Clinical Inquiries

Evidence summary

Chronic urticaria affects 1% of the general population and is usually defined as the presence of hives (with or without angioedema) for at least 6 weeks.¹

History and physical are key, a few tests may be useful
A systematic review of 29 studies involving 6462 patients done between 1966 and 2001 found no strong evidence for laboratory testing beyond a complete history and physical. However, the authors recommended that patients with chronic idiopathic urticaria have an erythrocyte sedimentation rate (ESR) measurement, white blood cell (WBC) count, and differential cell count.² A primarily expert-opinion-based guideline on chronic urticaria recommended a WBC count, ESR, urinalysis, and liver function tests to screen for underlying diseases.³

Is aggressive testing worth the effort?
It would appear not. A prospective study of 220 patients, representative of the studies included in the systematic review, compared 2 strategies to evaluate the cause of chronic urticaria:

• Detailed history taking and limited laboratory testing (hemoglobin, hematocrit, ESR, WBC count, and dermatographism test [hive associated with a scratch])
• Detailed history taking and extensive laboratory evaluation with 33 different tests, many of them special and invasive (radiographs, vaginal cultures, and skin biopsies).⁴

Detailed history taking and limited laboratory tests found a cause for urticaria in 45.9% of patients, compared with 52.7% of patients who underwent detailed history taking and extensive laboratory screening.⁴ This translates into testing 15 patients aggressively to diagnose one potentially reversible cause of chronic urticaria.

Among patients evaluated with a detailed history and extensive diagnostic work-up, 33.2% had physical urticaria (triggered by pressure, cold, heat, and light). Other diagnoses included adverse drug reactions (8.6%), adverse food reactions (6.8%), infection (1.8%), contact urticaria (0.9%), and internal disease (1.4%). No cause was identified in 47.3% of the patients.⁴

Recommendations

The British Association of Dermatologists has issued the following guidelines for evaluation and management of urticaria in adults and children:⁵

• The diagnosis of urticaria is primarily clinical.
• Diagnostic investigations should be guided by the history and shouldn’t be performed in all patients.

Acknowledgements

The opinions and assertions contained herein are the private views of the authors and should 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

Chronic urticaria affects 1% of the general population and is usually defined as the presence of hives (with or without angioedema) for at least 6 weeks.¹

History and physical are key, a few tests may be useful
A systematic review of 29 studies involving 6462 patients done between 1966 and 2001 found no strong evidence for laboratory testing beyond a complete history and physical. However, the authors recommended that patients with chronic idiopathic urticaria have an erythrocyte sedimentation rate (ESR) measurement, white blood cell (WBC) count, and differential cell count.² A primarily expert-opinion-based guideline on chronic urticaria recommended a WBC count, ESR, urinalysis, and liver function tests to screen for underlying diseases.³

Is aggressive testing worth the effort?
It would appear not. A prospective study of 220 patients, representative of the studies included in the systematic review, compared 2 strategies to evaluate the cause of chronic urticaria:

• Detailed history taking and limited laboratory testing (hemoglobin, hematocrit, ESR, WBC count, and dermatographism test [hive associated with a scratch])
• Detailed history taking and extensive laboratory evaluation with 33 different tests, many of them special and invasive (radiographs, vaginal cultures, and skin biopsies).⁴

Detailed history taking and limited laboratory tests found a cause for urticaria in 45.9% of patients, compared with 52.7% of patients who underwent detailed history taking and extensive laboratory screening.⁴ This translates into testing 15 patients aggressively to diagnose one potentially reversible cause of chronic urticaria.

Among patients evaluated with a detailed history and extensive diagnostic work-up, 33.2% had physical urticaria (triggered by pressure, cold, heat, and light). Other diagnoses included adverse drug reactions (8.6%), adverse food reactions (6.8%), infection (1.8%), contact urticaria (0.9%), and internal disease (1.4%). No cause was identified in 47.3% of the patients.⁴

Recommendations

The British Association of Dermatologists has issued the following guidelines for evaluation and management of urticaria in adults and children:⁵

• The diagnosis of urticaria is primarily clinical.
• Diagnostic investigations should be guided by the history and shouldn’t be performed in all patients.

Acknowledgements

The opinions and assertions contained herein are the private views of the authors and should 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

Chronic urticaria affects 1% of the general population and is usually defined as the presence of hives (with or without angioedema) for at least 6 weeks.¹

History and physical are key, a few tests may be useful
A systematic review of 29 studies involving 6462 patients done between 1966 and 2001 found no strong evidence for laboratory testing beyond a complete history and physical. However, the authors recommended that patients with chronic idiopathic urticaria have an erythrocyte sedimentation rate (ESR) measurement, white blood cell (WBC) count, and differential cell count.² A primarily expert-opinion-based guideline on chronic urticaria recommended a WBC count, ESR, urinalysis, and liver function tests to screen for underlying diseases.³

Is aggressive testing worth the effort?
It would appear not. A prospective study of 220 patients, representative of the studies included in the systematic review, compared 2 strategies to evaluate the cause of chronic urticaria:

• Detailed history taking and limited laboratory testing (hemoglobin, hematocrit, ESR, WBC count, and dermatographism test [hive associated with a scratch])
• Detailed history taking and extensive laboratory evaluation with 33 different tests, many of them special and invasive (radiographs, vaginal cultures, and skin biopsies).⁴

Detailed history taking and limited laboratory tests found a cause for urticaria in 45.9% of patients, compared with 52.7% of patients who underwent detailed history taking and extensive laboratory screening.⁴ This translates into testing 15 patients aggressively to diagnose one potentially reversible cause of chronic urticaria.

Among patients evaluated with a detailed history and extensive diagnostic work-up, 33.2% had physical urticaria (triggered by pressure, cold, heat, and light). Other diagnoses included adverse drug reactions (8.6%), adverse food reactions (6.8%), infection (1.8%), contact urticaria (0.9%), and internal disease (1.4%). No cause was identified in 47.3% of the patients.⁴

Recommendations

The British Association of Dermatologists has issued the following guidelines for evaluation and management of urticaria in adults and children:⁵

• The diagnosis of urticaria is primarily clinical.
• Diagnostic investigations should be guided by the history and shouldn’t be performed in all patients.

Acknowledgements

The opinions and assertions contained herein are the private views of the authors and should 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 systematic review of 4 double-blind RCTs found that the anti-inflammatory amlexanox 5% paste significantly reduced ulcer size compared with placebo (–1 vs 0 mm on Day 3 and–3 vs–1 mm on Day 5; P<.01). It also decreased healing time by Day 3 and duration of pain. When applied during the prodromal stage, amlexanox dramatically reduced the number of patients who progressed to full ulcers.¹

Clobetasol also reduces pain, ulcer size

A double-blind RCT comparing amlexanox 5% with another anti-inflammatory, clobetasol propionate 0.05% paste, showed equal declines in visual analog pain scores (an average drop from 6 to 2 points out of 10 for each treatment by Day 3 compared with baseline; P<.001). Both treatments produced equivalent reductions in ulcer size from baseline by Day 5 (P<.001).²

Corticosteroids decrease pain, number of ulcers

In a systematic review of 9 small RCTs comparing corticosteroids with placebo, 3 out of 4 RCTs reported significant pain relief. One trial found that 11 of 15 patients treated with beclomethasone spray experienced less pain compared with 0 of 15 patients in the placebo group (P<.05).³

Beclomethasone also decreased the ulcer index (number of ulcers each day over time) in 2 of 4 RCTS (eg, 13 of 15 patients [drug] vs 0 of 15 patients [placebo]; P<.01).³

Chemical cautery relieves pain but may not speed healing

An RCT of 97 patients compared physician-applied silver nitrate with placebo after patients in each group received a 2% lidocaine swab. Silver nitrate treatment was associated with more pain-free patients at Day 1 but didn’t decrease time until resolution.⁴

Debacterol reduced pain more than placebo by Day 3 (–45 vs–15 points on a visual analog scale; P<.001) and resolved symptoms more effectively by Day 6.⁵

Antiseptic mouthwashes lower ulcer index

A systematic review of 5 low-quality RCTs (small size, incomplete reporting of data, varied outcomes) comparing chlorhexidine with placebo reported that in 2 of 3 studies that measured the ulcer index, chlorhexidine reduced the index more than placebo (eg, 0.83 with chlorhexidine vs 1.66 with placebo; P<.05). Three of 4 studies that evaluated ulcer duration didn’t show a significant decrease (5.02 days with chlorhexidine vs 5.78 days with placebo; P>.05).²

One small RCT found that a commercial mouthwash (Listerine antiseptic containing menthol, thymol, methyl salicylate, and eucalyptol) modestly decreased ulcer duration (–1.43 days per outbreak for Listerine compared with–0.54 days for placebo; P<.001) and severity (–0.53 points on a 10-point pain scale with mouthwash vs 0 points with placebo; P<.001).⁶

Herbal preparation relieves pain, improves healing

In a small double-blind trial, Eupatorium laevigatum paste improved 5-day cure rates and pain relief more than triamcinolone 0.1% paste.⁷

Vitamin B₁₂, avoiding certain toothpastes, helps prevent recurrence

A small RCT comparing 1000 mcg of oral vitamin B₁₂ daily with placebo for prophylaxis found that B₁₂ reduced ulcer duration more than placebo (1.98±3.77 vs 4.84±5.71 days; P<.05), number of ulcers (3.88±7.98 vs 13.39±23; P<.05), and pain (0.64±1.45 vs 2.36 ±2.21 points; P<.01) by 5 months of treatment and increased the percentage of ulcer-free patients by 6 months.⁸

In 3 of 4 small trials, avoiding toothpaste containing sodium lauryl sulfate slightly but significantly reduced recurrent ulcers.⁹ Bioadherent agents such as Rincinol appear to have minimal benefit.⁹

“Magic mouthwashes” lack efficacy data

Clinicians often prescribe “magic mouthwashes” (various combinations of viscous lidocaine, benzocaine, milk of magnesia, kaolin pectate, chlorhexidine, or diphenhydramine), but we found no studies evaluating their effectiveness.

