Clinical Inquiries

Evidence summary

Organizational changes in physician offices can improve delivery of services for preventing and controlling disease.¹ Standing orders—typically defined as physician-approved protocols that authorize nurses or other staff members to perform procedures, such as immunizations without direct physician involvement¹—are readily applicable in ambulatory settings. However, only 30% of physicians use standing orders in their practices.²

Research on standing orders in ambulatory care has focused on immunizations and cancer screening (TABLE). Interventions implementing standing orders typically have multiple components and include staff education, chart flow sheets, and recall-reminders for patients.

TABLE
Effect of standing orders in ambulatory practice

Improvement in pneumococcal and flu vaccine rates
Three multicomponent RCTs of outpatient standing orders reported improved pneumococcal vaccination rates.^3-5 Similarly, 2 prospective, multicomponent cohort studies^6,7 and 1 retrospective study⁸ found improved rates of influenza vaccination with standing orders.

Childhood vaccination rates also show positive trends
Two controlled trials (1 randomized³ and 1 nonrandomized⁹) that incorporated standing orders examined their use in childhood immunizations (measles, mumps, and rubella [MMR]; oral polio vaccine [OPV]; Haemophilus influenzae, type b [HIB]; diphtheria and tetanus toxoids with acellular pertussis [DTaP]; and hepatitis B). One trial reported increased use of acute care immunization opportunities;⁹ the other showed a nonsignificant positive trend in vaccination rates.³

Standing orders increase 1 form of cancer screening, not another
A multicomponent RCT of standing orders for mammography and colorectal cancer screening found a statistically significant increase in screening for mammography, but not colorectal cancer.³

Recommendations

The Society of Adolescent Medicine recommends standing orders for administration of influenza vaccine during flu season.¹⁰

The Task Force on Community Preventive Services recommends standing orders for adult vaccinations based on “strong evidence,” but states that insufficient evidence exists to recommend standing orders for childhood vaccinations.¹¹ Vaccines examined include MMR, DTaP, HIB, hepatitis B, and varicella for young children; hepatitis B, varicella, MMR, and tetanus-diphtheria toxoids (Td) for adolescents; Td for adults up to 65 years of age; and influenza and pneumococcal vaccines for adults 65 years and older.

The Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention recommends standing orders for influenza and pneumococcal vaccines.¹²

Evidence summary

Organizational changes in physician offices can improve delivery of services for preventing and controlling disease.¹ Standing orders—typically defined as physician-approved protocols that authorize nurses or other staff members to perform procedures, such as immunizations without direct physician involvement¹—are readily applicable in ambulatory settings. However, only 30% of physicians use standing orders in their practices.²

Research on standing orders in ambulatory care has focused on immunizations and cancer screening (TABLE). Interventions implementing standing orders typically have multiple components and include staff education, chart flow sheets, and recall-reminders for patients.

TABLE
Effect of standing orders in ambulatory practice

Improvement in pneumococcal and flu vaccine rates
Three multicomponent RCTs of outpatient standing orders reported improved pneumococcal vaccination rates.^3-5 Similarly, 2 prospective, multicomponent cohort studies^6,7 and 1 retrospective study⁸ found improved rates of influenza vaccination with standing orders.

Childhood vaccination rates also show positive trends
Two controlled trials (1 randomized³ and 1 nonrandomized⁹) that incorporated standing orders examined their use in childhood immunizations (measles, mumps, and rubella [MMR]; oral polio vaccine [OPV]; Haemophilus influenzae, type b [HIB]; diphtheria and tetanus toxoids with acellular pertussis [DTaP]; and hepatitis B). One trial reported increased use of acute care immunization opportunities;⁹ the other showed a nonsignificant positive trend in vaccination rates.³

Standing orders increase 1 form of cancer screening, not another
A multicomponent RCT of standing orders for mammography and colorectal cancer screening found a statistically significant increase in screening for mammography, but not colorectal cancer.³

Recommendations

The Society of Adolescent Medicine recommends standing orders for administration of influenza vaccine during flu season.¹⁰

The Task Force on Community Preventive Services recommends standing orders for adult vaccinations based on “strong evidence,” but states that insufficient evidence exists to recommend standing orders for childhood vaccinations.¹¹ Vaccines examined include MMR, DTaP, HIB, hepatitis B, and varicella for young children; hepatitis B, varicella, MMR, and tetanus-diphtheria toxoids (Td) for adolescents; Td for adults up to 65 years of age; and influenza and pneumococcal vaccines for adults 65 years and older.

The Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention recommends standing orders for influenza and pneumococcal vaccines.¹²

Evidence summary

Organizational changes in physician offices can improve delivery of services for preventing and controlling disease.¹ Standing orders—typically defined as physician-approved protocols that authorize nurses or other staff members to perform procedures, such as immunizations without direct physician involvement¹—are readily applicable in ambulatory settings. However, only 30% of physicians use standing orders in their practices.²

Research on standing orders in ambulatory care has focused on immunizations and cancer screening (TABLE). Interventions implementing standing orders typically have multiple components and include staff education, chart flow sheets, and recall-reminders for patients.

TABLE
Effect of standing orders in ambulatory practice

Improvement in pneumococcal and flu vaccine rates
Three multicomponent RCTs of outpatient standing orders reported improved pneumococcal vaccination rates.^3-5 Similarly, 2 prospective, multicomponent cohort studies^6,7 and 1 retrospective study⁸ found improved rates of influenza vaccination with standing orders.

Childhood vaccination rates also show positive trends
Two controlled trials (1 randomized³ and 1 nonrandomized⁹) that incorporated standing orders examined their use in childhood immunizations (measles, mumps, and rubella [MMR]; oral polio vaccine [OPV]; Haemophilus influenzae, type b [HIB]; diphtheria and tetanus toxoids with acellular pertussis [DTaP]; and hepatitis B). One trial reported increased use of acute care immunization opportunities;⁹ the other showed a nonsignificant positive trend in vaccination rates.³

Standing orders increase 1 form of cancer screening, not another
A multicomponent RCT of standing orders for mammography and colorectal cancer screening found a statistically significant increase in screening for mammography, but not colorectal cancer.³

Recommendations

The Society of Adolescent Medicine recommends standing orders for administration of influenza vaccine during flu season.¹⁰

The Task Force on Community Preventive Services recommends standing orders for adult vaccinations based on “strong evidence,” but states that insufficient evidence exists to recommend standing orders for childhood vaccinations.¹¹ Vaccines examined include MMR, DTaP, HIB, hepatitis B, and varicella for young children; hepatitis B, varicella, MMR, and tetanus-diphtheria toxoids (Td) for adolescents; Td for adults up to 65 years of age; and influenza and pneumococcal vaccines for adults 65 years and older.

The Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention recommends standing orders for influenza and pneumococcal vaccines.¹²

Evidence summary

The largest meta-analysis to look at the effectiveness of clinical examination and patient history for diagnosing soft-tissue shoulder disorders evaluated 10 cohort studies.¹ Pooled results from 4 of the studies suggested that a composite clinical examination could safely rule out a full-thickness rotator cuff tear with a sensitivity of 0.9 (95% confidence interval [CI], 0.87-0.93). The standard in the studies cited was either arthrography or surgery. No single physical examination finding, when looked at in isolation, could reliably produce comparable accuracy.

The best 3 tests for full-thickness rotator cuff tear
A retrospective cohort study stratified 352 patients who had undergone operative evaluation by the degree of severity of rotator cuff pathology.² Investigators assessed 8 physical exam tests—the Neer impingement sign, Hawkins-Kennedy impingement sign, painful arc sign, supraspinatus muscle strength test, Speed’s test, cross-body adduction test, drop-arm sign, and infraspinatus muscle strength test—to determine their diagnostic utility. (See “A glossary of tests for rotator cuff injury”)

The combination of the painful arc sign, drop-arm sign, and weakness in external rotation (positive infraspinatus muscle strength test) produced the best likelihood ratios (LRs) for detecting full-thickness rotator cuff tears ( TABLE ).

TABLE
Comparison of tests to detect full-thickness rotator cuff tear

A winning diagnostic combination for any degree of impingement disease
The best combination of tests to diagnose any degree of impingement disease is:²

• a positive Hawkins-Kennedy impingement sign
• a positive painful arc sign
• a positive infraspinatus muscle strength test.

Negative supraspinatus test helps rule out massive tear
A third study evaluated the validity of the supraspinatus muscle strength test alone to diagnose patients with rotator cuff pathology using arthroscopy or open surgery as the reference standard.³ A negative supraspinatus test, when defined by weakness, significantly decreased the posttest probability (LR-=0.17) of detecting a massive rotator cuff tear.

Correlate test results with clinical history
Most individual tests for rotator cuff disease are not sensitive or specific enough to effectively rule in or rule out a rotator cuff tear. Many of the findings studied can be positive in the presence of other shoulder conditions and should be correlated with the clinical history of each patient. One of the noted limitations of these 3 studies is that none was carried out in a primary care setting.

Recommendations

A textbook published by The American Academy of Orthopedic Surgeons describes a continuum of injuries from impingement syndrome to full-thickness rotator cuff tears that commonly present as anterior and lateral shoulder pain and are often associated with night pain.⁴ Atrophy of the supraspinatus and infraspinatus muscles may indicate a longstanding rotator cuff tear. The authors’ recommendations for clinical examination include performing the Neer and Hawkins-Kennedy tests before and after subacromial injection with a local anesthetic and testing for weakness of the supraspinatus tendon.

Conservative management is recommended by the authors for patients with any rotator cuff tear, except for acute tears in younger patients.⁵

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Medical Department of the US Navy or the US Naval Service at large.

Evidence summary

The largest meta-analysis to look at the effectiveness of clinical examination and patient history for diagnosing soft-tissue shoulder disorders evaluated 10 cohort studies.¹ Pooled results from 4 of the studies suggested that a composite clinical examination could safely rule out a full-thickness rotator cuff tear with a sensitivity of 0.9 (95% confidence interval [CI], 0.87-0.93). The standard in the studies cited was either arthrography or surgery. No single physical examination finding, when looked at in isolation, could reliably produce comparable accuracy.

The best 3 tests for full-thickness rotator cuff tear
A retrospective cohort study stratified 352 patients who had undergone operative evaluation by the degree of severity of rotator cuff pathology.² Investigators assessed 8 physical exam tests—the Neer impingement sign, Hawkins-Kennedy impingement sign, painful arc sign, supraspinatus muscle strength test, Speed’s test, cross-body adduction test, drop-arm sign, and infraspinatus muscle strength test—to determine their diagnostic utility. (See “A glossary of tests for rotator cuff injury”)

The combination of the painful arc sign, drop-arm sign, and weakness in external rotation (positive infraspinatus muscle strength test) produced the best likelihood ratios (LRs) for detecting full-thickness rotator cuff tears ( TABLE ).

TABLE
Comparison of tests to detect full-thickness rotator cuff tear

A winning diagnostic combination for any degree of impingement disease
The best combination of tests to diagnose any degree of impingement disease is:²

• a positive Hawkins-Kennedy impingement sign
• a positive painful arc sign
• a positive infraspinatus muscle strength test.

Negative supraspinatus test helps rule out massive tear
A third study evaluated the validity of the supraspinatus muscle strength test alone to diagnose patients with rotator cuff pathology using arthroscopy or open surgery as the reference standard.³ A negative supraspinatus test, when defined by weakness, significantly decreased the posttest probability (LR-=0.17) of detecting a massive rotator cuff tear.

Correlate test results with clinical history
Most individual tests for rotator cuff disease are not sensitive or specific enough to effectively rule in or rule out a rotator cuff tear. Many of the findings studied can be positive in the presence of other shoulder conditions and should be correlated with the clinical history of each patient. One of the noted limitations of these 3 studies is that none was carried out in a primary care setting.

Recommendations

A textbook published by The American Academy of Orthopedic Surgeons describes a continuum of injuries from impingement syndrome to full-thickness rotator cuff tears that commonly present as anterior and lateral shoulder pain and are often associated with night pain.⁴ Atrophy of the supraspinatus and infraspinatus muscles may indicate a longstanding rotator cuff tear. The authors’ recommendations for clinical examination include performing the Neer and Hawkins-Kennedy tests before and after subacromial injection with a local anesthetic and testing for weakness of the supraspinatus tendon.

Conservative management is recommended by the authors for patients with any rotator cuff tear, except for acute tears in younger patients.⁵

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Medical Department of the US Navy or the US Naval Service at large.

Evidence summary

The largest meta-analysis to look at the effectiveness of clinical examination and patient history for diagnosing soft-tissue shoulder disorders evaluated 10 cohort studies.¹ Pooled results from 4 of the studies suggested that a composite clinical examination could safely rule out a full-thickness rotator cuff tear with a sensitivity of 0.9 (95% confidence interval [CI], 0.87-0.93). The standard in the studies cited was either arthrography or surgery. No single physical examination finding, when looked at in isolation, could reliably produce comparable accuracy.

