing have been ruled out, PPI nonresponders with typical symptoms should undergo upper endoscopy and subsequent pH monitoring. Normal esophageal acid exposure on pH testing suggests functional heartburn or functional dyspepsia. Negative pH testing in a patient with atypical symptoms suggests a non-GERD cause of symptoms, and referral to an otolaryngologist, pulmonologist, or allergist is often warranted.

While antireflux surgery can be considered for patients with nonacid reflux on impedance testing, it should again be noted that GERD in patients with no response to PPIs is less likely to respond to antireflux surgery.¹⁵

TAKE-HOME POINTS

• GERD is a common medical condition, affecting up to 40% of US adults at least once monthly.
• GERD can result in a wide variety of symptoms, including typical heartburn and regurgitation as well as atypical symptoms such as cough.
• On the other hand, keep in mind that multiple non-GERD causes of heartburn and regurgitation may exist.
• Testing for GERD includes endoscopy and pH testing as well as functional testing such as esophageal manometry.
• While in most patients GERD will respond to lifestyle changes and antisecretory therapy such as a PPI, careful attention must be given to patients with symptoms refractory to PPI therapy.
• For a subset of patients, antireflux surgery may be a reasonable option, but care must be taken to exclude patients with a lower likelihood of responding to surgery.

Gastroesophageal reflux disease (GERD) is a chronic and common medical problem, with up to 40% of the population experiencing its symptoms at least once per month.¹ The condition develops when the reflux of stomach contents causes troublesome symptoms or complications.²

GERD symptoms can range from heartburn and regurgitation to cough and hoarseness. While many patients’ symptoms respond to medical treatment, the diagnosis and treatment in those whose symptoms do not respond to a proton pump inhibitor (PPI) may be challenging.

This article reviews the diagnosis and treatment options for GERD.

SYMPTOMS: TYPICAL, ATYPICAL, AND ALARM

Symptoms of GERD (Table 1) can be classified as typical (heartburn and regurgitation) or atypical (cough, asthma, hoarseness, chronic laryngitis, throat-clearing, chest pain, dyspepsia, and nausea). Atypical symptoms are more likely to be due to GERD if patients also have typical symptoms and if the symptoms respond to a trial of a PPI.³

Alarm symptoms. Keep in mind that extraesophageal presentations may be multifactorial, and it may be difficult to establish that reflux, even if present, is actually the cause. While chest pain may be due to GERD, it is important to rule out cardiac chest pain before considering GERD as a cause. Similarly, dysphagia along with typical or atypical symptoms warrants investigation for potential complications such as underlying motility disorder, esophageal stricture, esophageal ring, or malignancy.⁴ Other alarm symptoms include odynophagia, bleeding, weight loss, and anemia.

DIAGNOSING GERD: RESPONSE TO A PPI IS DIAGNOSTIC

Patients with typical symptoms that respond to PPI therapy need no further evaluation for a diagnosis of GERD to be made.⁵ On the other hand, further testing should be undertaken in patients with typical symptoms that do not respond to PPI therapy, in patients presenting with atypical symptoms, and in patients in whom antireflux surgery is being considered. Figure 1 shows our proposed algorithm.

Try a PPI for 6–8 weeks

Relief of heartburn and regurgitation after a 6- to 8-week course of a PPI strongly suggests GERD.⁶ However, a negative trial of a PPI does not rule out GERD, as this approach has been found to have a sensitivity of 78% and specificity of 54%.⁶

Despite this limitation, a trial of PPI therapy should be offered to patients presenting with typical symptoms and no alarm features. This approach has been found to be more cost-effective than proceeding directly to diagnostic testing.⁷

Endoscopy

Endoscopic findings in GERD can include erosive esophagitis, peptic stricture, and Barrett esophagus. Esophageal erosions are a highly specific sign of GERD; the Los Angeles classification system, a standardized scale for grading the severity of erosive esophagitis (from A to D, with D the most severe) provides an objective way to assess severity.⁸ However, most patients with heartburn and regurgitation do not have erosive disease, thus limiting the sensitivity of upper endoscopy as an initial diagnostic test in patients with suspected GERD.⁹

We recommend endoscopy for patients who present with alarm symptoms, patients with noncardiac chest pain, PPI nonresponders, and patients with chronic GERD symptoms and multiple risk factors for Barrett esophagus besides GERD, such as older age, male sex, white race, overweight, and smoking.¹⁰

Ambulatory pH and impedance monitoring

Ambulatory pH monitoring is the gold standard test for pathologic acid exposure in the esophagus. pH testing is indicated in PPI nonresponders, patients presenting with atypical symptoms, and before antireflux surgery.

In general, pH testing should be performed after the patient has been off PPI therapy for at least 7 days, as the test is highly unlikely to be abnormal while a patient is on a PPI.¹¹ It is done either with a transnasal catheter for 24 hours, or with a wireless capsule (Bravo pH System, Given Imaging Ltd, Duluth, GA), which collects 48 to 96 hours of data. Studies of the wireless system have shown that its sensitivity increases 12% to 25% when it is performed for 48 hours compared with 24 hours.^12,13

The pH test can be combined with impedance testing to evaluate for nonacid reflux.¹⁴ However, the clinical significance of nonacid reflux remains controversial, and for this reason the Esophageal Diagnostic Advisory Panel recommends that the decision to perform antireflux surgery should not be based on abnormal impedance testing.¹⁵

Ambulatory pH monitoring is the gold-standard test

During pH and impedance testing, special software can calculate how closely the patient’s symptoms correlate with esophageal acid exposure. The symptom index (SI) and symptom association probability (SAP) are the symptom measurements most commonly used in practice. The SI measures the overall strength of the relationship, and an SI greater than 50% is considered positive.¹⁶ The SAP determines whether this relationship could have occurred by chance, and an SAP greater than 95% is statistically significant.¹⁷ In patients with normal levels of esophageal acid exposure, an elevated SI or SAP may indicate a component of esophageal hypersensitivity in symptom generation.

At our institution, we generally perform pH-only transnasal or wireless testing off PPI therapy to establish that the patient has pathologic acid exposure in the distal esophagus. Combined pH-impedance testing is typically reserved for patients with atypical symptoms unresponsive to PPI therapy and abnormal results on previous pH testing, which allows for correlation of nonacid reflux and symptoms.

Other tests

Esophageal manometry and barium esophagography have limited value in the primary diagnosis of GERD. However, they should be considered to rule out achalasia and other esophageal motility disorders in patients whose symptoms do not respond to PPIs. For this reason, esophageal manometry should be performed before considering antireflux surgery.

MANAGING GERD

Table 2 summarizes the various treatments for GERD.

Lifestyle modifications

Lifestyle modifications are the first-line therapy for GERD. Modifications that have been studied include weight loss, head-of-bed elevation, and avoidance of tobacco, alcohol, and late-night meals. Another modification that has been suggested is avoiding foods that can aggravate reflux symptoms—eg, caffeine, coffee, chocolate, spicy foods, highly acidic foods (oranges, tomatoes), and fatty foods. Of these, only weight loss and head-of-bed elevation have been proven effective.¹⁸

Three randomized controlled trials demonstrated that GERD symptoms and esophageal pH values improved with head-of-bed elevation using blocks or incline foam wedges.^19–21 Several cohort studies demonstrated reduction in GERD symptoms with weight loss.^22,23 Recently, a prospective cohort study also found that smoking cessation significantly improved GERD symptoms in patients with normal body mass index and severe symptoms.²⁴

Antacids

Several antacids (eg, sodium bicarbonate, calcium carbonate, magnesium hydroxide, aluminum hydroxide) are available over the counter.

Antacids were thought to control heartburn symptoms by increasing the pH of gastric contents that might subsequently reflux into the esophagus. However, well-controlled studies have shown that they relieve heartburn by neutralizing acid in the esophagus, with no significant effect on gastric pH.^25,26

Antacids provide rapid but short-lived relief from an existing episode of heartburn. Because they do not significantly raise the gastric pH, they do not prevent subsequent reflux episodes from repeatedly exposing the esophagus to gastric acid and causing heartburn. Additionally, antacids have not been shown to contribute to healing of erosive esophagitis.²⁷ Hence, they may not be optimal for treating frequent reflux heartburn.

Sodium alginate

Gastric acid pockets are unbuffered pools of acid that float on top of ingested food.²⁸ They develop as a result of poor mixing of newly secreted acid and food in the proximal stomach, which remains relatively quiescent after a meal compared with the distal stomach.²⁹ In GERD, proximal extension of the acid pocket above the diaphragm increases the risk of acid reflux.³⁰ The acid pocket is therefore an important source of postprandial acid in GERD and, as such, represents a unique therapeutic target.

Emerging evidence suggests that alginates may act directly on the acid pocket. Alginates are natural polysaccharide polymers that, on contact with gastric acid, precipitate within minutes into a low-density viscous gel of near-neutral pH. The change in pH triggers the sodium bicarbonate in the formulation to release carbon dioxide that becomes trapped in the alginate gel, causing it to float to the top of the gastric contents like a raft.³¹

A randomized controlled trial demonstrated that sodium alginate was as effective as omeprazole in relieving symptoms in patients with nonerosive reflux disease.³² Alginate has also been shown to provide more postprandial reflux relief than antacids.³³

Histamine-2 receptor antagonists

Histamine-2 receptor antagonists act more swiftly and increase postprandial gastric pH more rapidly than PPIs, thus making them a good option for prophylaxis against postprandial GERD.³⁴ Taking these drugs at bedtime may help in patients with objective nighttime reflux despite optimal PPI use. However, tachyphylaxis may occur as early as 1 week after starting combination therapy.³⁵

Proton pump inhibitors

There are currently seven available PPIs, including four that can be obtained over the counter (omeprazole, lansoprazole, esomeprazole, and omeprazole-sodium bicarbonate) and three available only by prescription (rabeprazole, pantoprazole, and dexlansoprazole). Studies have shown than a standard 6- to 8-week course of a PPI provides complete symptom relief in 70% to 80% of patients with erosive reflux disease and in 60% of patients with nonerosive reflux disease.^36,37 Clinically, PPIs all appear to be similar in their symptom relief.³⁸

Most PPIs should be taken 30 to 60 minutes before meals. Exceptions are omeprazole-sodium bicarbonate and dexlansoprazole, which can be taken without regard to meals. At our institution, we usually start PPIs in a once-daily standard dose for 6 to 8 weeks and consider increasing to twice-daily dosing if symptoms do not respond completely. Patients with mild intermittent GERD symptoms may benefit from “on-demand” use of PPIs. This approach is best suited for patients with nonerosive reflux disease without evidence of Barrett esophagus on endoscopy.

Safety and adverse effects of PPIs

PPIs are generally safe but can cause adverse effects (Table 3).

Osteoporosis. In 2010, the US Food and Drug Administration issued warnings regarding the potential for wrist, hip, and spine fractures in PPI users.²⁶ Most recent evidence suggests that PPIs may be associated with a small increase in risk of hip fractures in patients already at high risk.^39,40 However, the 2013 American College of Gastroenterology (ACG) guidelines say that patients with known osteoporosis can remain on PPI therapy, and concern for hip fractures and osteoporosis should not affect the decision to use PPIs long-term except in patients with other risk factors for hip fracture.⁴¹

Community-acquired pneumonia. An increased risk of community-acquired pneumonia cannot be clearly documented in association with PPI therapy. Multiple studies, including randomized controlled trials, investigated this potential correlation. However, evidence suggests that short-term but not long-term PPI use may be associated with an overall increased risk of community-acquired pneumonia.^42,43 Current guidelines suggest that in patients who need a PPI, the drug should not be withheld on the basis of a potential risk of community-acquired pneumonia.⁴¹

Clostridium difficile infection. In theory, PPIs may increase the risk of C difficile infection by increasing the ability of the spore to convert to the vegetative form and to survive intraluminally. In fact, studies and meta-analyses have suggested that PPIs do increase the risk of development and recurrence of C difficile infection.^44,45 Therefore, PPIs should be used with care in patients who are at risk.⁴¹

Interaction with clopidogrel. The antiplatelet activity of clopidogrel requires activation by CYP2C19, the same pathway required for metabolism of some PPIs. Concern was raised about decreased antiplatelet activity of clopidogrel in the presence of PPIs. This was extensively studied, and there now appears to be no increased risk of adverse cardiovascular events in patients on PPIs, based on data from well-controlled randomized trials.^46,47 A consensus panel of the American College of Cardiology Foundation, the American Heart Association, and the ACG said that PPIs may be used for appropriate indications in patients taking clopidogrel.⁴⁷

Hypomagnesemia. By an unknown molecular mechanism, PPIs are thought to reduce intestinal magnesium absorption, leading to hypomagnesemia. A meta-analysis published in 2011 showed that PPI-induced hypomagnesemia is a drug-class effect and occurred after a median of 5.5 years of PPI use. Stopping the PPI resulted in magnesium recovery in 4 days, and rechallenge led to recurrence within 4 days.⁴⁸

Hence, to avoid putting patients on long-term PPI therapy at risk, clinicians should anticipate this problem. Our practice is to check the magnesium level before starting a patient on long-term PPI therapy, and then to repeat the measurement every 1 to 2 years.

Baclofen

Transient lower esophageal sphincter relaxation has been shown to be a cause of reflux in healthy people and in patients with GERD.⁴⁹

Baclofen, a muscle relaxant with selective gamma-aminobutyric acid receptor class B agonist properties, reduces transient lower esophageal sphincter relaxation in humans.⁵⁰ In a well-designed, double-blind, randomized controlled trial, baclofen was associated with a significant decrease in upright reflux on 24-hour pH monitoring and significant improvement in belching and overall reflux symptoms.⁵¹ However, baclofen is not approved by the US Food and Drug Administration for the treatment of GERD, and its use may be limited by side effects such as somnolence and dizziness.

Antireflux surgery

Antireflux surgery is a reasonable option for selected patients with chronic GERD. The main types of surgery are laparoscopic fundoplication and, for obese patients, gastric bypass. Reasons to consider antireflux surgery include desire to stop PPI therapy, esophagitis not healed by PPIs, symptomatic hiatal hernia, and refractory reflux documented by pH testing.⁴¹

Clinically, all PPIs appear to be similar in their symptom relief

In general, surgical therapy may be considered in patients who respond to PPIs, but patients who do not respond to PPIs are less likely to respond to antireflux surgery.¹⁵ Other patients less likely to respond are those with symptoms of dyspepsia, such as nausea, vomiting, and epigastric pain.⁴¹

Common adverse effects of antireflux surgery include gas-bloat syndrome (up to 85% of patients), dysphagia (10% to 50% of patients), diarrhea (18% to 33% of patients), and recurrent heartburn (10% to 62% of patients).⁵²

Endoscopic and minimally invasive antireflux procedures include endoscopic plication of the lower esophageal sphincter, radiofrequency augmentation of the lower esophageal sphincter, and sphincter augmentation by a string of titanium beads. While some have shown promise, they are not recommended by the most recent ACG guidelines, given lack of long-term data.⁴¹

REFRACTORY GERD

There is no consensus on the definition of refractory GERD. However, for the sake of simplicity, we can define it as persistence of suspected GERD symptoms despite treatment with a PPI. This may vary from a partial response to PPI therapy to a complete absence of response.

It is extremely important to rule out non-GERD causes of the ongoing symptoms, such as achalasia, gastroparesis, eosinophilic esophagitis, rumination, and aerophagia. PPI nonresponders are more likely to be obese, poorly compliant, and have extraesophageal symptoms.^53–56 As previously discussed, PPIs should be taken 30 to 60 minutes before meals. For patients whose symptoms fail to respond to standard-dose daily PPI therapy, switching to another PPI or doubling the dose is common, although data to support this practice are limited. Of note, omeprazole-sodium bicarbonate has been shown to provide more symptom relief in nocturnal GERD.⁵⁷ Additionally, adding a nighttime histamine-2 receptor antagonist may also help in patients with objective nighttime reflux.⁴¹

After noncompliance and suboptimal PPI dosing have been ruled out, PPI nonresponders with typical symptoms should undergo upper endoscopy and subsequent pH monitoring. Normal esophageal acid exposure on pH testing suggests functional heartburn or functional dyspepsia. Negative pH testing in a patient with atypical symptoms suggests a non-GERD cause of symptoms, and referral to an otolaryngologist, pulmonologist, or allergist is often warranted.

While antireflux surgery can be considered for patients with nonacid reflux on impedance testing, it should again be noted that GERD in patients with no response to PPIs is less likely to respond to antireflux surgery.¹⁵

TAKE-HOME POINTS

• GERD is a common medical condition, affecting up to 40% of US adults at least once monthly.
• GERD can result in a wide variety of symptoms, including typical heartburn and regurgitation as well as atypical symptoms such as cough.
• On the other hand, keep in mind that multiple non-GERD causes of heartburn and regurgitation may exist.
• Testing for GERD includes endoscopy and pH testing as well as functional testing such as esophageal manometry.
• While in most patients GERD will respond to lifestyle changes and antisecretory therapy such as a PPI, careful attention must be given to patients with symptoms refractory to PPI therapy.
• For a subset of patients, antireflux surgery may be a reasonable option, but care must be taken to exclude patients with a lower likelihood of responding to surgery.

Gastroesophageal reflux disease (GERD) is a chronic and common medical problem, with up to 40% of the population experiencing its symptoms at least once per month.¹ The condition develops when the reflux of stomach contents causes troublesome symptoms or complications.²

GERD symptoms can range from heartburn and regurgitation to cough and hoarseness. While many patients’ symptoms respond to medical treatment, the diagnosis and treatment in those whose symptoms do not respond to a proton pump inhibitor (PPI) may be challenging.

This article reviews the diagnosis and treatment options for GERD.

SYMPTOMS: TYPICAL, ATYPICAL, AND ALARM

Symptoms of GERD (Table 1) can be classified as typical (heartburn and regurgitation) or atypical (cough, asthma, hoarseness, chronic laryngitis, throat-clearing, chest pain, dyspepsia, and nausea). Atypical symptoms are more likely to be due to GERD if patients also have typical symptoms and if the symptoms respond to a trial of a PPI.³

Alarm symptoms. Keep in mind that extraesophageal presentations may be multifactorial, and it may be difficult to establish that reflux, even if present, is actually the cause. While chest pain may be due to GERD, it is important to rule out cardiac chest pain before considering GERD as a cause. Similarly, dysphagia along with typical or atypical symptoms warrants investigation for potential complications such as underlying motility disorder, esophageal stricture, esophageal ring, or malignancy.⁴ Other alarm symptoms include odynophagia, bleeding, weight loss, and anemia.

DIAGNOSING GERD: RESPONSE TO A PPI IS DIAGNOSTIC

Patients with typical symptoms that respond to PPI therapy need no further evaluation for a diagnosis of GERD to be made.⁵ On the other hand, further testing should be undertaken in patients with typical symptoms that do not respond to PPI therapy, in patients presenting with atypical symptoms, and in patients in whom antireflux surgery is being considered. Figure 1 shows our proposed algorithm.

Try a PPI for 6–8 weeks

Relief of heartburn and regurgitation after a 6- to 8-week course of a PPI strongly suggests GERD.⁶ However, a negative trial of a PPI does not rule out GERD, as this approach has been found to have a sensitivity of 78% and specificity of 54%.⁶

Despite this limitation, a trial of PPI therapy should be offered to patients presenting with typical symptoms and no alarm features. This approach has been found to be more cost-effective than proceeding directly to diagnostic testing.⁷

Endoscopy

Endoscopic findings in GERD can include erosive esophagitis, peptic stricture, and Barrett esophagus. Esophageal erosions are a highly specific sign of GERD; the Los Angeles classification system, a standardized scale for grading the severity of erosive esophagitis (from A to D, with D the most severe) provides an objective way to assess severity.⁸ However, most patients with heartburn and regurgitation do not have erosive disease, thus limiting the sensitivity of upper endoscopy as an initial diagnostic test in patients with suspected GERD.⁹

We recommend endoscopy for patients who present with alarm symptoms, patients with noncardiac chest pain, PPI nonresponders, and patients with chronic GERD symptoms and multiple risk factors for Barrett esophagus besides GERD, such as older age, male sex, white race, overweight, and smoking.¹⁰

Ambulatory pH and impedance monitoring

Ambulatory pH monitoring is the gold standard test for pathologic acid exposure in the esophagus. pH testing is indicated in PPI nonresponders, patients presenting with atypical symptoms, and before antireflux surgery.

In general, pH testing should be performed after the patient has been off PPI therapy for at least 7 days, as the test is highly unlikely to be abnormal while a patient is on a PPI.¹¹ It is done either with a transnasal catheter for 24 hours, or with a wireless capsule (Bravo pH System, Given Imaging Ltd, Duluth, GA), which collects 48 to 96 hours of data. Studies of the wireless system have shown that its sensitivity increases 12% to 25% when it is performed for 48 hours compared with 24 hours.^12,13

The pH test can be combined with impedance testing to evaluate for nonacid reflux.¹⁴ However, the clinical significance of nonacid reflux remains controversial, and for this reason the Esophageal Diagnostic Advisory Panel recommends that the decision to perform antireflux surgery should not be based on abnormal impedance testing.¹⁵

Ambulatory pH monitoring is the gold-standard test

During pH and impedance testing, special software can calculate how closely the patient’s symptoms correlate with esophageal acid exposure. The symptom index (SI) and symptom association probability (SAP) are the symptom measurements most commonly used in practice. The SI measures the overall strength of the relationship, and an SI greater than 50% is considered positive.¹⁶ The SAP determines whether this relationship could have occurred by chance, and an SAP greater than 95% is statistically significant.¹⁷ In patients with normal levels of esophageal acid exposure, an elevated SI or SAP may indicate a component of esophageal hypersensitivity in symptom generation.

At our institution, we generally perform pH-only transnasal or wireless testing off PPI therapy to establish that the patient has pathologic acid exposure in the distal esophagus. Combined pH-impedance testing is typically reserved for patients with atypical symptoms unresponsive to PPI therapy and abnormal results on previous pH testing, which allows for correlation of nonacid reflux and symptoms.

Other tests

Esophageal manometry and barium esophagography have limited value in the primary diagnosis of GERD. However, they should be considered to rule out achalasia and other esophageal motility disorders in patients whose symptoms do not respond to PPIs. For this reason, esophageal manometry should be performed before considering antireflux surgery.

MANAGING GERD

Table 2 summarizes the various treatments for GERD.

Lifestyle modifications

Lifestyle modifications are the first-line therapy for GERD. Modifications that have been studied include weight loss, head-of-bed elevation, and avoidance of tobacco, alcohol, and late-night meals. Another modification that has been suggested is avoiding foods that can aggravate reflux symptoms—eg, caffeine, coffee, chocolate, spicy foods, highly acidic foods (oranges, tomatoes), and fatty foods. Of these, only weight loss and head-of-bed elevation have been proven effective.¹⁸

Three randomized controlled trials demonstrated that GERD symptoms and esophageal pH values improved with head-of-bed elevation using blocks or incline foam wedges.^19–21 Several cohort studies demonstrated reduction in GERD symptoms with weight loss.^22,23 Recently, a prospective cohort study also found that smoking cessation significantly improved GERD symptoms in patients with normal body mass index and severe symptoms.²⁴

Antacids

Several antacids (eg, sodium bicarbonate, calcium carbonate, magnesium hydroxide, aluminum hydroxide) are available over the counter.

Antacids were thought to control heartburn symptoms by increasing the pH of gastric contents that might subsequently reflux into the esophagus. However, well-controlled studies have shown that they relieve heartburn by neutralizing acid in the esophagus, with no significant effect on gastric pH.^25,26

Antacids provide rapid but short-lived relief from an existing episode of heartburn. Because they do not significantly raise the gastric pH, they do not prevent subsequent reflux episodes from repeatedly exposing the esophagus to gastric acid and causing heartburn. Additionally, antacids have not been shown to contribute to healing of erosive esophagitis.²⁷ Hence, they may not be optimal for treating frequent reflux heartburn.

Sodium alginate

Gastric acid pockets are unbuffered pools of acid that float on top of ingested food.²⁸ They develop as a result of poor mixing of newly secreted acid and food in the proximal stomach, which remains relatively quiescent after a meal compared with the distal stomach.²⁹ In GERD, proximal extension of the acid pocket above the diaphragm increases the risk of acid reflux.³⁰ The acid pocket is therefore an important source of postprandial acid in GERD and, as such, represents a unique therapeutic target.