Recommendations

The British National Health Service Clinical Knowledge Summary suggests avoiding aggravating foods and life stressors, using a soft bristled toothbrush, quitting smoking, and using topical corticosteroids, antimicrobial mouthwashes, and topical anesthetics.¹⁰

Evidence summary

A systematic review of 4 double-blind RCTs found that the anti-inflammatory amlexanox 5% paste significantly reduced ulcer size compared with placebo (–1 vs 0 mm on Day 3 and–3 vs–1 mm on Day 5; P<.01). It also decreased healing time by Day 3 and duration of pain. When applied during the prodromal stage, amlexanox dramatically reduced the number of patients who progressed to full ulcers.¹

Clobetasol also reduces pain, ulcer size

A double-blind RCT comparing amlexanox 5% with another anti-inflammatory, clobetasol propionate 0.05% paste, showed equal declines in visual analog pain scores (an average drop from 6 to 2 points out of 10 for each treatment by Day 3 compared with baseline; P<.001). Both treatments produced equivalent reductions in ulcer size from baseline by Day 5 (P<.001).²

Corticosteroids decrease pain, number of ulcers

In a systematic review of 9 small RCTs comparing corticosteroids with placebo, 3 out of 4 RCTs reported significant pain relief. One trial found that 11 of 15 patients treated with beclomethasone spray experienced less pain compared with 0 of 15 patients in the placebo group (P<.05).³

Beclomethasone also decreased the ulcer index (number of ulcers each day over time) in 2 of 4 RCTS (eg, 13 of 15 patients [drug] vs 0 of 15 patients [placebo]; P<.01).³

Chemical cautery relieves pain but may not speed healing

An RCT of 97 patients compared physician-applied silver nitrate with placebo after patients in each group received a 2% lidocaine swab. Silver nitrate treatment was associated with more pain-free patients at Day 1 but didn’t decrease time until resolution.⁴

Debacterol reduced pain more than placebo by Day 3 (–45 vs–15 points on a visual analog scale; P<.001) and resolved symptoms more effectively by Day 6.⁵

Antiseptic mouthwashes lower ulcer index

A systematic review of 5 low-quality RCTs (small size, incomplete reporting of data, varied outcomes) comparing chlorhexidine with placebo reported that in 2 of 3 studies that measured the ulcer index, chlorhexidine reduced the index more than placebo (eg, 0.83 with chlorhexidine vs 1.66 with placebo; P<.05). Three of 4 studies that evaluated ulcer duration didn’t show a significant decrease (5.02 days with chlorhexidine vs 5.78 days with placebo; P>.05).²

One small RCT found that a commercial mouthwash (Listerine antiseptic containing menthol, thymol, methyl salicylate, and eucalyptol) modestly decreased ulcer duration (–1.43 days per outbreak for Listerine compared with–0.54 days for placebo; P<.001) and severity (–0.53 points on a 10-point pain scale with mouthwash vs 0 points with placebo; P<.001).⁶

Herbal preparation relieves pain, improves healing

In a small double-blind trial, Eupatorium laevigatum paste improved 5-day cure rates and pain relief more than triamcinolone 0.1% paste.⁷

Vitamin B₁₂, avoiding certain toothpastes, helps prevent recurrence

A small RCT comparing 1000 mcg of oral vitamin B₁₂ daily with placebo for prophylaxis found that B₁₂ reduced ulcer duration more than placebo (1.98±3.77 vs 4.84±5.71 days; P<.05), number of ulcers (3.88±7.98 vs 13.39±23; P<.05), and pain (0.64±1.45 vs 2.36 ±2.21 points; P<.01) by 5 months of treatment and increased the percentage of ulcer-free patients by 6 months.⁸

In 3 of 4 small trials, avoiding toothpaste containing sodium lauryl sulfate slightly but significantly reduced recurrent ulcers.⁹ Bioadherent agents such as Rincinol appear to have minimal benefit.⁹

“Magic mouthwashes” lack efficacy data

Clinicians often prescribe “magic mouthwashes” (various combinations of viscous lidocaine, benzocaine, milk of magnesia, kaolin pectate, chlorhexidine, or diphenhydramine), but we found no studies evaluating their effectiveness.

Recommendations

The British National Health Service Clinical Knowledge Summary suggests avoiding aggravating foods and life stressors, using a soft bristled toothbrush, quitting smoking, and using topical corticosteroids, antimicrobial mouthwashes, and topical anesthetics.¹⁰

Evidence summary

A systematic review of 4 double-blind RCTs found that the anti-inflammatory amlexanox 5% paste significantly reduced ulcer size compared with placebo (–1 vs 0 mm on Day 3 and–3 vs–1 mm on Day 5; P<.01). It also decreased healing time by Day 3 and duration of pain. When applied during the prodromal stage, amlexanox dramatically reduced the number of patients who progressed to full ulcers.¹

Clobetasol also reduces pain, ulcer size

A double-blind RCT comparing amlexanox 5% with another anti-inflammatory, clobetasol propionate 0.05% paste, showed equal declines in visual analog pain scores (an average drop from 6 to 2 points out of 10 for each treatment by Day 3 compared with baseline; P<.001). Both treatments produced equivalent reductions in ulcer size from baseline by Day 5 (P<.001).²

Corticosteroids decrease pain, number of ulcers

In a systematic review of 9 small RCTs comparing corticosteroids with placebo, 3 out of 4 RCTs reported significant pain relief. One trial found that 11 of 15 patients treated with beclomethasone spray experienced less pain compared with 0 of 15 patients in the placebo group (P<.05).³

Beclomethasone also decreased the ulcer index (number of ulcers each day over time) in 2 of 4 RCTS (eg, 13 of 15 patients [drug] vs 0 of 15 patients [placebo]; P<.01).³

Chemical cautery relieves pain but may not speed healing

An RCT of 97 patients compared physician-applied silver nitrate with placebo after patients in each group received a 2% lidocaine swab. Silver nitrate treatment was associated with more pain-free patients at Day 1 but didn’t decrease time until resolution.⁴

Debacterol reduced pain more than placebo by Day 3 (–45 vs–15 points on a visual analog scale; P<.001) and resolved symptoms more effectively by Day 6.⁵

Antiseptic mouthwashes lower ulcer index

A systematic review of 5 low-quality RCTs (small size, incomplete reporting of data, varied outcomes) comparing chlorhexidine with placebo reported that in 2 of 3 studies that measured the ulcer index, chlorhexidine reduced the index more than placebo (eg, 0.83 with chlorhexidine vs 1.66 with placebo; P<.05). Three of 4 studies that evaluated ulcer duration didn’t show a significant decrease (5.02 days with chlorhexidine vs 5.78 days with placebo; P>.05).²

One small RCT found that a commercial mouthwash (Listerine antiseptic containing menthol, thymol, methyl salicylate, and eucalyptol) modestly decreased ulcer duration (–1.43 days per outbreak for Listerine compared with–0.54 days for placebo; P<.001) and severity (–0.53 points on a 10-point pain scale with mouthwash vs 0 points with placebo; P<.001).⁶

Herbal preparation relieves pain, improves healing

In a small double-blind trial, Eupatorium laevigatum paste improved 5-day cure rates and pain relief more than triamcinolone 0.1% paste.⁷

Vitamin B₁₂, avoiding certain toothpastes, helps prevent recurrence

A small RCT comparing 1000 mcg of oral vitamin B₁₂ daily with placebo for prophylaxis found that B₁₂ reduced ulcer duration more than placebo (1.98±3.77 vs 4.84±5.71 days; P<.05), number of ulcers (3.88±7.98 vs 13.39±23; P<.05), and pain (0.64±1.45 vs 2.36 ±2.21 points; P<.01) by 5 months of treatment and increased the percentage of ulcer-free patients by 6 months.⁸

In 3 of 4 small trials, avoiding toothpaste containing sodium lauryl sulfate slightly but significantly reduced recurrent ulcers.⁹ Bioadherent agents such as Rincinol appear to have minimal benefit.⁹

“Magic mouthwashes” lack efficacy data

Clinicians often prescribe “magic mouthwashes” (various combinations of viscous lidocaine, benzocaine, milk of magnesia, kaolin pectate, chlorhexidine, or diphenhydramine), but we found no studies evaluating their effectiveness.

Recommendations

The British National Health Service Clinical Knowledge Summary suggests avoiding aggravating foods and life stressors, using a soft bristled toothbrush, quitting smoking, and using topical corticosteroids, antimicrobial mouthwashes, and topical anesthetics.¹⁰

Evidence summary

Six studies have evaluated SUA levels in patients with acute gout. Despite variations in diagnostic approach (clinical criteria vs synovial crystal analysis) and definitions of normal SUA (based on laboratory methods and sex), all 6 studies found normal levels in 11% to 49% of patients with acute gout (TABLE 1).