The best 3 tests for full-thickness rotator cuff tear
A retrospective cohort study stratified 352 patients who had undergone operative evaluation by the degree of severity of rotator cuff pathology.² Investigators assessed 8 physical exam tests—the Neer impingement sign, Hawkins-Kennedy impingement sign, painful arc sign, supraspinatus muscle strength test, Speed’s test, cross-body adduction test, drop-arm sign, and infraspinatus muscle strength test—to determine their diagnostic utility. (See “A glossary of tests for rotator cuff injury”)

The combination of the painful arc sign, drop-arm sign, and weakness in external rotation (positive infraspinatus muscle strength test) produced the best likelihood ratios (LRs) for detecting full-thickness rotator cuff tears ( TABLE ).

TABLE
Comparison of tests to detect full-thickness rotator cuff tear

A winning diagnostic combination for any degree of impingement disease
The best combination of tests to diagnose any degree of impingement disease is:²

• a positive Hawkins-Kennedy impingement sign
• a positive painful arc sign
• a positive infraspinatus muscle strength test.

Negative supraspinatus test helps rule out massive tear
A third study evaluated the validity of the supraspinatus muscle strength test alone to diagnose patients with rotator cuff pathology using arthroscopy or open surgery as the reference standard.³ A negative supraspinatus test, when defined by weakness, significantly decreased the posttest probability (LR-=0.17) of detecting a massive rotator cuff tear.

Correlate test results with clinical history
Most individual tests for rotator cuff disease are not sensitive or specific enough to effectively rule in or rule out a rotator cuff tear. Many of the findings studied can be positive in the presence of other shoulder conditions and should be correlated with the clinical history of each patient. One of the noted limitations of these 3 studies is that none was carried out in a primary care setting.

Recommendations

A textbook published by The American Academy of Orthopedic Surgeons describes a continuum of injuries from impingement syndrome to full-thickness rotator cuff tears that commonly present as anterior and lateral shoulder pain and are often associated with night pain.⁴ Atrophy of the supraspinatus and infraspinatus muscles may indicate a longstanding rotator cuff tear. The authors’ recommendations for clinical examination include performing the Neer and Hawkins-Kennedy tests before and after subacromial injection with a local anesthetic and testing for weakness of the supraspinatus tendon.

Conservative management is recommended by the authors for patients with any rotator cuff tear, except for acute tears in younger patients.⁵

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Medical Department of the US Navy or the US Naval Service at large.

Evidence summary

Prospective studies and systematic reviews show increased risk for CRC in people with a significant family history. Little or no data are available regarding outcome improvements or head-to-head comparisons of the effects of different screening methods. Recommendations for screening rest largely on inference and consensus opinions.

Family history=higher risk at lower age
The US Preventive Services Task Force (USPSTF) recommends CRC screening starting at 50 years of age for patients with average risk, based on a 5.6% lifetime risk of developing CRC and good evidence that screening reduces morbidity and mortality.¹ Patients with a family history of CRC or AP have a risk of CRC at 40 years of age that approximates average risk at 50 years.² Right-sided colonic lesions are also more likely in patients with a family history of CRC (relative risk [RR]=2.25; 95% confidence interval [CI], 1.96-2.59).³

Risk increases with number of affected first-degree relatives
Moreover, systematic reviews show the RR of CRC to be 1.99 (95% CI, 1.55-2.55) in patients with a single first-degree relative with AP, 2.25 (95% CI, 2.00-2.53) with a single first-degree relative with CRC, and 4.25 (95% CI, 3.01-6.02) with 2 or more first-degree relatives with CRC.³

Younger age at diagnosis also increases risk
The effect of the relative’s age at diagnosis of CRC is demonstrated by an RR of 3.87 (95% CI, 2.40-6.22) if diagnosed at younger than 45 years, 2.25 (95% CI, 1.85-2.72) if diagnosed at 45 to 59 years, and 1.82 (95% CI, 1.47-2.25) if diagnosed at 60 years or older.³

Recommendations

Colonoscopy is the preferred screening option for most patients with family histories that put them at increased risk of CRC and right-sided colonic lesions.^4,5 The American Cancer Society (ACS) and the American Gastroenterological Association (AGA) recommend that patients with a first-degree relative diagnosed before the age of 60 years or 2 or more first-degree relatives with CRC are at highest risk and should undergo colonoscopy at age 40, or 10 years before the earliest relative’s age at diagnosis; colonoscopy should be repeated every 5 years.^4,5

Patients with a first-degree relative diagnosed with CRC or AP at 60 years or older or multiple second-degree relatives with CRC have an increased risk, but lower than the high-risk group.⁴ Such patients may start screening early, at 40 years, but using the same options as patients at average risk (see the Evidence-Based Answer).⁴ The TABLE summarizes these screening recommendations. Notably, the ACS recommends no screening change (from patients with average risk) for patients with CRC in second-degree relatives because of the modest increase in risk.⁵

TABLE
ACS and AGA guidelines for screening patients with a family history of colorectal cancer

The American Society for Gastrointestinal Endoscopy (ASGE) recommends screening colonoscopy for patients with a first-degree relative who was older than 60 when diagnosed with adenomas but notes that the timing of initial colonoscopy hasn’t been established and should be individualized. The interval for follow-up colonoscopy in these patients should be the same as for average-risk patients. Patients with a second- or third-degree relative with colonic neoplasia should adhere to average-risk screening recommendations.⁶ Otherwise, the ASGE recommendations agree with the ones described previously.

The most recent joint guidelines of the US Multisociety Task Force (USMSTF) on Colorectal Cancer, the American College of Radiology, and the ACS, released in 2008, make no recommendations regarding patients with a family history.⁷ The USMSTF defers to guidelines from the ACS and the AGA described earlier.

Acknowledgements
The opinions and assertions contained herein are the private views of the authors and not to be construed as official, or as reflecting the views of the US Air Force Medical Service or the US Air Force at large.

Evidence summary

Prospective studies and systematic reviews show increased risk for CRC in people with a significant family history. Little or no data are available regarding outcome improvements or head-to-head comparisons of the effects of different screening methods. Recommendations for screening rest largely on inference and consensus opinions.

Family history=higher risk at lower age
The US Preventive Services Task Force (USPSTF) recommends CRC screening starting at 50 years of age for patients with average risk, based on a 5.6% lifetime risk of developing CRC and good evidence that screening reduces morbidity and mortality.¹ Patients with a family history of CRC or AP have a risk of CRC at 40 years of age that approximates average risk at 50 years.² Right-sided colonic lesions are also more likely in patients with a family history of CRC (relative risk [RR]=2.25; 95% confidence interval [CI], 1.96-2.59).³

Risk increases with number of affected first-degree relatives
Moreover, systematic reviews show the RR of CRC to be 1.99 (95% CI, 1.55-2.55) in patients with a single first-degree relative with AP, 2.25 (95% CI, 2.00-2.53) with a single first-degree relative with CRC, and 4.25 (95% CI, 3.01-6.02) with 2 or more first-degree relatives with CRC.³

Younger age at diagnosis also increases risk
The effect of the relative’s age at diagnosis of CRC is demonstrated by an RR of 3.87 (95% CI, 2.40-6.22) if diagnosed at younger than 45 years, 2.25 (95% CI, 1.85-2.72) if diagnosed at 45 to 59 years, and 1.82 (95% CI, 1.47-2.25) if diagnosed at 60 years or older.³

Recommendations

Colonoscopy is the preferred screening option for most patients with family histories that put them at increased risk of CRC and right-sided colonic lesions.^4,5 The American Cancer Society (ACS) and the American Gastroenterological Association (AGA) recommend that patients with a first-degree relative diagnosed before the age of 60 years or 2 or more first-degree relatives with CRC are at highest risk and should undergo colonoscopy at age 40, or 10 years before the earliest relative’s age at diagnosis; colonoscopy should be repeated every 5 years.^4,5

Patients with a first-degree relative diagnosed with CRC or AP at 60 years or older or multiple second-degree relatives with CRC have an increased risk, but lower than the high-risk group.⁴ Such patients may start screening early, at 40 years, but using the same options as patients at average risk (see the Evidence-Based Answer).⁴ The TABLE summarizes these screening recommendations. Notably, the ACS recommends no screening change (from patients with average risk) for patients with CRC in second-degree relatives because of the modest increase in risk.⁵

TABLE
ACS and AGA guidelines for screening patients with a family history of colorectal cancer

The American Society for Gastrointestinal Endoscopy (ASGE) recommends screening colonoscopy for patients with a first-degree relative who was older than 60 when diagnosed with adenomas but notes that the timing of initial colonoscopy hasn’t been established and should be individualized. The interval for follow-up colonoscopy in these patients should be the same as for average-risk patients. Patients with a second- or third-degree relative with colonic neoplasia should adhere to average-risk screening recommendations.⁶ Otherwise, the ASGE recommendations agree with the ones described previously.

The most recent joint guidelines of the US Multisociety Task Force (USMSTF) on Colorectal Cancer, the American College of Radiology, and the ACS, released in 2008, make no recommendations regarding patients with a family history.⁷ The USMSTF defers to guidelines from the ACS and the AGA described earlier.

Acknowledgements
The opinions and assertions contained herein are the private views of the authors and not to be construed as official, or as reflecting the views of the US Air Force Medical Service or the US Air Force at large.

Evidence summary

Prospective studies and systematic reviews show increased risk for CRC in people with a significant family history. Little or no data are available regarding outcome improvements or head-to-head comparisons of the effects of different screening methods. Recommendations for screening rest largely on inference and consensus opinions.

Family history=higher risk at lower age
The US Preventive Services Task Force (USPSTF) recommends CRC screening starting at 50 years of age for patients with average risk, based on a 5.6% lifetime risk of developing CRC and good evidence that screening reduces morbidity and mortality.¹ Patients with a family history of CRC or AP have a risk of CRC at 40 years of age that approximates average risk at 50 years.² Right-sided colonic lesions are also more likely in patients with a family history of CRC (relative risk [RR]=2.25; 95% confidence interval [CI], 1.96-2.59).³

Risk increases with number of affected first-degree relatives
Moreover, systematic reviews show the RR of CRC to be 1.99 (95% CI, 1.55-2.55) in patients with a single first-degree relative with AP, 2.25 (95% CI, 2.00-2.53) with a single first-degree relative with CRC, and 4.25 (95% CI, 3.01-6.02) with 2 or more first-degree relatives with CRC.³

Younger age at diagnosis also increases risk
The effect of the relative’s age at diagnosis of CRC is demonstrated by an RR of 3.87 (95% CI, 2.40-6.22) if diagnosed at younger than 45 years, 2.25 (95% CI, 1.85-2.72) if diagnosed at 45 to 59 years, and 1.82 (95% CI, 1.47-2.25) if diagnosed at 60 years or older.³

Recommendations

Colonoscopy is the preferred screening option for most patients with family histories that put them at increased risk of CRC and right-sided colonic lesions.^4,5 The American Cancer Society (ACS) and the American Gastroenterological Association (AGA) recommend that patients with a first-degree relative diagnosed before the age of 60 years or 2 or more first-degree relatives with CRC are at highest risk and should undergo colonoscopy at age 40, or 10 years before the earliest relative’s age at diagnosis; colonoscopy should be repeated every 5 years.^4,5

Patients with a first-degree relative diagnosed with CRC or AP at 60 years or older or multiple second-degree relatives with CRC have an increased risk, but lower than the high-risk group.⁴ Such patients may start screening early, at 40 years, but using the same options as patients at average risk (see the Evidence-Based Answer).⁴ The TABLE summarizes these screening recommendations. Notably, the ACS recommends no screening change (from patients with average risk) for patients with CRC in second-degree relatives because of the modest increase in risk.⁵

TABLE
ACS and AGA guidelines for screening patients with a family history of colorectal cancer

The American Society for Gastrointestinal Endoscopy (ASGE) recommends screening colonoscopy for patients with a first-degree relative who was older than 60 when diagnosed with adenomas but notes that the timing of initial colonoscopy hasn’t been established and should be individualized. The interval for follow-up colonoscopy in these patients should be the same as for average-risk patients. Patients with a second- or third-degree relative with colonic neoplasia should adhere to average-risk screening recommendations.⁶ Otherwise, the ASGE recommendations agree with the ones described previously.

The most recent joint guidelines of the US Multisociety Task Force (USMSTF) on Colorectal Cancer, the American College of Radiology, and the ACS, released in 2008, make no recommendations regarding patients with a family history.⁷ The USMSTF defers to guidelines from the ACS and the AGA described earlier.

Acknowledgements
The opinions and assertions contained herein are the private views of the authors and not to be construed as official, or as reflecting the views of the US Air Force Medical Service or the US Air Force at large.

Evidence summary

A recent study randomized 158 patients who didn’t respond to antidepressant therapy to either cognitive therapy with clinical management or clinical management alone.¹ The cognitive therapy group had a 29% cumulative relapse rate at 68 weeks, compared with 47% in the clinical management control group (number needed to treat [NNT]=6).

A crossover RCT compared 12 weeks of the cognitive behavioral analysis system of psychotherapy (CBASP) in 61 patients who had failed to respond to a 12-week course of nefazodone with 12 weeks of nefazodone treatment in 79 patients who hadn’t responded to 12 weeks of CBASP.² Remission rates were comparable in the 2 crossover groups (28% for nefazodone vs 25% for CBASP; P=.92).