Emerging evidence suggests that alginates may act directly on the acid pocket. Alginates are natural polysaccharide polymers that, on contact with gastric acid, precipitate within minutes into a low-density viscous gel of near-neutral pH. The change in pH triggers the sodium bicarbonate in the formulation to release carbon dioxide that becomes trapped in the alginate gel, causing it to float to the top of the gastric contents like a raft.³¹

A randomized controlled trial demonstrated that sodium alginate was as effective as omeprazole in relieving symptoms in patients with nonerosive reflux disease.³² Alginate has also been shown to provide more postprandial reflux relief than antacids.³³

Histamine-2 receptor antagonists

Histamine-2 receptor antagonists act more swiftly and increase postprandial gastric pH more rapidly than PPIs, thus making them a good option for prophylaxis against postprandial GERD.³⁴ Taking these drugs at bedtime may help in patients with objective nighttime reflux despite optimal PPI use. However, tachyphylaxis may occur as early as 1 week after starting combination therapy.³⁵

Proton pump inhibitors

There are currently seven available PPIs, including four that can be obtained over the counter (omeprazole, lansoprazole, esomeprazole, and omeprazole-sodium bicarbonate) and three available only by prescription (rabeprazole, pantoprazole, and dexlansoprazole). Studies have shown than a standard 6- to 8-week course of a PPI provides complete symptom relief in 70% to 80% of patients with erosive reflux disease and in 60% of patients with nonerosive reflux disease.^36,37 Clinically, PPIs all appear to be similar in their symptom relief.³⁸

Most PPIs should be taken 30 to 60 minutes before meals. Exceptions are omeprazole-sodium bicarbonate and dexlansoprazole, which can be taken without regard to meals. At our institution, we usually start PPIs in a once-daily standard dose for 6 to 8 weeks and consider increasing to twice-daily dosing if symptoms do not respond completely. Patients with mild intermittent GERD symptoms may benefit from “on-demand” use of PPIs. This approach is best suited for patients with nonerosive reflux disease without evidence of Barrett esophagus on endoscopy.

Safety and adverse effects of PPIs

PPIs are generally safe but can cause adverse effects (Table 3).

Osteoporosis. In 2010, the US Food and Drug Administration issued warnings regarding the potential for wrist, hip, and spine fractures in PPI users.²⁶ Most recent evidence suggests that PPIs may be associated with a small increase in risk of hip fractures in patients already at high risk.^39,40 However, the 2013 American College of Gastroenterology (ACG) guidelines say that patients with known osteoporosis can remain on PPI therapy, and concern for hip fractures and osteoporosis should not affect the decision to use PPIs long-term except in patients with other risk factors for hip fracture.⁴¹

Community-acquired pneumonia. An increased risk of community-acquired pneumonia cannot be clearly documented in association with PPI therapy. Multiple studies, including randomized controlled trials, investigated this potential correlation. However, evidence suggests that short-term but not long-term PPI use may be associated with an overall increased risk of community-acquired pneumonia.^42,43 Current guidelines suggest that in patients who need a PPI, the drug should not be withheld on the basis of a potential risk of community-acquired pneumonia.⁴¹

Clostridium difficile infection. In theory, PPIs may increase the risk of C difficile infection by increasing the ability of the spore to convert to the vegetative form and to survive intraluminally. In fact, studies and meta-analyses have suggested that PPIs do increase the risk of development and recurrence of C difficile infection.^44,45 Therefore, PPIs should be used with care in patients who are at risk.⁴¹

Interaction with clopidogrel. The antiplatelet activity of clopidogrel requires activation by CYP2C19, the same pathway required for metabolism of some PPIs. Concern was raised about decreased antiplatelet activity of clopidogrel in the presence of PPIs. This was extensively studied, and there now appears to be no increased risk of adverse cardiovascular events in patients on PPIs, based on data from well-controlled randomized trials.^46,47 A consensus panel of the American College of Cardiology Foundation, the American Heart Association, and the ACG said that PPIs may be used for appropriate indications in patients taking clopidogrel.⁴⁷

Hypomagnesemia. By an unknown molecular mechanism, PPIs are thought to reduce intestinal magnesium absorption, leading to hypomagnesemia. A meta-analysis published in 2011 showed that PPI-induced hypomagnesemia is a drug-class effect and occurred after a median of 5.5 years of PPI use. Stopping the PPI resulted in magnesium recovery in 4 days, and rechallenge led to recurrence within 4 days.⁴⁸

Hence, to avoid putting patients on long-term PPI therapy at risk, clinicians should anticipate this problem. Our practice is to check the magnesium level before starting a patient on long-term PPI therapy, and then to repeat the measurement every 1 to 2 years.

Baclofen

Transient lower esophageal sphincter relaxation has been shown to be a cause of reflux in healthy people and in patients with GERD.⁴⁹

Baclofen, a muscle relaxant with selective gamma-aminobutyric acid receptor class B agonist properties, reduces transient lower esophageal sphincter relaxation in humans.⁵⁰ In a well-designed, double-blind, randomized controlled trial, baclofen was associated with a significant decrease in upright reflux on 24-hour pH monitoring and significant improvement in belching and overall reflux symptoms.⁵¹ However, baclofen is not approved by the US Food and Drug Administration for the treatment of GERD, and its use may be limited by side effects such as somnolence and dizziness.

Antireflux surgery

Antireflux surgery is a reasonable option for selected patients with chronic GERD. The main types of surgery are laparoscopic fundoplication and, for obese patients, gastric bypass. Reasons to consider antireflux surgery include desire to stop PPI therapy, esophagitis not healed by PPIs, symptomatic hiatal hernia, and refractory reflux documented by pH testing.⁴¹

Clinically, all PPIs appear to be similar in their symptom relief

In general, surgical therapy may be considered in patients who respond to PPIs, but patients who do not respond to PPIs are less likely to respond to antireflux surgery.¹⁵ Other patients less likely to respond are those with symptoms of dyspepsia, such as nausea, vomiting, and epigastric pain.⁴¹

Common adverse effects of antireflux surgery include gas-bloat syndrome (up to 85% of patients), dysphagia (10% to 50% of patients), diarrhea (18% to 33% of patients), and recurrent heartburn (10% to 62% of patients).⁵²

Endoscopic and minimally invasive antireflux procedures include endoscopic plication of the lower esophageal sphincter, radiofrequency augmentation of the lower esophageal sphincter, and sphincter augmentation by a string of titanium beads. While some have shown promise, they are not recommended by the most recent ACG guidelines, given lack of long-term data.⁴¹

REFRACTORY GERD

There is no consensus on the definition of refractory GERD. However, for the sake of simplicity, we can define it as persistence of suspected GERD symptoms despite treatment with a PPI. This may vary from a partial response to PPI therapy to a complete absence of response.

It is extremely important to rule out non-GERD causes of the ongoing symptoms, such as achalasia, gastroparesis, eosinophilic esophagitis, rumination, and aerophagia. PPI nonresponders are more likely to be obese, poorly compliant, and have extraesophageal symptoms.^53–56 As previously discussed, PPIs should be taken 30 to 60 minutes before meals. For patients whose symptoms fail to respond to standard-dose daily PPI therapy, switching to another PPI or doubling the dose is common, although data to support this practice are limited. Of note, omeprazole-sodium bicarbonate has been shown to provide more symptom relief in nocturnal GERD.⁵⁷ Additionally, adding a nighttime histamine-2 receptor antagonist may also help in patients with objective nighttime reflux.⁴¹

After noncompliance and suboptimal PPI dosing have been ruled out, PPI nonresponders with typical symptoms should undergo upper endoscopy and subsequent pH monitoring. Normal esophageal acid exposure on pH testing suggests functional heartburn or functional dyspepsia. Negative pH testing in a patient with atypical symptoms suggests a non-GERD cause of symptoms, and referral to an otolaryngologist, pulmonologist, or allergist is often warranted.

While antireflux surgery can be considered for patients with nonacid reflux on impedance testing, it should again be noted that GERD in patients with no response to PPIs is less likely to respond to antireflux surgery.¹⁵

TAKE-HOME POINTS

• GERD is a common medical condition, affecting up to 40% of US adults at least once monthly.
• GERD can result in a wide variety of symptoms, including typical heartburn and regurgitation as well as atypical symptoms such as cough.
• On the other hand, keep in mind that multiple non-GERD causes of heartburn and regurgitation may exist.
• Testing for GERD includes endoscopy and pH testing as well as functional testing such as esophageal manometry.
• While in most patients GERD will respond to lifestyle changes and antisecretory therapy such as a PPI, careful attention must be given to patients with symptoms refractory to PPI therapy.
• For a subset of patients, antireflux surgery may be a reasonable option, but care must be taken to exclude patients with a lower likelihood of responding to surgery.

In the United States, gynecologic surgeons remove approximately one uterus every minute of the year.¹ That rate translates to more than 525,000 hysterectomies annually in this country alone. Yet, despite the widespread availability of information on the benefits of a vaginal approach to hysterectomy, the great majority of these operations—close to 50%—are still performed via an open abdominal approach.²

As I pointed out last month in my “Update on Vaginal Hysterectomy,” the vaginal approach not only is more cosmetically pleasing than laparoscopic and robot-assisted hysterectomy (not to mention open abdominal surgery) but also has a lower complication rate.³

As I also noted, one reason for the low rate of vaginal hysterectomy may be the assumption, on the part of many gynecologic surgeons, that the techniques and tools they learned to use during training are still the only options available today. That assumption is wrong.

In this article, I describe the technique 
for vaginal hysterectomy using basic instru
mentation. This article is based on a master 
class in vaginal hysterectomy produced by 
the AAGL and co-sponsored by the Am
erican College of Obstetricians and Gynecologists and the Society of Gynecologic 
Surgeons. This master class offers continuing
 medical education credits and is avail
able at http://www.aagl.org/vaghystwebinar.

For a look at innovative tools for this procedure, see my “Update on Vaginal 
Hysterectomy” in the September 2015 issue of this journal at obgmanagement.com.

Vaginal hysterectomy has 
few contraindications

Many commonly cited contraindications to the vaginal approach are not, in fact, absolute contraindications. An open or laparoscopic approach is preferred when the patient has a known cancer, of course, and when deep infiltrating endometriosis is present at the rectovaginal septum. However, previous pelvic surgery, nulliparity, an enlarged uterus, or lack of a prior vaginal delivery need not exclude the vaginal approach. Nor does a narrow introitus necessarily mandate a laparoscopic or open abdominal approach. In fact, in this article, I describe my basic technique in a patient (a cadaver) with a very narrow pubic arch, and I offer strategies for gaining some needed mobility and avoiding complications (TABLES 1 and 2).

Next month, in the November issue of OBG Management, John B. Gebhart, MD, will describe his vaginal technique for right salpingectomy with ovarian preservation, as well as his technique for right salpingo-oophorectomy.

Proper patient positioning 
is key

You can simplify the operation by positioning the patient so that her buttocks are over the edge of the table fairly far—at least 1 inch beyond the edge of the table for optimal exposure and greater access. If the patient is thin, it then becomes important to pad the sacrum because, when she is positioned that far off the table, all her weight comes to rest on the sacrum. In overweight patients, this is not an issue, but for thin patients, I place a bit of egg crate or gel beneath the sacrum.

For the procedure, I prefer to place my instruments on a tray that is kept on my lap. This arrangement frees the scrub technician from having to hand tools over my shoulder—and it saves time. I use a narrow, covered Mayo stand, and I place a stepstool beneath my feet to keep my knees at right angles so that things don’t slip during the operation.

Surgical technique

Choose an appropriate retractor

In a woman with a narrow introitus, I find that a posterior weighted speculum takes up too much space. Once I place a clamp on the cervix with that speculum in place, I don’t have much room to work. However, if I substitute a small Deaver retractor, which is narrower, I gain more workspace.

Inject the uterosacral ligaments

Grasp the cervix using a Jacobs vulsellum tenaculum. Use of a single tenaculum allows for much more movement than the use of instruments placed anteriorly and posteriorly. The Jacobs tenaculum obtains a better purchase on the tissue than a single tooth and is considerably less likely to tear through the tissue.

Before beginning the hysterectomy, locate the uterosacral ligaments and inject each one at its junction with the cervix, aspirating slightly before infiltrating the ligament with 0.25% to 0.50% bupivacaine with epinephrine, with dilute vasopressin mixed in. (I place 1 unit in 20 mL of the local solution.) Injection of this solution achieves 2 goals:

• improved intraoperative hemostasis
• postoperative pain relief.

Use a short needle with a needle extender attached to a control syringe rather than a spinal needle for greater control.

Enter the posterior peritoneal cavity

Before entering the peritoneal cavity, create a right angle with the Jacobs tenaculum and Deaver retractor in relation to the surgical field (FIGURE 1). This right angle is difficult to achieve when you are using a weighted speculum in a tight vagina. Once you have a right angle, tent the vaginal tissue in the midline (FIGURE 2).

In a nulliparous patient or a woman with a tight pelvis, you may discover that the peritoneum is pulled up between the uterosacral ligaments. One common pitfall arises when the surgeon, having dissected the vaginal epithelium, continues cutting into the vaginal epithelium instead of reaching into the peritoneal cavity. Palpate the tissue to ensure that there is no bowel in the way and stay at right angles while confidently grasping the peritoneum with a toothed forceps.

I like to use a bit of electrosurgery to incise the vaginal wall. I don’t begin at the cervix but incise more distally into the vaginal epithelium approximately 2 cm from the cervicovaginal junction. This strategy prevents dissection into the cervix and/or rectovaginal septum rather than the posterior 
cul-de-sac (FIGURE 3).

Once the incision is made, it is possible to feel the posterior peritoneum. And as you tent the peritoneum, you can then very confidently extend the incision and enter the cavity posteriorly.

In a patient with significant adhesions such as this one, I feel around posteriorly to determine exactly where I am. One tactic I use is to release the tenaculum and regrasp the cervix with it. This allows for improved visualization and movement of the cervix as the procedure progresses. Depending on the case, it may be necessary to insert a sponge to hold bowel out of the way.

Avoid the bladder

Move the Deaver retractor to the anterior position, switch the Jacobs clamp to the anterior cervix, and pull straight down. Now that you have incised the vaginal epithelium posteriorly, the length of the cervix should be apparent to you, and you can easily determine the location of the bladder reflection.

Keep in mind that, in a postmenopausal patient, there will be fewer vaginal rugae to guide you. Place the Jacobs tenaculum as close to the midline as possible so that you can confidently grab the tissue without fear of grabbing the bladder. If you tilt the Jacobs clamp, you can feel the edge of the bladder reflection. Remember that postmenopausal patients with prolapse (or, occasionally, obese patients with cervical elongation but little actual descensus) may have altered anatomy.

You can create a bit more space in which to dissect by injecting the bupivacaine/ 
epinephrine solution into the vaginal epithelium. This technique also ensures that the vaginal epithelial incisions won’t bleed.

Now, tilt the Jacobs tenaculum downward and push the junction of the cervix with the bladder reflection toward you so that you have a good sense of how deeply to incise.

Once you’ve made the incision, reclamp the Jacobs tenaculum so that it holds all of that tissue, and repeat the maneuver, tilting the clamp downward and pushing the junction toward you. In this way, you create traction and countertraction, sweeping the tissue out of your way.

Always use sharp dissection. When adhesions are present, surgeons often get into trouble using blunt dissection and may inadvertently enter the bladder if they use a sponge-covered digit for dissection, because adhesions can be much denser than normal tissue. In such cases, the bladder tears open rather than the adhesions being swept away.

Consider this: You don’t need to enter the peritoneal cavity anteriorly in order to continue working on the procedure. You can safely protect the bladder throughout the case, until the very end, if necessary, in patients who have undergone multiple previous surgeries or cesarean deliveries.

Rather than enter the anterior peritoneum, I dissect as much of the vaginal epithelium as possible and place a second Deaver retractor posteriorly.

I massage the uterosacral ligament for about 10 seconds to lengthen it and create more descensus, then place a Ballantine Heaney clamp on the ligament.

Next, I cut the pedicle and suture it, maintaining a clamp on the uterosacral ligament suture so that I can use it later for repair of the vaginal cuff.

I recommend a vessel-sealing device to secure the major blood supply, but I do suture the uterosacral and round ligaments for attachment to the apex at the conclusion of the hysterectomy. I suggest that you place straight clamps to hold the uterosacral ligament sutures and curved clamps on the round ligament ties to help you keep track of what you’re doing.

I generally prefer to use a smaller vessel-sealing device, such as the LigaSure Max (Covidien), because it allows me to take very small bites of tissue. It is also less expensive because it uses a disposable electrode within a reusable Heaney-type clamp.

Many people have argued that we need to teach surgeons to suture vaginally and, for that reason, should avoid vessel sealing. My response: Why wouldn’t we want to use the very best technology available? Randomized trials have demonstrated a 50% reduction in pain relief postoperatively when we use vessel sealing.⁴ Less foreign material is left in the pelvis, lowering the risk of infection. And it really doesn’t matter which vessel-sealing technology you use, as long as you’re familiar with the specifics of the system you choose. Another advantage: There is no need to pass needles back and forth.

Take small bites of tissue

Because this patient has a very small uterus, a small bite of tissue will get you close to where you want to be. When you take a bite with the vessel sealer, try to protect the vaginal epithelium and vulva from the steam that is emitted. The clamp itself does not heat up, but the steam that is released from the tissue is 100° C, so place a finger between the clamp and the sidewall for protection. It is preferable to burn your own finger than to burn the patient.

Because you haven’t entered the peritoneal cavity anteriorly, it is important to ensure that you don’t take too big of a bite with the vessel sealer. Rather, stay where you know you’ve done your dissection, where things are safe.

One cardinal principle of surgery is that you shouldn’t operate where you can’t feel or see. One of the common errors in vaginal surgery is that surgeons start dissecting higher than they can see. It’s easy to get into trouble when you start pushing tissue or dissecting tissue that you can’t visualize.

At this point, the anterior Deaver retractor is not essential, so I remove it. If you don’t need it, don’t use it. I try to avoid metal when I can.

If I were using suture rather than vessel sealing, I would place a Heaney clamp on the uterosacral ligament and cut. Using a clamp-cut-tie technique, I would pull on the pedicle and cut just beyond the tip of the clamp to ensure that the suture will be secure 
(FIGURE 4). This approach would not be appropriate during use of a vessel sealer. In that case, you would want to cut to but not beyond the tip of the clamp.

One of the skills helpful in suturing is learning to move your elbow and wrist to achieve the proper angle. Determine where you want the suture to exit the tissue, and then angle your elbow and wrist so that the suture comes out where you want it. It’s easy to lose track of the needle tip, especially when you’re working in a limited space under the pubic symphysis, so use your shoulder, elbow, and wrist to control 
suture placement.

Protect the anterior epithelium

Because you have not yet entered the peritoneal cavity anteriorly, it is important to protect the anterior epithelium and bladder. Reinsert a narrow Deaver retractor anteriorly, remove the Jacobs clamp, and replace the clamp laterally so that the cervix can be pulled off to the side (FIGURE 5).

One nice thing about some vessel sealers is that the surgeon can twist them in any direction. It isn’t necessary to move your hand; you simply move the device itself.

Once you have taken at least the descending branch of the uterine artery, remove the posterior retractor and pull downward on the Jacobs tenaculum. You should have reached just about to the level of the uterine fundus, with the anatomy well visualized (FIGURE 6). Next, open the anterior peritoneum.

Pay attention to the surgical field

Now that you have entered the peritoneum anteriorly as well as posteriorly, identify the broad ligament, keeping in mind that the ureter is retroperitoneal, not intraperitoneal. If you were to place a clamp from the posterior leaf of the broad ligament across to the anterior leaf of the broad ligament, you would be grasping all the vessels but not the ureter. In fact, the anterior Deaver retractor is lifting both ureters up and out of the way. If you pull the cervix off to the opposite side, you create an additional couple of centimeters—a safe space for the vessel sealer 
(FIGURE 7).

In placing the vessel sealer, there is no need to move out laterally, as there is no need for space to place suture. Instead, hug the uterus. At this point, the main concern is the risk of damaging any small bowel behind the uterine fundus that might be coming down into the surgical field, obscured from vision. And because there may be steam emitted at the tip of the vessel-sealing clamp, keep a finger back there to protect anything that might be in the field.

Last steps

Before taking the last bite of tissue on the right-hand side, place the round ligament in a Heaney clamp. Now that the round and utero-ovarian ligaments have been skeletonized, you can grasp the pedicle in a clamp. If the pedicle is especially thick, it may be beneficial to close the clamp, leave it on for a few seconds, and then reapply it. In that way, you obtain a better purchase.

Next, free the rest of the tissue with a vessel sealer, or cut it. I prefer to use a vessel sealer, and I again protect the adjacent tissue with my fingers anteriorly and posteriorly.

With the clamp remaining on the round ligament and utero-ovarian ligament 
(FIGURE 8), which will be sutured, push the uterine tissue out of the way, back into the pelvis, to make room for suturing.

Because a postmenopausal vulva may be cut by the suture, it’s important to take pains not to abrade that tissue. Once you have finished suturing the round/utero-ovarian pedicle, leave the needle on the suture so that you can reconnect the round ligament to the anterior pubocervical ring to reconstruct the vaginal apex. For safekeeping, clamp the needle out of the way and tuck it beneath the drape.

Switching to the other side, use a Lahey clamp to flip the uterus, then clamp the pedicle and use the vessel sealer to separate it, again protecting the tissue beneath and ahead of the clamp. Sometimes, with an especially thick pedicle, the vessel sealer will signal that the tissue hasn’t been completely sealed. In that case, get another purchase of the pedicle, protect the adjacent tissue, and seal again.

Once the uterus has been removed completely, suture the utero-ovarian and round ligament on this side.
One tip to aid in the placement of suture is to move your clamped tissue in such a way as to prevent inadvertent suturing of other tissue (FIGURE 9).

An additional strategy for pain relief at this point is to infiltrate the round ligaments with local anesthetic. We know that we’re working with higher-level fibers—T10 to T12—through the round ligaments. By infiltrating them with anesthetic, you achieve denser pain relief for post- 
operative management.

Uterine reduction strategies facilitate vaginal removal of tissue

When the uterus is too large to remove intact through the vagina, there are a number of techniques for coring, wedging, and morcellating the tissue. As always, a complete knowledge of anatomy is essential, as well as an understanding that fibroids can frequently distort the uterus, twisting it to the left or right. It is important to anticipate such distortion to avoid the inadvertent destruction of anatomic landmarks or damage to the adnexae.

One straight-forward strategy is to debulk the uterus using a knife to core it, removing the central portion. In cases in which you need to keep the entire endometrial cavity intact, you can core the central portion of the uterus while grasping the cervix so that you can remove the endometrium intact for the pathologist (FIGURE).

For this strategy it is important to protect the vaginal sidewalls with metal. You can use another retractor to do that, pulling down on the cervix and beginning the morcellation. I generally prefer to use a short knife handle only because I want to be sure I’m not tempted to cut any higher than I can see.

For more on coring and wedging techniques, see the introductory video for the ACOG/SGS/AAGL master class on vaginal hysterectomy at http://www.aagl.org/vaghystwebinar.

Close the vaginal cuff

The reconstruction of the vaginal cuff is a critical component of any hysterectomy. My approach is to reattach the uterosacral ligaments to the posterior cuff and the round ligaments to the anterior cuff, thereby re- 
creating an intact pubocervical ring. It is not necessary to include the peritoneum in the cuff closure. In fact, kinking of the ureters is more likely when the peritoneum is closed.

Attach one uterosacral ligament, then place a running, full-thickness stitch across the posterior cuff, and attach the uterosacral ligament on the opposite side. Use the needle you left attached to the round ligament to bring the right pedicle to the anterior cuff at 10 o’clock (be sure you grasp the full thickness of the vaginal epithelium without compromising the bladder). Attach the left round-ligament pedicle at the 2 o’clock position. Then close the cuff side to side down to the uterosacral ligaments. This completely reconstructs the pubocervical ring and provides excellent support at the apex.