TABLE 1
Serum uric acid and acute gout: The evidence

Elevated SUA can be an indicator of gout—or not
A prospective cohort study of 82 patients at a Veterans Administration rheumatology clinic found elevated SUA to be the most sensitive indicator among various clinical criteria for diagnosing acute gout. However, 3 (11%) of the 28 patients who had crystal-proven gout also had a normal SUA.¹

A second prospective cohort study that evaluated 38 patients during 42 episodes of acute gout in various clinical settings reported a normal SUA in 43% of patients diagnosed on clinical grounds or by joint aspiration.²

Some patients become hyperuricemic after diagnosis
The largest retrospective cohort study evaluated 226 Korean inpatients with acute gout diagnosed either by synovial crystals or American College of Rheumatology (ACR) criteria (TABLE 2). It found that 12% (27) had a normal SUA at diagnosis. Interestingly, 81% became hyperuricemic some time after diagnosis.³

TABLE 2
American College of Rheumatology criteria for classifying acute gouty arthritis

What is a normal SUA value?
Another study reviewed SUA levels in a cohort derived from 2 large prospective RCTs of etoricoxib in patients diagnosed with acute gout by crystal analysis. The proportion of patients with a normal SUA varied substantially according to the definition of a normal value: 32% were normal using a value of 0.48 mmol/L; 11% had normal SUA levels when 0.36 mmol/L was used as the cutoff.⁴

A secondary analysis evaluated the effect of allopurinol on SUA. The proportion of patients on allopurinol with a normal SUA level compared with patients not taking allopurinol was 49% vs 29% using the higher normal cutoff value, and 29% vs 11% using the lower normal value (P<.001).⁴

Two studies find many gout patients with a normal SUA
A Japanese retrospective cohort study using ACR criteria found that nearly half of patients diagnosed with acute gout had a normal SUA level.⁵ A 1967 retrospective examination of Framingham Heart Study data found that one-third of patients clinically diagnosed with gout had a normal level. Some of the patients hadn’t been diagnosed at the time their SUA was measured, however.⁶

Recommendations

The ACR’s 1977 criteria for diagnosing gout include hyperuricemia as one potential indicator.⁷ The European League Against Rheumatism advises that normal SUA levels may accompany crystal-proven gout because uric acid either acts as a negative acute-phase reactant or increases in renal excretion during acute episodes. They conclude that SUA has “limited diagnostic value,” especially during acute gout.⁸

Evidence summary

Six studies have evaluated SUA levels in patients with acute gout. Despite variations in diagnostic approach (clinical criteria vs synovial crystal analysis) and definitions of normal SUA (based on laboratory methods and sex), all 6 studies found normal levels in 11% to 49% of patients with acute gout (TABLE 1).

TABLE 1
Serum uric acid and acute gout: The evidence

Elevated SUA can be an indicator of gout—or not
A prospective cohort study of 82 patients at a Veterans Administration rheumatology clinic found elevated SUA to be the most sensitive indicator among various clinical criteria for diagnosing acute gout. However, 3 (11%) of the 28 patients who had crystal-proven gout also had a normal SUA.¹

A second prospective cohort study that evaluated 38 patients during 42 episodes of acute gout in various clinical settings reported a normal SUA in 43% of patients diagnosed on clinical grounds or by joint aspiration.²

Some patients become hyperuricemic after diagnosis
The largest retrospective cohort study evaluated 226 Korean inpatients with acute gout diagnosed either by synovial crystals or American College of Rheumatology (ACR) criteria (TABLE 2). It found that 12% (27) had a normal SUA at diagnosis. Interestingly, 81% became hyperuricemic some time after diagnosis.³

TABLE 2
American College of Rheumatology criteria for classifying acute gouty arthritis

What is a normal SUA value?
Another study reviewed SUA levels in a cohort derived from 2 large prospective RCTs of etoricoxib in patients diagnosed with acute gout by crystal analysis. The proportion of patients with a normal SUA varied substantially according to the definition of a normal value: 32% were normal using a value of 0.48 mmol/L; 11% had normal SUA levels when 0.36 mmol/L was used as the cutoff.⁴

A secondary analysis evaluated the effect of allopurinol on SUA. The proportion of patients on allopurinol with a normal SUA level compared with patients not taking allopurinol was 49% vs 29% using the higher normal cutoff value, and 29% vs 11% using the lower normal value (P<.001).⁴

Two studies find many gout patients with a normal SUA
A Japanese retrospective cohort study using ACR criteria found that nearly half of patients diagnosed with acute gout had a normal SUA level.⁵ A 1967 retrospective examination of Framingham Heart Study data found that one-third of patients clinically diagnosed with gout had a normal level. Some of the patients hadn’t been diagnosed at the time their SUA was measured, however.⁶

Recommendations

The ACR’s 1977 criteria for diagnosing gout include hyperuricemia as one potential indicator.⁷ The European League Against Rheumatism advises that normal SUA levels may accompany crystal-proven gout because uric acid either acts as a negative acute-phase reactant or increases in renal excretion during acute episodes. They conclude that SUA has “limited diagnostic value,” especially during acute gout.⁸

Evidence summary

Six studies have evaluated SUA levels in patients with acute gout. Despite variations in diagnostic approach (clinical criteria vs synovial crystal analysis) and definitions of normal SUA (based on laboratory methods and sex), all 6 studies found normal levels in 11% to 49% of patients with acute gout (TABLE 1).

TABLE 1
Serum uric acid and acute gout: The evidence

Elevated SUA can be an indicator of gout—or not
A prospective cohort study of 82 patients at a Veterans Administration rheumatology clinic found elevated SUA to be the most sensitive indicator among various clinical criteria for diagnosing acute gout. However, 3 (11%) of the 28 patients who had crystal-proven gout also had a normal SUA.¹

A second prospective cohort study that evaluated 38 patients during 42 episodes of acute gout in various clinical settings reported a normal SUA in 43% of patients diagnosed on clinical grounds or by joint aspiration.²

Some patients become hyperuricemic after diagnosis
The largest retrospective cohort study evaluated 226 Korean inpatients with acute gout diagnosed either by synovial crystals or American College of Rheumatology (ACR) criteria (TABLE 2). It found that 12% (27) had a normal SUA at diagnosis. Interestingly, 81% became hyperuricemic some time after diagnosis.³

TABLE 2
American College of Rheumatology criteria for classifying acute gouty arthritis

What is a normal SUA value?
Another study reviewed SUA levels in a cohort derived from 2 large prospective RCTs of etoricoxib in patients diagnosed with acute gout by crystal analysis. The proportion of patients with a normal SUA varied substantially according to the definition of a normal value: 32% were normal using a value of 0.48 mmol/L; 11% had normal SUA levels when 0.36 mmol/L was used as the cutoff.⁴

A secondary analysis evaluated the effect of allopurinol on SUA. The proportion of patients on allopurinol with a normal SUA level compared with patients not taking allopurinol was 49% vs 29% using the higher normal cutoff value, and 29% vs 11% using the lower normal value (P<.001).⁴

Two studies find many gout patients with a normal SUA
A Japanese retrospective cohort study using ACR criteria found that nearly half of patients diagnosed with acute gout had a normal SUA level.⁵ A 1967 retrospective examination of Framingham Heart Study data found that one-third of patients clinically diagnosed with gout had a normal level. Some of the patients hadn’t been diagnosed at the time their SUA was measured, however.⁶

Recommendations

The ACR’s 1977 criteria for diagnosing gout include hyperuricemia as one potential indicator.⁷ The European League Against Rheumatism advises that normal SUA levels may accompany crystal-proven gout because uric acid either acts as a negative acute-phase reactant or increases in renal excretion during acute episodes. They conclude that SUA has “limited diagnostic value,” especially during acute gout.⁸

Evidence summary

An RCT compared breast emptying every 6 hours with no treatment for inflammatory breast symptoms in 213 women with 339 inflamed breasts.¹ Investigators classified symptoms into 3 groups based on milk leukocyte counts (MLC) and cultures from expressed milk (MC): infectious mastitis (MLC >10⁶/mL, MC >10³ bacteria/mL; n=165 breasts), noninfectious mastitis (MLC >10⁶/mL, MC <10³ bacteria/mL; n=48 breasts), and milk stasis, or inspissated milk (MLC <10⁶/mL, MC <10³ bacteria/mL; n=126 breasts).

Breast emptying reduced the mean duration of symptoms in women with infectious mastitis (4.2 vs 6.7 days with no treatment; P<.001) and noninfectious mastitis (3.2 vs 7.9 days with no treatment; P<.001). However, it didn’t shorten mean symptom duration in women with milk stasis (2.1 vs 2.3 days with no treatment; P not significant).¹

Moreover, breast emptying allowed more women with infectious and noninfectious mastitis to return to normal lactation within 2 weeks. Rates of return to lactation for women with infectious mastitis were 51% with breast emptying compared with 15% with no treatment (number needed to treat [NNT]=2; 95% confidence interval [CI], 2-5). For women with noninfectious mastitis, the rates of return to lactation within 2 weeks were 96% with breast emptying compared with 21% with no treatment (NNT=1; 95% CI, 1-2).¹

Antibiotics plus breast emptying work better than emptying alone
The investigators further randomized a subgroup of women (165 involved breasts) with culture-positive mastitis into 3 treatment groups, each with 55 women: culture-directed antibiotics for 6 days plus breast emptying every 6 hours, breast emptying alone, and no treatment.

Antibiotics improved the rate of return to normal lactation over breast emptying alone (96% vs 51% normal lactation at 2 weeks; NNT=3; 95% CI, 2-3) and no treatment (96% vs 15% normal lactation at 2 weeks, NNT=1; 95% CI, 1-2). They also reduced the mean duration of symptoms (2.1 days with antibiotics vs 4.2 days with breast emptying alone and 6.7 days with no treatment; P<.001 for each).¹

Amoxicillin and cephradine produce similar results
A smaller RCT (N=25) compared oral amoxicillin with oral cephradine for women with a clinical diagnosis of mastitis based on oral temperature above 37.6°C, breast tenderness, and erythema.² Investigators prescribed 7 days of amoxicillin (500 mg every 8 h; n=13) or cephradine (500 mg every 6 h; n=12) and instructed women to continue breastfeeding and apply warm, moist compresses to the involved breast every 4 to 6 hours. They also performed milk cultures on all women. The cultures grew penicillin-resistant staphylococci (15 women; 7 treated with amoxicillin and 8 with cephradine) and penicillin-sensitive streptococci (6 women).