Drugs may produce a faster response
The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial compared augmentation with as many as 16 sessions of cognitive therapy with pharmacologic augmentation and switch strategy among 65 patients who had failed to respond to 14 weeks of citalopram.³

The investigators concluded that augmentation with cognitive therapy or pharmacologic therapy was equally effective, but pharmacologic augmentation produced a more rapid response (mean time to first remission for cognitive therapy=53.3 days, compared with 40.1 days for pharmacologic therapy; P=.022). Patients who were switched to cognitive therapy had similar outcomes to patients who were switched to alternative antidepressants (remission rates=25% and 27.9%, respectively; P=.6881), but reported fewer adverse effects (0% vs 48%).

When an SSRI fails…
A recent systematic review of 8 RCTs (including STAR*D) and 23 open studies concluded that after a first failure of a selective serotonin reuptake inhibitor (SSRI), any switch within or between classes of antidepressant is legitimate and equally effective.⁴

Switching within the same class of antidepressant. The STAR*D study, an unblinded RCT, reported that patients (N=238; median age 41 years) who were switched to sertraline (as much as 200 mg per day for 14 weeks) when they didn’t tolerate or respond adequately to citalopram had remission rates of 17.6% on the Hamilton Rating Scale for Depression (HAM-D) and 26.6% on the Quick Inventory of Depressive Symptomatology (QIDS).⁵

Switching to a different class of antidepressant. In a multisite study, outpatients who failed to respond to 12-week, double-blind treatment with either sertraline (n=117) or imipramine (n=51) were randomized to an additional 12 weeks of double-blinded treatment with the alternate medication. Investigators reported a 60% response rate in the sertraline switch group and a 44% response rate in the imipramine switch group.⁶

In the STAR*D study, patients who didn’t tolerate or failed to respond to as many as 12 weeks of citalopram were switched to sustained-release (SR) bupropion, sertraline, or extended-release (ER) venlafaxine for as long as 14 weeks.⁵ The bupropion-SR switch group (n=239, up to 400 mg per day) had remission rates of 21.3% (HAM-D) and 25.5% (QIDS); the sertraline switch group (n=238, up to 200 mg per day) had remission rates of 17.6% (HAM-D) and 26.6% (QIDS); and the venlafaxine-ER switch group (n=250, up to 375 mg per day) had remission rates of 24.8% (HAM-D) and 25% (QIDS). There were no clinically or statistically significant differences among the groups.

Response declines with multiple switches
Patients who didn’t respond to this treatment arm and were switched again to either mirtazapine (n=114, as much as 60 mg per day) or nortriptyline (n=121, as much as 200 mg per day) had a much less favorable response (mirtazapine 12.3% vs nortriptyline 19.8%; NNT nortriptyline-mirtazapine=13).⁷

Patients who failed to respond to this treatment arm were randomized to either tranylcypromine (n=58, mean 36.9 mg per day) or venlafaxine plus mirtazapine (n=51, mean 210.3 and 35.7 mg per day, respectively). Both groups had low remission rates (tranylcypromine 6.9%, venlafaxine plus mirtazapine 13.7%; NNT venlafaxine plus mirtazapine-tranylcypromine=15).⁸

Lithium and T3 augmentation both work
A 1999 systematic review of 9 double-blind RCTs (N=234) reported that patients treated with lithium augmentation (250-1200 mg per day, or a serum level of ≥0.5 mmol/L for ≥2 weeks) had a 45% improvement in depressive symptoms (HAM-D), whereas the placebo group showed 18% improvement (NNT=3.7; 95% confidence interval [CI], 2.6-6.6).⁹ An updated meta-analysis of 10 RCTs confirmed the efficacy of lithium augmentation compared with placebo (41% vs 14.4% improvement; NNT=5).¹⁰

Recently, the STAR*D study (N=142) reported that augmentation with either lithium or triiodothyronine (T₃) after 2 antidepressant failures was equally effective (lithium response 15.9%; T₃ response 24.7%; NNT T₃-lithium=11; P=.43). However, lithium was more often associated with side effects (number needed to harm [NNH]=7; P=.045).¹¹

Bupropion and buspirone augmentation are comparable
An unblinded RCT found that patients who failed to respond to citalopram responded when augmented with either bupropion-SR or buspirone.¹² After 8 weeks of treatment, the bupropion-SR group (n=565, as much as 400 mg per day) had remission rates of 29.7% (HAM-D) and 39.9% (QIDS); the buspirone group (n=286, as much as 60 mg per day) had remission rates of 30.1% (HAM-D) and 26.9% (QIDS) (NNT buspirone-bupropion-SR=10). However, the bupropion-SR group had a lower dropout rate because of intolerance (12.5% vs 20.6%; NNH=12; P<.009).

Augmentation with atypical antipsychotics works
A recent meta-analysis of 10 RCTs (N=1500 outpatients) assessed the effectiveness of augmenting various antidepressants with atypical antipsychotic agents (olanzapine, risperidone, and quetiapine) for treatment-resistant major depressive disorder.¹³ The pooled remission and response rates favored augmentation with atypical antipsychotics over adjunctive placebo (47% vs 22.3% and 67.2% vs 35.4%, respectively).

Another randomized study of 362 patients with incomplete response to standard antidepressant treatment found adjunctive aripiprazole was effective and well tolerated (mean change in Montgomery-Åsberg Depression Rating Scale score: –8.8 in the aripiprazole group vs –5.8 in the placebo group; P<.001).¹⁴

Agents that aren’t recommended
Expert review doesn’t recommend routine use of other agents that have been studied for augmentation therapy, including dopaminergic drugs, pyschostimulants, modafinil, anticonvulsants, inositol, opiates, estrogen, dehydroepiandrosterone, folate and S-adenosylmethionine, tryptophan, omega-3 fatty acid, pindolol, and monoamine oxidase inhibitors.¹⁵

Recommendations

The Institute for Clinical Systems Improvement¹⁶ and the American Psychiatric Association¹⁷ recommend evaluating the dose and duration of medication, the patient’s adherence to medication, and the accuracy of diagnosis or impact of comorbidities for patients who don’t respond adequately to treatment. Physicians also may consider other strategies, including switch therapy, augmentation therapies, psychotherapy, and electroconvulsive therapy.

Evidence summary

A recent study randomized 158 patients who didn’t respond to antidepressant therapy to either cognitive therapy with clinical management or clinical management alone.¹ The cognitive therapy group had a 29% cumulative relapse rate at 68 weeks, compared with 47% in the clinical management control group (number needed to treat [NNT]=6).

A crossover RCT compared 12 weeks of the cognitive behavioral analysis system of psychotherapy (CBASP) in 61 patients who had failed to respond to a 12-week course of nefazodone with 12 weeks of nefazodone treatment in 79 patients who hadn’t responded to 12 weeks of CBASP.² Remission rates were comparable in the 2 crossover groups (28% for nefazodone vs 25% for CBASP; P=.92).

Drugs may produce a faster response
The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial compared augmentation with as many as 16 sessions of cognitive therapy with pharmacologic augmentation and switch strategy among 65 patients who had failed to respond to 14 weeks of citalopram.³

The investigators concluded that augmentation with cognitive therapy or pharmacologic therapy was equally effective, but pharmacologic augmentation produced a more rapid response (mean time to first remission for cognitive therapy=53.3 days, compared with 40.1 days for pharmacologic therapy; P=.022). Patients who were switched to cognitive therapy had similar outcomes to patients who were switched to alternative antidepressants (remission rates=25% and 27.9%, respectively; P=.6881), but reported fewer adverse effects (0% vs 48%).

When an SSRI fails…
A recent systematic review of 8 RCTs (including STAR*D) and 23 open studies concluded that after a first failure of a selective serotonin reuptake inhibitor (SSRI), any switch within or between classes of antidepressant is legitimate and equally effective.⁴

Switching within the same class of antidepressant. The STAR*D study, an unblinded RCT, reported that patients (N=238; median age 41 years) who were switched to sertraline (as much as 200 mg per day for 14 weeks) when they didn’t tolerate or respond adequately to citalopram had remission rates of 17.6% on the Hamilton Rating Scale for Depression (HAM-D) and 26.6% on the Quick Inventory of Depressive Symptomatology (QIDS).⁵

Switching to a different class of antidepressant. In a multisite study, outpatients who failed to respond to 12-week, double-blind treatment with either sertraline (n=117) or imipramine (n=51) were randomized to an additional 12 weeks of double-blinded treatment with the alternate medication. Investigators reported a 60% response rate in the sertraline switch group and a 44% response rate in the imipramine switch group.⁶

In the STAR*D study, patients who didn’t tolerate or failed to respond to as many as 12 weeks of citalopram were switched to sustained-release (SR) bupropion, sertraline, or extended-release (ER) venlafaxine for as long as 14 weeks.⁵ The bupropion-SR switch group (n=239, up to 400 mg per day) had remission rates of 21.3% (HAM-D) and 25.5% (QIDS); the sertraline switch group (n=238, up to 200 mg per day) had remission rates of 17.6% (HAM-D) and 26.6% (QIDS); and the venlafaxine-ER switch group (n=250, up to 375 mg per day) had remission rates of 24.8% (HAM-D) and 25% (QIDS). There were no clinically or statistically significant differences among the groups.

Response declines with multiple switches
Patients who didn’t respond to this treatment arm and were switched again to either mirtazapine (n=114, as much as 60 mg per day) or nortriptyline (n=121, as much as 200 mg per day) had a much less favorable response (mirtazapine 12.3% vs nortriptyline 19.8%; NNT nortriptyline-mirtazapine=13).⁷

Patients who failed to respond to this treatment arm were randomized to either tranylcypromine (n=58, mean 36.9 mg per day) or venlafaxine plus mirtazapine (n=51, mean 210.3 and 35.7 mg per day, respectively). Both groups had low remission rates (tranylcypromine 6.9%, venlafaxine plus mirtazapine 13.7%; NNT venlafaxine plus mirtazapine-tranylcypromine=15).⁸

Lithium and T3 augmentation both work
A 1999 systematic review of 9 double-blind RCTs (N=234) reported that patients treated with lithium augmentation (250-1200 mg per day, or a serum level of ≥0.5 mmol/L for ≥2 weeks) had a 45% improvement in depressive symptoms (HAM-D), whereas the placebo group showed 18% improvement (NNT=3.7; 95% confidence interval [CI], 2.6-6.6).⁹ An updated meta-analysis of 10 RCTs confirmed the efficacy of lithium augmentation compared with placebo (41% vs 14.4% improvement; NNT=5).¹⁰

Recently, the STAR*D study (N=142) reported that augmentation with either lithium or triiodothyronine (T₃) after 2 antidepressant failures was equally effective (lithium response 15.9%; T₃ response 24.7%; NNT T₃-lithium=11; P=.43). However, lithium was more often associated with side effects (number needed to harm [NNH]=7; P=.045).¹¹

Bupropion and buspirone augmentation are comparable
An unblinded RCT found that patients who failed to respond to citalopram responded when augmented with either bupropion-SR or buspirone.¹² After 8 weeks of treatment, the bupropion-SR group (n=565, as much as 400 mg per day) had remission rates of 29.7% (HAM-D) and 39.9% (QIDS); the buspirone group (n=286, as much as 60 mg per day) had remission rates of 30.1% (HAM-D) and 26.9% (QIDS) (NNT buspirone-bupropion-SR=10). However, the bupropion-SR group had a lower dropout rate because of intolerance (12.5% vs 20.6%; NNH=12; P<.009).

Augmentation with atypical antipsychotics works
A recent meta-analysis of 10 RCTs (N=1500 outpatients) assessed the effectiveness of augmenting various antidepressants with atypical antipsychotic agents (olanzapine, risperidone, and quetiapine) for treatment-resistant major depressive disorder.¹³ The pooled remission and response rates favored augmentation with atypical antipsychotics over adjunctive placebo (47% vs 22.3% and 67.2% vs 35.4%, respectively).

Another randomized study of 362 patients with incomplete response to standard antidepressant treatment found adjunctive aripiprazole was effective and well tolerated (mean change in Montgomery-Åsberg Depression Rating Scale score: –8.8 in the aripiprazole group vs –5.8 in the placebo group; P<.001).¹⁴

Agents that aren’t recommended
Expert review doesn’t recommend routine use of other agents that have been studied for augmentation therapy, including dopaminergic drugs, pyschostimulants, modafinil, anticonvulsants, inositol, opiates, estrogen, dehydroepiandrosterone, folate and S-adenosylmethionine, tryptophan, omega-3 fatty acid, pindolol, and monoamine oxidase inhibitors.¹⁵

Recommendations

The Institute for Clinical Systems Improvement¹⁶ and the American Psychiatric Association¹⁷ recommend evaluating the dose and duration of medication, the patient’s adherence to medication, and the accuracy of diagnosis or impact of comorbidities for patients who don’t respond adequately to treatment. Physicians also may consider other strategies, including switch therapy, augmentation therapies, psychotherapy, and electroconvulsive therapy.

Evidence summary

A recent study randomized 158 patients who didn’t respond to antidepressant therapy to either cognitive therapy with clinical management or clinical management alone.¹ The cognitive therapy group had a 29% cumulative relapse rate at 68 weeks, compared with 47% in the clinical management control group (number needed to treat [NNT]=6).

A crossover RCT compared 12 weeks of the cognitive behavioral analysis system of psychotherapy (CBASP) in 61 patients who had failed to respond to a 12-week course of nefazodone with 12 weeks of nefazodone treatment in 79 patients who hadn’t responded to 12 weeks of CBASP.² Remission rates were comparable in the 2 crossover groups (28% for nefazodone vs 25% for CBASP; P=.92).