Share your thoughts on this article! Send your Letter to the Editor to rbarbieri@frontlinemedcom.com. Please include your name and the city and state in which you practice.

In the United States, gynecologic surgeons remove approximately one uterus every minute of the year.¹ That rate translates to more than 525,000 hysterectomies annually in this country alone. Yet, despite the widespread availability of information on the benefits of a vaginal approach to hysterectomy, the great majority of these operations—close to 50%—are still performed via an open abdominal approach.²

As I pointed out last month in my “Update on Vaginal Hysterectomy,” the vaginal approach not only is more cosmetically pleasing than laparoscopic and robot-assisted hysterectomy (not to mention open abdominal surgery) but also has a lower complication rate.³

As I also noted, one reason for the low rate of vaginal hysterectomy may be the assumption, on the part of many gynecologic surgeons, that the techniques and tools they learned to use during training are still the only options available today. That assumption is wrong.

In this article, I describe the technique 
for vaginal hysterectomy using basic instru
mentation. This article is based on a master 
class in vaginal hysterectomy produced by 
the AAGL and co-sponsored by the Am
erican College of Obstetricians and Gynecologists and the Society of Gynecologic 
Surgeons. This master class offers continuing
 medical education credits and is avail
able at http://www.aagl.org/vaghystwebinar.

For a look at innovative tools for this procedure, see my “Update on Vaginal 
Hysterectomy” in the September 2015 issue of this journal at obgmanagement.com.

Vaginal hysterectomy has 
few contraindications

Many commonly cited contraindications to the vaginal approach are not, in fact, absolute contraindications. An open or laparoscopic approach is preferred when the patient has a known cancer, of course, and when deep infiltrating endometriosis is present at the rectovaginal septum. However, previous pelvic surgery, nulliparity, an enlarged uterus, or lack of a prior vaginal delivery need not exclude the vaginal approach. Nor does a narrow introitus necessarily mandate a laparoscopic or open abdominal approach. In fact, in this article, I describe my basic technique in a patient (a cadaver) with a very narrow pubic arch, and I offer strategies for gaining some needed mobility and avoiding complications (TABLES 1 and 2).

Next month, in the November issue of OBG Management, John B. Gebhart, MD, will describe his vaginal technique for right salpingectomy with ovarian preservation, as well as his technique for right salpingo-oophorectomy.

Proper patient positioning 
is key

You can simplify the operation by positioning the patient so that her buttocks are over the edge of the table fairly far—at least 1 inch beyond the edge of the table for optimal exposure and greater access. If the patient is thin, it then becomes important to pad the sacrum because, when she is positioned that far off the table, all her weight comes to rest on the sacrum. In overweight patients, this is not an issue, but for thin patients, I place a bit of egg crate or gel beneath the sacrum.

For the procedure, I prefer to place my instruments on a tray that is kept on my lap. This arrangement frees the scrub technician from having to hand tools over my shoulder—and it saves time. I use a narrow, covered Mayo stand, and I place a stepstool beneath my feet to keep my knees at right angles so that things don’t slip during the operation.

Surgical technique

Choose an appropriate retractor

In a woman with a narrow introitus, I find that a posterior weighted speculum takes up too much space. Once I place a clamp on the cervix with that speculum in place, I don’t have much room to work. However, if I substitute a small Deaver retractor, which is narrower, I gain more workspace.

Inject the uterosacral ligaments

Grasp the cervix using a Jacobs vulsellum tenaculum. Use of a single tenaculum allows for much more movement than the use of instruments placed anteriorly and posteriorly. The Jacobs tenaculum obtains a better purchase on the tissue than a single tooth and is considerably less likely to tear through the tissue.

Before beginning the hysterectomy, locate the uterosacral ligaments and inject each one at its junction with the cervix, aspirating slightly before infiltrating the ligament with 0.25% to 0.50% bupivacaine with epinephrine, with dilute vasopressin mixed in. (I place 1 unit in 20 mL of the local solution.) Injection of this solution achieves 2 goals:

• improved intraoperative hemostasis
• postoperative pain relief.

Use a short needle with a needle extender attached to a control syringe rather than a spinal needle for greater control.

Enter the posterior peritoneal cavity

Before entering the peritoneal cavity, create a right angle with the Jacobs tenaculum and Deaver retractor in relation to the surgical field (FIGURE 1). This right angle is difficult to achieve when you are using a weighted speculum in a tight vagina. Once you have a right angle, tent the vaginal tissue in the midline (FIGURE 2).

In a nulliparous patient or a woman with a tight pelvis, you may discover that the peritoneum is pulled up between the uterosacral ligaments. One common pitfall arises when the surgeon, having dissected the vaginal epithelium, continues cutting into the vaginal epithelium instead of reaching into the peritoneal cavity. Palpate the tissue to ensure that there is no bowel in the way and stay at right angles while confidently grasping the peritoneum with a toothed forceps.

I like to use a bit of electrosurgery to incise the vaginal wall. I don’t begin at the cervix but incise more distally into the vaginal epithelium approximately 2 cm from the cervicovaginal junction. This strategy prevents dissection into the cervix and/or rectovaginal septum rather than the posterior 
cul-de-sac (FIGURE 3).

Once the incision is made, it is possible to feel the posterior peritoneum. And as you tent the peritoneum, you can then very confidently extend the incision and enter the cavity posteriorly.

In a patient with significant adhesions such as this one, I feel around posteriorly to determine exactly where I am. One tactic I use is to release the tenaculum and regrasp the cervix with it. This allows for improved visualization and movement of the cervix as the procedure progresses. Depending on the case, it may be necessary to insert a sponge to hold bowel out of the way.

Avoid the bladder

Move the Deaver retractor to the anterior position, switch the Jacobs clamp to the anterior cervix, and pull straight down. Now that you have incised the vaginal epithelium posteriorly, the length of the cervix should be apparent to you, and you can easily determine the location of the bladder reflection.

Keep in mind that, in a postmenopausal patient, there will be fewer vaginal rugae to guide you. Place the Jacobs tenaculum as close to the midline as possible so that you can confidently grab the tissue without fear of grabbing the bladder. If you tilt the Jacobs clamp, you can feel the edge of the bladder reflection. Remember that postmenopausal patients with prolapse (or, occasionally, obese patients with cervical elongation but little actual descensus) may have altered anatomy.

You can create a bit more space in which to dissect by injecting the bupivacaine/ 
epinephrine solution into the vaginal epithelium. This technique also ensures that the vaginal epithelial incisions won’t bleed.

Now, tilt the Jacobs tenaculum downward and push the junction of the cervix with the bladder reflection toward you so that you have a good sense of how deeply to incise.

Once you’ve made the incision, reclamp the Jacobs tenaculum so that it holds all of that tissue, and repeat the maneuver, tilting the clamp downward and pushing the junction toward you. In this way, you create traction and countertraction, sweeping the tissue out of your way.

Always use sharp dissection. When adhesions are present, surgeons often get into trouble using blunt dissection and may inadvertently enter the bladder if they use a sponge-covered digit for dissection, because adhesions can be much denser than normal tissue. In such cases, the bladder tears open rather than the adhesions being swept away.

Consider this: You don’t need to enter the peritoneal cavity anteriorly in order to continue working on the procedure. You can safely protect the bladder throughout the case, until the very end, if necessary, in patients who have undergone multiple previous surgeries or cesarean deliveries.

Rather than enter the anterior peritoneum, I dissect as much of the vaginal epithelium as possible and place a second Deaver retractor posteriorly.

I massage the uterosacral ligament for about 10 seconds to lengthen it and create more descensus, then place a Ballantine Heaney clamp on the ligament.

Next, I cut the pedicle and suture it, maintaining a clamp on the uterosacral ligament suture so that I can use it later for repair of the vaginal cuff.

I recommend a vessel-sealing device to secure the major blood supply, but I do suture the uterosacral and round ligaments for attachment to the apex at the conclusion of the hysterectomy. I suggest that you place straight clamps to hold the uterosacral ligament sutures and curved clamps on the round ligament ties to help you keep track of what you’re doing.

I generally prefer to use a smaller vessel-sealing device, such as the LigaSure Max (Covidien), because it allows me to take very small bites of tissue. It is also less expensive because it uses a disposable electrode within a reusable Heaney-type clamp.

Many people have argued that we need to teach surgeons to suture vaginally and, for that reason, should avoid vessel sealing. My response: Why wouldn’t we want to use the very best technology available? Randomized trials have demonstrated a 50% reduction in pain relief postoperatively when we use vessel sealing.⁴ Less foreign material is left in the pelvis, lowering the risk of infection. And it really doesn’t matter which vessel-sealing technology you use, as long as you’re familiar with the specifics of the system you choose. Another advantage: There is no need to pass needles back and forth.

Take small bites of tissue

Because this patient has a very small uterus, a small bite of tissue will get you close to where you want to be. When you take a bite with the vessel sealer, try to protect the vaginal epithelium and vulva from the steam that is emitted. The clamp itself does not heat up, but the steam that is released from the tissue is 100° C, so place a finger between the clamp and the sidewall for protection. It is preferable to burn your own finger than to burn the patient.

Because you haven’t entered the peritoneal cavity anteriorly, it is important to ensure that you don’t take too big of a bite with the vessel sealer. Rather, stay where you know you’ve done your dissection, where things are safe.

One cardinal principle of surgery is that you shouldn’t operate where you can’t feel or see. One of the common errors in vaginal surgery is that surgeons start dissecting higher than they can see. It’s easy to get into trouble when you start pushing tissue or dissecting tissue that you can’t visualize.

At this point, the anterior Deaver retractor is not essential, so I remove it. If you don’t need it, don’t use it. I try to avoid metal when I can.

If I were using suture rather than vessel sealing, I would place a Heaney clamp on the uterosacral ligament and cut. Using a clamp-cut-tie technique, I would pull on the pedicle and cut just beyond the tip of the clamp to ensure that the suture will be secure 
(FIGURE 4). This approach would not be appropriate during use of a vessel sealer. In that case, you would want to cut to but not beyond the tip of the clamp.

One of the skills helpful in suturing is learning to move your elbow and wrist to achieve the proper angle. Determine where you want the suture to exit the tissue, and then angle your elbow and wrist so that the suture comes out where you want it. It’s easy to lose track of the needle tip, especially when you’re working in a limited space under the pubic symphysis, so use your shoulder, elbow, and wrist to control 
suture placement.

Protect the anterior epithelium

Because you have not yet entered the peritoneal cavity anteriorly, it is important to protect the anterior epithelium and bladder. Reinsert a narrow Deaver retractor anteriorly, remove the Jacobs clamp, and replace the clamp laterally so that the cervix can be pulled off to the side (FIGURE 5).

One nice thing about some vessel sealers is that the surgeon can twist them in any direction. It isn’t necessary to move your hand; you simply move the device itself.

Once you have taken at least the descending branch of the uterine artery, remove the posterior retractor and pull downward on the Jacobs tenaculum. You should have reached just about to the level of the uterine fundus, with the anatomy well visualized (FIGURE 6). Next, open the anterior peritoneum.

Pay attention to the surgical field

Now that you have entered the peritoneum anteriorly as well as posteriorly, identify the broad ligament, keeping in mind that the ureter is retroperitoneal, not intraperitoneal. If you were to place a clamp from the posterior leaf of the broad ligament across to the anterior leaf of the broad ligament, you would be grasping all the vessels but not the ureter. In fact, the anterior Deaver retractor is lifting both ureters up and out of the way. If you pull the cervix off to the opposite side, you create an additional couple of centimeters—a safe space for the vessel sealer 
(FIGURE 7).

In placing the vessel sealer, there is no need to move out laterally, as there is no need for space to place suture. Instead, hug the uterus. At this point, the main concern is the risk of damaging any small bowel behind the uterine fundus that might be coming down into the surgical field, obscured from vision. And because there may be steam emitted at the tip of the vessel-sealing clamp, keep a finger back there to protect anything that might be in the field.

Last steps

Before taking the last bite of tissue on the right-hand side, place the round ligament in a Heaney clamp. Now that the round and utero-ovarian ligaments have been skeletonized, you can grasp the pedicle in a clamp. If the pedicle is especially thick, it may be beneficial to close the clamp, leave it on for a few seconds, and then reapply it. In that way, you obtain a better purchase.

Next, free the rest of the tissue with a vessel sealer, or cut it. I prefer to use a vessel sealer, and I again protect the adjacent tissue with my fingers anteriorly and posteriorly.

With the clamp remaining on the round ligament and utero-ovarian ligament 
(FIGURE 8), which will be sutured, push the uterine tissue out of the way, back into the pelvis, to make room for suturing.

Because a postmenopausal vulva may be cut by the suture, it’s important to take pains not to abrade that tissue. Once you have finished suturing the round/utero-ovarian pedicle, leave the needle on the suture so that you can reconnect the round ligament to the anterior pubocervical ring to reconstruct the vaginal apex. For safekeeping, clamp the needle out of the way and tuck it beneath the drape.

Switching to the other side, use a Lahey clamp to flip the uterus, then clamp the pedicle and use the vessel sealer to separate it, again protecting the tissue beneath and ahead of the clamp. Sometimes, with an especially thick pedicle, the vessel sealer will signal that the tissue hasn’t been completely sealed. In that case, get another purchase of the pedicle, protect the adjacent tissue, and seal again.

Once the uterus has been removed completely, suture the utero-ovarian and round ligament on this side.
One tip to aid in the placement of suture is to move your clamped tissue in such a way as to prevent inadvertent suturing of other tissue (FIGURE 9).

An additional strategy for pain relief at this point is to infiltrate the round ligaments with local anesthetic. We know that we’re working with higher-level fibers—T10 to T12—through the round ligaments. By infiltrating them with anesthetic, you achieve denser pain relief for post- 
operative management.

Uterine reduction strategies facilitate vaginal removal of tissue

When the uterus is too large to remove intact through the vagina, there are a number of techniques for coring, wedging, and morcellating the tissue. As always, a complete knowledge of anatomy is essential, as well as an understanding that fibroids can frequently distort the uterus, twisting it to the left or right. It is important to anticipate such distortion to avoid the inadvertent destruction of anatomic landmarks or damage to the adnexae.

One straight-forward strategy is to debulk the uterus using a knife to core it, removing the central portion. In cases in which you need to keep the entire endometrial cavity intact, you can core the central portion of the uterus while grasping the cervix so that you can remove the endometrium intact for the pathologist (FIGURE).

For this strategy it is important to protect the vaginal sidewalls with metal. You can use another retractor to do that, pulling down on the cervix and beginning the morcellation. I generally prefer to use a short knife handle only because I want to be sure I’m not tempted to cut any higher than I can see.

For more on coring and wedging techniques, see the introductory video for the ACOG/SGS/AAGL master class on vaginal hysterectomy at http://www.aagl.org/vaghystwebinar.

Close the vaginal cuff

The reconstruction of the vaginal cuff is a critical component of any hysterectomy. My approach is to reattach the uterosacral ligaments to the posterior cuff and the round ligaments to the anterior cuff, thereby re- 
creating an intact pubocervical ring. It is not necessary to include the peritoneum in the cuff closure. In fact, kinking of the ureters is more likely when the peritoneum is closed.

Attach one uterosacral ligament, then place a running, full-thickness stitch across the posterior cuff, and attach the uterosacral ligament on the opposite side. Use the needle you left attached to the round ligament to bring the right pedicle to the anterior cuff at 10 o’clock (be sure you grasp the full thickness of the vaginal epithelium without compromising the bladder). Attach the left round-ligament pedicle at the 2 o’clock position. Then close the cuff side to side down to the uterosacral ligaments. This completely reconstructs the pubocervical ring and provides excellent support at the apex.

Share your thoughts on this article! Send your Letter to the Editor to rbarbieri@frontlinemedcom.com. Please include your name and the city and state in which you practice.

In the United States, gynecologic surgeons remove approximately one uterus every minute of the year.¹ That rate translates to more than 525,000 hysterectomies annually in this country alone. Yet, despite the widespread availability of information on the benefits of a vaginal approach to hysterectomy, the great majority of these operations—close to 50%—are still performed via an open abdominal approach.²

As I pointed out last month in my “Update on Vaginal Hysterectomy,” the vaginal approach not only is more cosmetically pleasing than laparoscopic and robot-assisted hysterectomy (not to mention open abdominal surgery) but also has a lower complication rate.³

As I also noted, one reason for the low rate of vaginal hysterectomy may be the assumption, on the part of many gynecologic surgeons, that the techniques and tools they learned to use during training are still the only options available today. That assumption is wrong.

In this article, I describe the technique 
for vaginal hysterectomy using basic instru
mentation. This article is based on a master 
class in vaginal hysterectomy produced by 
the AAGL and co-sponsored by the Am
erican College of Obstetricians and Gynecologists and the Society of Gynecologic 
Surgeons. This master class offers continuing
 medical education credits and is avail
able at http://www.aagl.org/vaghystwebinar.

For a look at innovative tools for this procedure, see my “Update on Vaginal 
Hysterectomy” in the September 2015 issue of this journal at obgmanagement.com.

Vaginal hysterectomy has 
few contraindications

Many commonly cited contraindications to the vaginal approach are not, in fact, absolute contraindications. An open or laparoscopic approach is preferred when the patient has a known cancer, of course, and when deep infiltrating endometriosis is present at the rectovaginal septum. However, previous pelvic surgery, nulliparity, an enlarged uterus, or lack of a prior vaginal delivery need not exclude the vaginal approach. Nor does a narrow introitus necessarily mandate a laparoscopic or open abdominal approach. In fact, in this article, I describe my basic technique in a patient (a cadaver) with a very narrow pubic arch, and I offer strategies for gaining some needed mobility and avoiding complications (TABLES 1 and 2).

Next month, in the November issue of OBG Management, John B. Gebhart, MD, will describe his vaginal technique for right salpingectomy with ovarian preservation, as well as his technique for right salpingo-oophorectomy.

Proper patient positioning 
is key

You can simplify the operation by positioning the patient so that her buttocks are over the edge of the table fairly far—at least 1 inch beyond the edge of the table for optimal exposure and greater access. If the patient is thin, it then becomes important to pad the sacrum because, when she is positioned that far off the table, all her weight comes to rest on the sacrum. In overweight patients, this is not an issue, but for thin patients, I place a bit of egg crate or gel beneath the sacrum.

For the procedure, I prefer to place my instruments on a tray that is kept on my lap. This arrangement frees the scrub technician from having to hand tools over my shoulder—and it saves time. I use a narrow, covered Mayo stand, and I place a stepstool beneath my feet to keep my knees at right angles so that things don’t slip during the operation.

Surgical technique

Choose an appropriate retractor

In a woman with a narrow introitus, I find that a posterior weighted speculum takes up too much space. Once I place a clamp on the cervix with that speculum in place, I don’t have much room to work. However, if I substitute a small Deaver retractor, which is narrower, I gain more workspace.

Inject the uterosacral ligaments

Grasp the cervix using a Jacobs vulsellum tenaculum. Use of a single tenaculum allows for much more movement than the use of instruments placed anteriorly and posteriorly. The Jacobs tenaculum obtains a better purchase on the tissue than a single tooth and is considerably less likely to tear through the tissue.

Before beginning the hysterectomy, locate the uterosacral ligaments and inject each one at its junction with the cervix, aspirating slightly before infiltrating the ligament with 0.25% to 0.50% bupivacaine with epinephrine, with dilute vasopressin mixed in. (I place 1 unit in 20 mL of the local solution.) Injection of this solution achieves 2 goals:

• improved intraoperative hemostasis
• postoperative pain relief.

Use a short needle with a needle extender attached to a control syringe rather than a spinal needle for greater control.

Enter the posterior peritoneal cavity

Before entering the peritoneal cavity, create a right angle with the Jacobs tenaculum and Deaver retractor in relation to the surgical field (FIGURE 1). This right angle is difficult to achieve when you are using a weighted speculum in a tight vagina. Once you have a right angle, tent the vaginal tissue in the midline (FIGURE 2).

In a nulliparous patient or a woman with a tight pelvis, you may discover that the peritoneum is pulled up between the uterosacral ligaments. One common pitfall arises when the surgeon, having dissected the vaginal epithelium, continues cutting into the vaginal epithelium instead of reaching into the peritoneal cavity. Palpate the tissue to ensure that there is no bowel in the way and stay at right angles while confidently grasping the peritoneum with a toothed forceps.

I like to use a bit of electrosurgery to incise the vaginal wall. I don’t begin at the cervix but incise more distally into the vaginal epithelium approximately 2 cm from the cervicovaginal junction. This strategy prevents dissection into the cervix and/or rectovaginal septum rather than the posterior 
cul-de-sac (FIGURE 3).

Once the incision is made, it is possible to feel the posterior peritoneum. And as you tent the peritoneum, you can then very confidently extend the incision and enter the cavity posteriorly.

In a patient with significant adhesions such as this one, I feel around posteriorly to determine exactly where I am. One tactic I use is to release the tenaculum and regrasp the cervix with it. This allows for improved visualization and movement of the cervix as the procedure progresses. Depending on the case, it may be necessary to insert a sponge to hold bowel out of the way.

Avoid the bladder

Move the Deaver retractor to the anterior position, switch the Jacobs clamp to the anterior cervix, and pull straight down. Now that you have incised the vaginal epithelium posteriorly, the length of the cervix should be apparent to you, and you can easily determine the location of the bladder reflection.

Keep in mind that, in a postmenopausal patient, there will be fewer vaginal rugae to guide you. Place the Jacobs tenaculum as close to the midline as possible so that you can confidently grab the tissue without fear of grabbing the bladder. If you tilt the Jacobs clamp, you can feel the edge of the bladder reflection. Remember that postmenopausal patients with prolapse (or, occasionally, obese patients with cervical elongation but little actual descensus) may have altered anatomy.

You can create a bit more space in which to dissect by injecting the bupivacaine/ 
epinephrine solution into the vaginal epithelium. This technique also ensures that the vaginal epithelial incisions won’t bleed.

Now, tilt the Jacobs tenaculum downward and push the junction of the cervix with the bladder reflection toward you so that you have a good sense of how deeply to incise.

Once you’ve made the incision, reclamp the Jacobs tenaculum so that it holds all of that tissue, and repeat the maneuver, tilting the clamp downward and pushing the junction toward you. In this way, you create traction and countertraction, sweeping the tissue out of your way.

Always use sharp dissection. When adhesions are present, surgeons often get into trouble using blunt dissection and may inadvertently enter the bladder if they use a sponge-covered digit for dissection, because adhesions can be much denser than normal tissue. In such cases, the bladder tears open rather than the adhesions being swept away.

Consider this: You don’t need to enter the peritoneal cavity anteriorly in order to continue working on the procedure. You can safely protect the bladder throughout the case, until the very end, if necessary, in patients who have undergone multiple previous surgeries or cesarean deliveries.

Rather than enter the anterior peritoneum, I dissect as much of the vaginal epithelium as possible and place a second Deaver retractor posteriorly.

I massage the uterosacral ligament for about 10 seconds to lengthen it and create more descensus, then place a Ballantine Heaney clamp on the ligament.

Next, I cut the pedicle and suture it, maintaining a clamp on the uterosacral ligament suture so that I can use it later for repair of the vaginal cuff.

I recommend a vessel-sealing device to secure the major blood supply, but I do suture the uterosacral and round ligaments for attachment to the apex at the conclusion of the hysterectomy. I suggest that you place straight clamps to hold the uterosacral ligament sutures and curved clamps on the round ligament ties to help you keep track of what you’re doing.

I generally prefer to use a smaller vessel-sealing device, such as the LigaSure Max (Covidien), because it allows me to take very small bites of tissue. It is also less expensive because it uses a disposable electrode within a reusable Heaney-type clamp.

Many people have argued that we need to teach surgeons to suture vaginally and, for that reason, should avoid vessel sealing. My response: Why wouldn’t we want to use the very best technology available? Randomized trials have demonstrated a 50% reduction in pain relief postoperatively when we use vessel sealing.⁴ Less foreign material is left in the pelvis, lowering the risk of infection. And it really doesn’t matter which vessel-sealing technology you use, as long as you’re familiar with the specifics of the system you choose. Another advantage: There is no need to pass needles back and forth.