After 7 days, the investigators found no significant differences in fever, breast tenderness, or erythema between the 2 groups (relative risk=0.85; 95% CI, 0.65-1.12, favoring cephradine).^2,3 Two treatment failures occurred in the amoxicillin-treated group, both in women who had positive cultures for Staphylococcus aureus.²

A Cochrane systematic review of the 2 RCTs described here concluded that insufficient evidence exists to support or refute antibiotic therapy for treating lactational mastitis.³

Recommendations

The World Health Organization (WHO) recommends continued breastfeeding and improving breastfeeding technique for women with mastitis. WHO advises ensuring proper infant attachment, frequent breastfeeding, and expression of breast milk by hand or pump, if necessary. They urge clinicians to promote continued breastfeeding by providing encouragement and reassurance about its value.⁴

The Academy of Breastfeeding Medicine (ABM) recommends anti-inflammatory medications for analgesia to allow women with mastitis to continue to breastfeed. Hot packs or a hot shower also may alleviate symptoms.⁵

WHO and ABM both recommend prescribing a 14-day course of antibiotics effective against S aureus for women whose symptoms don’t improve after breast emptying for 12 to 24 hours.^4,5

Evidence summary

An RCT compared breast emptying every 6 hours with no treatment for inflammatory breast symptoms in 213 women with 339 inflamed breasts.¹ Investigators classified symptoms into 3 groups based on milk leukocyte counts (MLC) and cultures from expressed milk (MC): infectious mastitis (MLC >10⁶/mL, MC >10³ bacteria/mL; n=165 breasts), noninfectious mastitis (MLC >10⁶/mL, MC <10³ bacteria/mL; n=48 breasts), and milk stasis, or inspissated milk (MLC <10⁶/mL, MC <10³ bacteria/mL; n=126 breasts).

Breast emptying reduced the mean duration of symptoms in women with infectious mastitis (4.2 vs 6.7 days with no treatment; P<.001) and noninfectious mastitis (3.2 vs 7.9 days with no treatment; P<.001). However, it didn’t shorten mean symptom duration in women with milk stasis (2.1 vs 2.3 days with no treatment; P not significant).¹

Moreover, breast emptying allowed more women with infectious and noninfectious mastitis to return to normal lactation within 2 weeks. Rates of return to lactation for women with infectious mastitis were 51% with breast emptying compared with 15% with no treatment (number needed to treat [NNT]=2; 95% confidence interval [CI], 2-5). For women with noninfectious mastitis, the rates of return to lactation within 2 weeks were 96% with breast emptying compared with 21% with no treatment (NNT=1; 95% CI, 1-2).¹

Antibiotics plus breast emptying work better than emptying alone
The investigators further randomized a subgroup of women (165 involved breasts) with culture-positive mastitis into 3 treatment groups, each with 55 women: culture-directed antibiotics for 6 days plus breast emptying every 6 hours, breast emptying alone, and no treatment.

Antibiotics improved the rate of return to normal lactation over breast emptying alone (96% vs 51% normal lactation at 2 weeks; NNT=3; 95% CI, 2-3) and no treatment (96% vs 15% normal lactation at 2 weeks, NNT=1; 95% CI, 1-2). They also reduced the mean duration of symptoms (2.1 days with antibiotics vs 4.2 days with breast emptying alone and 6.7 days with no treatment; P<.001 for each).¹

Amoxicillin and cephradine produce similar results
A smaller RCT (N=25) compared oral amoxicillin with oral cephradine for women with a clinical diagnosis of mastitis based on oral temperature above 37.6°C, breast tenderness, and erythema.² Investigators prescribed 7 days of amoxicillin (500 mg every 8 h; n=13) or cephradine (500 mg every 6 h; n=12) and instructed women to continue breastfeeding and apply warm, moist compresses to the involved breast every 4 to 6 hours. They also performed milk cultures on all women. The cultures grew penicillin-resistant staphylococci (15 women; 7 treated with amoxicillin and 8 with cephradine) and penicillin-sensitive streptococci (6 women).

After 7 days, the investigators found no significant differences in fever, breast tenderness, or erythema between the 2 groups (relative risk=0.85; 95% CI, 0.65-1.12, favoring cephradine).^2,3 Two treatment failures occurred in the amoxicillin-treated group, both in women who had positive cultures for Staphylococcus aureus.²

A Cochrane systematic review of the 2 RCTs described here concluded that insufficient evidence exists to support or refute antibiotic therapy for treating lactational mastitis.³

Recommendations

The World Health Organization (WHO) recommends continued breastfeeding and improving breastfeeding technique for women with mastitis. WHO advises ensuring proper infant attachment, frequent breastfeeding, and expression of breast milk by hand or pump, if necessary. They urge clinicians to promote continued breastfeeding by providing encouragement and reassurance about its value.⁴

The Academy of Breastfeeding Medicine (ABM) recommends anti-inflammatory medications for analgesia to allow women with mastitis to continue to breastfeed. Hot packs or a hot shower also may alleviate symptoms.⁵

WHO and ABM both recommend prescribing a 14-day course of antibiotics effective against S aureus for women whose symptoms don’t improve after breast emptying for 12 to 24 hours.^4,5

Evidence summary

An RCT compared breast emptying every 6 hours with no treatment for inflammatory breast symptoms in 213 women with 339 inflamed breasts.¹ Investigators classified symptoms into 3 groups based on milk leukocyte counts (MLC) and cultures from expressed milk (MC): infectious mastitis (MLC >10⁶/mL, MC >10³ bacteria/mL; n=165 breasts), noninfectious mastitis (MLC >10⁶/mL, MC <10³ bacteria/mL; n=48 breasts), and milk stasis, or inspissated milk (MLC <10⁶/mL, MC <10³ bacteria/mL; n=126 breasts).

Breast emptying reduced the mean duration of symptoms in women with infectious mastitis (4.2 vs 6.7 days with no treatment; P<.001) and noninfectious mastitis (3.2 vs 7.9 days with no treatment; P<.001). However, it didn’t shorten mean symptom duration in women with milk stasis (2.1 vs 2.3 days with no treatment; P not significant).¹

Moreover, breast emptying allowed more women with infectious and noninfectious mastitis to return to normal lactation within 2 weeks. Rates of return to lactation for women with infectious mastitis were 51% with breast emptying compared with 15% with no treatment (number needed to treat [NNT]=2; 95% confidence interval [CI], 2-5). For women with noninfectious mastitis, the rates of return to lactation within 2 weeks were 96% with breast emptying compared with 21% with no treatment (NNT=1; 95% CI, 1-2).¹

Antibiotics plus breast emptying work better than emptying alone
The investigators further randomized a subgroup of women (165 involved breasts) with culture-positive mastitis into 3 treatment groups, each with 55 women: culture-directed antibiotics for 6 days plus breast emptying every 6 hours, breast emptying alone, and no treatment.

Antibiotics improved the rate of return to normal lactation over breast emptying alone (96% vs 51% normal lactation at 2 weeks; NNT=3; 95% CI, 2-3) and no treatment (96% vs 15% normal lactation at 2 weeks, NNT=1; 95% CI, 1-2). They also reduced the mean duration of symptoms (2.1 days with antibiotics vs 4.2 days with breast emptying alone and 6.7 days with no treatment; P<.001 for each).¹

Amoxicillin and cephradine produce similar results
A smaller RCT (N=25) compared oral amoxicillin with oral cephradine for women with a clinical diagnosis of mastitis based on oral temperature above 37.6°C, breast tenderness, and erythema.² Investigators prescribed 7 days of amoxicillin (500 mg every 8 h; n=13) or cephradine (500 mg every 6 h; n=12) and instructed women to continue breastfeeding and apply warm, moist compresses to the involved breast every 4 to 6 hours. They also performed milk cultures on all women. The cultures grew penicillin-resistant staphylococci (15 women; 7 treated with amoxicillin and 8 with cephradine) and penicillin-sensitive streptococci (6 women).

After 7 days, the investigators found no significant differences in fever, breast tenderness, or erythema between the 2 groups (relative risk=0.85; 95% CI, 0.65-1.12, favoring cephradine).^2,3 Two treatment failures occurred in the amoxicillin-treated group, both in women who had positive cultures for Staphylococcus aureus.²

A Cochrane systematic review of the 2 RCTs described here concluded that insufficient evidence exists to support or refute antibiotic therapy for treating lactational mastitis.³

Recommendations

The World Health Organization (WHO) recommends continued breastfeeding and improving breastfeeding technique for women with mastitis. WHO advises ensuring proper infant attachment, frequent breastfeeding, and expression of breast milk by hand or pump, if necessary. They urge clinicians to promote continued breastfeeding by providing encouragement and reassurance about its value.⁴

The Academy of Breastfeeding Medicine (ABM) recommends anti-inflammatory medications for analgesia to allow women with mastitis to continue to breastfeed. Hot packs or a hot shower also may alleviate symptoms.⁵

WHO and ABM both recommend prescribing a 14-day course of antibiotics effective against S aureus for women whose symptoms don’t improve after breast emptying for 12 to 24 hours.^4,5

Evidence summary

The TABLE summarizes the results of 6 RCTs that evaluated physician counseling for weight loss. The largest RCT, which included patients with elevated serum low-density lipoprotein levels (>75th percentile), randomized participants to 3 groups: physician counseling plus office support (dietary assessment tools, counseling algorithms, and in-office prompts), physician counseling alone, or usual care.¹

Patients who received physician counseling with office support lost 2.3 kg (P<.001 vs usual care), whereas patients who received physician counseling alone lost 1.0 kg and patients who received usual care didn’t lose any weight.

TABLE
The effectiveness of weight loss counseling by physicians: What the RCTs reveal

Other large studies show mixed results
The second largest RCT randomized participants from community health centers in Colorado to receive either physician counseling (in which physicians reviewed nutritional and physical activity goals generated by a computer in response to a survey) or usual care (patient handouts alone).² Although the physician-counseled group didn’t lose more total weight, more people in this group had lost 2.7 kg or more at the 12-month follow-up (32% vs 19% for usual care; P=.006).

The third largest RCT assigned low-income women from primary care clinics in Louisiana to either a 6-month tailored weight loss intervention or usual care.³ The intervention included monthly 15-minute visits with physician counseling about weight loss, fat intake, physical activity, barriers to weight loss, and weight loss maintenance. Women who received counseling lost 1.5 kg at the 9-month follow-up compared with a loss of 0.6 kg for women who received usual care. Both groups showed no net loss at the 12- and 18-month follow-up.