Drugs may produce a faster response
The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial compared augmentation with as many as 16 sessions of cognitive therapy with pharmacologic augmentation and switch strategy among 65 patients who had failed to respond to 14 weeks of citalopram.³

The investigators concluded that augmentation with cognitive therapy or pharmacologic therapy was equally effective, but pharmacologic augmentation produced a more rapid response (mean time to first remission for cognitive therapy=53.3 days, compared with 40.1 days for pharmacologic therapy; P=.022). Patients who were switched to cognitive therapy had similar outcomes to patients who were switched to alternative antidepressants (remission rates=25% and 27.9%, respectively; P=.6881), but reported fewer adverse effects (0% vs 48%).

When an SSRI fails…
A recent systematic review of 8 RCTs (including STAR*D) and 23 open studies concluded that after a first failure of a selective serotonin reuptake inhibitor (SSRI), any switch within or between classes of antidepressant is legitimate and equally effective.⁴

Switching within the same class of antidepressant. The STAR*D study, an unblinded RCT, reported that patients (N=238; median age 41 years) who were switched to sertraline (as much as 200 mg per day for 14 weeks) when they didn’t tolerate or respond adequately to citalopram had remission rates of 17.6% on the Hamilton Rating Scale for Depression (HAM-D) and 26.6% on the Quick Inventory of Depressive Symptomatology (QIDS).⁵

Switching to a different class of antidepressant. In a multisite study, outpatients who failed to respond to 12-week, double-blind treatment with either sertraline (n=117) or imipramine (n=51) were randomized to an additional 12 weeks of double-blinded treatment with the alternate medication. Investigators reported a 60% response rate in the sertraline switch group and a 44% response rate in the imipramine switch group.⁶

In the STAR*D study, patients who didn’t tolerate or failed to respond to as many as 12 weeks of citalopram were switched to sustained-release (SR) bupropion, sertraline, or extended-release (ER) venlafaxine for as long as 14 weeks.⁵ The bupropion-SR switch group (n=239, up to 400 mg per day) had remission rates of 21.3% (HAM-D) and 25.5% (QIDS); the sertraline switch group (n=238, up to 200 mg per day) had remission rates of 17.6% (HAM-D) and 26.6% (QIDS); and the venlafaxine-ER switch group (n=250, up to 375 mg per day) had remission rates of 24.8% (HAM-D) and 25% (QIDS). There were no clinically or statistically significant differences among the groups.

Response declines with multiple switches
Patients who didn’t respond to this treatment arm and were switched again to either mirtazapine (n=114, as much as 60 mg per day) or nortriptyline (n=121, as much as 200 mg per day) had a much less favorable response (mirtazapine 12.3% vs nortriptyline 19.8%; NNT nortriptyline-mirtazapine=13).⁷

Patients who failed to respond to this treatment arm were randomized to either tranylcypromine (n=58, mean 36.9 mg per day) or venlafaxine plus mirtazapine (n=51, mean 210.3 and 35.7 mg per day, respectively). Both groups had low remission rates (tranylcypromine 6.9%, venlafaxine plus mirtazapine 13.7%; NNT venlafaxine plus mirtazapine-tranylcypromine=15).⁸

Lithium and T3 augmentation both work
A 1999 systematic review of 9 double-blind RCTs (N=234) reported that patients treated with lithium augmentation (250-1200 mg per day, or a serum level of ≥0.5 mmol/L for ≥2 weeks) had a 45% improvement in depressive symptoms (HAM-D), whereas the placebo group showed 18% improvement (NNT=3.7; 95% confidence interval [CI], 2.6-6.6).⁹ An updated meta-analysis of 10 RCTs confirmed the efficacy of lithium augmentation compared with placebo (41% vs 14.4% improvement; NNT=5).¹⁰

Recently, the STAR*D study (N=142) reported that augmentation with either lithium or triiodothyronine (T₃) after 2 antidepressant failures was equally effective (lithium response 15.9%; T₃ response 24.7%; NNT T₃-lithium=11; P=.43). However, lithium was more often associated with side effects (number needed to harm [NNH]=7; P=.045).¹¹

Bupropion and buspirone augmentation are comparable
An unblinded RCT found that patients who failed to respond to citalopram responded when augmented with either bupropion-SR or buspirone.¹² After 8 weeks of treatment, the bupropion-SR group (n=565, as much as 400 mg per day) had remission rates of 29.7% (HAM-D) and 39.9% (QIDS); the buspirone group (n=286, as much as 60 mg per day) had remission rates of 30.1% (HAM-D) and 26.9% (QIDS) (NNT buspirone-bupropion-SR=10). However, the bupropion-SR group had a lower dropout rate because of intolerance (12.5% vs 20.6%; NNH=12; P<.009).

Augmentation with atypical antipsychotics works
A recent meta-analysis of 10 RCTs (N=1500 outpatients) assessed the effectiveness of augmenting various antidepressants with atypical antipsychotic agents (olanzapine, risperidone, and quetiapine) for treatment-resistant major depressive disorder.¹³ The pooled remission and response rates favored augmentation with atypical antipsychotics over adjunctive placebo (47% vs 22.3% and 67.2% vs 35.4%, respectively).

Another randomized study of 362 patients with incomplete response to standard antidepressant treatment found adjunctive aripiprazole was effective and well tolerated (mean change in Montgomery-Åsberg Depression Rating Scale score: –8.8 in the aripiprazole group vs –5.8 in the placebo group; P<.001).¹⁴

Agents that aren’t recommended
Expert review doesn’t recommend routine use of other agents that have been studied for augmentation therapy, including dopaminergic drugs, pyschostimulants, modafinil, anticonvulsants, inositol, opiates, estrogen, dehydroepiandrosterone, folate and S-adenosylmethionine, tryptophan, omega-3 fatty acid, pindolol, and monoamine oxidase inhibitors.¹⁵

Recommendations

The Institute for Clinical Systems Improvement¹⁶ and the American Psychiatric Association¹⁷ recommend evaluating the dose and duration of medication, the patient’s adherence to medication, and the accuracy of diagnosis or impact of comorbidities for patients who don’t respond adequately to treatment. Physicians also may consider other strategies, including switch therapy, augmentation therapies, psychotherapy, and electroconvulsive therapy.

Evidence summary

Ligamentous knee injuries from trauma are common. In 2003, patients made about 19.4 million visits to the doctor because of knee problems.¹ The ACL is the most often injured knee ligament. The incidence of ACL injury is approximately 200,000 annually in the United States; 100,000 ACL reconstructions are performed each year.^2,3 A complete tear of the ACL can lead to significant knee instability and, unless repaired, may limit physical activity and quality of life.

In contrast, PCL injuries don’t often cause significant instability and generally respond to nonsurgical treatment; they have less impact on a patient’s quality of life. Surgery for PCL injury is usually reserved for elite athletes and unstable injuries. MCL and LCL injuries also are generally treated nonsurgically with rehabilitation and bracing; they normally don’t require arthroscopic evaluation and repair.

An effective alternative to arthroscopy
Arthroscopy with direct visualization of the ligamentous structures is considered the gold standard for diagnosing intra-articular ligamentous knee injuries, but it’s invasive and costly. Although clinical examination is helpful in identifying injured ligaments, it may lead to unnecessary arthroscopies when used alone because of the high false-positive rate. MRI has been shown to be an effective tool for accurately diagnosing ligamentous knee injury.^2,3

MRI offers high sensitivity for detecting ACL, PCL tears
Several prospective studies have compared MRI with arthroscopy for diagnosing ACL and PCL tears (TABLE).^4-8 All enrolled patients had sustained knee trauma and had had a clinical exam that suggested ligamentous injury. MRI and arthroscopy were performed regardless of MRI findings. The surgeons performing arthroscopy were blinded to the MRI results.

Although MRI equipment and techniques varied in all the studies, the sensitivity and specificity remained consistently high for detecting ACL injuries. Thin-slice views, special oblique views, and a fast spin-echo technique didn’t improve either the sensitivity or specificity compared with conventional techniques or views, nor did decreasing the time interval from injury to imaging.^5,9,10 Prospective studies of PCL injuries also revealed high sensitivity and specificity with MRI.

No data on MRI for MCL and LCL injuries
No prospective studies are available to assess the accuracy of MRI for suspected MCL and LCL injuries; however, MRI would likely not affect treatment or clinical outcomes, as both of these injuries are typically treated nonsurgically.

Recommendations

The American Academy of Orthopaedic Surgeons supports MRI as an effective tool for evaluating knee injury,¹ and offers recommendations and guidelines for treating ligamentous knee injury based on the findings of clinical examination and MRI. The Academy states that MRI is invaluable in preventing unnecessary surgery, and recommends it whenever ligamentous injury is suspected.

Acknowledgements 

The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the US Air Force medical Service or the US Air Force at large.

Evidence summary

Ligamentous knee injuries from trauma are common. In 2003, patients made about 19.4 million visits to the doctor because of knee problems.¹ The ACL is the most often injured knee ligament. The incidence of ACL injury is approximately 200,000 annually in the United States; 100,000 ACL reconstructions are performed each year.^2,3 A complete tear of the ACL can lead to significant knee instability and, unless repaired, may limit physical activity and quality of life.

In contrast, PCL injuries don’t often cause significant instability and generally respond to nonsurgical treatment; they have less impact on a patient’s quality of life. Surgery for PCL injury is usually reserved for elite athletes and unstable injuries. MCL and LCL injuries also are generally treated nonsurgically with rehabilitation and bracing; they normally don’t require arthroscopic evaluation and repair.

An effective alternative to arthroscopy
Arthroscopy with direct visualization of the ligamentous structures is considered the gold standard for diagnosing intra-articular ligamentous knee injuries, but it’s invasive and costly. Although clinical examination is helpful in identifying injured ligaments, it may lead to unnecessary arthroscopies when used alone because of the high false-positive rate. MRI has been shown to be an effective tool for accurately diagnosing ligamentous knee injury.^2,3

MRI offers high sensitivity for detecting ACL, PCL tears
Several prospective studies have compared MRI with arthroscopy for diagnosing ACL and PCL tears (TABLE).^4-8 All enrolled patients had sustained knee trauma and had had a clinical exam that suggested ligamentous injury. MRI and arthroscopy were performed regardless of MRI findings. The surgeons performing arthroscopy were blinded to the MRI results.

Although MRI equipment and techniques varied in all the studies, the sensitivity and specificity remained consistently high for detecting ACL injuries. Thin-slice views, special oblique views, and a fast spin-echo technique didn’t improve either the sensitivity or specificity compared with conventional techniques or views, nor did decreasing the time interval from injury to imaging.^5,9,10 Prospective studies of PCL injuries also revealed high sensitivity and specificity with MRI.

No data on MRI for MCL and LCL injuries
No prospective studies are available to assess the accuracy of MRI for suspected MCL and LCL injuries; however, MRI would likely not affect treatment or clinical outcomes, as both of these injuries are typically treated nonsurgically.

Recommendations

The American Academy of Orthopaedic Surgeons supports MRI as an effective tool for evaluating knee injury,¹ and offers recommendations and guidelines for treating ligamentous knee injury based on the findings of clinical examination and MRI. The Academy states that MRI is invaluable in preventing unnecessary surgery, and recommends it whenever ligamentous injury is suspected.

Acknowledgements 

The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the US Air Force medical Service or the US Air Force at large.

Evidence summary

Ligamentous knee injuries from trauma are common. In 2003, patients made about 19.4 million visits to the doctor because of knee problems.¹ The ACL is the most often injured knee ligament. The incidence of ACL injury is approximately 200,000 annually in the United States; 100,000 ACL reconstructions are performed each year.^2,3 A complete tear of the ACL can lead to significant knee instability and, unless repaired, may limit physical activity and quality of life.

In contrast, PCL injuries don’t often cause significant instability and generally respond to nonsurgical treatment; they have less impact on a patient’s quality of life. Surgery for PCL injury is usually reserved for elite athletes and unstable injuries. MCL and LCL injuries also are generally treated nonsurgically with rehabilitation and bracing; they normally don’t require arthroscopic evaluation and repair.

An effective alternative to arthroscopy
Arthroscopy with direct visualization of the ligamentous structures is considered the gold standard for diagnosing intra-articular ligamentous knee injuries, but it’s invasive and costly. Although clinical examination is helpful in identifying injured ligaments, it may lead to unnecessary arthroscopies when used alone because of the high false-positive rate. MRI has been shown to be an effective tool for accurately diagnosing ligamentous knee injury.^2,3

MRI offers high sensitivity for detecting ACL, PCL tears
Several prospective studies have compared MRI with arthroscopy for diagnosing ACL and PCL tears (TABLE).^4-8 All enrolled patients had sustained knee trauma and had had a clinical exam that suggested ligamentous injury. MRI and arthroscopy were performed regardless of MRI findings. The surgeons performing arthroscopy were blinded to the MRI results.

Although MRI equipment and techniques varied in all the studies, the sensitivity and specificity remained consistently high for detecting ACL injuries. Thin-slice views, special oblique views, and a fast spin-echo technique didn’t improve either the sensitivity or specificity compared with conventional techniques or views, nor did decreasing the time interval from injury to imaging.^5,9,10 Prospective studies of PCL injuries also revealed high sensitivity and specificity with MRI.