Take small bites of tissue

Because this patient has a very small uterus, a small bite of tissue will get you close to where you want to be. When you take a bite with the vessel sealer, try to protect the vaginal epithelium and vulva from the steam that is emitted. The clamp itself does not heat up, but the steam that is released from the tissue is 100° C, so place a finger between the clamp and the sidewall for protection. It is preferable to burn your own finger than to burn the patient.

Because you haven’t entered the peritoneal cavity anteriorly, it is important to ensure that you don’t take too big of a bite with the vessel sealer. Rather, stay where you know you’ve done your dissection, where things are safe.

One cardinal principle of surgery is that you shouldn’t operate where you can’t feel or see. One of the common errors in vaginal surgery is that surgeons start dissecting higher than they can see. It’s easy to get into trouble when you start pushing tissue or dissecting tissue that you can’t visualize.

At this point, the anterior Deaver retractor is not essential, so I remove it. If you don’t need it, don’t use it. I try to avoid metal when I can.

If I were using suture rather than vessel sealing, I would place a Heaney clamp on the uterosacral ligament and cut. Using a clamp-cut-tie technique, I would pull on the pedicle and cut just beyond the tip of the clamp to ensure that the suture will be secure 
(FIGURE 4). This approach would not be appropriate during use of a vessel sealer. In that case, you would want to cut to but not beyond the tip of the clamp.

One of the skills helpful in suturing is learning to move your elbow and wrist to achieve the proper angle. Determine where you want the suture to exit the tissue, and then angle your elbow and wrist so that the suture comes out where you want it. It’s easy to lose track of the needle tip, especially when you’re working in a limited space under the pubic symphysis, so use your shoulder, elbow, and wrist to control 
suture placement.

Protect the anterior epithelium

Because you have not yet entered the peritoneal cavity anteriorly, it is important to protect the anterior epithelium and bladder. Reinsert a narrow Deaver retractor anteriorly, remove the Jacobs clamp, and replace the clamp laterally so that the cervix can be pulled off to the side (FIGURE 5).

One nice thing about some vessel sealers is that the surgeon can twist them in any direction. It isn’t necessary to move your hand; you simply move the device itself.

Once you have taken at least the descending branch of the uterine artery, remove the posterior retractor and pull downward on the Jacobs tenaculum. You should have reached just about to the level of the uterine fundus, with the anatomy well visualized (FIGURE 6). Next, open the anterior peritoneum.

Pay attention to the surgical field

Now that you have entered the peritoneum anteriorly as well as posteriorly, identify the broad ligament, keeping in mind that the ureter is retroperitoneal, not intraperitoneal. If you were to place a clamp from the posterior leaf of the broad ligament across to the anterior leaf of the broad ligament, you would be grasping all the vessels but not the ureter. In fact, the anterior Deaver retractor is lifting both ureters up and out of the way. If you pull the cervix off to the opposite side, you create an additional couple of centimeters—a safe space for the vessel sealer 
(FIGURE 7).

In placing the vessel sealer, there is no need to move out laterally, as there is no need for space to place suture. Instead, hug the uterus. At this point, the main concern is the risk of damaging any small bowel behind the uterine fundus that might be coming down into the surgical field, obscured from vision. And because there may be steam emitted at the tip of the vessel-sealing clamp, keep a finger back there to protect anything that might be in the field.

Last steps

Before taking the last bite of tissue on the right-hand side, place the round ligament in a Heaney clamp. Now that the round and utero-ovarian ligaments have been skeletonized, you can grasp the pedicle in a clamp. If the pedicle is especially thick, it may be beneficial to close the clamp, leave it on for a few seconds, and then reapply it. In that way, you obtain a better purchase.

Next, free the rest of the tissue with a vessel sealer, or cut it. I prefer to use a vessel sealer, and I again protect the adjacent tissue with my fingers anteriorly and posteriorly.

With the clamp remaining on the round ligament and utero-ovarian ligament 
(FIGURE 8), which will be sutured, push the uterine tissue out of the way, back into the pelvis, to make room for suturing.

Because a postmenopausal vulva may be cut by the suture, it’s important to take pains not to abrade that tissue. Once you have finished suturing the round/utero-ovarian pedicle, leave the needle on the suture so that you can reconnect the round ligament to the anterior pubocervical ring to reconstruct the vaginal apex. For safekeeping, clamp the needle out of the way and tuck it beneath the drape.

Switching to the other side, use a Lahey clamp to flip the uterus, then clamp the pedicle and use the vessel sealer to separate it, again protecting the tissue beneath and ahead of the clamp. Sometimes, with an especially thick pedicle, the vessel sealer will signal that the tissue hasn’t been completely sealed. In that case, get another purchase of the pedicle, protect the adjacent tissue, and seal again.

Once the uterus has been removed completely, suture the utero-ovarian and round ligament on this side.
One tip to aid in the placement of suture is to move your clamped tissue in such a way as to prevent inadvertent suturing of other tissue (FIGURE 9).

An additional strategy for pain relief at this point is to infiltrate the round ligaments with local anesthetic. We know that we’re working with higher-level fibers—T10 to T12—through the round ligaments. By infiltrating them with anesthetic, you achieve denser pain relief for post- 
operative management.

Uterine reduction strategies facilitate vaginal removal of tissue

When the uterus is too large to remove intact through the vagina, there are a number of techniques for coring, wedging, and morcellating the tissue. As always, a complete knowledge of anatomy is essential, as well as an understanding that fibroids can frequently distort the uterus, twisting it to the left or right. It is important to anticipate such distortion to avoid the inadvertent destruction of anatomic landmarks or damage to the adnexae.

One straight-forward strategy is to debulk the uterus using a knife to core it, removing the central portion. In cases in which you need to keep the entire endometrial cavity intact, you can core the central portion of the uterus while grasping the cervix so that you can remove the endometrium intact for the pathologist (FIGURE).

For this strategy it is important to protect the vaginal sidewalls with metal. You can use another retractor to do that, pulling down on the cervix and beginning the morcellation. I generally prefer to use a short knife handle only because I want to be sure I’m not tempted to cut any higher than I can see.

For more on coring and wedging techniques, see the introductory video for the ACOG/SGS/AAGL master class on vaginal hysterectomy at http://www.aagl.org/vaghystwebinar.

Close the vaginal cuff

The reconstruction of the vaginal cuff is a critical component of any hysterectomy. My approach is to reattach the uterosacral ligaments to the posterior cuff and the round ligaments to the anterior cuff, thereby re- 
creating an intact pubocervical ring. It is not necessary to include the peritoneum in the cuff closure. In fact, kinking of the ureters is more likely when the peritoneum is closed.

Attach one uterosacral ligament, then place a running, full-thickness stitch across the posterior cuff, and attach the uterosacral ligament on the opposite side. Use the needle you left attached to the round ligament to bring the right pedicle to the anterior cuff at 10 o’clock (be sure you grasp the full thickness of the vaginal epithelium without compromising the bladder). Attach the left round-ligament pedicle at the 2 o’clock position. Then close the cuff side to side down to the uterosacral ligaments. This completely reconstructs the pubocervical ring and provides excellent support at the apex.

Share your thoughts on this article! Send your Letter to the Editor to rbarbieri@frontlinemedcom.com. Please include your name and the city and state in which you practice.

Clinicians work to maximize the quality of life and longevity of every patient. For women with moderate to severe menopausal symptoms, oral estrogen therapy can improve quality of life, but at the cost of significant adverse effects. The Women’s Health Initiative
 (WHI) reported that for postmenopausal women with a uterus, 
conjugated estrogen plus medroxyprogesterone acetate (CEE+MPA) hormone therapy (HT) versus placebo 
significantly increased the risk of 
cardiovascular events (relative risk 
[RR], 1.13), breast cancer (RR, 1.24), 
stroke (RR, 1.37), deep vein thrombosis (RR, 1.87), and pulmonary 
embolism (RR, 1.98).¹ In postmeno
pausal women without a uterus, CEE 
HT did not increase the risk of breast 
cancer (RR, 0.79), compared with 
placebo, but it did significantly in
crease the risk of cardiovascular 
events (RR, 1.11), stroke (RR, 1.35), 
deep vein thrombosis (RR, 1.48), and 
pulmonary embolism (RR, 1.35).¹

Clinicians prescribing estrogen must individualize therapy according
 to its benefits and risks. An important issue that has received insufficient at
tention is, “What is the effect of HT 
on mortality in recently menopausal women?” Here, I examine this issue.

HT reduces mortality in 
recently menopausal women

Pooling the results of the WHI CEE+MPA and CEE-only trials 
reveals that there were 70 deaths in the HT-treated groups and 98 deaths in the placebo groups among women aged 50 to 59 years.¹ With 4,706 and 4,259 women alive at the conclusion of the study in the HT and placebo groups, respectively, the women in the placebo group had significantly more deaths than the women in the HT-treated groups (Fisher exact test, P = .0194, χ² test with Yates correction, P = .0226).

Using pooled data from the WHI, the RR of death in the HT versus placebo group was estimated at 0.70 (95% confidence interval [CI], 0.51−0.96), representing approximately 5 fewer deaths per 
1,000 women per 5 years of therapy.² In women aged 60 to 69 years and 70 to 79 years there were no significant differences in death rates between the HT- and placebo-treated women.

My interpretation of these results is that HT likely is associated with a reduced risk of death in recently menopausal women, but not in 
women distant from menopause onset.

Cochrane review of 
HT and mortality

Consistent with the WHI findings, authors of a recent Cochrane 
meta-analysis of 19 randomized trials including 40,410 menopausal women reported that HT significantly increased the risk of stroke (RR, 1.24; 95% CI, 1.10−1.41), venous thromboembolism (RR, 1.92; 95% CI, 1.36−2.69), and pulmonary emboli (RR, 1.81; 95% CI, 1.32−2.48).³ However, among women treated with oral HT within 10 years after the start of menopause, there was a reduced risk of coronary heart disease (RR, 0.52; 95% CI, 0.29−0.96). Using data from 5 clinical trials, the Cochrane meta-analysis researchers reported that, compared with placebo, HT reduced mortality (RR, 0.70; 95% CI, 0.52−0.95).³

Results of the Cochrane meta-analysis are consistent with those of a previous meta-analysis of 
19 randomized trials involving 16,000 women. In this analysis, investigators found a reduced risk of death in recently menopausal women treated with hormone therapy (RR, 0.73; 95% CI, 0.52−0.96).⁴

Early menopause, 
HT, and mortality

Authors of multiple large epidemiologic studies have reported that early menopause is associated with an increased risk of death if HT is not initiated.⁵⁻⁷ For example, results of a study of women in Olmsted County, Minnesota, conducted from 1950 to 1987, indicated that, for women younger than age 45 years who underwent bilateral oophorectomy, the risk of death was increased among those who did not initiate HT, compared with women who did not undergo oophorectomy (hazard ratio [HR], 1.84; 95% CI, 1.27−2.68; 
P = .001).⁷

By contrast, women younger than 45 years who underwent bilateral oophorectomy and initiated estrogen therapy did not have an increased risk of death compared with women who did not undergo oophorectomy (HR, 0.65; 95% CI, 0.30−1.41; P = .28).⁷ An excess number of cardiovascular events appeared to account for the increased mortality among women with early surgical menopause who did not initiate HT.

The “timing hypothesis” proposes that the initiation of HT soon after the onset of menopause is associated with beneficial cardiovascular effects, but initiation more than 10 years after the onset of menopause is not associated with beneficial cardiovascular effects. The timing hypothesis is supported by the finding that, in recently menopausal women, HT is associated with reduced carotid intima-media thickness (CIMT), compared with placebo.⁸ Greater CIMT thickness is associated with an increased risk of cardiovascular events.

In my experience, few primary care clinicians are aware of these data. Often, these clinicians over-emphasize the risks and withhold HT in this vulnerable group of women.

HT: Minimizing the risks 
of stroke, deep vein thrombosis, pulmonary embolism, and 
breast cancer

Results of multiple studies have shown that certain HT regimens increase the risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer. Is it possible to prescribe HT in a way that reduces these risks?

Results of observational studies indicate that, compared with oral estrogen therapy, transdermal HT is associated with a lower risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer 
(TABLE).⁹⁻¹⁵

Reducing the risk of stroke caused by HT is an important goal. In a study of 15,710 women who had stroke and 59,958 control women aged 50 to 79 years, transdermal estradiol at a dose of 50 µg or less daily was not associated with an increased risk of stroke, compared with HT nonuse (rate ratio, 0.81; 95% CI, 0.62−1.05).⁹ Compared with HT nonuse, the use of oral estrogen (rate ratio, 1.28; 95% CI, 1.15−1.42) or transdermal estradiol 50 µg or greater daily (rate ratio, 1.89; 95% CI, 1.15−3.11) was associated with an increased risk of stroke.⁹

Reducing the risks of deep venous thromboembolism (VTE) and pulmonary embolism caused by HT is an important goal. In a meta-analysis of the risk of VTE with HT, compared with nonusers, oral estrogen therapy was associated with a significantly increased risk of VTE (odds ratio [OR], 2.5; 95% CI, 1.9−3.4). Compared with nonuse, transdermal estrogen therapy was not associated with an increased risk of VTE (OR, 1.2; 95% CI, 0.9−1.7).¹¹ In a study comparing oral versus transdermal estradiol, transdermal estradiol was associated with a reduced risk of pulmonary embolism (0.46 [95% CI, 0.22−0.97]).¹³

Reducing the risk of breast cancer caused by HT is an important goal. Results of one study showed that the combination of oral estrogen plus synthetic progestin was associated with an increased risk of breast cancer, compared with nonuse (RR, 1.5; 95% CI, 1.1−1.9). By contrast, the combination of transdermal estradiol plus micronized progesterone was not associated with an increased risk of breast cancer, compared with nonuse (RR, 0.9; 95% CI, 0.7−1.2).¹⁵

INSTANT POLL
Many health insurers use pharmacy benefit managers to control the cost of prescription medicines. These managers often develop formulary algorithms that favor the use of oral estrogen and medroxyprogesterone acetate over transdermal estradiol and micronized progesterone. When you prescribe transdermal estradiol and micronized progesterone, have your patients had difficulty filling the prescription?

Tell us! Send your Letter to the Editor!

The bottom line

In recently menopausal women with moderate to severe hot flashes, HT improves quality of life and appears to decrease mortality. However, HT with oral estrogen plus synthetic progestin is associated with an increased risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer. Compared with oral estrogen, transdermal estradiol treatment is associated with a lower risk of stroke, deep vein thrombosis, and pulmonary embolism. Compared with oral estrogen plus a synthetic progestin, transdermal estradiol plus micronized progesterone is associated with a lower risk of breast cancer. The benefits of HT are likely maximized by initiating therapy in the perimenopause transition or early in the postmenopause, and the risks are minimized by using transdermal estradiol.¹⁶⁻¹⁸

Share your thoughts on this article! Send your Letter to the Editor to rbarbieri@frontlinemedcom.com. Please include your name and the city and state in which you practice.

Clinicians work to maximize the quality of life and longevity of every patient. For women with moderate to severe menopausal symptoms, oral estrogen therapy can improve quality of life, but at the cost of significant adverse effects. The Women’s Health Initiative
 (WHI) reported that for postmenopausal women with a uterus, 
conjugated estrogen plus medroxyprogesterone acetate (CEE+MPA) hormone therapy (HT) versus placebo 
significantly increased the risk of 
cardiovascular events (relative risk 
[RR], 1.13), breast cancer (RR, 1.24), 
stroke (RR, 1.37), deep vein thrombosis (RR, 1.87), and pulmonary 
embolism (RR, 1.98).¹ In postmeno
pausal women without a uterus, CEE 
HT did not increase the risk of breast 
cancer (RR, 0.79), compared with 
placebo, but it did significantly in
crease the risk of cardiovascular 
events (RR, 1.11), stroke (RR, 1.35), 
deep vein thrombosis (RR, 1.48), and 
pulmonary embolism (RR, 1.35).¹

Clinicians prescribing estrogen must individualize therapy according
 to its benefits and risks. An important issue that has received insufficient at
tention is, “What is the effect of HT 
on mortality in recently menopausal women?” Here, I examine this issue.

HT reduces mortality in 
recently menopausal women

Pooling the results of the WHI CEE+MPA and CEE-only trials 
reveals that there were 70 deaths in the HT-treated groups and 98 deaths in the placebo groups among women aged 50 to 59 years.¹ With 4,706 and 4,259 women alive at the conclusion of the study in the HT and placebo groups, respectively, the women in the placebo group had significantly more deaths than the women in the HT-treated groups (Fisher exact test, P = .0194, χ² test with Yates correction, P = .0226).

Using pooled data from the WHI, the RR of death in the HT versus placebo group was estimated at 0.70 (95% confidence interval [CI], 0.51−0.96), representing approximately 5 fewer deaths per 
1,000 women per 5 years of therapy.² In women aged 60 to 69 years and 70 to 79 years there were no significant differences in death rates between the HT- and placebo-treated women.

My interpretation of these results is that HT likely is associated with a reduced risk of death in recently menopausal women, but not in 
women distant from menopause onset.

Cochrane review of 
HT and mortality

Consistent with the WHI findings, authors of a recent Cochrane 
meta-analysis of 19 randomized trials including 40,410 menopausal women reported that HT significantly increased the risk of stroke (RR, 1.24; 95% CI, 1.10−1.41), venous thromboembolism (RR, 1.92; 95% CI, 1.36−2.69), and pulmonary emboli (RR, 1.81; 95% CI, 1.32−2.48).³ However, among women treated with oral HT within 10 years after the start of menopause, there was a reduced risk of coronary heart disease (RR, 0.52; 95% CI, 0.29−0.96). Using data from 5 clinical trials, the Cochrane meta-analysis researchers reported that, compared with placebo, HT reduced mortality (RR, 0.70; 95% CI, 0.52−0.95).³

Results of the Cochrane meta-analysis are consistent with those of a previous meta-analysis of 
19 randomized trials involving 16,000 women. In this analysis, investigators found a reduced risk of death in recently menopausal women treated with hormone therapy (RR, 0.73; 95% CI, 0.52−0.96).⁴

Early menopause, 
HT, and mortality

Authors of multiple large epidemiologic studies have reported that early menopause is associated with an increased risk of death if HT is not initiated.⁵⁻⁷ For example, results of a study of women in Olmsted County, Minnesota, conducted from 1950 to 1987, indicated that, for women younger than age 45 years who underwent bilateral oophorectomy, the risk of death was increased among those who did not initiate HT, compared with women who did not undergo oophorectomy (hazard ratio [HR], 1.84; 95% CI, 1.27−2.68; 
P = .001).⁷

By contrast, women younger than 45 years who underwent bilateral oophorectomy and initiated estrogen therapy did not have an increased risk of death compared with women who did not undergo oophorectomy (HR, 0.65; 95% CI, 0.30−1.41; P = .28).⁷ An excess number of cardiovascular events appeared to account for the increased mortality among women with early surgical menopause who did not initiate HT.

The “timing hypothesis” proposes that the initiation of HT soon after the onset of menopause is associated with beneficial cardiovascular effects, but initiation more than 10 years after the onset of menopause is not associated with beneficial cardiovascular effects. The timing hypothesis is supported by the finding that, in recently menopausal women, HT is associated with reduced carotid intima-media thickness (CIMT), compared with placebo.⁸ Greater CIMT thickness is associated with an increased risk of cardiovascular events.

In my experience, few primary care clinicians are aware of these data. Often, these clinicians over-emphasize the risks and withhold HT in this vulnerable group of women.

HT: Minimizing the risks 
of stroke, deep vein thrombosis, pulmonary embolism, and 
breast cancer

Results of multiple studies have shown that certain HT regimens increase the risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer. Is it possible to prescribe HT in a way that reduces these risks?

Results of observational studies indicate that, compared with oral estrogen therapy, transdermal HT is associated with a lower risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer 
(TABLE).⁹⁻¹⁵

Reducing the risk of stroke caused by HT is an important goal. In a study of 15,710 women who had stroke and 59,958 control women aged 50 to 79 years, transdermal estradiol at a dose of 50 µg or less daily was not associated with an increased risk of stroke, compared with HT nonuse (rate ratio, 0.81; 95% CI, 0.62−1.05).⁹ Compared with HT nonuse, the use of oral estrogen (rate ratio, 1.28; 95% CI, 1.15−1.42) or transdermal estradiol 50 µg or greater daily (rate ratio, 1.89; 95% CI, 1.15−3.11) was associated with an increased risk of stroke.⁹

Reducing the risks of deep venous thromboembolism (VTE) and pulmonary embolism caused by HT is an important goal. In a meta-analysis of the risk of VTE with HT, compared with nonusers, oral estrogen therapy was associated with a significantly increased risk of VTE (odds ratio [OR], 2.5; 95% CI, 1.9−3.4). Compared with nonuse, transdermal estrogen therapy was not associated with an increased risk of VTE (OR, 1.2; 95% CI, 0.9−1.7).¹¹ In a study comparing oral versus transdermal estradiol, transdermal estradiol was associated with a reduced risk of pulmonary embolism (0.46 [95% CI, 0.22−0.97]).¹³

Reducing the risk of breast cancer caused by HT is an important goal. Results of one study showed that the combination of oral estrogen plus synthetic progestin was associated with an increased risk of breast cancer, compared with nonuse (RR, 1.5; 95% CI, 1.1−1.9). By contrast, the combination of transdermal estradiol plus micronized progesterone was not associated with an increased risk of breast cancer, compared with nonuse (RR, 0.9; 95% CI, 0.7−1.2).¹⁵

INSTANT POLL
Many health insurers use pharmacy benefit managers to control the cost of prescription medicines. These managers often develop formulary algorithms that favor the use of oral estrogen and medroxyprogesterone acetate over transdermal estradiol and micronized progesterone. When you prescribe transdermal estradiol and micronized progesterone, have your patients had difficulty filling the prescription?

Tell us! Send your Letter to the Editor!

The bottom line

In recently menopausal women with moderate to severe hot flashes, HT improves quality of life and appears to decrease mortality. However, HT with oral estrogen plus synthetic progestin is associated with an increased risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer. Compared with oral estrogen, transdermal estradiol treatment is associated with a lower risk of stroke, deep vein thrombosis, and pulmonary embolism. Compared with oral estrogen plus a synthetic progestin, transdermal estradiol plus micronized progesterone is associated with a lower risk of breast cancer. The benefits of HT are likely maximized by initiating therapy in the perimenopause transition or early in the postmenopause, and the risks are minimized by using transdermal estradiol.¹⁶⁻¹⁸

Share your thoughts on this article! Send your Letter to the Editor to rbarbieri@frontlinemedcom.com. Please include your name and the city and state in which you practice.

Clinicians work to maximize the quality of life and longevity of every patient. For women with moderate to severe menopausal symptoms, oral estrogen therapy can improve quality of life, but at the cost of significant adverse effects. The Women’s Health Initiative
 (WHI) reported that for postmenopausal women with a uterus, 
conjugated estrogen plus medroxyprogesterone acetate (CEE+MPA) hormone therapy (HT) versus placebo 
significantly increased the risk of 
cardiovascular events (relative risk 
[RR], 1.13), breast cancer (RR, 1.24), 
stroke (RR, 1.37), deep vein thrombosis (RR, 1.87), and pulmonary 
embolism (RR, 1.98).¹ In postmeno
pausal women without a uterus, CEE 
HT did not increase the risk of breast 
cancer (RR, 0.79), compared with 
placebo, but it did significantly in
crease the risk of cardiovascular 
events (RR, 1.11), stroke (RR, 1.35), 
deep vein thrombosis (RR, 1.48), and 
pulmonary embolism (RR, 1.35).¹

Clinicians prescribing estrogen must individualize therapy according
 to its benefits and risks. An important issue that has received insufficient at
tention is, “What is the effect of HT 
on mortality in recently menopausal women?” Here, I examine this issue.

HT reduces mortality in 
recently menopausal women

Pooling the results of the WHI CEE+MPA and CEE-only trials 
reveals that there were 70 deaths in the HT-treated groups and 98 deaths in the placebo groups among women aged 50 to 59 years.¹ With 4,706 and 4,259 women alive at the conclusion of the study in the HT and placebo groups, respectively, the women in the placebo group had significantly more deaths than the women in the HT-treated groups (Fisher exact test, P = .0194, χ² test with Yates correction, P = .0226).