Counseling with follow-up leads to drop in BMI
Physician counseling in an Italian RCT included a 1-minute, patient-centered assessment of readiness for change, 2 to 5 minutes of exercise counseling by a physician for patients in active and maintenance stages, and phone or mail follow-up at 2 to 3 weeks.⁴ The reported decreases in body mass index (BMI) in the counseling group would translate to 2.4 and 1.0 kg of weight loss for men and women of average height, respectively.

No significant weight loss in a pediatric study
An RCT of children brought to a pediatric clinic for well-child visits recruited children who had either a BMI in the 85th to 95th percentile or obese parents (BMI ≥30 kg/m²).⁵ Parents were randomized to intensive counseling, minimal counseling, or usual care. The intensive intervention group participated in a 10- to 15-minute motivational interview with the pediatrician, followed by 2 45-minute sessions with a dietician at months 1 and 3 of the 6-month program; the minimal intervention group only participated in the motivational interview. No significant weight loss occurred in any of the 3 study groups.

The smallest RCT compared counseling by a family medicine resident with usual care in 30 adult patients.⁶ At 6 months, the counseling group had lost 0.9 kg compared with a gain of 1.3 kg in the usual care group, but follow-up at 12 months found no difference between the groups.

Recommendations

The US Preventive Services Task Force (USPSTF) says that intensive counseling (person-to-person meetings at least monthly, combined with diet, exercise, and behavioral interventions plus longer-term maintenance) can promote modest sustained weight loss and improve clinical outcomes.⁷ They recommend screening adults for obesity and offering intensive counseling and behavioral interventions for obese adults.

USPSTF notes, however, that evidence is insufficient to recommend for or against low- or moderate-intensity counseling and behavioral interventions in obese or overweight adults because the trials showed mixed results, typically had small sample sizes and high dropout rates, and reported average weight change rather than frequency of response.⁸

Evidence summary

The TABLE summarizes the results of 6 RCTs that evaluated physician counseling for weight loss. The largest RCT, which included patients with elevated serum low-density lipoprotein levels (>75th percentile), randomized participants to 3 groups: physician counseling plus office support (dietary assessment tools, counseling algorithms, and in-office prompts), physician counseling alone, or usual care.¹

Patients who received physician counseling with office support lost 2.3 kg (P<.001 vs usual care), whereas patients who received physician counseling alone lost 1.0 kg and patients who received usual care didn’t lose any weight.

TABLE
The effectiveness of weight loss counseling by physicians: What the RCTs reveal

Other large studies show mixed results
The second largest RCT randomized participants from community health centers in Colorado to receive either physician counseling (in which physicians reviewed nutritional and physical activity goals generated by a computer in response to a survey) or usual care (patient handouts alone).² Although the physician-counseled group didn’t lose more total weight, more people in this group had lost 2.7 kg or more at the 12-month follow-up (32% vs 19% for usual care; P=.006).

The third largest RCT assigned low-income women from primary care clinics in Louisiana to either a 6-month tailored weight loss intervention or usual care.³ The intervention included monthly 15-minute visits with physician counseling about weight loss, fat intake, physical activity, barriers to weight loss, and weight loss maintenance. Women who received counseling lost 1.5 kg at the 9-month follow-up compared with a loss of 0.6 kg for women who received usual care. Both groups showed no net loss at the 12- and 18-month follow-up.

Counseling with follow-up leads to drop in BMI
Physician counseling in an Italian RCT included a 1-minute, patient-centered assessment of readiness for change, 2 to 5 minutes of exercise counseling by a physician for patients in active and maintenance stages, and phone or mail follow-up at 2 to 3 weeks.⁴ The reported decreases in body mass index (BMI) in the counseling group would translate to 2.4 and 1.0 kg of weight loss for men and women of average height, respectively.

No significant weight loss in a pediatric study
An RCT of children brought to a pediatric clinic for well-child visits recruited children who had either a BMI in the 85th to 95th percentile or obese parents (BMI ≥30 kg/m²).⁵ Parents were randomized to intensive counseling, minimal counseling, or usual care. The intensive intervention group participated in a 10- to 15-minute motivational interview with the pediatrician, followed by 2 45-minute sessions with a dietician at months 1 and 3 of the 6-month program; the minimal intervention group only participated in the motivational interview. No significant weight loss occurred in any of the 3 study groups.

The smallest RCT compared counseling by a family medicine resident with usual care in 30 adult patients.⁶ At 6 months, the counseling group had lost 0.9 kg compared with a gain of 1.3 kg in the usual care group, but follow-up at 12 months found no difference between the groups.

Recommendations

The US Preventive Services Task Force (USPSTF) says that intensive counseling (person-to-person meetings at least monthly, combined with diet, exercise, and behavioral interventions plus longer-term maintenance) can promote modest sustained weight loss and improve clinical outcomes.⁷ They recommend screening adults for obesity and offering intensive counseling and behavioral interventions for obese adults.

USPSTF notes, however, that evidence is insufficient to recommend for or against low- or moderate-intensity counseling and behavioral interventions in obese or overweight adults because the trials showed mixed results, typically had small sample sizes and high dropout rates, and reported average weight change rather than frequency of response.⁸

Evidence summary

The TABLE summarizes the results of 6 RCTs that evaluated physician counseling for weight loss. The largest RCT, which included patients with elevated serum low-density lipoprotein levels (>75th percentile), randomized participants to 3 groups: physician counseling plus office support (dietary assessment tools, counseling algorithms, and in-office prompts), physician counseling alone, or usual care.¹

Patients who received physician counseling with office support lost 2.3 kg (P<.001 vs usual care), whereas patients who received physician counseling alone lost 1.0 kg and patients who received usual care didn’t lose any weight.

TABLE
The effectiveness of weight loss counseling by physicians: What the RCTs reveal

Other large studies show mixed results
The second largest RCT randomized participants from community health centers in Colorado to receive either physician counseling (in which physicians reviewed nutritional and physical activity goals generated by a computer in response to a survey) or usual care (patient handouts alone).² Although the physician-counseled group didn’t lose more total weight, more people in this group had lost 2.7 kg or more at the 12-month follow-up (32% vs 19% for usual care; P=.006).

The third largest RCT assigned low-income women from primary care clinics in Louisiana to either a 6-month tailored weight loss intervention or usual care.³ The intervention included monthly 15-minute visits with physician counseling about weight loss, fat intake, physical activity, barriers to weight loss, and weight loss maintenance. Women who received counseling lost 1.5 kg at the 9-month follow-up compared with a loss of 0.6 kg for women who received usual care. Both groups showed no net loss at the 12- and 18-month follow-up.

Counseling with follow-up leads to drop in BMI
Physician counseling in an Italian RCT included a 1-minute, patient-centered assessment of readiness for change, 2 to 5 minutes of exercise counseling by a physician for patients in active and maintenance stages, and phone or mail follow-up at 2 to 3 weeks.⁴ The reported decreases in body mass index (BMI) in the counseling group would translate to 2.4 and 1.0 kg of weight loss for men and women of average height, respectively.

No significant weight loss in a pediatric study
An RCT of children brought to a pediatric clinic for well-child visits recruited children who had either a BMI in the 85th to 95th percentile or obese parents (BMI ≥30 kg/m²).⁵ Parents were randomized to intensive counseling, minimal counseling, or usual care. The intensive intervention group participated in a 10- to 15-minute motivational interview with the pediatrician, followed by 2 45-minute sessions with a dietician at months 1 and 3 of the 6-month program; the minimal intervention group only participated in the motivational interview. No significant weight loss occurred in any of the 3 study groups.

The smallest RCT compared counseling by a family medicine resident with usual care in 30 adult patients.⁶ At 6 months, the counseling group had lost 0.9 kg compared with a gain of 1.3 kg in the usual care group, but follow-up at 12 months found no difference between the groups.

Recommendations

The US Preventive Services Task Force (USPSTF) says that intensive counseling (person-to-person meetings at least monthly, combined with diet, exercise, and behavioral interventions plus longer-term maintenance) can promote modest sustained weight loss and improve clinical outcomes.⁷ They recommend screening adults for obesity and offering intensive counseling and behavioral interventions for obese adults.

USPSTF notes, however, that evidence is insufficient to recommend for or against low- or moderate-intensity counseling and behavioral interventions in obese or overweight adults because the trials showed mixed results, typically had small sample sizes and high dropout rates, and reported average weight change rather than frequency of response.⁸

Evidence summary

One systematic review found increased risk of CDAD in patients taking cephalosporins, penicillin, or clindamycin (TABLE 1).¹ A subsequent retrospective cohort investigation of 5619 patients during a CDAD epidemic in Quebec, Canada reported that quinolone antibiotics were most strongly associated with CDAD, whereas other antibiotics posed an intermediate risk.²

A prospective cohort study of 101,796 admissions over a 5-year period at a tertiary medical center defined a group of high-risk antibiotics before starting research.³ They included fluoroquinolones, cephalosporins, intravenous β-lactam/β-lactamase inhibitors, macrolides, clindamycin, and carbapenems. All other antibiotics were considered low risk. High-risk antibiotics were associated with a 3-fold increase in CDAD compared with low-risk drugs (odds ratio [OR]=3.37; 95% confidence interval [CI], 2.64-4.31); number needed to harm [NNH]= 10).³

TABLE 1
Medications associated with C difficile diarrhea

The number of antibiotics is a factor
The number of antibiotics used also may influence the risk of CDAD. A retrospective cohort of 2859 patients from a community hospital found that an increased number of antibiotics was a risk factor for CDAD (OR=1.49; 95% CI, 1.23-1.81; NNH=44).⁴ Another retrospective cohort study of 1187 inpatients at a Montreal hospital found 3 or more antibiotics increased the risk (adjusted OR=2.1; 95% CI, 1.3-3.4; NNH=20).⁵