No data on MRI for MCL and LCL injuries
No prospective studies are available to assess the accuracy of MRI for suspected MCL and LCL injuries; however, MRI would likely not affect treatment or clinical outcomes, as both of these injuries are typically treated nonsurgically.

Recommendations

The American Academy of Orthopaedic Surgeons supports MRI as an effective tool for evaluating knee injury,¹ and offers recommendations and guidelines for treating ligamentous knee injury based on the findings of clinical examination and MRI. The Academy states that MRI is invaluable in preventing unnecessary surgery, and recommends it whenever ligamentous injury is suspected.

Acknowledgements 

The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the US Air Force medical Service or the US Air Force at large.

Evidence Summary

One RCT examined PE rates in 400 patients with acute proximal DVT who were randomized to receive or not receive a permanent IVCF and also randomized to receive either unfractionated heparin or low-molecular-weight heparin for at least the first 3 months.^1,2 Patients with a contraindication to anticoagulation or history of anticoagulation failure were excluded.

After 8 years of follow-up, symptomatic PE occurred less often in the filter group than the nonfilter group (6.2% vs 15.1%; P=.008; hazard ratio [HR]=0.36, 95% confidence interval [CI], 0.17-0.77; number needed to treat [NNT]=11.2). The filter group had a higher incidence of recurrent DVT than the nonfilter group (35.7% vs 27.5%; HR=1.52, 95% CI, 1.02- 2.27; number needed to harm=12.2).^1,2

The study lacked statistical power to draw any conclusion about the efficacy of IVCFs in preventing PE over shorter time periods or in reducing PE-related or overall mortality.³ Further research, including RCTs, needs to be done to determine how the efficacy of endovascular filters compares with standard PE prophylaxis.

How often does PE occur in patients with filters?
Patients with DVT generally have associated PE 10% of the time.⁴ Several cohort studies have examined the prevalence of recurrent PE in pa- tients with IVCFs, but none compared preva- lence in patients with and without filters.

A prospective cohort study followed 481 patients who received an IVCF because of ei- ther a contraindication to anticoagulation or sustained recurrent embolization despite ad- equate anticoagulation. Of the patients who had a filter for 6 months or longer, 2% had clinically suspected PE, but PE was confirmed in only 0.5%.⁵

Another multicenter, prospective cohort study (N=222) found radiographically con- firmed PE after filter placement in only 2% of patients with IVCFs after a mean follow-up of 15 months.⁶

A retrospective cohort study (N=318) concluded that 3.1% of the patients with IVCFs had a recurrent PE, diagnosed radiographically.⁷

A single-center retrospective cohort study of 1731 patients with IVCFs placed for various indications showed PE in 5.6% of patients. Some embolisms were clinically suspected and not confirmed.⁸

Complications of filter placement
Complications from IVCF placement generally occur less than 3% of the time. The most common complication is postthrombotic syndrome (70%). Risks associated with IVCF placement include DVT, postthrombotic syndrome, maldeployed filter, caval thrombosis, retroperitoneal hemorrhage, malposition, filter migration, arrhythmia, insertion site complications (such as infection or hematoma), PE, myocardial infarction, and death.^1,2,5-12

Recommendations

The American College of Chest Physicians recommends considering an IVCF for patients with DVT who have a contraindication to anticoagulation, complication of anticoagulation, or recurrent thromboembolism despite adequate anticoagulation.¹²

Evidence Summary

One RCT examined PE rates in 400 patients with acute proximal DVT who were randomized to receive or not receive a permanent IVCF and also randomized to receive either unfractionated heparin or low-molecular-weight heparin for at least the first 3 months.^1,2 Patients with a contraindication to anticoagulation or history of anticoagulation failure were excluded.

After 8 years of follow-up, symptomatic PE occurred less often in the filter group than the nonfilter group (6.2% vs 15.1%; P=.008; hazard ratio [HR]=0.36, 95% confidence interval [CI], 0.17-0.77; number needed to treat [NNT]=11.2). The filter group had a higher incidence of recurrent DVT than the nonfilter group (35.7% vs 27.5%; HR=1.52, 95% CI, 1.02- 2.27; number needed to harm=12.2).^1,2

The study lacked statistical power to draw any conclusion about the efficacy of IVCFs in preventing PE over shorter time periods or in reducing PE-related or overall mortality.³ Further research, including RCTs, needs to be done to determine how the efficacy of endovascular filters compares with standard PE prophylaxis.

How often does PE occur in patients with filters?
Patients with DVT generally have associated PE 10% of the time.⁴ Several cohort studies have examined the prevalence of recurrent PE in pa- tients with IVCFs, but none compared preva- lence in patients with and without filters.

A prospective cohort study followed 481 patients who received an IVCF because of ei- ther a contraindication to anticoagulation or sustained recurrent embolization despite ad- equate anticoagulation. Of the patients who had a filter for 6 months or longer, 2% had clinically suspected PE, but PE was confirmed in only 0.5%.⁵

Another multicenter, prospective cohort study (N=222) found radiographically con- firmed PE after filter placement in only 2% of patients with IVCFs after a mean follow-up of 15 months.⁶

A retrospective cohort study (N=318) concluded that 3.1% of the patients with IVCFs had a recurrent PE, diagnosed radiographically.⁷

A single-center retrospective cohort study of 1731 patients with IVCFs placed for various indications showed PE in 5.6% of patients. Some embolisms were clinically suspected and not confirmed.⁸

Complications of filter placement
Complications from IVCF placement generally occur less than 3% of the time. The most common complication is postthrombotic syndrome (70%). Risks associated with IVCF placement include DVT, postthrombotic syndrome, maldeployed filter, caval thrombosis, retroperitoneal hemorrhage, malposition, filter migration, arrhythmia, insertion site complications (such as infection or hematoma), PE, myocardial infarction, and death.^1,2,5-12

Recommendations

The American College of Chest Physicians recommends considering an IVCF for patients with DVT who have a contraindication to anticoagulation, complication of anticoagulation, or recurrent thromboembolism despite adequate anticoagulation.¹²

Evidence Summary

One RCT examined PE rates in 400 patients with acute proximal DVT who were randomized to receive or not receive a permanent IVCF and also randomized to receive either unfractionated heparin or low-molecular-weight heparin for at least the first 3 months.^1,2 Patients with a contraindication to anticoagulation or history of anticoagulation failure were excluded.

After 8 years of follow-up, symptomatic PE occurred less often in the filter group than the nonfilter group (6.2% vs 15.1%; P=.008; hazard ratio [HR]=0.36, 95% confidence interval [CI], 0.17-0.77; number needed to treat [NNT]=11.2). The filter group had a higher incidence of recurrent DVT than the nonfilter group (35.7% vs 27.5%; HR=1.52, 95% CI, 1.02- 2.27; number needed to harm=12.2).^1,2

The study lacked statistical power to draw any conclusion about the efficacy of IVCFs in preventing PE over shorter time periods or in reducing PE-related or overall mortality.³ Further research, including RCTs, needs to be done to determine how the efficacy of endovascular filters compares with standard PE prophylaxis.

How often does PE occur in patients with filters?
Patients with DVT generally have associated PE 10% of the time.⁴ Several cohort studies have examined the prevalence of recurrent PE in pa- tients with IVCFs, but none compared preva- lence in patients with and without filters.

A prospective cohort study followed 481 patients who received an IVCF because of ei- ther a contraindication to anticoagulation or sustained recurrent embolization despite ad- equate anticoagulation. Of the patients who had a filter for 6 months or longer, 2% had clinically suspected PE, but PE was confirmed in only 0.5%.⁵

Another multicenter, prospective cohort study (N=222) found radiographically con- firmed PE after filter placement in only 2% of patients with IVCFs after a mean follow-up of 15 months.⁶

A retrospective cohort study (N=318) concluded that 3.1% of the patients with IVCFs had a recurrent PE, diagnosed radiographically.⁷

A single-center retrospective cohort study of 1731 patients with IVCFs placed for various indications showed PE in 5.6% of patients. Some embolisms were clinically suspected and not confirmed.⁸

Complications of filter placement
Complications from IVCF placement generally occur less than 3% of the time. The most common complication is postthrombotic syndrome (70%). Risks associated with IVCF placement include DVT, postthrombotic syndrome, maldeployed filter, caval thrombosis, retroperitoneal hemorrhage, malposition, filter migration, arrhythmia, insertion site complications (such as infection or hematoma), PE, myocardial infarction, and death.^1,2,5-12

Recommendations

The American College of Chest Physicians recommends considering an IVCF for patients with DVT who have a contraindication to anticoagulation, complication of anticoagulation, or recurrent thromboembolism despite adequate anticoagulation.¹²

Evidence summary

Patients with acute DVT have traditionally been treated with immobilization and bed rest, combined with anticoagulation, for days. This approach is motivated by fear of dislodging an unstable thrombus and causing a pulmonary embolism (PE) and by the belief that inactivity relieves local pain and swelling. On the other hand, bed rest promotes stasis, an element in Virchow’s triad.

Early ambulation doesn’t raise risk of PE
We performed a structured literature review, which found 6 RCTs and 3 cohort studies that address this problem. All 6 RCTs included patients with acute DVT but without life-threatening conditions.^1-6 They assessed various outcomes, including incidence of new PE, change in leg circumference, leg pain, patient well-being, and progression of DVT.

The studies consistently found that early ambulation, along with compression, is safe when compared with bed rest ( TABLE ). Although the sample size of all the RCTs was small, the RCTs showed consistent trends in favor of ambulation and compression.

A prospective cohort study of new PE in patients treated with ambulation and compression plus anticoagulation found that the incidence of PE was significantly lower than historical incidence rates in patients managed with bed rest.⁷

Another study using the RIETE registry, a Spanish registry of consecutively enrolled patients with objectively confirmed acute DVT or PE, found no significant difference in occurrence of new PE between immobilized and mobilized patients.⁸ Patients with DVT who were immobilized were generally sicker, more likely to have PaO₂ <60, and more likely to have received lower doses of low-molecular-weight heparin (LMWH) compared with the group that walked (P<.005).

TABLE
Early ambulation and compression: What RCTs show

Does ambulation affect thrombus propagation?
A multicenter RCT showed that thrombus progression occurred more often in patients who were treated with bed rest compared with patients treated with ambulation and compression (P<.01).²

Another RCT revealed a similar trend, though the difference didn’t reach statistical significance because of small sample size.⁴ The clinical importance of these phlebographic studies isn’t clear.

Is it the walking, or compression, that works?
RCTs have shown that ambulation with leg compression, compared with bed rest without compression, can effectively decrease leg swelling and pain^1,2,4 The difference was detectable 2 years after DVT.⁷

In contrast, RCTs in which both ambulating and resting patients received compression therapy showed no significant difference in leg circumference at 1 or 6 months.³ This finding suggests that the benefit on local symptoms may result from compression rather than ambulation.

Reduced mortality? Evidence is weak
Estimates of the possible effect on mortality of ambulation compared with bed rest are based on cohort studies. A cohort study in which 691 patients were kept walking with compression therapy reported a mortality rate of 0.2%.⁹ In another cohort, the mortality rate was also 0.2%, and all deaths occurred in patients older than 70 years.¹⁰

This rate is lower than rates reported in the historic literature, which typically are 1% among patients treated with unfractionated heparin and bed rest.^9,10 A retrospective, multicenter cohort of 1647 patients treated with unfractionated heparin and bed rest in different German hospitals reported a rate of fatal PE of 2.33%.¹¹

Data from the RIETE registry indicated that overall mortality was significantly higher in immobilized patients with a PE (3.6% vs 0.5% in mobile patients; P=.01).⁸ Notably, immobilized patients with a PE were more likely to be hypoxic and also tended to receive lower doses of LMWH. No differences were found in outcomes for patients with DVT.

Recommendations

The American College of Chest Physicians (ACCP) doesn’t recommend bed rest in its guidelines for treating acute venous thromboembolism, but rather ambulation as tolerated after starting anticoagulation. Patients who are not hemodynamically stable should be stabilized first.

The ACCP also recommends wearing an elastic compression stocking with a pressure of 30 to 40 mm Hg at the ankle for 2 years after an episode of DVT and a course of intermittent pneumatic compression for patients with severe edema of the leg resulting from post-thrombotic syndrome.¹²

A joint guideline from the American College of Physicians and the American Academy of Family Physicians doesn’t make recommendations about ambulation for therapy of DVT and PE.¹³

Evidence summary

Patients with acute DVT have traditionally been treated with immobilization and bed rest, combined with anticoagulation, for days. This approach is motivated by fear of dislodging an unstable thrombus and causing a pulmonary embolism (PE) and by the belief that inactivity relieves local pain and swelling. On the other hand, bed rest promotes stasis, an element in Virchow’s triad.

Early ambulation doesn’t raise risk of PE
We performed a structured literature review, which found 6 RCTs and 3 cohort studies that address this problem. All 6 RCTs included patients with acute DVT but without life-threatening conditions.^1-6 They assessed various outcomes, including incidence of new PE, change in leg circumference, leg pain, patient well-being, and progression of DVT.

The studies consistently found that early ambulation, along with compression, is safe when compared with bed rest ( TABLE ). Although the sample size of all the RCTs was small, the RCTs showed consistent trends in favor of ambulation and compression.