Using pooled data from the WHI, the RR of death in the HT versus placebo group was estimated at 0.70 (95% confidence interval [CI], 0.51−0.96), representing approximately 5 fewer deaths per 
1,000 women per 5 years of therapy.² In women aged 60 to 69 years and 70 to 79 years there were no significant differences in death rates between the HT- and placebo-treated women.

My interpretation of these results is that HT likely is associated with a reduced risk of death in recently menopausal women, but not in 
women distant from menopause onset.

Cochrane review of 
HT and mortality

Consistent with the WHI findings, authors of a recent Cochrane 
meta-analysis of 19 randomized trials including 40,410 menopausal women reported that HT significantly increased the risk of stroke (RR, 1.24; 95% CI, 1.10−1.41), venous thromboembolism (RR, 1.92; 95% CI, 1.36−2.69), and pulmonary emboli (RR, 1.81; 95% CI, 1.32−2.48).³ However, among women treated with oral HT within 10 years after the start of menopause, there was a reduced risk of coronary heart disease (RR, 0.52; 95% CI, 0.29−0.96). Using data from 5 clinical trials, the Cochrane meta-analysis researchers reported that, compared with placebo, HT reduced mortality (RR, 0.70; 95% CI, 0.52−0.95).³

Results of the Cochrane meta-analysis are consistent with those of a previous meta-analysis of 
19 randomized trials involving 16,000 women. In this analysis, investigators found a reduced risk of death in recently menopausal women treated with hormone therapy (RR, 0.73; 95% CI, 0.52−0.96).⁴

Early menopause, 
HT, and mortality

Authors of multiple large epidemiologic studies have reported that early menopause is associated with an increased risk of death if HT is not initiated.⁵⁻⁷ For example, results of a study of women in Olmsted County, Minnesota, conducted from 1950 to 1987, indicated that, for women younger than age 45 years who underwent bilateral oophorectomy, the risk of death was increased among those who did not initiate HT, compared with women who did not undergo oophorectomy (hazard ratio [HR], 1.84; 95% CI, 1.27−2.68; 
P = .001).⁷

By contrast, women younger than 45 years who underwent bilateral oophorectomy and initiated estrogen therapy did not have an increased risk of death compared with women who did not undergo oophorectomy (HR, 0.65; 95% CI, 0.30−1.41; P = .28).⁷ An excess number of cardiovascular events appeared to account for the increased mortality among women with early surgical menopause who did not initiate HT.

The “timing hypothesis” proposes that the initiation of HT soon after the onset of menopause is associated with beneficial cardiovascular effects, but initiation more than 10 years after the onset of menopause is not associated with beneficial cardiovascular effects. The timing hypothesis is supported by the finding that, in recently menopausal women, HT is associated with reduced carotid intima-media thickness (CIMT), compared with placebo.⁸ Greater CIMT thickness is associated with an increased risk of cardiovascular events.

In my experience, few primary care clinicians are aware of these data. Often, these clinicians over-emphasize the risks and withhold HT in this vulnerable group of women.

HT: Minimizing the risks 
of stroke, deep vein thrombosis, pulmonary embolism, and 
breast cancer

Results of multiple studies have shown that certain HT regimens increase the risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer. Is it possible to prescribe HT in a way that reduces these risks?

Results of observational studies indicate that, compared with oral estrogen therapy, transdermal HT is associated with a lower risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer 
(TABLE).⁹⁻¹⁵

Reducing the risk of stroke caused by HT is an important goal. In a study of 15,710 women who had stroke and 59,958 control women aged 50 to 79 years, transdermal estradiol at a dose of 50 µg or less daily was not associated with an increased risk of stroke, compared with HT nonuse (rate ratio, 0.81; 95% CI, 0.62−1.05).⁹ Compared with HT nonuse, the use of oral estrogen (rate ratio, 1.28; 95% CI, 1.15−1.42) or transdermal estradiol 50 µg or greater daily (rate ratio, 1.89; 95% CI, 1.15−3.11) was associated with an increased risk of stroke.⁹

Reducing the risks of deep venous thromboembolism (VTE) and pulmonary embolism caused by HT is an important goal. In a meta-analysis of the risk of VTE with HT, compared with nonusers, oral estrogen therapy was associated with a significantly increased risk of VTE (odds ratio [OR], 2.5; 95% CI, 1.9−3.4). Compared with nonuse, transdermal estrogen therapy was not associated with an increased risk of VTE (OR, 1.2; 95% CI, 0.9−1.7).¹¹ In a study comparing oral versus transdermal estradiol, transdermal estradiol was associated with a reduced risk of pulmonary embolism (0.46 [95% CI, 0.22−0.97]).¹³

Reducing the risk of breast cancer caused by HT is an important goal. Results of one study showed that the combination of oral estrogen plus synthetic progestin was associated with an increased risk of breast cancer, compared with nonuse (RR, 1.5; 95% CI, 1.1−1.9). By contrast, the combination of transdermal estradiol plus micronized progesterone was not associated with an increased risk of breast cancer, compared with nonuse (RR, 0.9; 95% CI, 0.7−1.2).¹⁵

INSTANT POLL
Many health insurers use pharmacy benefit managers to control the cost of prescription medicines. These managers often develop formulary algorithms that favor the use of oral estrogen and medroxyprogesterone acetate over transdermal estradiol and micronized progesterone. When you prescribe transdermal estradiol and micronized progesterone, have your patients had difficulty filling the prescription?

Tell us! Send your Letter to the Editor!

The bottom line

In recently menopausal women with moderate to severe hot flashes, HT improves quality of life and appears to decrease mortality. However, HT with oral estrogen plus synthetic progestin is associated with an increased risk of stroke, deep vein thrombosis, pulmonary embolism, and breast cancer. Compared with oral estrogen, transdermal estradiol treatment is associated with a lower risk of stroke, deep vein thrombosis, and pulmonary embolism. Compared with oral estrogen plus a synthetic progestin, transdermal estradiol plus micronized progesterone is associated with a lower risk of breast cancer. The benefits of HT are likely maximized by initiating therapy in the perimenopause transition or early in the postmenopause, and the risks are minimized by using transdermal estradiol.¹⁶⁻¹⁸

Share your thoughts on this article! Send your Letter to the Editor to rbarbieri@frontlinemedcom.com. Please include your name and the city and state in which you practice.

Click here to download the PDF.
Every surgeon knows that, if you operate, complications will follow. Surgeons are perfectionists and strive to reduce complications through years of diligent personal training and continuous quality improvement. Surgeons hate complications, especially those that might be preventable (such as retained foreign bodies, wrong site surgery, and medication errors).

In this special issue of OBG Management, world-renowned experts focus on topical issues in safety in minimally invasive gynecologic surgery (MIGS). In a roundtable, Drs. Neal Lonky, John Gebhart, Rosanne Kho, and Malcolm Munro discuss important issues in MIGS, including the need to prioritize the vaginal and laparoscopic routes of hysterectomy and the role of single-port surgery. In a concise yet detailed discussion of safety issues related to radiofrequency and ultrasound energy devices, Dr. Munro alerts us to the dangers of heat injury and direct and capacitative coupling. Drs. Andrew Sokol and Katelyn Smithling provide guidance on preventing apical prolapse of the vagina following hysterectomy. And Dr. Antonio Gargiulo explores the rapidly expanding role of simulation training with computer-assisted (robotic) surgical simulators and predicts that, instead of learning surgery on patients, future trainees will gain skills in a simulation environment.

We thank the master surgeons who participated in this special issue for providing guidance and helping us to reduce complications. Our patients are the beneficiaries of the wisdom provided herein.

>>Robert L. Barbieri, MD
Editor in Chief, OBG Management

To view the video, "McCall" culdoplasty technique by Mickey Karram, MD, that accompanies the article by Sokol et al in this supplement, click here.

Click here to download the PDF.
Every surgeon knows that, if you operate, complications will follow. Surgeons are perfectionists and strive to reduce complications through years of diligent personal training and continuous quality improvement. Surgeons hate complications, especially those that might be preventable (such as retained foreign bodies, wrong site surgery, and medication errors).

In this special issue of OBG Management, world-renowned experts focus on topical issues in safety in minimally invasive gynecologic surgery (MIGS). In a roundtable, Drs. Neal Lonky, John Gebhart, Rosanne Kho, and Malcolm Munro discuss important issues in MIGS, including the need to prioritize the vaginal and laparoscopic routes of hysterectomy and the role of single-port surgery. In a concise yet detailed discussion of safety issues related to radiofrequency and ultrasound energy devices, Dr. Munro alerts us to the dangers of heat injury and direct and capacitative coupling. Drs. Andrew Sokol and Katelyn Smithling provide guidance on preventing apical prolapse of the vagina following hysterectomy. And Dr. Antonio Gargiulo explores the rapidly expanding role of simulation training with computer-assisted (robotic) surgical simulators and predicts that, instead of learning surgery on patients, future trainees will gain skills in a simulation environment.

We thank the master surgeons who participated in this special issue for providing guidance and helping us to reduce complications. Our patients are the beneficiaries of the wisdom provided herein.

>>Robert L. Barbieri, MD
Editor in Chief, OBG Management

To view the video, "McCall" culdoplasty technique by Mickey Karram, MD, that accompanies the article by Sokol et al in this supplement, click here.

Click here to download the PDF.
Every surgeon knows that, if you operate, complications will follow. Surgeons are perfectionists and strive to reduce complications through years of diligent personal training and continuous quality improvement. Surgeons hate complications, especially those that might be preventable (such as retained foreign bodies, wrong site surgery, and medication errors).

In this special issue of OBG Management, world-renowned experts focus on topical issues in safety in minimally invasive gynecologic surgery (MIGS). In a roundtable, Drs. Neal Lonky, John Gebhart, Rosanne Kho, and Malcolm Munro discuss important issues in MIGS, including the need to prioritize the vaginal and laparoscopic routes of hysterectomy and the role of single-port surgery. In a concise yet detailed discussion of safety issues related to radiofrequency and ultrasound energy devices, Dr. Munro alerts us to the dangers of heat injury and direct and capacitative coupling. Drs. Andrew Sokol and Katelyn Smithling provide guidance on preventing apical prolapse of the vagina following hysterectomy. And Dr. Antonio Gargiulo explores the rapidly expanding role of simulation training with computer-assisted (robotic) surgical simulators and predicts that, instead of learning surgery on patients, future trainees will gain skills in a simulation environment.

We thank the master surgeons who participated in this special issue for providing guidance and helping us to reduce complications. Our patients are the beneficiaries of the wisdom provided herein.

>>Robert L. Barbieri, MD
Editor in Chief, OBG Management

To view the video, "McCall" culdoplasty technique by Mickey Karram, MD, that accompanies the article by Sokol et al in this supplement, click here.

CHICAGO – Despite exciting new advances in the understanding of chronic lymphocytic leukemia, particularly with respect to prognostic features that predict risk for relapse, a watch-and-wait approach remains appropriate for asymptomatic disease pending outcomes data for newer approaches, according to Dr. John G. Gribben.

“When the disease is diagnosed, if it is asymptomatic, the correct approach – of course – is to continue to watch and wait,” Dr. Gribben of Barts Cancer Institute, Queen Mary University of London, said at the American Society of Hematology Meeting on Hematologic Malignancies.

Numerous clinical trials have demonstrated no advantage of early treatment vs. watch and wait, he said, adding that all of the trials published to date have used treatment of all-comers, and have used chlorambucil (CLB) as the treatment.

“There has been a whole variety of more modern trials that have used select prognostic features to identify subgroups of people who are at higher risk of relapse, who then go on to receive earlier treatment with either FCR [fludarabine, cyclophosphamide, rituximab], or more recently, ibrutinib,” he said.

These treatments are interesting, and the trials have demonstrated that prognostic features can identify patients who will progress more rapidly, but none have reported, he explained.

“In the absence of any published trial, I continue to ‘watch and wait’ patients, and there are no high-risk features which will make me alter that approach. Even the highest-risk features of complex karyotype and p53 abnormalities are not indications to treat patients until they become symptomatic,” he said.

It is striking how white counts vary widely in both asymptomatic and symptomatic patients, he noted.

“I don’t personally have any particular white count which is the number at which I’ll treat a patient. I don’t treat white counts, I treat patients,” he said.

When patients become symptomatic, the treatment of choice is now immunochemotherapies, irrespective of performance status, he said.

“Within the past year we have seen approval of obinutuzumab and ofatumumab for treatment of previously untreated CLL, as well as ibrutinib and idelalisib plus rituximab for treatment of both previously untreated CLL and those with 17p deletions for relapsed/refractory disease, as well as for up-front treatment,” he said, adding that these new agents greatly increase the available options for treating CLL.

Dr. Gribben said he considers these questions when it comes to treating CLL:

• Does the patient require treatment, or is watching and waiting appropriate?

• What is the goal of therapy? This is determined through conversation with the patient and the patient’s family regarding the side-effect profile they are willing to tolerate vs. the potential longer duration of response.

• What comorbidities are present to determine “fitness” for specific immunochemotherapy? Specifically, is the patient fit for an FCR-type approach, or is an alternative more appropriate?

• Is there a 17p deletion or P53 mutation that would make chemotherapy a less attractive option, and use of novel agents a more attractive option?

His approach, based on the answers to these questions, is as follows:

• In Rai stage 0-II patients with inactive disease, fitness and 17p deletion or p53 mutation status is irrelevant; no therapy is given.

• For active disease or Rai III-IV disease, a “go-go” patient, (or patient in good physical condition) is treated based on the presence or absence of 17p deletion or p53 mutation status. Those without 17p deletions or p53 mutations can be treated with FCR (his preference), or fludarabine-rituximab (FR). Bendamustine-rituximab (BR) is also an attractive option in certain cases, he said.

• For patients with active disease or Rai stage III-IV disease who do have a 17p deletion, his treatment of choice is either ibrutinib or idelalisib plus rituximab, depending on the patient.

• In “slow-go” patients (those with poorer physical condition) treatment is again based on mutational analysis. Those without mutation receive either FR, BR, or CLB plus obinutuzumab. These are very good options, and represent a spectrum to choose from based on the patient’s core abilities and ability to withstand particular types of treatments, he said.

“If they do have a 17p deletion, these patients are just as eligible for ibrutinib or idelalisib plus rituximab as the younger patients,” he noted.

His choices are based largely on the findings from the CLL8 trial (Lancet 2010 Oct;376[9747]:1164-74) which demonstrated an overall survival advantage with chemoimmunotherapy for front-line therapy vs. chemotherapy alone (hazard ratio, 0.68).

Over time, the advantage has become even more pronounced, according to follow-up data.

Starting with something “gentle” and saving the best treatment for later in the event of relapse was recently considered a reasonable approach, but in the wake of the CLL8 findings, this is no longer an acceptable plan, Dr. Gribben said.

“That’s why for my choice, FCR remains the treatment of choice for those patients who are fit enough to tolerate this type of approach,” he said.

In those with 17p deletions or P53 mutations, the CLL8 trial showed poor outcomes with FCR.

“This is a group of patients whom I believe chemoimmunotherapy would no longer be the treatment of choice,” he said, adding that newer findings suggest outcomes in these patients are better with novel agents.

He also noted that patients with 11q abnormalities, which were previously associated with a poor prognosis, were found in CLL8 to respond well to chemoimmunotherapy when used front line.

While there are special considerations in the elderly, and different strategies in relapsed and refractory disease, the future of CLL treatment is promising. The benefit of adding rituximab to combination chemotherapy is well established, the benefit of novel agents is also now established, and the future likely involves combining targeted therapies with each other and with immunochemotherapies, and combining targeted therapies across different pathways.

“And of course the hope is that we’re going to use the biology of the disease to decide what specific therapy is ideal for that patient. Better understanding of biology and genetics is facilitating rational development of new treatments,” he said, adding that whenever possible, patients should be treated within clinical trials.

Dr. Gribben has received research funding from the NIH, Cancer Research UK, MRC, and Wellcome Trust. He has received honoraria from Roche/Genentech, Celgene, Janssen, Pharmacyclics, Gilead, Mundipharma, Infinity, TG Therapeutics, and Ascerta, and he has a patent or receives royalties from Celgene. He also has been the principal investigator on a clinical trial for Roche, Takeda, Pharmacyclics, Gilead, and Infinity.

sworcester@frontlinemedcom.com

CHICAGO – Despite exciting new advances in the understanding of chronic lymphocytic leukemia, particularly with respect to prognostic features that predict risk for relapse, a watch-and-wait approach remains appropriate for asymptomatic disease pending outcomes data for newer approaches, according to Dr. John G. Gribben.

“When the disease is diagnosed, if it is asymptomatic, the correct approach – of course – is to continue to watch and wait,” Dr. Gribben of Barts Cancer Institute, Queen Mary University of London, said at the American Society of Hematology Meeting on Hematologic Malignancies.

Numerous clinical trials have demonstrated no advantage of early treatment vs. watch and wait, he said, adding that all of the trials published to date have used treatment of all-comers, and have used chlorambucil (CLB) as the treatment.

“There has been a whole variety of more modern trials that have used select prognostic features to identify subgroups of people who are at higher risk of relapse, who then go on to receive earlier treatment with either FCR [fludarabine, cyclophosphamide, rituximab], or more recently, ibrutinib,” he said.

These treatments are interesting, and the trials have demonstrated that prognostic features can identify patients who will progress more rapidly, but none have reported, he explained.

“In the absence of any published trial, I continue to ‘watch and wait’ patients, and there are no high-risk features which will make me alter that approach. Even the highest-risk features of complex karyotype and p53 abnormalities are not indications to treat patients until they become symptomatic,” he said.

It is striking how white counts vary widely in both asymptomatic and symptomatic patients, he noted.

“I don’t personally have any particular white count which is the number at which I’ll treat a patient. I don’t treat white counts, I treat patients,” he said.

When patients become symptomatic, the treatment of choice is now immunochemotherapies, irrespective of performance status, he said.

“Within the past year we have seen approval of obinutuzumab and ofatumumab for treatment of previously untreated CLL, as well as ibrutinib and idelalisib plus rituximab for treatment of both previously untreated CLL and those with 17p deletions for relapsed/refractory disease, as well as for up-front treatment,” he said, adding that these new agents greatly increase the available options for treating CLL.

Dr. Gribben said he considers these questions when it comes to treating CLL:

• Does the patient require treatment, or is watching and waiting appropriate?

• What is the goal of therapy? This is determined through conversation with the patient and the patient’s family regarding the side-effect profile they are willing to tolerate vs. the potential longer duration of response.

• What comorbidities are present to determine “fitness” for specific immunochemotherapy? Specifically, is the patient fit for an FCR-type approach, or is an alternative more appropriate?

• Is there a 17p deletion or P53 mutation that would make chemotherapy a less attractive option, and use of novel agents a more attractive option?

His approach, based on the answers to these questions, is as follows:

• In Rai stage 0-II patients with inactive disease, fitness and 17p deletion or p53 mutation status is irrelevant; no therapy is given.

• For active disease or Rai III-IV disease, a “go-go” patient, (or patient in good physical condition) is treated based on the presence or absence of 17p deletion or p53 mutation status. Those without 17p deletions or p53 mutations can be treated with FCR (his preference), or fludarabine-rituximab (FR). Bendamustine-rituximab (BR) is also an attractive option in certain cases, he said.

• For patients with active disease or Rai stage III-IV disease who do have a 17p deletion, his treatment of choice is either ibrutinib or idelalisib plus rituximab, depending on the patient.

• In “slow-go” patients (those with poorer physical condition) treatment is again based on mutational analysis. Those without mutation receive either FR, BR, or CLB plus obinutuzumab. These are very good options, and represent a spectrum to choose from based on the patient’s core abilities and ability to withstand particular types of treatments, he said.

“If they do have a 17p deletion, these patients are just as eligible for ibrutinib or idelalisib plus rituximab as the younger patients,” he noted.

His choices are based largely on the findings from the CLL8 trial (Lancet 2010 Oct;376[9747]:1164-74) which demonstrated an overall survival advantage with chemoimmunotherapy for front-line therapy vs. chemotherapy alone (hazard ratio, 0.68).

Over time, the advantage has become even more pronounced, according to follow-up data.

Starting with something “gentle” and saving the best treatment for later in the event of relapse was recently considered a reasonable approach, but in the wake of the CLL8 findings, this is no longer an acceptable plan, Dr. Gribben said.

“That’s why for my choice, FCR remains the treatment of choice for those patients who are fit enough to tolerate this type of approach,” he said.

In those with 17p deletions or P53 mutations, the CLL8 trial showed poor outcomes with FCR.

“This is a group of patients whom I believe chemoimmunotherapy would no longer be the treatment of choice,” he said, adding that newer findings suggest outcomes in these patients are better with novel agents.

He also noted that patients with 11q abnormalities, which were previously associated with a poor prognosis, were found in CLL8 to respond well to chemoimmunotherapy when used front line.

While there are special considerations in the elderly, and different strategies in relapsed and refractory disease, the future of CLL treatment is promising. The benefit of adding rituximab to combination chemotherapy is well established, the benefit of novel agents is also now established, and the future likely involves combining targeted therapies with each other and with immunochemotherapies, and combining targeted therapies across different pathways.

“And of course the hope is that we’re going to use the biology of the disease to decide what specific therapy is ideal for that patient. Better understanding of biology and genetics is facilitating rational development of new treatments,” he said, adding that whenever possible, patients should be treated within clinical trials.

Dr. Gribben has received research funding from the NIH, Cancer Research UK, MRC, and Wellcome Trust. He has received honoraria from Roche/Genentech, Celgene, Janssen, Pharmacyclics, Gilead, Mundipharma, Infinity, TG Therapeutics, and Ascerta, and he has a patent or receives royalties from Celgene. He also has been the principal investigator on a clinical trial for Roche, Takeda, Pharmacyclics, Gilead, and Infinity.

sworcester@frontlinemedcom.com

CHICAGO – Despite exciting new advances in the understanding of chronic lymphocytic leukemia, particularly with respect to prognostic features that predict risk for relapse, a watch-and-wait approach remains appropriate for asymptomatic disease pending outcomes data for newer approaches, according to Dr. John G. Gribben.

“When the disease is diagnosed, if it is asymptomatic, the correct approach – of course – is to continue to watch and wait,” Dr. Gribben of Barts Cancer Institute, Queen Mary University of London, said at the American Society of Hematology Meeting on Hematologic Malignancies.

Numerous clinical trials have demonstrated no advantage of early treatment vs. watch and wait, he said, adding that all of the trials published to date have used treatment of all-comers, and have used chlorambucil (CLB) as the treatment.

“There has been a whole variety of more modern trials that have used select prognostic features to identify subgroups of people who are at higher risk of relapse, who then go on to receive earlier treatment with either FCR [fludarabine, cyclophosphamide, rituximab], or more recently, ibrutinib,” he said.

These treatments are interesting, and the trials have demonstrated that prognostic features can identify patients who will progress more rapidly, but none have reported, he explained.

“In the absence of any published trial, I continue to ‘watch and wait’ patients, and there are no high-risk features which will make me alter that approach. Even the highest-risk features of complex karyotype and p53 abnormalities are not indications to treat patients until they become symptomatic,” he said.

It is striking how white counts vary widely in both asymptomatic and symptomatic patients, he noted.

“I don’t personally have any particular white count which is the number at which I’ll treat a patient. I don’t treat white counts, I treat patients,” he said.

When patients become symptomatic, the treatment of choice is now immunochemotherapies, irrespective of performance status, he said.

“Within the past year we have seen approval of obinutuzumab and ofatumumab for treatment of previously untreated CLL, as well as ibrutinib and idelalisib plus rituximab for treatment of both previously untreated CLL and those with 17p deletions for relapsed/refractory disease, as well as for up-front treatment,” he said, adding that these new agents greatly increase the available options for treating CLL.

Dr. Gribben said he considers these questions when it comes to treating CLL:

• Does the patient require treatment, or is watching and waiting appropriate?

• What is the goal of therapy? This is determined through conversation with the patient and the patient’s family regarding the side-effect profile they are willing to tolerate vs. the potential longer duration of response.

• What comorbidities are present to determine “fitness” for specific immunochemotherapy? Specifically, is the patient fit for an FCR-type approach, or is an alternative more appropriate?

• Is there a 17p deletion or P53 mutation that would make chemotherapy a less attractive option, and use of novel agents a more attractive option?