Hospital risks: Proximity to an infected patient, length of stay
A prospective cohort of 252 patients and a retrospective cohort of 1187 patients show that recent hospitalization puts patients at risk for CDAD (TABLE 2).^5,6 Several retrospective cohort studies have shown that patients in close proximity to a C difficile-positive patient in the hospital (roommate, neighbor, or subsequent tenant) are at risk for CDAD.^4,7

Length of hospitalization is also a risk factor.² A retrospective cohort study of 2859 patients found that patients with CDAD had spent more time in the hospital—a mean of 19 days compared with 8 days for patients without diarrhea (P<.001).⁴ A prospective cohort study of 101,796 admissions reported that the mean length of stay was 15 days (range=8.0-26.0) for CDAD patients compared with 5 days (range=3.0-8.0) for patients without CDAD (P<.001).³

TABLE 2
Hospital risk factors for C difficile diarrhea

Patient risk factors: Age and comorbid disease

Two cohort studies found that CDAD patients were about 10 years older than patients without CDAD.^3,8 Among 535 patients in Jerusalem, patients positive for C difficile toxin had a mean age of 76±20 years compared with 66±26 years in toxin-negative patients (P<.001).⁸ In the previously mentioned study of 101,796 patients, the average age for patients with CDAD was 65.4±16.9 years compared with 56.5±19.9 years for patients without CDAD (P<.001).³

The patients in this study also showed significant associations between CDAD and comorbid conditions, including myocardial infarction, heart failure, chronic pulmonary disease, peripheral vascular disease, complicated diabetes, fluid and electrolyte disorders, chronic renal failure, cancer, coagulopathy, and methicillin-resistant Staphylococcus aureus infection.³

Acid suppression therapy is another risk
A systematic review that included a total of 2948 patients in 12 studies (cross-sectional, case-control, and cohort) evaluated acid suppression therapy and found an association between CDAD and use of histamine₂-receptor antagonists (H₂RAs) (OR=1.48; 95% CI, 1.06-2.06; NNH=45) and between CDAD and proton-pump inhibitors (PPIs) (OR=2.05; 95% CI, 1.47-2.85; NNH=21).⁹ Significant heterogeneity among the studies limited the interpretation of results, however.

The prospective cohort study of 101,796 patients also reported an increased risk of CDAD with H₂RAs and PPIs.³ The risk of CDAD rose with progression from no acid suppression to H₂RA use to daily PPI use to more frequent PPI use.³ Another cohort study of 1187 patients found an association between PPIs and CDAD (adjusted OR=2.1; 95% CI, 1.2-3.5).⁵

Using a score to gauge risk
Researchers studying a cohort of 54,226 patients developed a risk score using clinical characteristics associated with CDAD.¹⁰ The patients were older than 18 years, hospitalized longer than 48 hours, and had received broad spectrum antibiotics (intravenous glycopeptides, fluoroquinolones, penicillins, cephalosporins, or carbapenems). When the researchers tested their clinical risk index on a validation cohort of 13,002 patients, they found that increasing scores were significantly associated with increasing risk for C difficile colitis (OR=3.31; 95% CI, 2.61-4.91; area under the receiver operating characteristic curve=0.712).¹⁰

Recommendations

Clinical practice guidelines by the Society for Healthcare Epidemiology of America and the Infectious Diseases Society of America recommend minimizing the frequency and duration of antibiotics and the total number of antibiotics used.¹¹ They also suggest private rooms, chlorine cleaning products, and contact precautions (gloves, hand hygiene, and disposable thermometers) to reduce risk.

The authors of the guidelines propose antimicrobial stewardship programs based on the local epidemiology of C difficile strains, including restricted use of cephalosporins and clindamycin, except for surgical prophylaxis.

Evidence summary

One systematic review found increased risk of CDAD in patients taking cephalosporins, penicillin, or clindamycin (TABLE 1).¹ A subsequent retrospective cohort investigation of 5619 patients during a CDAD epidemic in Quebec, Canada reported that quinolone antibiotics were most strongly associated with CDAD, whereas other antibiotics posed an intermediate risk.²

A prospective cohort study of 101,796 admissions over a 5-year period at a tertiary medical center defined a group of high-risk antibiotics before starting research.³ They included fluoroquinolones, cephalosporins, intravenous β-lactam/β-lactamase inhibitors, macrolides, clindamycin, and carbapenems. All other antibiotics were considered low risk. High-risk antibiotics were associated with a 3-fold increase in CDAD compared with low-risk drugs (odds ratio [OR]=3.37; 95% confidence interval [CI], 2.64-4.31); number needed to harm [NNH]= 10).³

TABLE 1
Medications associated with C difficile diarrhea

The number of antibiotics is a factor
The number of antibiotics used also may influence the risk of CDAD. A retrospective cohort of 2859 patients from a community hospital found that an increased number of antibiotics was a risk factor for CDAD (OR=1.49; 95% CI, 1.23-1.81; NNH=44).⁴ Another retrospective cohort study of 1187 inpatients at a Montreal hospital found 3 or more antibiotics increased the risk (adjusted OR=2.1; 95% CI, 1.3-3.4; NNH=20).⁵

Hospital risks: Proximity to an infected patient, length of stay
A prospective cohort of 252 patients and a retrospective cohort of 1187 patients show that recent hospitalization puts patients at risk for CDAD (TABLE 2).^5,6 Several retrospective cohort studies have shown that patients in close proximity to a C difficile-positive patient in the hospital (roommate, neighbor, or subsequent tenant) are at risk for CDAD.^4,7

Length of hospitalization is also a risk factor.² A retrospective cohort study of 2859 patients found that patients with CDAD had spent more time in the hospital—a mean of 19 days compared with 8 days for patients without diarrhea (P<.001).⁴ A prospective cohort study of 101,796 admissions reported that the mean length of stay was 15 days (range=8.0-26.0) for CDAD patients compared with 5 days (range=3.0-8.0) for patients without CDAD (P<.001).³

TABLE 2
Hospital risk factors for C difficile diarrhea

Patient risk factors: Age and comorbid disease

Two cohort studies found that CDAD patients were about 10 years older than patients without CDAD.^3,8 Among 535 patients in Jerusalem, patients positive for C difficile toxin had a mean age of 76±20 years compared with 66±26 years in toxin-negative patients (P<.001).⁸ In the previously mentioned study of 101,796 patients, the average age for patients with CDAD was 65.4±16.9 years compared with 56.5±19.9 years for patients without CDAD (P<.001).³

The patients in this study also showed significant associations between CDAD and comorbid conditions, including myocardial infarction, heart failure, chronic pulmonary disease, peripheral vascular disease, complicated diabetes, fluid and electrolyte disorders, chronic renal failure, cancer, coagulopathy, and methicillin-resistant Staphylococcus aureus infection.³

Acid suppression therapy is another risk
A systematic review that included a total of 2948 patients in 12 studies (cross-sectional, case-control, and cohort) evaluated acid suppression therapy and found an association between CDAD and use of histamine₂-receptor antagonists (H₂RAs) (OR=1.48; 95% CI, 1.06-2.06; NNH=45) and between CDAD and proton-pump inhibitors (PPIs) (OR=2.05; 95% CI, 1.47-2.85; NNH=21).⁹ Significant heterogeneity among the studies limited the interpretation of results, however.

The prospective cohort study of 101,796 patients also reported an increased risk of CDAD with H₂RAs and PPIs.³ The risk of CDAD rose with progression from no acid suppression to H₂RA use to daily PPI use to more frequent PPI use.³ Another cohort study of 1187 patients found an association between PPIs and CDAD (adjusted OR=2.1; 95% CI, 1.2-3.5).⁵

Using a score to gauge risk
Researchers studying a cohort of 54,226 patients developed a risk score using clinical characteristics associated with CDAD.¹⁰ The patients were older than 18 years, hospitalized longer than 48 hours, and had received broad spectrum antibiotics (intravenous glycopeptides, fluoroquinolones, penicillins, cephalosporins, or carbapenems). When the researchers tested their clinical risk index on a validation cohort of 13,002 patients, they found that increasing scores were significantly associated with increasing risk for C difficile colitis (OR=3.31; 95% CI, 2.61-4.91; area under the receiver operating characteristic curve=0.712).¹⁰

Recommendations

Clinical practice guidelines by the Society for Healthcare Epidemiology of America and the Infectious Diseases Society of America recommend minimizing the frequency and duration of antibiotics and the total number of antibiotics used.¹¹ They also suggest private rooms, chlorine cleaning products, and contact precautions (gloves, hand hygiene, and disposable thermometers) to reduce risk.

The authors of the guidelines propose antimicrobial stewardship programs based on the local epidemiology of C difficile strains, including restricted use of cephalosporins and clindamycin, except for surgical prophylaxis.