A prospective cohort study of new PE in patients treated with ambulation and compression plus anticoagulation found that the incidence of PE was significantly lower than historical incidence rates in patients managed with bed rest.⁷

Another study using the RIETE registry, a Spanish registry of consecutively enrolled patients with objectively confirmed acute DVT or PE, found no significant difference in occurrence of new PE between immobilized and mobilized patients.⁸ Patients with DVT who were immobilized were generally sicker, more likely to have PaO₂ <60, and more likely to have received lower doses of low-molecular-weight heparin (LMWH) compared with the group that walked (P<.005).

TABLE
Early ambulation and compression: What RCTs show

Does ambulation affect thrombus propagation?
A multicenter RCT showed that thrombus progression occurred more often in patients who were treated with bed rest compared with patients treated with ambulation and compression (P<.01).²

Another RCT revealed a similar trend, though the difference didn’t reach statistical significance because of small sample size.⁴ The clinical importance of these phlebographic studies isn’t clear.

Is it the walking, or compression, that works?
RCTs have shown that ambulation with leg compression, compared with bed rest without compression, can effectively decrease leg swelling and pain^1,2,4 The difference was detectable 2 years after DVT.⁷

In contrast, RCTs in which both ambulating and resting patients received compression therapy showed no significant difference in leg circumference at 1 or 6 months.³ This finding suggests that the benefit on local symptoms may result from compression rather than ambulation.

Reduced mortality? Evidence is weak
Estimates of the possible effect on mortality of ambulation compared with bed rest are based on cohort studies. A cohort study in which 691 patients were kept walking with compression therapy reported a mortality rate of 0.2%.⁹ In another cohort, the mortality rate was also 0.2%, and all deaths occurred in patients older than 70 years.¹⁰

This rate is lower than rates reported in the historic literature, which typically are 1% among patients treated with unfractionated heparin and bed rest.^9,10 A retrospective, multicenter cohort of 1647 patients treated with unfractionated heparin and bed rest in different German hospitals reported a rate of fatal PE of 2.33%.¹¹

Data from the RIETE registry indicated that overall mortality was significantly higher in immobilized patients with a PE (3.6% vs 0.5% in mobile patients; P=.01).⁸ Notably, immobilized patients with a PE were more likely to be hypoxic and also tended to receive lower doses of LMWH. No differences were found in outcomes for patients with DVT.

Recommendations

The American College of Chest Physicians (ACCP) doesn’t recommend bed rest in its guidelines for treating acute venous thromboembolism, but rather ambulation as tolerated after starting anticoagulation. Patients who are not hemodynamically stable should be stabilized first.

The ACCP also recommends wearing an elastic compression stocking with a pressure of 30 to 40 mm Hg at the ankle for 2 years after an episode of DVT and a course of intermittent pneumatic compression for patients with severe edema of the leg resulting from post-thrombotic syndrome.¹²

A joint guideline from the American College of Physicians and the American Academy of Family Physicians doesn’t make recommendations about ambulation for therapy of DVT and PE.¹³

Evidence summary

Patients with acute DVT have traditionally been treated with immobilization and bed rest, combined with anticoagulation, for days. This approach is motivated by fear of dislodging an unstable thrombus and causing a pulmonary embolism (PE) and by the belief that inactivity relieves local pain and swelling. On the other hand, bed rest promotes stasis, an element in Virchow’s triad.

Early ambulation doesn’t raise risk of PE
We performed a structured literature review, which found 6 RCTs and 3 cohort studies that address this problem. All 6 RCTs included patients with acute DVT but without life-threatening conditions.^1-6 They assessed various outcomes, including incidence of new PE, change in leg circumference, leg pain, patient well-being, and progression of DVT.

The studies consistently found that early ambulation, along with compression, is safe when compared with bed rest ( TABLE ). Although the sample size of all the RCTs was small, the RCTs showed consistent trends in favor of ambulation and compression.

A prospective cohort study of new PE in patients treated with ambulation and compression plus anticoagulation found that the incidence of PE was significantly lower than historical incidence rates in patients managed with bed rest.⁷

Another study using the RIETE registry, a Spanish registry of consecutively enrolled patients with objectively confirmed acute DVT or PE, found no significant difference in occurrence of new PE between immobilized and mobilized patients.⁸ Patients with DVT who were immobilized were generally sicker, more likely to have PaO₂ <60, and more likely to have received lower doses of low-molecular-weight heparin (LMWH) compared with the group that walked (P<.005).

TABLE
Early ambulation and compression: What RCTs show

Does ambulation affect thrombus propagation?
A multicenter RCT showed that thrombus progression occurred more often in patients who were treated with bed rest compared with patients treated with ambulation and compression (P<.01).²

Another RCT revealed a similar trend, though the difference didn’t reach statistical significance because of small sample size.⁴ The clinical importance of these phlebographic studies isn’t clear.

Is it the walking, or compression, that works?
RCTs have shown that ambulation with leg compression, compared with bed rest without compression, can effectively decrease leg swelling and pain^1,2,4 The difference was detectable 2 years after DVT.⁷

In contrast, RCTs in which both ambulating and resting patients received compression therapy showed no significant difference in leg circumference at 1 or 6 months.³ This finding suggests that the benefit on local symptoms may result from compression rather than ambulation.

Reduced mortality? Evidence is weak
Estimates of the possible effect on mortality of ambulation compared with bed rest are based on cohort studies. A cohort study in which 691 patients were kept walking with compression therapy reported a mortality rate of 0.2%.⁹ In another cohort, the mortality rate was also 0.2%, and all deaths occurred in patients older than 70 years.¹⁰

This rate is lower than rates reported in the historic literature, which typically are 1% among patients treated with unfractionated heparin and bed rest.^9,10 A retrospective, multicenter cohort of 1647 patients treated with unfractionated heparin and bed rest in different German hospitals reported a rate of fatal PE of 2.33%.¹¹

Data from the RIETE registry indicated that overall mortality was significantly higher in immobilized patients with a PE (3.6% vs 0.5% in mobile patients; P=.01).⁸ Notably, immobilized patients with a PE were more likely to be hypoxic and also tended to receive lower doses of LMWH. No differences were found in outcomes for patients with DVT.

Recommendations

The American College of Chest Physicians (ACCP) doesn’t recommend bed rest in its guidelines for treating acute venous thromboembolism, but rather ambulation as tolerated after starting anticoagulation. Patients who are not hemodynamically stable should be stabilized first.

The ACCP also recommends wearing an elastic compression stocking with a pressure of 30 to 40 mm Hg at the ankle for 2 years after an episode of DVT and a course of intermittent pneumatic compression for patients with severe edema of the leg resulting from post-thrombotic syndrome.¹²

A joint guideline from the American College of Physicians and the American Academy of Family Physicians doesn’t make recommendations about ambulation for therapy of DVT and PE.¹³

Evidence summary

Treating partners of patients with sexually transmitted infection has been a core component of therapy since the 1940s. Traditionally, partners have been referred to a health care provider (by the index patient, the provider, or a public health officer) for evaluation before being treated. Current methods of partner referral reach only 40% to 60% of named sexual partners.¹

Expedited partner therapy vs traditional patient referral
Success of treatment is most readily measured by a reduction in the persistence or recurrence of infection in the index patient. Four RCTs and 1 observational cohort study have compared traditional patient referral with expedited partner treatment.^2-6 The primary outcome measure in all studies was reduction of persistent or recurrent infection in the index patient ( TABLE 1 ).

Chlamydia. Of the 4 studies that evaluated expedited partner treatment for chlamydia, 1 cohort study showed a statistically significant decrease in recurrent or persistent chlamydial infection in index patients.² One RCT showed a statistically significant reduction in recurrent or persistent urethritis, but didn’t report persistent and recurrent gonorrheal and chlamydial infections separately.³ Two RCTs showed a decrease in recurrent or persistent chlamydial infection in the index patient, but the difference didn’t reach statistical significance.^4,5

Gonorrhea. Two RCTs evaluated expedited partner treatment for gonorrhea compared with patient referral. One demonstrated a statistically significant decrease in persistent or recurrent gonococcal infection.⁵ The other showed a statistically significant decrease in recurrent or persistent urethritis, but without identifying recurrent gonorrheal and chlamydial infections separately.³

Trichomoniasis. One RCT compared expedited partner therapy with patient referral for patients with trichomoniasis. The study didn’t show a statistically significant difference in recurrent or persistent infection.

TABLE 1
Traditional patient referral vs expedited partner treatment: How the 2 compare

The verdict: Expedited partner therapy works better
A meta-analysis of the above studies evaluated the effect of expedited partner therapy compared with patient referral on the rate of recurrent or persistent gonorrhea, chlamydia, and trichomoniasis and the number of partners treated per index patient.¹ Empiric therapy was associated with a lower rate of recurrent or persistent infections (risk ratio [RR]=0.73; 95% confidence interval [CI], 0.57-0.93) and a higher number of partners treated per patient (RR=1.44; 95% CI, 1.12-1.86).

Take state law into account
Providers need to consider their state’s laws regarding empiric partner therapy. A state-by-state evaluation of the legal status of expedited partner therapy is available on the Centers for Disease Control and Prevention’s Web site, and is summarized in TABLE 2.⁷

TABLE 2
What’s the status of expedited partner therapy (EPT) in your state?⁷

Recommendations

A review of expedited partner therapy by the Centers for Disease Control and Prevention concluded: “The evidence indicates that expedited partner therapy should be available to clinicians as an option for partner management…[but it] does not replace other strategies such as standard patient referral or provider-assisted referral, when available…. Expedited partner therapy should be accompanied by information [advising] recipients to seek personal health care in addition to expedited partner therapy. Expedited partner therapy has a limited role in partner management for trichomoniasis. No data support its use in the routine management of syphilis, and there is no experience with expedited partner therapy for gonorrhea or chlamydial infection among men who have sex with men.”⁸

Neither the American Academy of Family Physicians nor the American College of Obstetricians and Gynecologists has issued a policy statement on expedited partner therapy.

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Medical Department of the United States Navy or the US Naval Service at large.

Evidence summary

Treating partners of patients with sexually transmitted infection has been a core component of therapy since the 1940s. Traditionally, partners have been referred to a health care provider (by the index patient, the provider, or a public health officer) for evaluation before being treated. Current methods of partner referral reach only 40% to 60% of named sexual partners.¹

Expedited partner therapy vs traditional patient referral
Success of treatment is most readily measured by a reduction in the persistence or recurrence of infection in the index patient. Four RCTs and 1 observational cohort study have compared traditional patient referral with expedited partner treatment.^2-6 The primary outcome measure in all studies was reduction of persistent or recurrent infection in the index patient ( TABLE 1 ).

Chlamydia. Of the 4 studies that evaluated expedited partner treatment for chlamydia, 1 cohort study showed a statistically significant decrease in recurrent or persistent chlamydial infection in index patients.² One RCT showed a statistically significant reduction in recurrent or persistent urethritis, but didn’t report persistent and recurrent gonorrheal and chlamydial infections separately.³ Two RCTs showed a decrease in recurrent or persistent chlamydial infection in the index patient, but the difference didn’t reach statistical significance.^4,5

Gonorrhea. Two RCTs evaluated expedited partner treatment for gonorrhea compared with patient referral. One demonstrated a statistically significant decrease in persistent or recurrent gonococcal infection.⁵ The other showed a statistically significant decrease in recurrent or persistent urethritis, but without identifying recurrent gonorrheal and chlamydial infections separately.³

Trichomoniasis. One RCT compared expedited partner therapy with patient referral for patients with trichomoniasis. The study didn’t show a statistically significant difference in recurrent or persistent infection.

TABLE 1
Traditional patient referral vs expedited partner treatment: How the 2 compare

The verdict: Expedited partner therapy works better
A meta-analysis of the above studies evaluated the effect of expedited partner therapy compared with patient referral on the rate of recurrent or persistent gonorrhea, chlamydia, and trichomoniasis and the number of partners treated per index patient.¹ Empiric therapy was associated with a lower rate of recurrent or persistent infections (risk ratio [RR]=0.73; 95% confidence interval [CI], 0.57-0.93) and a higher number of partners treated per patient (RR=1.44; 95% CI, 1.12-1.86).

Take state law into account
Providers need to consider their state’s laws regarding empiric partner therapy. A state-by-state evaluation of the legal status of expedited partner therapy is available on the Centers for Disease Control and Prevention’s Web site, and is summarized in TABLE 2.⁷

TABLE 2
What’s the status of expedited partner therapy (EPT) in your state?⁷

Recommendations

A review of expedited partner therapy by the Centers for Disease Control and Prevention concluded: “The evidence indicates that expedited partner therapy should be available to clinicians as an option for partner management…[but it] does not replace other strategies such as standard patient referral or provider-assisted referral, when available…. Expedited partner therapy should be accompanied by information [advising] recipients to seek personal health care in addition to expedited partner therapy. Expedited partner therapy has a limited role in partner management for trichomoniasis. No data support its use in the routine management of syphilis, and there is no experience with expedited partner therapy for gonorrhea or chlamydial infection among men who have sex with men.”⁸

Neither the American Academy of Family Physicians nor the American College of Obstetricians and Gynecologists has issued a policy statement on expedited partner therapy.

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Medical Department of the United States Navy or the US Naval Service at large.