His approach, based on the answers to these questions, is as follows:

• In Rai stage 0-II patients with inactive disease, fitness and 17p deletion or p53 mutation status is irrelevant; no therapy is given.

• For active disease or Rai III-IV disease, a “go-go” patient, (or patient in good physical condition) is treated based on the presence or absence of 17p deletion or p53 mutation status. Those without 17p deletions or p53 mutations can be treated with FCR (his preference), or fludarabine-rituximab (FR). Bendamustine-rituximab (BR) is also an attractive option in certain cases, he said.

• For patients with active disease or Rai stage III-IV disease who do have a 17p deletion, his treatment of choice is either ibrutinib or idelalisib plus rituximab, depending on the patient.

• In “slow-go” patients (those with poorer physical condition) treatment is again based on mutational analysis. Those without mutation receive either FR, BR, or CLB plus obinutuzumab. These are very good options, and represent a spectrum to choose from based on the patient’s core abilities and ability to withstand particular types of treatments, he said.

“If they do have a 17p deletion, these patients are just as eligible for ibrutinib or idelalisib plus rituximab as the younger patients,” he noted.

His choices are based largely on the findings from the CLL8 trial (Lancet 2010 Oct;376[9747]:1164-74) which demonstrated an overall survival advantage with chemoimmunotherapy for front-line therapy vs. chemotherapy alone (hazard ratio, 0.68).

Over time, the advantage has become even more pronounced, according to follow-up data.

Starting with something “gentle” and saving the best treatment for later in the event of relapse was recently considered a reasonable approach, but in the wake of the CLL8 findings, this is no longer an acceptable plan, Dr. Gribben said.

“That’s why for my choice, FCR remains the treatment of choice for those patients who are fit enough to tolerate this type of approach,” he said.

In those with 17p deletions or P53 mutations, the CLL8 trial showed poor outcomes with FCR.

“This is a group of patients whom I believe chemoimmunotherapy would no longer be the treatment of choice,” he said, adding that newer findings suggest outcomes in these patients are better with novel agents.

He also noted that patients with 11q abnormalities, which were previously associated with a poor prognosis, were found in CLL8 to respond well to chemoimmunotherapy when used front line.

While there are special considerations in the elderly, and different strategies in relapsed and refractory disease, the future of CLL treatment is promising. The benefit of adding rituximab to combination chemotherapy is well established, the benefit of novel agents is also now established, and the future likely involves combining targeted therapies with each other and with immunochemotherapies, and combining targeted therapies across different pathways.

“And of course the hope is that we’re going to use the biology of the disease to decide what specific therapy is ideal for that patient. Better understanding of biology and genetics is facilitating rational development of new treatments,” he said, adding that whenever possible, patients should be treated within clinical trials.

Dr. Gribben has received research funding from the NIH, Cancer Research UK, MRC, and Wellcome Trust. He has received honoraria from Roche/Genentech, Celgene, Janssen, Pharmacyclics, Gilead, Mundipharma, Infinity, TG Therapeutics, and Ascerta, and he has a patent or receives royalties from Celgene. He also has been the principal investigator on a clinical trial for Roche, Takeda, Pharmacyclics, Gilead, and Infinity.

sworcester@frontlinemedcom.com

Introduction

The use of antipsychotic medications has become more common in children and adolescents over the past 2 decades (Arch Gen Psychiatry. 2012 Dec;69[12]:1247-56). Whether or not one agrees that the trend in prescribing these agents is problematic (and I do), often the prescription and monitoring of antipsychotic medication falls to the primary care clinician who may have concerns about how to manage them. Here, we present a case to illustrate just such an issue.

Case summary

James is a 17-year-old young man with a diagnosis of bipolar disorder made earlier this year after an inpatient day at a local hospital. James had a history of attention-deficit/hyperactivity disorder (ADHD) growing up, but also had a strong family history of bipolar disorder. He began having increasing difficulty during the summer after his junior year, and then had a clear manic episode with elation and grandiosity that necessitated the hospitalization. During the relatively short stay on the psychiatric unit, he was placed on lithium carbonate and risperidone to treat the mania, which did respond. Although it was recommended that he follow up with a child and adolescent psychiatrist, through a series of happenstances he missed two appointments for an intake. He is now only able to get a new intake that is 4 months off.

You have continued to prescribe for him, waiting for what always seemed to be an imminent appointment. But now he comes to the office for a follow-up visit that can no longer wait. His mood symptoms are actually doing quite well. However, you have always known James to have a relatively thin build in the 25th percentile for weight and the 75th percentile for height. Now (is it possible?) 6 months after starting the risperidone and lithium, he returns to the office in the 50th percentile for weight and remains in the 75th percentile for height. There is nothing else noted to be concerning on physical or mental status examination, but you wonder what should be done for monitoring of his weight, should you be doing it, and whether there are other metabolic parameters that you should be measuring?

Case discussion

Regardless of your stance on pediatric bipolar disorder and the treatments for it, James has been placed on medications that need to be monitored. While it appears that James will, eventually, have a prescriber who can monitor his medication for side effects, it is incumbent on all of his providers to make sure that monitoring is occurring. Recent studies have demonstrated that guidelines for monitoring of antipsychotic use are not consistently being adhered to. For example, Rettew et al. (Pediatrics. 2015 Apr;135[4]:658-65) recently reported that metabolic monitoring that included laboratory tests was reported in only 57.2% of cases where an antipsychotic was prescribed.

Children and adolescents placed on these agents should be monitored in a number of ways. First, height and weight should be taken at baseline and at follow-up visits – at least every 6 months. The American Academy of Child and Adolescent Psychiatry practice parameters state that “consideration of weight management interventions and increased regularity of blood glucose and lipid levels should be implemented if [atypical antipsychotic agent]–induced weight gain exceeds 90th percentile body mass index (BMI) for age, or a change of five BMI units in those youths who were obese at the beginning of treatment.” Fasting blood sugar, fasting triglyceride, and cholesterol panels should be done at baseline and at approximately 6-month intervals. Screening for dystonic movements with an Abnormal Involuntary Movement Scale (AIMS) should be done at 6-month intervals. In the case of risperidone, if there are any questions about gynecomastia, galactorrhea, and/or sexual dysfunction, a prolactin level should be considered. Certain other antipsychotics require specific monitoring (for example, ECG for prolonged QT interval for ziprasidone, CBC for clozapine, and an eye exam for quetiapine).

The most important kinds of monitoring are for dose and efficacy. While the antipsychotic medications may have a role, they are not a panacea and do carry longer-term risks of metabolic problems and obesity, among others. If it is possible to provide interventions to reduce the dose or duration of use, that is preferable. If not, it’s best to work with the psychiatrist (if available) to determine who will perform the monitoring and how often it will be done to stave off metabolic problems as early as possible.

Dr. Althoff is associate professor of psychiatry, psychology, and pediatrics at the University of Vermont, Burlington. He is director of the division of behavioral genetics and conducts research on the development of self-regulation in children. Dr. Althoff receives no funding from pharmaceutical companies or industry. He has grant funding from the National Institute of General Medical Sciences and the Klingenstein Third Generation Foundation and is employed, in part, by the nonprofit Research Center for Children, Youth, and Families that develops the Child Behavior Checklist and associated instruments. E-mail him at pdnews@frontlinemedcom.com.

Introduction

The use of antipsychotic medications has become more common in children and adolescents over the past 2 decades (Arch Gen Psychiatry. 2012 Dec;69[12]:1247-56). Whether or not one agrees that the trend in prescribing these agents is problematic (and I do), often the prescription and monitoring of antipsychotic medication falls to the primary care clinician who may have concerns about how to manage them. Here, we present a case to illustrate just such an issue.

Case summary

James is a 17-year-old young man with a diagnosis of bipolar disorder made earlier this year after an inpatient day at a local hospital. James had a history of attention-deficit/hyperactivity disorder (ADHD) growing up, but also had a strong family history of bipolar disorder. He began having increasing difficulty during the summer after his junior year, and then had a clear manic episode with elation and grandiosity that necessitated the hospitalization. During the relatively short stay on the psychiatric unit, he was placed on lithium carbonate and risperidone to treat the mania, which did respond. Although it was recommended that he follow up with a child and adolescent psychiatrist, through a series of happenstances he missed two appointments for an intake. He is now only able to get a new intake that is 4 months off.

You have continued to prescribe for him, waiting for what always seemed to be an imminent appointment. But now he comes to the office for a follow-up visit that can no longer wait. His mood symptoms are actually doing quite well. However, you have always known James to have a relatively thin build in the 25th percentile for weight and the 75th percentile for height. Now (is it possible?) 6 months after starting the risperidone and lithium, he returns to the office in the 50th percentile for weight and remains in the 75th percentile for height. There is nothing else noted to be concerning on physical or mental status examination, but you wonder what should be done for monitoring of his weight, should you be doing it, and whether there are other metabolic parameters that you should be measuring?

Case discussion

Regardless of your stance on pediatric bipolar disorder and the treatments for it, James has been placed on medications that need to be monitored. While it appears that James will, eventually, have a prescriber who can monitor his medication for side effects, it is incumbent on all of his providers to make sure that monitoring is occurring. Recent studies have demonstrated that guidelines for monitoring of antipsychotic use are not consistently being adhered to. For example, Rettew et al. (Pediatrics. 2015 Apr;135[4]:658-65) recently reported that metabolic monitoring that included laboratory tests was reported in only 57.2% of cases where an antipsychotic was prescribed.

Children and adolescents placed on these agents should be monitored in a number of ways. First, height and weight should be taken at baseline and at follow-up visits – at least every 6 months. The American Academy of Child and Adolescent Psychiatry practice parameters state that “consideration of weight management interventions and increased regularity of blood glucose and lipid levels should be implemented if [atypical antipsychotic agent]–induced weight gain exceeds 90th percentile body mass index (BMI) for age, or a change of five BMI units in those youths who were obese at the beginning of treatment.” Fasting blood sugar, fasting triglyceride, and cholesterol panels should be done at baseline and at approximately 6-month intervals. Screening for dystonic movements with an Abnormal Involuntary Movement Scale (AIMS) should be done at 6-month intervals. In the case of risperidone, if there are any questions about gynecomastia, galactorrhea, and/or sexual dysfunction, a prolactin level should be considered. Certain other antipsychotics require specific monitoring (for example, ECG for prolonged QT interval for ziprasidone, CBC for clozapine, and an eye exam for quetiapine).

The most important kinds of monitoring are for dose and efficacy. While the antipsychotic medications may have a role, they are not a panacea and do carry longer-term risks of metabolic problems and obesity, among others. If it is possible to provide interventions to reduce the dose or duration of use, that is preferable. If not, it’s best to work with the psychiatrist (if available) to determine who will perform the monitoring and how often it will be done to stave off metabolic problems as early as possible.

Dr. Althoff is associate professor of psychiatry, psychology, and pediatrics at the University of Vermont, Burlington. He is director of the division of behavioral genetics and conducts research on the development of self-regulation in children. Dr. Althoff receives no funding from pharmaceutical companies or industry. He has grant funding from the National Institute of General Medical Sciences and the Klingenstein Third Generation Foundation and is employed, in part, by the nonprofit Research Center for Children, Youth, and Families that develops the Child Behavior Checklist and associated instruments. E-mail him at pdnews@frontlinemedcom.com.

Introduction

The use of antipsychotic medications has become more common in children and adolescents over the past 2 decades (Arch Gen Psychiatry. 2012 Dec;69[12]:1247-56). Whether or not one agrees that the trend in prescribing these agents is problematic (and I do), often the prescription and monitoring of antipsychotic medication falls to the primary care clinician who may have concerns about how to manage them. Here, we present a case to illustrate just such an issue.

Case summary

James is a 17-year-old young man with a diagnosis of bipolar disorder made earlier this year after an inpatient day at a local hospital. James had a history of attention-deficit/hyperactivity disorder (ADHD) growing up, but also had a strong family history of bipolar disorder. He began having increasing difficulty during the summer after his junior year, and then had a clear manic episode with elation and grandiosity that necessitated the hospitalization. During the relatively short stay on the psychiatric unit, he was placed on lithium carbonate and risperidone to treat the mania, which did respond. Although it was recommended that he follow up with a child and adolescent psychiatrist, through a series of happenstances he missed two appointments for an intake. He is now only able to get a new intake that is 4 months off.

You have continued to prescribe for him, waiting for what always seemed to be an imminent appointment. But now he comes to the office for a follow-up visit that can no longer wait. His mood symptoms are actually doing quite well. However, you have always known James to have a relatively thin build in the 25th percentile for weight and the 75th percentile for height. Now (is it possible?) 6 months after starting the risperidone and lithium, he returns to the office in the 50th percentile for weight and remains in the 75th percentile for height. There is nothing else noted to be concerning on physical or mental status examination, but you wonder what should be done for monitoring of his weight, should you be doing it, and whether there are other metabolic parameters that you should be measuring?

Case discussion

Regardless of your stance on pediatric bipolar disorder and the treatments for it, James has been placed on medications that need to be monitored. While it appears that James will, eventually, have a prescriber who can monitor his medication for side effects, it is incumbent on all of his providers to make sure that monitoring is occurring. Recent studies have demonstrated that guidelines for monitoring of antipsychotic use are not consistently being adhered to. For example, Rettew et al. (Pediatrics. 2015 Apr;135[4]:658-65) recently reported that metabolic monitoring that included laboratory tests was reported in only 57.2% of cases where an antipsychotic was prescribed.

Children and adolescents placed on these agents should be monitored in a number of ways. First, height and weight should be taken at baseline and at follow-up visits – at least every 6 months. The American Academy of Child and Adolescent Psychiatry practice parameters state that “consideration of weight management interventions and increased regularity of blood glucose and lipid levels should be implemented if [atypical antipsychotic agent]–induced weight gain exceeds 90th percentile body mass index (BMI) for age, or a change of five BMI units in those youths who were obese at the beginning of treatment.” Fasting blood sugar, fasting triglyceride, and cholesterol panels should be done at baseline and at approximately 6-month intervals. Screening for dystonic movements with an Abnormal Involuntary Movement Scale (AIMS) should be done at 6-month intervals. In the case of risperidone, if there are any questions about gynecomastia, galactorrhea, and/or sexual dysfunction, a prolactin level should be considered. Certain other antipsychotics require specific monitoring (for example, ECG for prolonged QT interval for ziprasidone, CBC for clozapine, and an eye exam for quetiapine).

The most important kinds of monitoring are for dose and efficacy. While the antipsychotic medications may have a role, they are not a panacea and do carry longer-term risks of metabolic problems and obesity, among others. If it is possible to provide interventions to reduce the dose or duration of use, that is preferable. If not, it’s best to work with the psychiatrist (if available) to determine who will perform the monitoring and how often it will be done to stave off metabolic problems as early as possible.

Dr. Althoff is associate professor of psychiatry, psychology, and pediatrics at the University of Vermont, Burlington. He is director of the division of behavioral genetics and conducts research on the development of self-regulation in children. Dr. Althoff receives no funding from pharmaceutical companies or industry. He has grant funding from the National Institute of General Medical Sciences and the Klingenstein Third Generation Foundation and is employed, in part, by the nonprofit Research Center for Children, Youth, and Families that develops the Child Behavior Checklist and associated instruments. E-mail him at pdnews@frontlinemedcom.com.

Veterans Health Administration (VA) hospitals have received notoriety due to episodes of misdiagnosis, poor management, and negligent care described in many recent reports and news articles.^1-3 While veterans are appropriately the primary focus of these investigative reports, physicians are also challenged in this setting, as they often meet resistance when advocating for patients and attempting to improve a flawed system.² Although my residency training includes 6 months at a VA hospital mired in controversy, the hospital has played a critical role in my training.³

Despite my many frustrations with the VA and the daily stresses incurred because of barriers impeding the timing and quality of care, I have several reasons to see the glass as “half full” when reflecting on my experiences as an orthopedic surgery resident at a VA medical center. This editorial will focus on the most important of these reasons—the special opportunity and pride associated with caring for veterans and these patients’ extremely appreciative nature.

The VA is one of the largest integrated health care systems in the United States, offering both inpatient and outpatient care to eligible veterans. Although eligibility has historically been based on military service–related medical conditions, disability, and financial need, reforms from 1996 to 2002 expanded enrollment to veteran populations previously deemed ineligible for VA care.^4,5 Despite this, studies suggest that some uninsured veterans do not seek VA care, even when eligible for VA coverage. This troubling notion is further complicated by research suggesting that veterans who use the VA for all of their health care are more likely to be from poor, less-educated, and minority populations, and are more likely to report fair or poor health and seek more disability days.⁶

Such disheartening realities can mask the most important attributes of VA patients, which pertain to their selfless commitment to our country. Orthopedic surgery residents must appreciate these attributes as well as the tremendous need for musculoskeletal care in this setting, as musculoskeletal conditions are some of the most common reasons for patient visits at the VA.⁷ Although combat-related high-energy blast injuries and the reconstructive procedures used to treat them have received a lot of attention, it is the more common musculoskeletal disorders that are most responsible for the tremendous burden of musculoskeletal disease in the VA. In a study by Dominick and colleagues,⁸ veterans had significantly greater odds of reporting doctor-diagnosed arthritis compared with nonveterans. Furthermore, veterans are also more vulnerable to overuse injuries, a finding attributed to the intense physical activity associated with military training and service.⁹

The busy orthopedic surgery clinic at my VA hospital is a fulfilling experience and a reminder of the large demand for musculoskeletal care. However, it is the patient population that makes it most gratifying. Most of the veterans seeking care are appreciative, regularly expressing their gratitude. They view me and the other residents as their physicians, not simply as doctors in training, like so many other non-VA patients do. Despite the fact that VA patients sometimes have to wait several hours to be seen in clinic and several months for surgery, I have never been subjected to their inevitable disdain or frustration. This is true in even the most trying and infuriating times, such as when an operation is cancelled on the day of surgery for reasons that many surgeons in non-VA hospitals would consider trivial. And even when witness to my visible irritation with the VA system, the veterans remain respectful and understanding; if they ever share similar feelings, they most certainly never voice them to me.

I cannot refute the notion that the VA must change and that the veterans deserve an improved health care system. However, this editorial is not written as a call to action. Instead, I hope it helps to humanize the patients of the VA, serving as a reminder to residents and other providers that the VA is a unique and extraordinary opportunity to give back and say thank you to veterans.

This editorial is dedicated to CPT David Huskie, USAR (Ret.), a veteran of Operation Desert Storm and orthopedic nurse at my VA hospital. It was he who first reminded me, and the other orthopedic residents, of the importance of our time at the VA. The Figure depicts the letter he gives to orthopedic residents at our program, along with a pewter coin, after their first VA rotation.

Veterans Health Administration (VA) hospitals have received notoriety due to episodes of misdiagnosis, poor management, and negligent care described in many recent reports and news articles.^1-3 While veterans are appropriately the primary focus of these investigative reports, physicians are also challenged in this setting, as they often meet resistance when advocating for patients and attempting to improve a flawed system.² Although my residency training includes 6 months at a VA hospital mired in controversy, the hospital has played a critical role in my training.³

Despite my many frustrations with the VA and the daily stresses incurred because of barriers impeding the timing and quality of care, I have several reasons to see the glass as “half full” when reflecting on my experiences as an orthopedic surgery resident at a VA medical center. This editorial will focus on the most important of these reasons—the special opportunity and pride associated with caring for veterans and these patients’ extremely appreciative nature.

The VA is one of the largest integrated health care systems in the United States, offering both inpatient and outpatient care to eligible veterans. Although eligibility has historically been based on military service–related medical conditions, disability, and financial need, reforms from 1996 to 2002 expanded enrollment to veteran populations previously deemed ineligible for VA care.^4,5 Despite this, studies suggest that some uninsured veterans do not seek VA care, even when eligible for VA coverage. This troubling notion is further complicated by research suggesting that veterans who use the VA for all of their health care are more likely to be from poor, less-educated, and minority populations, and are more likely to report fair or poor health and seek more disability days.⁶

Such disheartening realities can mask the most important attributes of VA patients, which pertain to their selfless commitment to our country. Orthopedic surgery residents must appreciate these attributes as well as the tremendous need for musculoskeletal care in this setting, as musculoskeletal conditions are some of the most common reasons for patient visits at the VA.⁷ Although combat-related high-energy blast injuries and the reconstructive procedures used to treat them have received a lot of attention, it is the more common musculoskeletal disorders that are most responsible for the tremendous burden of musculoskeletal disease in the VA. In a study by Dominick and colleagues,⁸ veterans had significantly greater odds of reporting doctor-diagnosed arthritis compared with nonveterans. Furthermore, veterans are also more vulnerable to overuse injuries, a finding attributed to the intense physical activity associated with military training and service.⁹

The busy orthopedic surgery clinic at my VA hospital is a fulfilling experience and a reminder of the large demand for musculoskeletal care. However, it is the patient population that makes it most gratifying. Most of the veterans seeking care are appreciative, regularly expressing their gratitude. They view me and the other residents as their physicians, not simply as doctors in training, like so many other non-VA patients do. Despite the fact that VA patients sometimes have to wait several hours to be seen in clinic and several months for surgery, I have never been subjected to their inevitable disdain or frustration. This is true in even the most trying and infuriating times, such as when an operation is cancelled on the day of surgery for reasons that many surgeons in non-VA hospitals would consider trivial. And even when witness to my visible irritation with the VA system, the veterans remain respectful and understanding; if they ever share similar feelings, they most certainly never voice them to me.

I cannot refute the notion that the VA must change and that the veterans deserve an improved health care system. However, this editorial is not written as a call to action. Instead, I hope it helps to humanize the patients of the VA, serving as a reminder to residents and other providers that the VA is a unique and extraordinary opportunity to give back and say thank you to veterans.

This editorial is dedicated to CPT David Huskie, USAR (Ret.), a veteran of Operation Desert Storm and orthopedic nurse at my VA hospital. It was he who first reminded me, and the other orthopedic residents, of the importance of our time at the VA. The Figure depicts the letter he gives to orthopedic residents at our program, along with a pewter coin, after their first VA rotation.

Veterans Health Administration (VA) hospitals have received notoriety due to episodes of misdiagnosis, poor management, and negligent care described in many recent reports and news articles.^1-3 While veterans are appropriately the primary focus of these investigative reports, physicians are also challenged in this setting, as they often meet resistance when advocating for patients and attempting to improve a flawed system.² Although my residency training includes 6 months at a VA hospital mired in controversy, the hospital has played a critical role in my training.³

Despite my many frustrations with the VA and the daily stresses incurred because of barriers impeding the timing and quality of care, I have several reasons to see the glass as “half full” when reflecting on my experiences as an orthopedic surgery resident at a VA medical center. This editorial will focus on the most important of these reasons—the special opportunity and pride associated with caring for veterans and these patients’ extremely appreciative nature.

The VA is one of the largest integrated health care systems in the United States, offering both inpatient and outpatient care to eligible veterans. Although eligibility has historically been based on military service–related medical conditions, disability, and financial need, reforms from 1996 to 2002 expanded enrollment to veteran populations previously deemed ineligible for VA care.^4,5 Despite this, studies suggest that some uninsured veterans do not seek VA care, even when eligible for VA coverage. This troubling notion is further complicated by research suggesting that veterans who use the VA for all of their health care are more likely to be from poor, less-educated, and minority populations, and are more likely to report fair or poor health and seek more disability days.⁶

Such disheartening realities can mask the most important attributes of VA patients, which pertain to their selfless commitment to our country. Orthopedic surgery residents must appreciate these attributes as well as the tremendous need for musculoskeletal care in this setting, as musculoskeletal conditions are some of the most common reasons for patient visits at the VA.⁷ Although combat-related high-energy blast injuries and the reconstructive procedures used to treat them have received a lot of attention, it is the more common musculoskeletal disorders that are most responsible for the tremendous burden of musculoskeletal disease in the VA. In a study by Dominick and colleagues,⁸ veterans had significantly greater odds of reporting doctor-diagnosed arthritis compared with nonveterans. Furthermore, veterans are also more vulnerable to overuse injuries, a finding attributed to the intense physical activity associated with military training and service.⁹

The busy orthopedic surgery clinic at my VA hospital is a fulfilling experience and a reminder of the large demand for musculoskeletal care. However, it is the patient population that makes it most gratifying. Most of the veterans seeking care are appreciative, regularly expressing their gratitude. They view me and the other residents as their physicians, not simply as doctors in training, like so many other non-VA patients do. Despite the fact that VA patients sometimes have to wait several hours to be seen in clinic and several months for surgery, I have never been subjected to their inevitable disdain or frustration. This is true in even the most trying and infuriating times, such as when an operation is cancelled on the day of surgery for reasons that many surgeons in non-VA hospitals would consider trivial. And even when witness to my visible irritation with the VA system, the veterans remain respectful and understanding; if they ever share similar feelings, they most certainly never voice them to me.