Evidence summary

One systematic review found increased risk of CDAD in patients taking cephalosporins, penicillin, or clindamycin (TABLE 1).¹ A subsequent retrospective cohort investigation of 5619 patients during a CDAD epidemic in Quebec, Canada reported that quinolone antibiotics were most strongly associated with CDAD, whereas other antibiotics posed an intermediate risk.²

A prospective cohort study of 101,796 admissions over a 5-year period at a tertiary medical center defined a group of high-risk antibiotics before starting research.³ They included fluoroquinolones, cephalosporins, intravenous β-lactam/β-lactamase inhibitors, macrolides, clindamycin, and carbapenems. All other antibiotics were considered low risk. High-risk antibiotics were associated with a 3-fold increase in CDAD compared with low-risk drugs (odds ratio [OR]=3.37; 95% confidence interval [CI], 2.64-4.31); number needed to harm [NNH]= 10).³

TABLE 1
Medications associated with C difficile diarrhea

The number of antibiotics is a factor
The number of antibiotics used also may influence the risk of CDAD. A retrospective cohort of 2859 patients from a community hospital found that an increased number of antibiotics was a risk factor for CDAD (OR=1.49; 95% CI, 1.23-1.81; NNH=44).⁴ Another retrospective cohort study of 1187 inpatients at a Montreal hospital found 3 or more antibiotics increased the risk (adjusted OR=2.1; 95% CI, 1.3-3.4; NNH=20).⁵

Hospital risks: Proximity to an infected patient, length of stay
A prospective cohort of 252 patients and a retrospective cohort of 1187 patients show that recent hospitalization puts patients at risk for CDAD (TABLE 2).^5,6 Several retrospective cohort studies have shown that patients in close proximity to a C difficile-positive patient in the hospital (roommate, neighbor, or subsequent tenant) are at risk for CDAD.^4,7

Length of hospitalization is also a risk factor.² A retrospective cohort study of 2859 patients found that patients with CDAD had spent more time in the hospital—a mean of 19 days compared with 8 days for patients without diarrhea (P<.001).⁴ A prospective cohort study of 101,796 admissions reported that the mean length of stay was 15 days (range=8.0-26.0) for CDAD patients compared with 5 days (range=3.0-8.0) for patients without CDAD (P<.001).³

TABLE 2
Hospital risk factors for C difficile diarrhea

Patient risk factors: Age and comorbid disease

Two cohort studies found that CDAD patients were about 10 years older than patients without CDAD.^3,8 Among 535 patients in Jerusalem, patients positive for C difficile toxin had a mean age of 76±20 years compared with 66±26 years in toxin-negative patients (P<.001).⁸ In the previously mentioned study of 101,796 patients, the average age for patients with CDAD was 65.4±16.9 years compared with 56.5±19.9 years for patients without CDAD (P<.001).³

The patients in this study also showed significant associations between CDAD and comorbid conditions, including myocardial infarction, heart failure, chronic pulmonary disease, peripheral vascular disease, complicated diabetes, fluid and electrolyte disorders, chronic renal failure, cancer, coagulopathy, and methicillin-resistant Staphylococcus aureus infection.³

Acid suppression therapy is another risk
A systematic review that included a total of 2948 patients in 12 studies (cross-sectional, case-control, and cohort) evaluated acid suppression therapy and found an association between CDAD and use of histamine₂-receptor antagonists (H₂RAs) (OR=1.48; 95% CI, 1.06-2.06; NNH=45) and between CDAD and proton-pump inhibitors (PPIs) (OR=2.05; 95% CI, 1.47-2.85; NNH=21).⁹ Significant heterogeneity among the studies limited the interpretation of results, however.

The prospective cohort study of 101,796 patients also reported an increased risk of CDAD with H₂RAs and PPIs.³ The risk of CDAD rose with progression from no acid suppression to H₂RA use to daily PPI use to more frequent PPI use.³ Another cohort study of 1187 patients found an association between PPIs and CDAD (adjusted OR=2.1; 95% CI, 1.2-3.5).⁵

Using a score to gauge risk
Researchers studying a cohort of 54,226 patients developed a risk score using clinical characteristics associated with CDAD.¹⁰ The patients were older than 18 years, hospitalized longer than 48 hours, and had received broad spectrum antibiotics (intravenous glycopeptides, fluoroquinolones, penicillins, cephalosporins, or carbapenems). When the researchers tested their clinical risk index on a validation cohort of 13,002 patients, they found that increasing scores were significantly associated with increasing risk for C difficile colitis (OR=3.31; 95% CI, 2.61-4.91; area under the receiver operating characteristic curve=0.712).¹⁰

Recommendations

Clinical practice guidelines by the Society for Healthcare Epidemiology of America and the Infectious Diseases Society of America recommend minimizing the frequency and duration of antibiotics and the total number of antibiotics used.¹¹ They also suggest private rooms, chlorine cleaning products, and contact precautions (gloves, hand hygiene, and disposable thermometers) to reduce risk.

The authors of the guidelines propose antimicrobial stewardship programs based on the local epidemiology of C difficile strains, including restricted use of cephalosporins and clindamycin, except for surgical prophylaxis.

Evidence summary

A Cochrane review of 7 RCTs, with a total of 46,885 children, examined the effect of pneumococcal vaccine on the incidence of otitis media.¹ The authors didn’t pool the results because of large heterogeneity among the studies. Only 2 trials used the licensed 7-valent PCV Prevnar; both enrolled infants vaccinated according to the routine schedule before 12 months of age.

The largest study (37,686 children followed to 42 months of age) showed a 7.8% reduction in the number of office visits for otitis media after pneumococcal vaccination (95% CI, 5.4%-10.1%).² The incidence of ROM—defined as at least 3 episodes within 6 months or at least 4 episodes within 1 year—decreased 9% (95% CI, 4%-14%) among vaccine recipients.² The study also demonstrated a 24% decrease in the need for tympanostomy tubes (95% CI, 12%-35%).² The other 7-valent PCV trial (1662 children) showed a 6% risk reduction in AOM (95% CI, -4% to 16%) in the vaccinated group.¹

A retrospective cohort study in New York and Tennessee that enrolled 176,000 children vaccinated according to the recommended schedule starting at 2 months of age found reductions in ROM of 28% (95% CI, 11%-33%) for New York’s vaccinated cohort and 17% (95% CI, 6%-19%) for the Tennessee cohort. ROM was defined as at least 3 episodes within 6 months or at least 4 episodes within 1 year. The study also showed a 23% decrease in tympanostomy tube placement in New York (95% CI, 1%-35%) and a 16% decrease in Tennessee (95% CI, 3%-21%).³

A simplified vaccination schedule linked to fewer cases of otitis media
A single-blinded prospective cohort study of 1571 children examined the effectiveness of a simplified PCV-7 schedule. Children vaccinated with the licensed 7-valent PCV at 3, 5, and 11 months of age showed a 17% reduction in AOM (95% CI, 2%-39%) compared with unvaccinated controls.⁴

Vaccination starting at 12 months shows mixed results
Results from 3 trials that vaccinated children older than 1 year were mixed.¹ An analysis of 264 healthy 12- to 35-month-olds given a 9-valent PCV found a 17% reduction in AOM (95% CI, -2% to 33%). However, in 2 other trials, of 74 and 383 children, the vaccine didn’t decrease the incidence of AOM.¹

Children in these 2 trials were 1 to 7 years of age and all had had at least 2 episodes of otitis media in the year before enrollment. Both trials employed 2 doses of the licensed 7-valent PCV in children younger than 2 years and 1 dose of the 7-valent vaccine followed by a dose of the 23-valent vaccine in children older than 2 years.¹

Recommendations

A clinical practice guideline issued jointly by the American Academy of Pediatrics and American Academy of Family Physicians recognizes the benefit of PCV vaccination for preventing AOM.⁵ Although preventing AOM is not the primary indication, PCV vaccination at 2, 4, 6, and 12 to 15 months of age is part of the routine childhood immunization series recommended by the Advisory Committee on Immunization Practices.⁶

Evidence summary

A Cochrane review of 7 RCTs, with a total of 46,885 children, examined the effect of pneumococcal vaccine on the incidence of otitis media.¹ The authors didn’t pool the results because of large heterogeneity among the studies. Only 2 trials used the licensed 7-valent PCV Prevnar; both enrolled infants vaccinated according to the routine schedule before 12 months of age.

The largest study (37,686 children followed to 42 months of age) showed a 7.8% reduction in the number of office visits for otitis media after pneumococcal vaccination (95% CI, 5.4%-10.1%).² The incidence of ROM—defined as at least 3 episodes within 6 months or at least 4 episodes within 1 year—decreased 9% (95% CI, 4%-14%) among vaccine recipients.² The study also demonstrated a 24% decrease in the need for tympanostomy tubes (95% CI, 12%-35%).² The other 7-valent PCV trial (1662 children) showed a 6% risk reduction in AOM (95% CI, -4% to 16%) in the vaccinated group.¹

A retrospective cohort study in New York and Tennessee that enrolled 176,000 children vaccinated according to the recommended schedule starting at 2 months of age found reductions in ROM of 28% (95% CI, 11%-33%) for New York’s vaccinated cohort and 17% (95% CI, 6%-19%) for the Tennessee cohort. ROM was defined as at least 3 episodes within 6 months or at least 4 episodes within 1 year. The study also showed a 23% decrease in tympanostomy tube placement in New York (95% CI, 1%-35%) and a 16% decrease in Tennessee (95% CI, 3%-21%).³

A simplified vaccination schedule linked to fewer cases of otitis media
A single-blinded prospective cohort study of 1571 children examined the effectiveness of a simplified PCV-7 schedule. Children vaccinated with the licensed 7-valent PCV at 3, 5, and 11 months of age showed a 17% reduction in AOM (95% CI, 2%-39%) compared with unvaccinated controls.⁴

Vaccination starting at 12 months shows mixed results
Results from 3 trials that vaccinated children older than 1 year were mixed.¹ An analysis of 264 healthy 12- to 35-month-olds given a 9-valent PCV found a 17% reduction in AOM (95% CI, -2% to 33%). However, in 2 other trials, of 74 and 383 children, the vaccine didn’t decrease the incidence of AOM.¹

Children in these 2 trials were 1 to 7 years of age and all had had at least 2 episodes of otitis media in the year before enrollment. Both trials employed 2 doses of the licensed 7-valent PCV in children younger than 2 years and 1 dose of the 7-valent vaccine followed by a dose of the 23-valent vaccine in children older than 2 years.¹

Recommendations

A clinical practice guideline issued jointly by the American Academy of Pediatrics and American Academy of Family Physicians recognizes the benefit of PCV vaccination for preventing AOM.⁵ Although preventing AOM is not the primary indication, PCV vaccination at 2, 4, 6, and 12 to 15 months of age is part of the routine childhood immunization series recommended by the Advisory Committee on Immunization Practices.⁶

Evidence summary

A Cochrane review of 7 RCTs, with a total of 46,885 children, examined the effect of pneumococcal vaccine on the incidence of otitis media.¹ The authors didn’t pool the results because of large heterogeneity among the studies. Only 2 trials used the licensed 7-valent PCV Prevnar; both enrolled infants vaccinated according to the routine schedule before 12 months of age.