Evidence summary

Treating partners of patients with sexually transmitted infection has been a core component of therapy since the 1940s. Traditionally, partners have been referred to a health care provider (by the index patient, the provider, or a public health officer) for evaluation before being treated. Current methods of partner referral reach only 40% to 60% of named sexual partners.¹

Expedited partner therapy vs traditional patient referral
Success of treatment is most readily measured by a reduction in the persistence or recurrence of infection in the index patient. Four RCTs and 1 observational cohort study have compared traditional patient referral with expedited partner treatment.^2-6 The primary outcome measure in all studies was reduction of persistent or recurrent infection in the index patient ( TABLE 1 ).

Chlamydia. Of the 4 studies that evaluated expedited partner treatment for chlamydia, 1 cohort study showed a statistically significant decrease in recurrent or persistent chlamydial infection in index patients.² One RCT showed a statistically significant reduction in recurrent or persistent urethritis, but didn’t report persistent and recurrent gonorrheal and chlamydial infections separately.³ Two RCTs showed a decrease in recurrent or persistent chlamydial infection in the index patient, but the difference didn’t reach statistical significance.^4,5

Gonorrhea. Two RCTs evaluated expedited partner treatment for gonorrhea compared with patient referral. One demonstrated a statistically significant decrease in persistent or recurrent gonococcal infection.⁵ The other showed a statistically significant decrease in recurrent or persistent urethritis, but without identifying recurrent gonorrheal and chlamydial infections separately.³

Trichomoniasis. One RCT compared expedited partner therapy with patient referral for patients with trichomoniasis. The study didn’t show a statistically significant difference in recurrent or persistent infection.

TABLE 1
Traditional patient referral vs expedited partner treatment: How the 2 compare

The verdict: Expedited partner therapy works better
A meta-analysis of the above studies evaluated the effect of expedited partner therapy compared with patient referral on the rate of recurrent or persistent gonorrhea, chlamydia, and trichomoniasis and the number of partners treated per index patient.¹ Empiric therapy was associated with a lower rate of recurrent or persistent infections (risk ratio [RR]=0.73; 95% confidence interval [CI], 0.57-0.93) and a higher number of partners treated per patient (RR=1.44; 95% CI, 1.12-1.86).

Take state law into account
Providers need to consider their state’s laws regarding empiric partner therapy. A state-by-state evaluation of the legal status of expedited partner therapy is available on the Centers for Disease Control and Prevention’s Web site, and is summarized in TABLE 2.⁷

TABLE 2
What’s the status of expedited partner therapy (EPT) in your state?⁷

Recommendations

A review of expedited partner therapy by the Centers for Disease Control and Prevention concluded: “The evidence indicates that expedited partner therapy should be available to clinicians as an option for partner management…[but it] does not replace other strategies such as standard patient referral or provider-assisted referral, when available…. Expedited partner therapy should be accompanied by information [advising] recipients to seek personal health care in addition to expedited partner therapy. Expedited partner therapy has a limited role in partner management for trichomoniasis. No data support its use in the routine management of syphilis, and there is no experience with expedited partner therapy for gonorrhea or chlamydial infection among men who have sex with men.”⁸

Neither the American Academy of Family Physicians nor the American College of Obstetricians and Gynecologists has issued a policy statement on expedited partner therapy.

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Medical Department of the United States Navy or the US Naval Service at large.

Evidence summary

The Healthy People 2010 report calls for increasing the proportion of Americans who engage in moderate physical activity (activities that use large muscle groups and are at least equivalent to brisk walking) from 15% to 30%.¹ The report doesn’t describe how best to achieve this objective.

Systematic review reveals approaches worth trying
The US Department of Health and Human Services (DHHS) and the Centers for Disease Control and Prevention (CDC) conducted a systematic review of 94 qualifying trials and assigned interventions to 1 of 3 approaches: “information based,” “behavioral and social,” and “facilities and activities.”²

Behavioral and social interventions have the best data support.² Within this category, strong evidence backed school-based physical education and accountability partners or exercise support groups. School-based physical education resulted in a median net increase in physical activity time of 50.3% (range 6.0%-125.3%); accountability partners or support groups produced a mean net increase of 44.2% (interquartile range 19.9%-45.6%).

“Green” prescriptions are primary care behavioral interventions that include measurable goals, self-reward, structured problem-solving, social network reinforcement, and relapse prevention counseling. In the DHHS review, 10 trials studied green prescriptions; the median net increase in physical activity time was 35.4% (interquartile range 16.7%-83.3%).² A trial in 42 rural and urban New Zealand general practices that added 3 telephone follow-up sessions to the green prescription showed a 10% increase in achieving 150 minutes of vigorous exercise weekly among green prescription participants compared with controls (number needed to treat=10).³

Pedometers. A systematic review using meta-regression to calculate summary effects evaluated the use of pedometers by study participants for an average of 18 weeks.⁴ Pedometer users increased their physical activity significantly, by 2491 steps per day compared with controls (95% confidence interval [CI], 1098-3885 steps per day).⁴ In adults, walking normally and walking briskly for an average of 2500 steps burns 100 and 150 kcal, respectively.⁵

Here’s what works with kids
A British systematic review of 24 high-quality controlled trials involving adolescents and children reported significant improvements with interventions that were school-based and either community- or family-based. Multidimensional outcomes included a 42% increase in participation in regular physical activity and an increase of 83 minutes weekly in moderate-to-vigorous physical activity.⁶

A US meta-analysis of 11 after-school programs with an average contact time of 275 minutes per week showed a positive standardized mean difference effect size for physical activity (0.44; 95% CI, 0.28-0.60).⁷

Evidence for other interventions is lacking
Insufficient evidence exists to support other interventions, such as classroom-based informational health education, mass media campaigns, college-based health and physical education, and classroom-based education focused on reducing television viewing and video-game playing.²

Recommendations

The British National Institute for Health and Clinical Excellence (NICE) has found sufficient evidence to recommend brief interventions in primary care. They include:

• using a validated tool to identify inactive patients
• recommending at least 30 minutes of patient-specific exercise at least 5 days per week
• establishing exercise goals
• presenting patients with written material on the benefits of exercise and local exercise opportunities
• following up several times over a 3-to 6-month period.⁸

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and not to be construed as official, or as reflecting the views of the US Air Force Medical Service or the US Air Force at large.

Evidence summary

The Healthy People 2010 report calls for increasing the proportion of Americans who engage in moderate physical activity (activities that use large muscle groups and are at least equivalent to brisk walking) from 15% to 30%.¹ The report doesn’t describe how best to achieve this objective.

Systematic review reveals approaches worth trying
The US Department of Health and Human Services (DHHS) and the Centers for Disease Control and Prevention (CDC) conducted a systematic review of 94 qualifying trials and assigned interventions to 1 of 3 approaches: “information based,” “behavioral and social,” and “facilities and activities.”²

Behavioral and social interventions have the best data support.² Within this category, strong evidence backed school-based physical education and accountability partners or exercise support groups. School-based physical education resulted in a median net increase in physical activity time of 50.3% (range 6.0%-125.3%); accountability partners or support groups produced a mean net increase of 44.2% (interquartile range 19.9%-45.6%).

“Green” prescriptions are primary care behavioral interventions that include measurable goals, self-reward, structured problem-solving, social network reinforcement, and relapse prevention counseling. In the DHHS review, 10 trials studied green prescriptions; the median net increase in physical activity time was 35.4% (interquartile range 16.7%-83.3%).² A trial in 42 rural and urban New Zealand general practices that added 3 telephone follow-up sessions to the green prescription showed a 10% increase in achieving 150 minutes of vigorous exercise weekly among green prescription participants compared with controls (number needed to treat=10).³

Pedometers. A systematic review using meta-regression to calculate summary effects evaluated the use of pedometers by study participants for an average of 18 weeks.⁴ Pedometer users increased their physical activity significantly, by 2491 steps per day compared with controls (95% confidence interval [CI], 1098-3885 steps per day).⁴ In adults, walking normally and walking briskly for an average of 2500 steps burns 100 and 150 kcal, respectively.⁵

Here’s what works with kids
A British systematic review of 24 high-quality controlled trials involving adolescents and children reported significant improvements with interventions that were school-based and either community- or family-based. Multidimensional outcomes included a 42% increase in participation in regular physical activity and an increase of 83 minutes weekly in moderate-to-vigorous physical activity.⁶

A US meta-analysis of 11 after-school programs with an average contact time of 275 minutes per week showed a positive standardized mean difference effect size for physical activity (0.44; 95% CI, 0.28-0.60).⁷

Evidence for other interventions is lacking
Insufficient evidence exists to support other interventions, such as classroom-based informational health education, mass media campaigns, college-based health and physical education, and classroom-based education focused on reducing television viewing and video-game playing.²

Recommendations

The British National Institute for Health and Clinical Excellence (NICE) has found sufficient evidence to recommend brief interventions in primary care. They include:

• using a validated tool to identify inactive patients
• recommending at least 30 minutes of patient-specific exercise at least 5 days per week
• establishing exercise goals
• presenting patients with written material on the benefits of exercise and local exercise opportunities
• following up several times over a 3-to 6-month period.⁸

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and not to be construed as official, or as reflecting the views of the US Air Force Medical Service or the US Air Force at large.

Evidence summary

The Healthy People 2010 report calls for increasing the proportion of Americans who engage in moderate physical activity (activities that use large muscle groups and are at least equivalent to brisk walking) from 15% to 30%.¹ The report doesn’t describe how best to achieve this objective.

Systematic review reveals approaches worth trying
The US Department of Health and Human Services (DHHS) and the Centers for Disease Control and Prevention (CDC) conducted a systematic review of 94 qualifying trials and assigned interventions to 1 of 3 approaches: “information based,” “behavioral and social,” and “facilities and activities.”²

Behavioral and social interventions have the best data support.² Within this category, strong evidence backed school-based physical education and accountability partners or exercise support groups. School-based physical education resulted in a median net increase in physical activity time of 50.3% (range 6.0%-125.3%); accountability partners or support groups produced a mean net increase of 44.2% (interquartile range 19.9%-45.6%).

“Green” prescriptions are primary care behavioral interventions that include measurable goals, self-reward, structured problem-solving, social network reinforcement, and relapse prevention counseling. In the DHHS review, 10 trials studied green prescriptions; the median net increase in physical activity time was 35.4% (interquartile range 16.7%-83.3%).² A trial in 42 rural and urban New Zealand general practices that added 3 telephone follow-up sessions to the green prescription showed a 10% increase in achieving 150 minutes of vigorous exercise weekly among green prescription participants compared with controls (number needed to treat=10).³

Pedometers. A systematic review using meta-regression to calculate summary effects evaluated the use of pedometers by study participants for an average of 18 weeks.⁴ Pedometer users increased their physical activity significantly, by 2491 steps per day compared with controls (95% confidence interval [CI], 1098-3885 steps per day).⁴ In adults, walking normally and walking briskly for an average of 2500 steps burns 100 and 150 kcal, respectively.⁵

Here’s what works with kids
A British systematic review of 24 high-quality controlled trials involving adolescents and children reported significant improvements with interventions that were school-based and either community- or family-based. Multidimensional outcomes included a 42% increase in participation in regular physical activity and an increase of 83 minutes weekly in moderate-to-vigorous physical activity.⁶

A US meta-analysis of 11 after-school programs with an average contact time of 275 minutes per week showed a positive standardized mean difference effect size for physical activity (0.44; 95% CI, 0.28-0.60).⁷

Evidence for other interventions is lacking
Insufficient evidence exists to support other interventions, such as classroom-based informational health education, mass media campaigns, college-based health and physical education, and classroom-based education focused on reducing television viewing and video-game playing.²

Recommendations

The British National Institute for Health and Clinical Excellence (NICE) has found sufficient evidence to recommend brief interventions in primary care. They include:

• using a validated tool to identify inactive patients
• recommending at least 30 minutes of patient-specific exercise at least 5 days per week
• establishing exercise goals
• presenting patients with written material on the benefits of exercise and local exercise opportunities
• following up several times over a 3-to 6-month period.⁸

Acknowledgments
The opinions and assertions contained herein are the private views of the authors and not to be construed as official, or as reflecting the views of the US Air Force Medical Service or the US Air Force at large.

Evidence summary

Little evidence directly compares treatments for pyogenic granuloma. Most studies examine multiple treatment methods without comparing results statistically, combine data from adults and children, or comprise case series using a single treatment method. The TABLE summarizes outcomes for different therapies.

TABLE
How treatment outcomes for pyogenic granuloma compare

Surgical excision: Low recurrence, but scarring is common
A retrospective cohort study audited recurrence rates in 408 patients (mean age 41 years, range 5 months to 90 years) whose pyogenic granulomas were treated with either surgical excision or combinations of curettage, shave, and cautery.¹ Investigators identified cases of histopathologically confirmed pyogenic granuloma over a 10-year period from a hospital database. Thirty-six percent of granulomas were located on the head and neck, 33% on the arm, 15% on the trunk, and 8% on the leg.