I cannot refute the notion that the VA must change and that the veterans deserve an improved health care system. However, this editorial is not written as a call to action. Instead, I hope it helps to humanize the patients of the VA, serving as a reminder to residents and other providers that the VA is a unique and extraordinary opportunity to give back and say thank you to veterans.

This editorial is dedicated to CPT David Huskie, USAR (Ret.), a veteran of Operation Desert Storm and orthopedic nurse at my VA hospital. It was he who first reminded me, and the other orthopedic residents, of the importance of our time at the VA. The Figure depicts the letter he gives to orthopedic residents at our program, along with a pewter coin, after their first VA rotation.

Fractures of the clavicle, which account for 2.6% of all fractures, are displaced in 70% of cases and are mid-diaphyseal in 80% of cases.^1-3 Historically, both displaced and nondisplaced fractures were treated nonoperatively with excellent outcomes reported in the majority of patients.^1-3 Traditionally, the indications for surgical fixation of a clavicular fracture include open fractures, which occur infrequently, accounting for only 3.2% of clavicle fractures.⁴ Other indications include floating shoulder girdle or scapulothoracic dissociation, neurovascular injury, and skin “tenting” by the fracture fragments.^3,5 Recently, both meta-analyses and randomized clinical trials have reported reduced malunion rates and improved patient outcomes with open reduction and internal fixation (ORIF).^6-9 Consequently, operative fixation could be considered in patients with 100% displacement or greater than 1.5 cm shortening.^6-9 Open reduction and internal fixation of the clavicle has been demonstrated to have excellent outcomes in pediatric populations as well.¹⁰

The clavicle is subcutaneous for much of its length and, thus, displaced clavicular fractures often result in a visible deformity with a stretch of the soft-tissue envelope over the fracture. While this has been suggested as an operative indication, several recent sources indicate that this concern may only be theoretical. According to the fourth edition of Skeletal Trauma, “It is often stated that open reduction and internal fixation should be considered if the skin is threatened by pressure from a prominent clavicle fracture fragment; however, it is extremely rare of the skin to be perforated from within.”⁵ The most recent Journal of Bone and Joint Surgery Current Concepts Review on the subject stated that “open fractures or soft-tissue tenting sufficient to produce skin necrosis is uncommon.”³ To the best of our knowledge, there is no reported case of a displaced midshaft clavicle fracture with secondary skin necrosis and conversion into an open fracture, validating the conclusion that this complication may be only theoretical. Given that surgical fixation carries a risk of complications including wound complications, infection, nonunion, malunion, and damage to the nearby neurovascular structures and pleural apices,¹¹ some surgeons may be uncertain how to proceed in cases at risk for disturbance of the soft tissues.

We report 2 adolescent cases of displaced, comminuted clavicle fractures in which the skin was initially intact. Both were managed nonoperatively and both secondarily presented with open lesions at the fracture site requiring urgent irrigation and débridement (I&D) and ORIF. The patients and their guardians provided written informed consent for print and electronic publication of these case reports.

Case Reports

Case 1

A 15-year-old boy with no significant medical or surgical history flipped over the handlebars of his bicycle the day prior to presentation and sustained a clavicle fracture on his left nondominant upper extremity. This was an isolated injury. On examination, his skin was intact with an area of tender mild osseous protuberance at the midclavicle with associated surrounding edema. He was neurovascularly intact. Radiographs showed a displaced fracture of the midshaft of the clavicle with 20% shortening with a vertically angulated piece of comminution (Figure 1A). After a discussion of the treatment options with the family, the decision was made to pursue nonoperative treatment with sling immobilization as needed and restriction from gym and sports.

Two and a half weeks later, the patient presented at follow-up with significant reduction but persistence of his pain and a new complaint of drainage from the area of the fracture. On examination, he was found to have a puncture wound of the skin with exposed clavicle protruding through the wound with a 1-cm circumferential area of erythema without purulence present or expressible. The patient denied reinjury and endorsed compliance with sling immobilization. He was taken for urgent I&D and ORIF. After excision of the eschar surrounding the open lesion and full I&D of the soft tissues, the protruding spike was partially excised and the fracture site was débrided. The fracture was reduced and fixated with a lag screw and neutralization plate technique using an anatomically contoured locking clavicle plate (Synthes). Vancomycin powder was sprinkled into the wound at the completion of the procedure to reduce the chance of infection.¹²

Postoperatively, the patient was prescribed oral clindamycin but was subsequently switched to oral cephalexin because of mild signs of an allergic reaction, for a total course of antibiotics of 1 week. The patient was immobilized in a sling for comfort for the first 9 weeks postoperatively until radiographic union occurred. The patient’s wound healed uneventfully and with acceptable cosmesis. He was released to full activities at 10 weeks postoperatively. At final follow-up 6 months after surgery, the patient had returned to all of his regular activities without pain, and with full range of motion and no demonstrable deficits with radiographic union (Figure 1B).

Case 2

An 11-year-old boy with no significant medical or surgical history fell onto his right dominant upper extremity while doing a jump on his dirt bike 1 week prior to presentation, sustaining a clavicle fracture. This was an isolated injury. He was seen and evaluated by an outside orthopedist who noted that the soft-tissue envelope was intact and the patient was neurovascularly intact. Radiographs showed a displaced fracture of the midshaft of the clavicle with 15% shortening and with a vertically angulated piece of comminution (Figure 1C). Nonoperative treatment with a figure-of-8 brace was recommended. The patient’s discomfort completely resolved.

One week later, when he presented to the outside orthopedist for follow-up, the development of a wound overlying the fracture site was noted, and the patient was started on oral trimethoprim/sulfamethoxazole and referred to our office for treatment (Figure 1D). The patient denied reinjury and endorsed compliance with brace immobilization. On examination, the patient was afebrile and was noted to have a puncture wound at the fracture site with a protruding spike of bone and surrounding erythema but without present or expressible discharge (Figure 2). The patient was taken urgently for I&D and ORIF, using a similar technique to case 1, except that no lag screw was employed.

Postoperatively, the patient did well with no complications; he was prescribed oral cephalexin for 1 week. The patient was immobilized in a sling for the first 5 weeks after surgery until radiographic union had occurred, after which the sling was discontinued. The patient’s wound healed uneventfully and with acceptable cosmesis. The patient was released from activity restrictions at 6 weeks postoperatively. At final follow-up 5 weeks after surgery, the patient had full painless range of motion, no tenderness at the fracture site, no signs of infection on examination, and radiographic union (Figure 1D).

Discussion

Optimal treatment of displaced clavicle fractures is controversial. While nonoperative treatment has been recommended,^1-3 especially in skeletally immature populations with a capacity for remodeling,^7-9 2 recent randomized clinical trials have demonstrated improved patient outcomes with ORIF.^6,8,9 Traditionally, ORIF was recommended with tenting of the skin because of concern for an impending open fracture. However, recent review materials have implied that this complication may only be theoretical.^3,5 Indeed, in 2 randomized trials, sufficient displacement to cause concern for impending violation of the skin envelope was not listed as an exclusion criteria.^8,9 We report 2 cases of displaced comminuted clavicle fractures that were initially managed nonoperatively but developed open lesions at the fracture site. This complication, while rare, is possible, and surgeons must consider it as a possibility when assessing patients with displaced clavicle fractures. To the best of the authors’ knowledge, no guidelines exist to direct antibiotic choice and duration in secondarily open fractures.

These 2 cases have several features in common that may serve as risk factors for impending violation of the skin envelope. Both fractures had a vertically angulated segmental piece of comminution with a sharp spike. This feature has been identified as a potential risk factor for subsequent development of an open fracture in a case report of fragment excision without reduction or fixation to allow rapid return to play in a professional jockey.¹³ Both patients in these cases presented with high-velocity mechanisms of injury and significant displacement, both of which may serve as risk factors. In the only similar case the authors could identify, Strauss and colleagues¹⁴ described a distal clavicle fracture with significant displacement and with secondary ulceration of the skin complicated by infection presenting with purulent discharge, cultured positive for methicillin-sensitive Staphylococcus aureus, requiring management with an external fixator and 6 weeks of intravenous antibiotics. Because both cases presented here occurred in healthy adolescent patients who were taken urgently for I&D and ORIF as soon as the wound was discovered, deep infection was avoided in these cases. Finally, in 1 case, a figure-of-8 brace was employed, which may also have placed pressure on the skin overlying the fracture and may have predisposed this patient to this complication.

Conclusion

In displaced midshaft clavicle fractures, tenting of the skin sufficient to cause subsequent violation of the soft-tissue envelope is possible and is more than a theoretical risk. At-risk patients, ie, those with a vertically angulated sharp fragment of comminution, should be counseled appropriately and observed closely or considered for primary ORIF.

Fractures of the clavicle, which account for 2.6% of all fractures, are displaced in 70% of cases and are mid-diaphyseal in 80% of cases.^1-3 Historically, both displaced and nondisplaced fractures were treated nonoperatively with excellent outcomes reported in the majority of patients.^1-3 Traditionally, the indications for surgical fixation of a clavicular fracture include open fractures, which occur infrequently, accounting for only 3.2% of clavicle fractures.⁴ Other indications include floating shoulder girdle or scapulothoracic dissociation, neurovascular injury, and skin “tenting” by the fracture fragments.^3,5 Recently, both meta-analyses and randomized clinical trials have reported reduced malunion rates and improved patient outcomes with open reduction and internal fixation (ORIF).^6-9 Consequently, operative fixation could be considered in patients with 100% displacement or greater than 1.5 cm shortening.^6-9 Open reduction and internal fixation of the clavicle has been demonstrated to have excellent outcomes in pediatric populations as well.¹⁰

The clavicle is subcutaneous for much of its length and, thus, displaced clavicular fractures often result in a visible deformity with a stretch of the soft-tissue envelope over the fracture. While this has been suggested as an operative indication, several recent sources indicate that this concern may only be theoretical. According to the fourth edition of Skeletal Trauma, “It is often stated that open reduction and internal fixation should be considered if the skin is threatened by pressure from a prominent clavicle fracture fragment; however, it is extremely rare of the skin to be perforated from within.”⁵ The most recent Journal of Bone and Joint Surgery Current Concepts Review on the subject stated that “open fractures or soft-tissue tenting sufficient to produce skin necrosis is uncommon.”³ To the best of our knowledge, there is no reported case of a displaced midshaft clavicle fracture with secondary skin necrosis and conversion into an open fracture, validating the conclusion that this complication may be only theoretical. Given that surgical fixation carries a risk of complications including wound complications, infection, nonunion, malunion, and damage to the nearby neurovascular structures and pleural apices,¹¹ some surgeons may be uncertain how to proceed in cases at risk for disturbance of the soft tissues.

We report 2 adolescent cases of displaced, comminuted clavicle fractures in which the skin was initially intact. Both were managed nonoperatively and both secondarily presented with open lesions at the fracture site requiring urgent irrigation and débridement (I&D) and ORIF. The patients and their guardians provided written informed consent for print and electronic publication of these case reports.

Case Reports

Case 1

A 15-year-old boy with no significant medical or surgical history flipped over the handlebars of his bicycle the day prior to presentation and sustained a clavicle fracture on his left nondominant upper extremity. This was an isolated injury. On examination, his skin was intact with an area of tender mild osseous protuberance at the midclavicle with associated surrounding edema. He was neurovascularly intact. Radiographs showed a displaced fracture of the midshaft of the clavicle with 20% shortening with a vertically angulated piece of comminution (Figure 1A). After a discussion of the treatment options with the family, the decision was made to pursue nonoperative treatment with sling immobilization as needed and restriction from gym and sports.

Two and a half weeks later, the patient presented at follow-up with significant reduction but persistence of his pain and a new complaint of drainage from the area of the fracture. On examination, he was found to have a puncture wound of the skin with exposed clavicle protruding through the wound with a 1-cm circumferential area of erythema without purulence present or expressible. The patient denied reinjury and endorsed compliance with sling immobilization. He was taken for urgent I&D and ORIF. After excision of the eschar surrounding the open lesion and full I&D of the soft tissues, the protruding spike was partially excised and the fracture site was débrided. The fracture was reduced and fixated with a lag screw and neutralization plate technique using an anatomically contoured locking clavicle plate (Synthes). Vancomycin powder was sprinkled into the wound at the completion of the procedure to reduce the chance of infection.¹²

Postoperatively, the patient was prescribed oral clindamycin but was subsequently switched to oral cephalexin because of mild signs of an allergic reaction, for a total course of antibiotics of 1 week. The patient was immobilized in a sling for comfort for the first 9 weeks postoperatively until radiographic union occurred. The patient’s wound healed uneventfully and with acceptable cosmesis. He was released to full activities at 10 weeks postoperatively. At final follow-up 6 months after surgery, the patient had returned to all of his regular activities without pain, and with full range of motion and no demonstrable deficits with radiographic union (Figure 1B).

Case 2

An 11-year-old boy with no significant medical or surgical history fell onto his right dominant upper extremity while doing a jump on his dirt bike 1 week prior to presentation, sustaining a clavicle fracture. This was an isolated injury. He was seen and evaluated by an outside orthopedist who noted that the soft-tissue envelope was intact and the patient was neurovascularly intact. Radiographs showed a displaced fracture of the midshaft of the clavicle with 15% shortening and with a vertically angulated piece of comminution (Figure 1C). Nonoperative treatment with a figure-of-8 brace was recommended. The patient’s discomfort completely resolved.

One week later, when he presented to the outside orthopedist for follow-up, the development of a wound overlying the fracture site was noted, and the patient was started on oral trimethoprim/sulfamethoxazole and referred to our office for treatment (Figure 1D). The patient denied reinjury and endorsed compliance with brace immobilization. On examination, the patient was afebrile and was noted to have a puncture wound at the fracture site with a protruding spike of bone and surrounding erythema but without present or expressible discharge (Figure 2). The patient was taken urgently for I&D and ORIF, using a similar technique to case 1, except that no lag screw was employed.

Postoperatively, the patient did well with no complications; he was prescribed oral cephalexin for 1 week. The patient was immobilized in a sling for the first 5 weeks after surgery until radiographic union had occurred, after which the sling was discontinued. The patient’s wound healed uneventfully and with acceptable cosmesis. The patient was released from activity restrictions at 6 weeks postoperatively. At final follow-up 5 weeks after surgery, the patient had full painless range of motion, no tenderness at the fracture site, no signs of infection on examination, and radiographic union (Figure 1D).

Discussion

Optimal treatment of displaced clavicle fractures is controversial. While nonoperative treatment has been recommended,^1-3 especially in skeletally immature populations with a capacity for remodeling,^7-9 2 recent randomized clinical trials have demonstrated improved patient outcomes with ORIF.^6,8,9 Traditionally, ORIF was recommended with tenting of the skin because of concern for an impending open fracture. However, recent review materials have implied that this complication may only be theoretical.^3,5 Indeed, in 2 randomized trials, sufficient displacement to cause concern for impending violation of the skin envelope was not listed as an exclusion criteria.^8,9 We report 2 cases of displaced comminuted clavicle fractures that were initially managed nonoperatively but developed open lesions at the fracture site. This complication, while rare, is possible, and surgeons must consider it as a possibility when assessing patients with displaced clavicle fractures. To the best of the authors’ knowledge, no guidelines exist to direct antibiotic choice and duration in secondarily open fractures.

These 2 cases have several features in common that may serve as risk factors for impending violation of the skin envelope. Both fractures had a vertically angulated segmental piece of comminution with a sharp spike. This feature has been identified as a potential risk factor for subsequent development of an open fracture in a case report of fragment excision without reduction or fixation to allow rapid return to play in a professional jockey.¹³ Both patients in these cases presented with high-velocity mechanisms of injury and significant displacement, both of which may serve as risk factors. In the only similar case the authors could identify, Strauss and colleagues¹⁴ described a distal clavicle fracture with significant displacement and with secondary ulceration of the skin complicated by infection presenting with purulent discharge, cultured positive for methicillin-sensitive Staphylococcus aureus, requiring management with an external fixator and 6 weeks of intravenous antibiotics. Because both cases presented here occurred in healthy adolescent patients who were taken urgently for I&D and ORIF as soon as the wound was discovered, deep infection was avoided in these cases. Finally, in 1 case, a figure-of-8 brace was employed, which may also have placed pressure on the skin overlying the fracture and may have predisposed this patient to this complication.

Conclusion

In displaced midshaft clavicle fractures, tenting of the skin sufficient to cause subsequent violation of the soft-tissue envelope is possible and is more than a theoretical risk. At-risk patients, ie, those with a vertically angulated sharp fragment of comminution, should be counseled appropriately and observed closely or considered for primary ORIF.

Fractures of the clavicle, which account for 2.6% of all fractures, are displaced in 70% of cases and are mid-diaphyseal in 80% of cases.^1-3 Historically, both displaced and nondisplaced fractures were treated nonoperatively with excellent outcomes reported in the majority of patients.^1-3 Traditionally, the indications for surgical fixation of a clavicular fracture include open fractures, which occur infrequently, accounting for only 3.2% of clavicle fractures.⁴ Other indications include floating shoulder girdle or scapulothoracic dissociation, neurovascular injury, and skin “tenting” by the fracture fragments.^3,5 Recently, both meta-analyses and randomized clinical trials have reported reduced malunion rates and improved patient outcomes with open reduction and internal fixation (ORIF).^6-9 Consequently, operative fixation could be considered in patients with 100% displacement or greater than 1.5 cm shortening.^6-9 Open reduction and internal fixation of the clavicle has been demonstrated to have excellent outcomes in pediatric populations as well.¹⁰

The clavicle is subcutaneous for much of its length and, thus, displaced clavicular fractures often result in a visible deformity with a stretch of the soft-tissue envelope over the fracture. While this has been suggested as an operative indication, several recent sources indicate that this concern may only be theoretical. According to the fourth edition of Skeletal Trauma, “It is often stated that open reduction and internal fixation should be considered if the skin is threatened by pressure from a prominent clavicle fracture fragment; however, it is extremely rare of the skin to be perforated from within.”⁵ The most recent Journal of Bone and Joint Surgery Current Concepts Review on the subject stated that “open fractures or soft-tissue tenting sufficient to produce skin necrosis is uncommon.”³ To the best of our knowledge, there is no reported case of a displaced midshaft clavicle fracture with secondary skin necrosis and conversion into an open fracture, validating the conclusion that this complication may be only theoretical. Given that surgical fixation carries a risk of complications including wound complications, infection, nonunion, malunion, and damage to the nearby neurovascular structures and pleural apices,¹¹ some surgeons may be uncertain how to proceed in cases at risk for disturbance of the soft tissues.

We report 2 adolescent cases of displaced, comminuted clavicle fractures in which the skin was initially intact. Both were managed nonoperatively and both secondarily presented with open lesions at the fracture site requiring urgent irrigation and débridement (I&D) and ORIF. The patients and their guardians provided written informed consent for print and electronic publication of these case reports.

Case Reports

Case 1

A 15-year-old boy with no significant medical or surgical history flipped over the handlebars of his bicycle the day prior to presentation and sustained a clavicle fracture on his left nondominant upper extremity. This was an isolated injury. On examination, his skin was intact with an area of tender mild osseous protuberance at the midclavicle with associated surrounding edema. He was neurovascularly intact. Radiographs showed a displaced fracture of the midshaft of the clavicle with 20% shortening with a vertically angulated piece of comminution (Figure 1A). After a discussion of the treatment options with the family, the decision was made to pursue nonoperative treatment with sling immobilization as needed and restriction from gym and sports.

Two and a half weeks later, the patient presented at follow-up with significant reduction but persistence of his pain and a new complaint of drainage from the area of the fracture. On examination, he was found to have a puncture wound of the skin with exposed clavicle protruding through the wound with a 1-cm circumferential area of erythema without purulence present or expressible. The patient denied reinjury and endorsed compliance with sling immobilization. He was taken for urgent I&D and ORIF. After excision of the eschar surrounding the open lesion and full I&D of the soft tissues, the protruding spike was partially excised and the fracture site was débrided. The fracture was reduced and fixated with a lag screw and neutralization plate technique using an anatomically contoured locking clavicle plate (Synthes). Vancomycin powder was sprinkled into the wound at the completion of the procedure to reduce the chance of infection.¹²

Postoperatively, the patient was prescribed oral clindamycin but was subsequently switched to oral cephalexin because of mild signs of an allergic reaction, for a total course of antibiotics of 1 week. The patient was immobilized in a sling for comfort for the first 9 weeks postoperatively until radiographic union occurred. The patient’s wound healed uneventfully and with acceptable cosmesis. He was released to full activities at 10 weeks postoperatively. At final follow-up 6 months after surgery, the patient had returned to all of his regular activities without pain, and with full range of motion and no demonstrable deficits with radiographic union (Figure 1B).

Case 2

An 11-year-old boy with no significant medical or surgical history fell onto his right dominant upper extremity while doing a jump on his dirt bike 1 week prior to presentation, sustaining a clavicle fracture. This was an isolated injury. He was seen and evaluated by an outside orthopedist who noted that the soft-tissue envelope was intact and the patient was neurovascularly intact. Radiographs showed a displaced fracture of the midshaft of the clavicle with 15% shortening and with a vertically angulated piece of comminution (Figure 1C). Nonoperative treatment with a figure-of-8 brace was recommended. The patient’s discomfort completely resolved.

One week later, when he presented to the outside orthopedist for follow-up, the development of a wound overlying the fracture site was noted, and the patient was started on oral trimethoprim/sulfamethoxazole and referred to our office for treatment (Figure 1D). The patient denied reinjury and endorsed compliance with brace immobilization. On examination, the patient was afebrile and was noted to have a puncture wound at the fracture site with a protruding spike of bone and surrounding erythema but without present or expressible discharge (Figure 2). The patient was taken urgently for I&D and ORIF, using a similar technique to case 1, except that no lag screw was employed.

Postoperatively, the patient did well with no complications; he was prescribed oral cephalexin for 1 week. The patient was immobilized in a sling for the first 5 weeks after surgery until radiographic union had occurred, after which the sling was discontinued. The patient’s wound healed uneventfully and with acceptable cosmesis. The patient was released from activity restrictions at 6 weeks postoperatively. At final follow-up 5 weeks after surgery, the patient had full painless range of motion, no tenderness at the fracture site, no signs of infection on examination, and radiographic union (Figure 1D).

Discussion

Optimal treatment of displaced clavicle fractures is controversial. While nonoperative treatment has been recommended,^1-3 especially in skeletally immature populations with a capacity for remodeling,^7-9 2 recent randomized clinical trials have demonstrated improved patient outcomes with ORIF.^6,8,9 Traditionally, ORIF was recommended with tenting of the skin because of concern for an impending open fracture. However, recent review materials have implied that this complication may only be theoretical.^3,5 Indeed, in 2 randomized trials, sufficient displacement to cause concern for impending violation of the skin envelope was not listed as an exclusion criteria.^8,9 We report 2 cases of displaced comminuted clavicle fractures that were initially managed nonoperatively but developed open lesions at the fracture site. This complication, while rare, is possible, and surgeons must consider it as a possibility when assessing patients with displaced clavicle fractures. To the best of the authors’ knowledge, no guidelines exist to direct antibiotic choice and duration in secondarily open fractures.