The largest study (37,686 children followed to 42 months of age) showed a 7.8% reduction in the number of office visits for otitis media after pneumococcal vaccination (95% CI, 5.4%-10.1%).² The incidence of ROM—defined as at least 3 episodes within 6 months or at least 4 episodes within 1 year—decreased 9% (95% CI, 4%-14%) among vaccine recipients.² The study also demonstrated a 24% decrease in the need for tympanostomy tubes (95% CI, 12%-35%).² The other 7-valent PCV trial (1662 children) showed a 6% risk reduction in AOM (95% CI, -4% to 16%) in the vaccinated group.¹

A retrospective cohort study in New York and Tennessee that enrolled 176,000 children vaccinated according to the recommended schedule starting at 2 months of age found reductions in ROM of 28% (95% CI, 11%-33%) for New York’s vaccinated cohort and 17% (95% CI, 6%-19%) for the Tennessee cohort. ROM was defined as at least 3 episodes within 6 months or at least 4 episodes within 1 year. The study also showed a 23% decrease in tympanostomy tube placement in New York (95% CI, 1%-35%) and a 16% decrease in Tennessee (95% CI, 3%-21%).³

A simplified vaccination schedule linked to fewer cases of otitis media
A single-blinded prospective cohort study of 1571 children examined the effectiveness of a simplified PCV-7 schedule. Children vaccinated with the licensed 7-valent PCV at 3, 5, and 11 months of age showed a 17% reduction in AOM (95% CI, 2%-39%) compared with unvaccinated controls.⁴

Vaccination starting at 12 months shows mixed results
Results from 3 trials that vaccinated children older than 1 year were mixed.¹ An analysis of 264 healthy 12- to 35-month-olds given a 9-valent PCV found a 17% reduction in AOM (95% CI, -2% to 33%). However, in 2 other trials, of 74 and 383 children, the vaccine didn’t decrease the incidence of AOM.¹

Children in these 2 trials were 1 to 7 years of age and all had had at least 2 episodes of otitis media in the year before enrollment. Both trials employed 2 doses of the licensed 7-valent PCV in children younger than 2 years and 1 dose of the 7-valent vaccine followed by a dose of the 23-valent vaccine in children older than 2 years.¹

Recommendations

A clinical practice guideline issued jointly by the American Academy of Pediatrics and American Academy of Family Physicians recognizes the benefit of PCV vaccination for preventing AOM.⁵ Although preventing AOM is not the primary indication, PCV vaccination at 2, 4, 6, and 12 to 15 months of age is part of the routine childhood immunization series recommended by the Advisory Committee on Immunization Practices.⁶

Evidence summary

Pregnant women without immunity to varicella are at risk of developing chickenpox, which can cause congenital varicella syndrome. An estimated 44 cases of congenital varicella occurred each year in the prevaccine era.¹

Varicella vaccine contains live attenuated virus. Approximately 2% to 3% of vaccinees develop either a localized rash around the injection site or a generalized rash.¹ The vaccine virus can, theoretically, spread from vaccinees who develop a rash to other people. Nevertheless, the probability of contracting varicella after contact with a healthy vaccinee is very low.

Minimal transmission, no infection from contact with healthy vaccinees
A prospective vaccine efficacy study found that 3 of 446 (0.67%) contacts of healthy vaccinees seroconverted, but had no clinical evidence of varicella.² In a smaller study, 30 immunocompromised siblings of 37 healthy children who received varicella vaccine showed no clinical or serological evidence of the virus.³

Five case reports document varicella infection in people who had contact with healthy vaccinees.¹ One of these was a pregnant woman who chose to terminate the pregnancy, but subsequent tests showed no virus in the fetus.⁴ We couldn’t find any reports of congenital varicella attributable to infection of the mother from a recent vaccinee.

Transmission by immunocompromised vaccinees is slightly higher
The risk of contracting vaccine-associated varicella from contact with an immunocompromised vaccinee is slightly higher than for a healthy vaccinee. The National Institute of Allergy and Infectious Diseases Varicella Vaccine Collaborative Study evaluated transmission and infectivity of the varicella vaccine virus in the close contacts of 482 vaccinated children with leukemia.⁵ One hundred fifty-six vaccinees developed a rash approximately one month after vaccination. Among 88 healthy susceptible siblings in close contact with the 156 vaccinees, 15 (17%) showed evidence of virus transmission. Of the 15, 4 had subclinical infection and the other 11 had a mild rash.

Recommendations

The American Academy of Pediatrics, Advisory Committee on Immunization Practices, and Centers for Disease Control and Prevention say that no precautions are necessary after varicella vaccination of family members in households with pregnant women. If a vaccinee develops a rash, precautions such as separating the vaccinee and the pregnant woman until the rash resolves are advisable. Giving Varicella zoster immune globulin to pregnant women without immunity who are exposed to varicella should be considered. Varicella vaccines are contraindicated in people with malignancies, immunodeficiencies (congenital or acquired), and immunosuppression caused by medications.^1,3,6,7

Evidence summary

Pregnant women without immunity to varicella are at risk of developing chickenpox, which can cause congenital varicella syndrome. An estimated 44 cases of congenital varicella occurred each year in the prevaccine era.¹

Varicella vaccine contains live attenuated virus. Approximately 2% to 3% of vaccinees develop either a localized rash around the injection site or a generalized rash.¹ The vaccine virus can, theoretically, spread from vaccinees who develop a rash to other people. Nevertheless, the probability of contracting varicella after contact with a healthy vaccinee is very low.

Minimal transmission, no infection from contact with healthy vaccinees
A prospective vaccine efficacy study found that 3 of 446 (0.67%) contacts of healthy vaccinees seroconverted, but had no clinical evidence of varicella.² In a smaller study, 30 immunocompromised siblings of 37 healthy children who received varicella vaccine showed no clinical or serological evidence of the virus.³

Five case reports document varicella infection in people who had contact with healthy vaccinees.¹ One of these was a pregnant woman who chose to terminate the pregnancy, but subsequent tests showed no virus in the fetus.⁴ We couldn’t find any reports of congenital varicella attributable to infection of the mother from a recent vaccinee.

Transmission by immunocompromised vaccinees is slightly higher
The risk of contracting vaccine-associated varicella from contact with an immunocompromised vaccinee is slightly higher than for a healthy vaccinee. The National Institute of Allergy and Infectious Diseases Varicella Vaccine Collaborative Study evaluated transmission and infectivity of the varicella vaccine virus in the close contacts of 482 vaccinated children with leukemia.⁵ One hundred fifty-six vaccinees developed a rash approximately one month after vaccination. Among 88 healthy susceptible siblings in close contact with the 156 vaccinees, 15 (17%) showed evidence of virus transmission. Of the 15, 4 had subclinical infection and the other 11 had a mild rash.

Recommendations

The American Academy of Pediatrics, Advisory Committee on Immunization Practices, and Centers for Disease Control and Prevention say that no precautions are necessary after varicella vaccination of family members in households with pregnant women. If a vaccinee develops a rash, precautions such as separating the vaccinee and the pregnant woman until the rash resolves are advisable. Giving Varicella zoster immune globulin to pregnant women without immunity who are exposed to varicella should be considered. Varicella vaccines are contraindicated in people with malignancies, immunodeficiencies (congenital or acquired), and immunosuppression caused by medications.^1,3,6,7

Evidence summary

Pregnant women without immunity to varicella are at risk of developing chickenpox, which can cause congenital varicella syndrome. An estimated 44 cases of congenital varicella occurred each year in the prevaccine era.¹

Varicella vaccine contains live attenuated virus. Approximately 2% to 3% of vaccinees develop either a localized rash around the injection site or a generalized rash.¹ The vaccine virus can, theoretically, spread from vaccinees who develop a rash to other people. Nevertheless, the probability of contracting varicella after contact with a healthy vaccinee is very low.

Minimal transmission, no infection from contact with healthy vaccinees
A prospective vaccine efficacy study found that 3 of 446 (0.67%) contacts of healthy vaccinees seroconverted, but had no clinical evidence of varicella.² In a smaller study, 30 immunocompromised siblings of 37 healthy children who received varicella vaccine showed no clinical or serological evidence of the virus.³

Five case reports document varicella infection in people who had contact with healthy vaccinees.¹ One of these was a pregnant woman who chose to terminate the pregnancy, but subsequent tests showed no virus in the fetus.⁴ We couldn’t find any reports of congenital varicella attributable to infection of the mother from a recent vaccinee.

Transmission by immunocompromised vaccinees is slightly higher
The risk of contracting vaccine-associated varicella from contact with an immunocompromised vaccinee is slightly higher than for a healthy vaccinee. The National Institute of Allergy and Infectious Diseases Varicella Vaccine Collaborative Study evaluated transmission and infectivity of the varicella vaccine virus in the close contacts of 482 vaccinated children with leukemia.⁵ One hundred fifty-six vaccinees developed a rash approximately one month after vaccination. Among 88 healthy susceptible siblings in close contact with the 156 vaccinees, 15 (17%) showed evidence of virus transmission. Of the 15, 4 had subclinical infection and the other 11 had a mild rash.

Recommendations

The American Academy of Pediatrics, Advisory Committee on Immunization Practices, and Centers for Disease Control and Prevention say that no precautions are necessary after varicella vaccination of family members in households with pregnant women. If a vaccinee develops a rash, precautions such as separating the vaccinee and the pregnant woman until the rash resolves are advisable. Giving Varicella zoster immune globulin to pregnant women without immunity who are exposed to varicella should be considered. Varicella vaccines are contraindicated in people with malignancies, immunodeficiencies (congenital or acquired), and immunosuppression caused by medications.^1,3,6,7