Of 326 lesions treated with surgical excision, 4 (3.7%) recurred. The overall recurrence rate was 10.3% for 82 lesions removed by combinations of curettage, shave, and cautery (the lesions weren’t differentiated by removal method or location). Investigators didn’t report on residual scarring with any method. However, expert opinion states that surgical excision often results in a conspicuous linear scar.²

Surgery vs laser therapy or no treatment
Another retrospective cohort study described treatment, recurrence rate, residual scarring, and patient satisfaction in 76 patients with pyogenic granuloma (mean age 6 years; range 4 months to 17 years). Outcomes were assessed by telephone follow-up.³

Fifty-eight lesions were removed by surgical excision and cautery with no recurrences (55% of patients had subtle scarring). Nine lesions were treated with pulsed dye laser (33% recurrence, 44% subtle scarring); 3 lesions were removed by CO₂ laser (100% recurrence, 33% subtle scarring). Four patients were followed but not treated (no explanation given); all untreated pyogenic granulomas disappeared spontaneously within 6 to 18 months with no recurrences; 1 patient had subtle scarring.

Cryotherapy may require more treatments than curettage
A prospective trial randomized 89 patients (mean age 34 years; range 11-88 years) with pyogenic granulomas that were 1.5 cm or smaller to receive either curettage or cryotherapy, then evaluated the number of treatments required and whether scarring occurred. Follow-up was 85%.⁴

A single curettage resolved pyogenic granuloma in 35 of 36 patients (97%); 9 of the 36 patients (31%) had residual scarring. Twenty-five of 40 pyogenic granulomas (63%) responded to 1 cryotherapy treatment, 13 lesions (32%) resolved after 2 treatments, and 2 (5%) resolved after 3 treatments; 17 of the 40 patients (42.5%) had a residual scar. Curettage required fewer treatments overall than cryotherapy (P<.001), but no significant difference in residual scarring was noted between the 2 treatments.

A case series reported on 135 patients (mean age 26 years; range 4 months to 70 years) whose pyogenic granulomas were treated with cryotherapy.⁵ Seventy-eight (58%) had complete resolution with 1 session, 30% needed 2 sessions, 8% needed 3 sessions, and 5% needed 4 sessions (mean 1.58 sessions). Ninety-four percent had an excellent cosmetic result (including 12% with a small flat scar); 5% had residual hypopigmentation.

CO₂ laser usually removes lesions in 1 session
Another case series of 100 patients (mean age 27 years; range 6 months to 84 years) treated with CO₂ laser reported that the pyogenic granuloma was removed completely in 1 session in 98 patients. Twelve percent of patients had visible scarring and another 10% had slight textural skin changes. All patients reported satisfaction with the results.⁶

Pulsed dye laser works for small lesions
A case series of 22 patients (mean age 3.4 years; range 6 months to 16 years) treated with pulsed dye laser for mostly small lesions (average diameter 4 mm) on the face reported successful removal in 20 children in 1 to 6 treatment sessions (average 2.25) with no residual scarring. Two children with larger lesions required shave excision with cautery (scarring was not assessed).⁷

Sclerotherapy: No recurrence, inconspicuous scars
A case series reported results in 9 patients (median age 18 years; range 1-57 years) with pyogenic granuloma treated with a single injection of the sclerosing agent monoethanolamine oleate.² All lesions disappeared without recurrence; the authors described remaining scars as inconspicuous. One patient reported residual pain lasting 4 weeks after injection of the sclerosing agent into a 1.5 cm × 1.0 cm pyogenic granuloma that hadn’t responded to previous cryotherapy.

Recommendations

A standard dermatology textbook recommends curettage with cautery, and reports that spontaneous regression is common after 6 months.⁸ A standard pediatric textbook recommends surgical excision with or without cautery, adding that small pyogenic granuloma lesions (<5 mm) may be removed with pulsed dye laser.⁹

An online textbook recommends either excision or shave (with or without curettage), but advises surgical excision with histologic confirmation for pyogenic granuloma lesions that can’t be differentiated with certainty from amelanotic melanoma, which typically grows more slowly.¹⁰

Evidence summary

Little evidence directly compares treatments for pyogenic granuloma. Most studies examine multiple treatment methods without comparing results statistically, combine data from adults and children, or comprise case series using a single treatment method. The TABLE summarizes outcomes for different therapies.

TABLE
How treatment outcomes for pyogenic granuloma compare

Surgical excision: Low recurrence, but scarring is common
A retrospective cohort study audited recurrence rates in 408 patients (mean age 41 years, range 5 months to 90 years) whose pyogenic granulomas were treated with either surgical excision or combinations of curettage, shave, and cautery.¹ Investigators identified cases of histopathologically confirmed pyogenic granuloma over a 10-year period from a hospital database. Thirty-six percent of granulomas were located on the head and neck, 33% on the arm, 15% on the trunk, and 8% on the leg.

Of 326 lesions treated with surgical excision, 4 (3.7%) recurred. The overall recurrence rate was 10.3% for 82 lesions removed by combinations of curettage, shave, and cautery (the lesions weren’t differentiated by removal method or location). Investigators didn’t report on residual scarring with any method. However, expert opinion states that surgical excision often results in a conspicuous linear scar.²

Surgery vs laser therapy or no treatment
Another retrospective cohort study described treatment, recurrence rate, residual scarring, and patient satisfaction in 76 patients with pyogenic granuloma (mean age 6 years; range 4 months to 17 years). Outcomes were assessed by telephone follow-up.³

Fifty-eight lesions were removed by surgical excision and cautery with no recurrences (55% of patients had subtle scarring). Nine lesions were treated with pulsed dye laser (33% recurrence, 44% subtle scarring); 3 lesions were removed by CO₂ laser (100% recurrence, 33% subtle scarring). Four patients were followed but not treated (no explanation given); all untreated pyogenic granulomas disappeared spontaneously within 6 to 18 months with no recurrences; 1 patient had subtle scarring.

Cryotherapy may require more treatments than curettage
A prospective trial randomized 89 patients (mean age 34 years; range 11-88 years) with pyogenic granulomas that were 1.5 cm or smaller to receive either curettage or cryotherapy, then evaluated the number of treatments required and whether scarring occurred. Follow-up was 85%.⁴

A single curettage resolved pyogenic granuloma in 35 of 36 patients (97%); 9 of the 36 patients (31%) had residual scarring. Twenty-five of 40 pyogenic granulomas (63%) responded to 1 cryotherapy treatment, 13 lesions (32%) resolved after 2 treatments, and 2 (5%) resolved after 3 treatments; 17 of the 40 patients (42.5%) had a residual scar. Curettage required fewer treatments overall than cryotherapy (P<.001), but no significant difference in residual scarring was noted between the 2 treatments.

A case series reported on 135 patients (mean age 26 years; range 4 months to 70 years) whose pyogenic granulomas were treated with cryotherapy.⁵ Seventy-eight (58%) had complete resolution with 1 session, 30% needed 2 sessions, 8% needed 3 sessions, and 5% needed 4 sessions (mean 1.58 sessions). Ninety-four percent had an excellent cosmetic result (including 12% with a small flat scar); 5% had residual hypopigmentation.

CO₂ laser usually removes lesions in 1 session
Another case series of 100 patients (mean age 27 years; range 6 months to 84 years) treated with CO₂ laser reported that the pyogenic granuloma was removed completely in 1 session in 98 patients. Twelve percent of patients had visible scarring and another 10% had slight textural skin changes. All patients reported satisfaction with the results.⁶

Pulsed dye laser works for small lesions
A case series of 22 patients (mean age 3.4 years; range 6 months to 16 years) treated with pulsed dye laser for mostly small lesions (average diameter 4 mm) on the face reported successful removal in 20 children in 1 to 6 treatment sessions (average 2.25) with no residual scarring. Two children with larger lesions required shave excision with cautery (scarring was not assessed).⁷

Sclerotherapy: No recurrence, inconspicuous scars
A case series reported results in 9 patients (median age 18 years; range 1-57 years) with pyogenic granuloma treated with a single injection of the sclerosing agent monoethanolamine oleate.² All lesions disappeared without recurrence; the authors described remaining scars as inconspicuous. One patient reported residual pain lasting 4 weeks after injection of the sclerosing agent into a 1.5 cm × 1.0 cm pyogenic granuloma that hadn’t responded to previous cryotherapy.

Recommendations

A standard dermatology textbook recommends curettage with cautery, and reports that spontaneous regression is common after 6 months.⁸ A standard pediatric textbook recommends surgical excision with or without cautery, adding that small pyogenic granuloma lesions (<5 mm) may be removed with pulsed dye laser.⁹

An online textbook recommends either excision or shave (with or without curettage), but advises surgical excision with histologic confirmation for pyogenic granuloma lesions that can’t be differentiated with certainty from amelanotic melanoma, which typically grows more slowly.¹⁰

Evidence summary

Little evidence directly compares treatments for pyogenic granuloma. Most studies examine multiple treatment methods without comparing results statistically, combine data from adults and children, or comprise case series using a single treatment method. The TABLE summarizes outcomes for different therapies.

TABLE
How treatment outcomes for pyogenic granuloma compare

Surgical excision: Low recurrence, but scarring is common
A retrospective cohort study audited recurrence rates in 408 patients (mean age 41 years, range 5 months to 90 years) whose pyogenic granulomas were treated with either surgical excision or combinations of curettage, shave, and cautery.¹ Investigators identified cases of histopathologically confirmed pyogenic granuloma over a 10-year period from a hospital database. Thirty-six percent of granulomas were located on the head and neck, 33% on the arm, 15% on the trunk, and 8% on the leg.

Of 326 lesions treated with surgical excision, 4 (3.7%) recurred. The overall recurrence rate was 10.3% for 82 lesions removed by combinations of curettage, shave, and cautery (the lesions weren’t differentiated by removal method or location). Investigators didn’t report on residual scarring with any method. However, expert opinion states that surgical excision often results in a conspicuous linear scar.²

Surgery vs laser therapy or no treatment
Another retrospective cohort study described treatment, recurrence rate, residual scarring, and patient satisfaction in 76 patients with pyogenic granuloma (mean age 6 years; range 4 months to 17 years). Outcomes were assessed by telephone follow-up.³

Fifty-eight lesions were removed by surgical excision and cautery with no recurrences (55% of patients had subtle scarring). Nine lesions were treated with pulsed dye laser (33% recurrence, 44% subtle scarring); 3 lesions were removed by CO₂ laser (100% recurrence, 33% subtle scarring). Four patients were followed but not treated (no explanation given); all untreated pyogenic granulomas disappeared spontaneously within 6 to 18 months with no recurrences; 1 patient had subtle scarring.

Cryotherapy may require more treatments than curettage
A prospective trial randomized 89 patients (mean age 34 years; range 11-88 years) with pyogenic granulomas that were 1.5 cm or smaller to receive either curettage or cryotherapy, then evaluated the number of treatments required and whether scarring occurred. Follow-up was 85%.⁴

A single curettage resolved pyogenic granuloma in 35 of 36 patients (97%); 9 of the 36 patients (31%) had residual scarring. Twenty-five of 40 pyogenic granulomas (63%) responded to 1 cryotherapy treatment, 13 lesions (32%) resolved after 2 treatments, and 2 (5%) resolved after 3 treatments; 17 of the 40 patients (42.5%) had a residual scar. Curettage required fewer treatments overall than cryotherapy (P<.001), but no significant difference in residual scarring was noted between the 2 treatments.

A case series reported on 135 patients (mean age 26 years; range 4 months to 70 years) whose pyogenic granulomas were treated with cryotherapy.⁵ Seventy-eight (58%) had complete resolution with 1 session, 30% needed 2 sessions, 8% needed 3 sessions, and 5% needed 4 sessions (mean 1.58 sessions). Ninety-four percent had an excellent cosmetic result (including 12% with a small flat scar); 5% had residual hypopigmentation.

CO₂ laser usually removes lesions in 1 session
Another case series of 100 patients (mean age 27 years; range 6 months to 84 years) treated with CO₂ laser reported that the pyogenic granuloma was removed completely in 1 session in 98 patients. Twelve percent of patients had visible scarring and another 10% had slight textural skin changes. All patients reported satisfaction with the results.⁶

Pulsed dye laser works for small lesions
A case series of 22 patients (mean age 3.4 years; range 6 months to 16 years) treated with pulsed dye laser for mostly small lesions (average diameter 4 mm) on the face reported successful removal in 20 children in 1 to 6 treatment sessions (average 2.25) with no residual scarring. Two children with larger lesions required shave excision with cautery (scarring was not assessed).⁷

Sclerotherapy: No recurrence, inconspicuous scars
A case series reported results in 9 patients (median age 18 years; range 1-57 years) with pyogenic granuloma treated with a single injection of the sclerosing agent monoethanolamine oleate.² All lesions disappeared without recurrence; the authors described remaining scars as inconspicuous. One patient reported residual pain lasting 4 weeks after injection of the sclerosing agent into a 1.5 cm × 1.0 cm pyogenic granuloma that hadn’t responded to previous cryotherapy.

Recommendations

A standard dermatology textbook recommends curettage with cautery, and reports that spontaneous regression is common after 6 months.⁸ A standard pediatric textbook recommends surgical excision with or without cautery, adding that small pyogenic granuloma lesions (<5 mm) may be removed with pulsed dye laser.⁹

An online textbook recommends either excision or shave (with or without curettage), but advises surgical excision with histologic confirmation for pyogenic granuloma lesions that can’t be differentiated with certainty from amelanotic melanoma, which typically grows more slowly.¹⁰