These 2 cases have several features in common that may serve as risk factors for impending violation of the skin envelope. Both fractures had a vertically angulated segmental piece of comminution with a sharp spike. This feature has been identified as a potential risk factor for subsequent development of an open fracture in a case report of fragment excision without reduction or fixation to allow rapid return to play in a professional jockey.¹³ Both patients in these cases presented with high-velocity mechanisms of injury and significant displacement, both of which may serve as risk factors. In the only similar case the authors could identify, Strauss and colleagues¹⁴ described a distal clavicle fracture with significant displacement and with secondary ulceration of the skin complicated by infection presenting with purulent discharge, cultured positive for methicillin-sensitive Staphylococcus aureus, requiring management with an external fixator and 6 weeks of intravenous antibiotics. Because both cases presented here occurred in healthy adolescent patients who were taken urgently for I&D and ORIF as soon as the wound was discovered, deep infection was avoided in these cases. Finally, in 1 case, a figure-of-8 brace was employed, which may also have placed pressure on the skin overlying the fracture and may have predisposed this patient to this complication.

Conclusion

In displaced midshaft clavicle fractures, tenting of the skin sufficient to cause subsequent violation of the soft-tissue envelope is possible and is more than a theoretical risk. At-risk patients, ie, those with a vertically angulated sharp fragment of comminution, should be counseled appropriately and observed closely or considered for primary ORIF.

Adamantinomas are rare primary malignant bone tumors (less than 1% of all bone tumors) that arise most commonly in the tibia.¹ There is a predilection for adult men aged 20 to 50 years, with rare occurrences in children. These tumors are malignant, highly invasive, and have significant metastatic potential.² A rarely seen, benign variant, known as osteofibrous dysplasia–like adamantinoma, is described in the literature, with fewer than 135 cases reported.^3-5 This variant predominantly has benign characteristics of an osteofibrous dysplasia lesion but has the potential to transform into an adamantinoma.⁶ Osteofibrous dysplasia–like adamantinoma has been observed to regress with age and is also referred to as a regressing adamantinoma or differentiated adamantinoma.⁷ 

We report an uncommon case of an osteofibrous dysplasia–like adamantinoma of the tibia in a 15-year-old girl. We decided to observe the tumor with regular 3- to 6-month follow-ups. Osteofibrous dysplasia–like adamantinoma in our patient has remained stable for 2 years and has an excellent prognosis.⁸ We report this case for its rarity, its short-term stability, and significant treatment implications due to its potential to regress or develop into a malignant form. The patient and the patient’s guardian provided written informed consent for print and electronic publication of this case report. 

Case Report

A healthy 15-year-old girl was referred to our institution for evaluation of anterior left knee pain. She had sustained a fall while playing basketball 3 months earlier and had been having left knee pain since that time. She did not have any swelling, catching, or locking in her left knee. She denied any recent fever, chills, night sweats, weight loss, nausea, vomiting, or diarrhea. On physical examination, her gait was normal and no swelling, erythema, or tenderness was noticed around the left knee.

Plain radiographs revealed a heterogeneous lesion with sclerosis and thickening of the anteromedial cortex of the proximal left tibia (Figures 1A, 1B). A computed tomography (CT) scan of the abdomen, pelvis, and chest showed no osseous abnormalities. A whole-body bone scan showed activity in the anterior aspect of the left proximal tibia. No other areas of activity were noted. Magnetic resonance imaging of the left leg showed an elongated, multiloculated, enhancing mass arising from the anterolateral cortex and extending from the tibial tuberosity to the mid-diaphysis of the left tibia. Histologic examination of the CT-guided core needle biopsy specimen showed that the lesion was composed of dense fibrocollagenous tissue separating irregular bony trabeculae with osteoblastic and osteoclastic activity. There was no evidence of any atypical cells, necrosis, or significant mitotic activity. No epithelial cells were identified on hematoxylin-eosin (H&E) stain (Figure 2). However, immunohistochemical staining was positive for focal cytokeratin-positive epithelial cells (Figure 3). The lesion was diagnosed as an osteofibrous dysplasia–like adamantinoma on the basis of the radiographic and histologic findings. We elected nonoperative intervention given the benign nature of the lesion and its potential to regress. Given the possibility of sampling error and potential for progression, the patient was followed regularly at 3- to 6-month intervals over a 2-year period without disease progression. 

Discussion

Osteofibrous dysplasia, osteofibrous dysplasia–like adamantinoma, and adamantinoma are rare fibro-osseous lesions that largely involve the midshaft of the tibia. Osteofibrous dysplasia accounts for 0.2% of primary bone tumors, whereas adamantinoma accounts for 0.1% to 0.5% of malignant bone tumors.⁹ Osteofibrous dysplasia is a benign lesion composed primarily of fibro-osseous tissue. Adamantinoma, however, is a slow-growing, low-grade, malignant biphasic tumor with intermingled epithelial and fibro-osseous components. It is an aggressive tumor that is locally invasive and can metastasize.² Osteofibrous dysplasia–like adamantinoma (also known as differentiated or regressing adamantinoma) is a benign lesion like osteofibrous dysplasia but has features of both osteofibrous dysplasia and adamantinoma. Osteofibrous dysplasia–like adamantinoma may progress and become a malignant adamantinoma.^6,10

The radiologic features of the 3 lesions are quite similar. It is not possible to distinguish between osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma based on imaging alone.⁹ Adamantinoma, being highly invasive, can be distinguished from osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma according to the extent of involvement of the medullary cavity seen on magnetic resonance imaging.⁹ Complete involvement of the medullary cavity is almost always seen in an adamantinoma. Involvement of the medullary cavity is minimal or absent in osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma lesions. 

Tissue confirmation through biopsy is crucial for accurate diagnosis. A biopsy should always be performed on any suspicious lesion,^3,6 and the fibro-osseous lesion should be treated as an adamantinoma if findings are equivocal. A biopsy also distinctly distinguishes these lesions from benign fibrous cortical defects, which have a similar radiographic appearance. While open biopsy is the gold standard, minimally invasive techniques such as core needle biopsy and fine needle biopsy are increasingly used.⁶ Because of the higher risk of misdiagnosis with minimally invasive techniques, radiographic confirmation is highly recommended.⁵

Histologically, both osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma do not stain for cytokeratin on H&E stain. However, they can be differentiated based on immunohistochemical staining for cytokeratin. Osteofibrous dysplasia lesions exhibit diffuse staining whereas osteofibrous dysplasia–like adamantinoma lesions show focal staining of small nests of epithelial cells. Adamantinoma, in comparison, exhibits a biphasic pattern on H&E stain, representing areas of epithelial and osteofibrous cells. Immunohistochemical staining for cytokeratin of an adamantinoma reveals large nests of epithelial cells.  

The association between osteofibrous dysplasia, osteofibrous dysplasia–like adamantinoma, and adamantinoma is not clearly established. However, it is widely believed that these 3 lesions represent a spectrum of the same disease and are linearly related in disease progression, with osteofibrous dysplasia at the benign end of the spectrum, osteofibrous dysplasia–like adamantinoma the intermediate form, and adamantinoma at the malignant end of the spectrum.¹¹

Hazelbag and colleagues⁶ and Springfield and colleagues¹⁰ point out that osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma could be precursor lesions of adamantinoma. We found several studies in the literature that support and document progression from osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma to an adamantinoma.^4,6,10,12 Other studies, however, showed no progression from either a benign osteofibrous dysplasia or an osteofibrous dysplasia–like adamantinoma lesion to a malignant adamantinoma. Park and colleagues¹³ described 41 cases of osteofibrous dysplasia that did not progress to adamantinoma. Kuruvilla and Steiner⁸ described 5 cases of osteofibrous dysplasia–like adamantinoma that showed no progression to adamantinoma. Additionally, our case has not progressed and has remained radiographically stable over a 2-year follow-up. Czerniak and colleagues⁷ and Ueda and colleagues¹⁴ postulated, based on histologic and immunohistochemical studies, that osteofibrous dysplasia–like adamantinoma might be a regressing form of an adamantinoma that is undergoing reparative processes that could result in complete elimination of all tumor cells.

In general, any lesion with absent to low malignant potential could be managed nonoperatively with periodic observation and without the need for surgical intervention. Thus, identification of a stable or nonprogressing osteofibrous dysplasia–like adamantinoma lesion has significant treatment implications. Campanacci and Laus¹⁵ at the Rizzoli Institute in Milan, through long term follow-up of their patients with osteofibrous dysplasia, found that most lesions had a tendency to regress spontaneously by puberty. They recommended that nonextensive osteofibrous dysplasia lesions should be observed, and surgery should be delayed until puberty. Gleason and colleagues¹⁶ also recommended nonoperative management of osteofibrous dysplasia lesions, with surgery used only for large, deforming, and highly invasive lesions. We recommend a similar treatment approach for osteofibrous dysplasia–like adamantinoma lesions.

Adamantinomas, however, are usually symptomatic, are highly invasive, have a high recurrence rate, and can metastasize.9 In these patients, a wide en bloc resection or amputation should be performed as soon as possible.11 Our case highlights that osteofibrous dysplasia–like adamantinoma lesions can occur in children and can remain stable, especially over the short term. Such lesions can be observed without surgical intervention.

Adamantinomas are rare primary malignant bone tumors (less than 1% of all bone tumors) that arise most commonly in the tibia.¹ There is a predilection for adult men aged 20 to 50 years, with rare occurrences in children. These tumors are malignant, highly invasive, and have significant metastatic potential.² A rarely seen, benign variant, known as osteofibrous dysplasia–like adamantinoma, is described in the literature, with fewer than 135 cases reported.^3-5 This variant predominantly has benign characteristics of an osteofibrous dysplasia lesion but has the potential to transform into an adamantinoma.⁶ Osteofibrous dysplasia–like adamantinoma has been observed to regress with age and is also referred to as a regressing adamantinoma or differentiated adamantinoma.⁷ 

We report an uncommon case of an osteofibrous dysplasia–like adamantinoma of the tibia in a 15-year-old girl. We decided to observe the tumor with regular 3- to 6-month follow-ups. Osteofibrous dysplasia–like adamantinoma in our patient has remained stable for 2 years and has an excellent prognosis.⁸ We report this case for its rarity, its short-term stability, and significant treatment implications due to its potential to regress or develop into a malignant form. The patient and the patient’s guardian provided written informed consent for print and electronic publication of this case report. 

Case Report

A healthy 15-year-old girl was referred to our institution for evaluation of anterior left knee pain. She had sustained a fall while playing basketball 3 months earlier and had been having left knee pain since that time. She did not have any swelling, catching, or locking in her left knee. She denied any recent fever, chills, night sweats, weight loss, nausea, vomiting, or diarrhea. On physical examination, her gait was normal and no swelling, erythema, or tenderness was noticed around the left knee.

Plain radiographs revealed a heterogeneous lesion with sclerosis and thickening of the anteromedial cortex of the proximal left tibia (Figures 1A, 1B). A computed tomography (CT) scan of the abdomen, pelvis, and chest showed no osseous abnormalities. A whole-body bone scan showed activity in the anterior aspect of the left proximal tibia. No other areas of activity were noted. Magnetic resonance imaging of the left leg showed an elongated, multiloculated, enhancing mass arising from the anterolateral cortex and extending from the tibial tuberosity to the mid-diaphysis of the left tibia. Histologic examination of the CT-guided core needle biopsy specimen showed that the lesion was composed of dense fibrocollagenous tissue separating irregular bony trabeculae with osteoblastic and osteoclastic activity. There was no evidence of any atypical cells, necrosis, or significant mitotic activity. No epithelial cells were identified on hematoxylin-eosin (H&E) stain (Figure 2). However, immunohistochemical staining was positive for focal cytokeratin-positive epithelial cells (Figure 3). The lesion was diagnosed as an osteofibrous dysplasia–like adamantinoma on the basis of the radiographic and histologic findings. We elected nonoperative intervention given the benign nature of the lesion and its potential to regress. Given the possibility of sampling error and potential for progression, the patient was followed regularly at 3- to 6-month intervals over a 2-year period without disease progression. 

Discussion

Osteofibrous dysplasia, osteofibrous dysplasia–like adamantinoma, and adamantinoma are rare fibro-osseous lesions that largely involve the midshaft of the tibia. Osteofibrous dysplasia accounts for 0.2% of primary bone tumors, whereas adamantinoma accounts for 0.1% to 0.5% of malignant bone tumors.⁹ Osteofibrous dysplasia is a benign lesion composed primarily of fibro-osseous tissue. Adamantinoma, however, is a slow-growing, low-grade, malignant biphasic tumor with intermingled epithelial and fibro-osseous components. It is an aggressive tumor that is locally invasive and can metastasize.² Osteofibrous dysplasia–like adamantinoma (also known as differentiated or regressing adamantinoma) is a benign lesion like osteofibrous dysplasia but has features of both osteofibrous dysplasia and adamantinoma. Osteofibrous dysplasia–like adamantinoma may progress and become a malignant adamantinoma.^6,10

The radiologic features of the 3 lesions are quite similar. It is not possible to distinguish between osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma based on imaging alone.⁹ Adamantinoma, being highly invasive, can be distinguished from osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma according to the extent of involvement of the medullary cavity seen on magnetic resonance imaging.⁹ Complete involvement of the medullary cavity is almost always seen in an adamantinoma. Involvement of the medullary cavity is minimal or absent in osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma lesions. 

Tissue confirmation through biopsy is crucial for accurate diagnosis. A biopsy should always be performed on any suspicious lesion,^3,6 and the fibro-osseous lesion should be treated as an adamantinoma if findings are equivocal. A biopsy also distinctly distinguishes these lesions from benign fibrous cortical defects, which have a similar radiographic appearance. While open biopsy is the gold standard, minimally invasive techniques such as core needle biopsy and fine needle biopsy are increasingly used.⁶ Because of the higher risk of misdiagnosis with minimally invasive techniques, radiographic confirmation is highly recommended.⁵

Histologically, both osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma do not stain for cytokeratin on H&E stain. However, they can be differentiated based on immunohistochemical staining for cytokeratin. Osteofibrous dysplasia lesions exhibit diffuse staining whereas osteofibrous dysplasia–like adamantinoma lesions show focal staining of small nests of epithelial cells. Adamantinoma, in comparison, exhibits a biphasic pattern on H&E stain, representing areas of epithelial and osteofibrous cells. Immunohistochemical staining for cytokeratin of an adamantinoma reveals large nests of epithelial cells.  

The association between osteofibrous dysplasia, osteofibrous dysplasia–like adamantinoma, and adamantinoma is not clearly established. However, it is widely believed that these 3 lesions represent a spectrum of the same disease and are linearly related in disease progression, with osteofibrous dysplasia at the benign end of the spectrum, osteofibrous dysplasia–like adamantinoma the intermediate form, and adamantinoma at the malignant end of the spectrum.¹¹

Hazelbag and colleagues⁶ and Springfield and colleagues¹⁰ point out that osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma could be precursor lesions of adamantinoma. We found several studies in the literature that support and document progression from osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma to an adamantinoma.^4,6,10,12 Other studies, however, showed no progression from either a benign osteofibrous dysplasia or an osteofibrous dysplasia–like adamantinoma lesion to a malignant adamantinoma. Park and colleagues¹³ described 41 cases of osteofibrous dysplasia that did not progress to adamantinoma. Kuruvilla and Steiner⁸ described 5 cases of osteofibrous dysplasia–like adamantinoma that showed no progression to adamantinoma. Additionally, our case has not progressed and has remained radiographically stable over a 2-year follow-up. Czerniak and colleagues⁷ and Ueda and colleagues¹⁴ postulated, based on histologic and immunohistochemical studies, that osteofibrous dysplasia–like adamantinoma might be a regressing form of an adamantinoma that is undergoing reparative processes that could result in complete elimination of all tumor cells.

In general, any lesion with absent to low malignant potential could be managed nonoperatively with periodic observation and without the need for surgical intervention. Thus, identification of a stable or nonprogressing osteofibrous dysplasia–like adamantinoma lesion has significant treatment implications. Campanacci and Laus¹⁵ at the Rizzoli Institute in Milan, through long term follow-up of their patients with osteofibrous dysplasia, found that most lesions had a tendency to regress spontaneously by puberty. They recommended that nonextensive osteofibrous dysplasia lesions should be observed, and surgery should be delayed until puberty. Gleason and colleagues¹⁶ also recommended nonoperative management of osteofibrous dysplasia lesions, with surgery used only for large, deforming, and highly invasive lesions. We recommend a similar treatment approach for osteofibrous dysplasia–like adamantinoma lesions.

Adamantinomas, however, are usually symptomatic, are highly invasive, have a high recurrence rate, and can metastasize.9 In these patients, a wide en bloc resection or amputation should be performed as soon as possible.11 Our case highlights that osteofibrous dysplasia–like adamantinoma lesions can occur in children and can remain stable, especially over the short term. Such lesions can be observed without surgical intervention.

Adamantinomas are rare primary malignant bone tumors (less than 1% of all bone tumors) that arise most commonly in the tibia.¹ There is a predilection for adult men aged 20 to 50 years, with rare occurrences in children. These tumors are malignant, highly invasive, and have significant metastatic potential.² A rarely seen, benign variant, known as osteofibrous dysplasia–like adamantinoma, is described in the literature, with fewer than 135 cases reported.^3-5 This variant predominantly has benign characteristics of an osteofibrous dysplasia lesion but has the potential to transform into an adamantinoma.⁶ Osteofibrous dysplasia–like adamantinoma has been observed to regress with age and is also referred to as a regressing adamantinoma or differentiated adamantinoma.⁷ 

We report an uncommon case of an osteofibrous dysplasia–like adamantinoma of the tibia in a 15-year-old girl. We decided to observe the tumor with regular 3- to 6-month follow-ups. Osteofibrous dysplasia–like adamantinoma in our patient has remained stable for 2 years and has an excellent prognosis.⁸ We report this case for its rarity, its short-term stability, and significant treatment implications due to its potential to regress or develop into a malignant form. The patient and the patient’s guardian provided written informed consent for print and electronic publication of this case report. 

Case Report

A healthy 15-year-old girl was referred to our institution for evaluation of anterior left knee pain. She had sustained a fall while playing basketball 3 months earlier and had been having left knee pain since that time. She did not have any swelling, catching, or locking in her left knee. She denied any recent fever, chills, night sweats, weight loss, nausea, vomiting, or diarrhea. On physical examination, her gait was normal and no swelling, erythema, or tenderness was noticed around the left knee.

Plain radiographs revealed a heterogeneous lesion with sclerosis and thickening of the anteromedial cortex of the proximal left tibia (Figures 1A, 1B). A computed tomography (CT) scan of the abdomen, pelvis, and chest showed no osseous abnormalities. A whole-body bone scan showed activity in the anterior aspect of the left proximal tibia. No other areas of activity were noted. Magnetic resonance imaging of the left leg showed an elongated, multiloculated, enhancing mass arising from the anterolateral cortex and extending from the tibial tuberosity to the mid-diaphysis of the left tibia. Histologic examination of the CT-guided core needle biopsy specimen showed that the lesion was composed of dense fibrocollagenous tissue separating irregular bony trabeculae with osteoblastic and osteoclastic activity. There was no evidence of any atypical cells, necrosis, or significant mitotic activity. No epithelial cells were identified on hematoxylin-eosin (H&E) stain (Figure 2). However, immunohistochemical staining was positive for focal cytokeratin-positive epithelial cells (Figure 3). The lesion was diagnosed as an osteofibrous dysplasia–like adamantinoma on the basis of the radiographic and histologic findings. We elected nonoperative intervention given the benign nature of the lesion and its potential to regress. Given the possibility of sampling error and potential for progression, the patient was followed regularly at 3- to 6-month intervals over a 2-year period without disease progression. 

Discussion

Osteofibrous dysplasia, osteofibrous dysplasia–like adamantinoma, and adamantinoma are rare fibro-osseous lesions that largely involve the midshaft of the tibia. Osteofibrous dysplasia accounts for 0.2% of primary bone tumors, whereas adamantinoma accounts for 0.1% to 0.5% of malignant bone tumors.⁹ Osteofibrous dysplasia is a benign lesion composed primarily of fibro-osseous tissue. Adamantinoma, however, is a slow-growing, low-grade, malignant biphasic tumor with intermingled epithelial and fibro-osseous components. It is an aggressive tumor that is locally invasive and can metastasize.² Osteofibrous dysplasia–like adamantinoma (also known as differentiated or regressing adamantinoma) is a benign lesion like osteofibrous dysplasia but has features of both osteofibrous dysplasia and adamantinoma. Osteofibrous dysplasia–like adamantinoma may progress and become a malignant adamantinoma.^6,10

The radiologic features of the 3 lesions are quite similar. It is not possible to distinguish between osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma based on imaging alone.⁹ Adamantinoma, being highly invasive, can be distinguished from osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma according to the extent of involvement of the medullary cavity seen on magnetic resonance imaging.⁹ Complete involvement of the medullary cavity is almost always seen in an adamantinoma. Involvement of the medullary cavity is minimal or absent in osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma lesions. 

Tissue confirmation through biopsy is crucial for accurate diagnosis. A biopsy should always be performed on any suspicious lesion,^3,6 and the fibro-osseous lesion should be treated as an adamantinoma if findings are equivocal. A biopsy also distinctly distinguishes these lesions from benign fibrous cortical defects, which have a similar radiographic appearance. While open biopsy is the gold standard, minimally invasive techniques such as core needle biopsy and fine needle biopsy are increasingly used.⁶ Because of the higher risk of misdiagnosis with minimally invasive techniques, radiographic confirmation is highly recommended.⁵

Histologically, both osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma do not stain for cytokeratin on H&E stain. However, they can be differentiated based on immunohistochemical staining for cytokeratin. Osteofibrous dysplasia lesions exhibit diffuse staining whereas osteofibrous dysplasia–like adamantinoma lesions show focal staining of small nests of epithelial cells. Adamantinoma, in comparison, exhibits a biphasic pattern on H&E stain, representing areas of epithelial and osteofibrous cells. Immunohistochemical staining for cytokeratin of an adamantinoma reveals large nests of epithelial cells.  

The association between osteofibrous dysplasia, osteofibrous dysplasia–like adamantinoma, and adamantinoma is not clearly established. However, it is widely believed that these 3 lesions represent a spectrum of the same disease and are linearly related in disease progression, with osteofibrous dysplasia at the benign end of the spectrum, osteofibrous dysplasia–like adamantinoma the intermediate form, and adamantinoma at the malignant end of the spectrum.¹¹

Hazelbag and colleagues⁶ and Springfield and colleagues¹⁰ point out that osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma could be precursor lesions of adamantinoma. We found several studies in the literature that support and document progression from osteofibrous dysplasia and osteofibrous dysplasia–like adamantinoma to an adamantinoma.^4,6,10,12 Other studies, however, showed no progression from either a benign osteofibrous dysplasia or an osteofibrous dysplasia–like adamantinoma lesion to a malignant adamantinoma. Park and colleagues¹³ described 41 cases of osteofibrous dysplasia that did not progress to adamantinoma. Kuruvilla and Steiner⁸ described 5 cases of osteofibrous dysplasia–like adamantinoma that showed no progression to adamantinoma. Additionally, our case has not progressed and has remained radiographically stable over a 2-year follow-up. Czerniak and colleagues⁷ and Ueda and colleagues¹⁴ postulated, based on histologic and immunohistochemical studies, that osteofibrous dysplasia–like adamantinoma might be a regressing form of an adamantinoma that is undergoing reparative processes that could result in complete elimination of all tumor cells.

In general, any lesion with absent to low malignant potential could be managed nonoperatively with periodic observation and without the need for surgical intervention. Thus, identification of a stable or nonprogressing osteofibrous dysplasia–like adamantinoma lesion has significant treatment implications. Campanacci and Laus¹⁵ at the Rizzoli Institute in Milan, through long term follow-up of their patients with osteofibrous dysplasia, found that most lesions had a tendency to regress spontaneously by puberty. They recommended that nonextensive osteofibrous dysplasia lesions should be observed, and surgery should be delayed until puberty. Gleason and colleagues¹⁶ also recommended nonoperative management of osteofibrous dysplasia lesions, with surgery used only for large, deforming, and highly invasive lesions. We recommend a similar treatment approach for osteofibrous dysplasia–like adamantinoma lesions.

Adamantinomas, however, are usually symptomatic, are highly invasive, have a high recurrence rate, and can metastasize.9 In these patients, a wide en bloc resection or amputation should be performed as soon as possible.11 Our case highlights that osteofibrous dysplasia–like adamantinoma lesions can occur in children and can remain stable, especially over the short term. Such lesions can be observed without surgical intervention.