Nichole Zehnder, MD

Clinical question: What is the impact, and sustainability, of chlorhexidine bathing on central-venous-catheter-associated bloodstream infections?

Background: Chlorhexidine bathing has been associated with reductions in healthcare-associated bloodstream infections, including vancomycin-resistant enterococci and methicillin-resistant Staphylococcus aureus. No prospective studies have evaluated the impact and sustainability of chlorhexidine bathing.

Study design: Prospective, three-phase study.

Setting: Medical-surgical ICUs and respiratory-care units at five New York hospitals.

Synopsis: In the pre-intervention phase (six to nine months, 1,808 admissions), patients were bathed with soap and water or nonmedicated bathing cloths. In the intervention phase (eight months, 1,832 admissions), patients were bathed with 2% chlorhexidine cloths. In the post-intervention phase (12 months, 2,834 admissions), chlorhexidine bathing was continued without oversight by researchers.

During the intervention phase, there were significantly fewer central-venous-catheter-associated bloodstream infections (2.6/1,000 catheter days vs. 6.4/1,000 pre-intervention). The reductions in bloodstream infections were sustained during the post-intervention period (2.9/1,000 catheter days). Compliance with chlorhexidine bathing was 82% and 88% during the intervention and post-intervention phases, and was well tolerated by the patients.

Limitations of this study include lack of patient-specific data and severity of illness data, as well as lack of randomization and blinding. Although not evaluated in this study, the savings associated with decreased bloodstream infections likely outweigh the cost of chlorhexidine bathing.

Bottom line: Chlorhexidine bathing is a well-tolerated, sustainable intervention that significantly reduces central-venous-catheter-associated bloodstream infections.

Citation: Montecalvo MA, McKenna D, Yarrish R, et al. Chlorhexidine bathing to reduce central venous catheter-associated bloodstream infection: impact and sustainability.Am J Med. 2012;125(5):505-511.

For more physician reviews of recent HM-relevant literature, visit our website.

Clinical question: What is the impact, and sustainability, of chlorhexidine bathing on central-venous-catheter-associated bloodstream infections?

Background: Chlorhexidine bathing has been associated with reductions in healthcare-associated bloodstream infections, including vancomycin-resistant enterococci and methicillin-resistant Staphylococcus aureus. No prospective studies have evaluated the impact and sustainability of chlorhexidine bathing.

Study design: Prospective, three-phase study.

Setting: Medical-surgical ICUs and respiratory-care units at five New York hospitals.

Synopsis: In the pre-intervention phase (six to nine months, 1,808 admissions), patients were bathed with soap and water or nonmedicated bathing cloths. In the intervention phase (eight months, 1,832 admissions), patients were bathed with 2% chlorhexidine cloths. In the post-intervention phase (12 months, 2,834 admissions), chlorhexidine bathing was continued without oversight by researchers.

During the intervention phase, there were significantly fewer central-venous-catheter-associated bloodstream infections (2.6/1,000 catheter days vs. 6.4/1,000 pre-intervention). The reductions in bloodstream infections were sustained during the post-intervention period (2.9/1,000 catheter days). Compliance with chlorhexidine bathing was 82% and 88% during the intervention and post-intervention phases, and was well tolerated by the patients.

Limitations of this study include lack of patient-specific data and severity of illness data, as well as lack of randomization and blinding. Although not evaluated in this study, the savings associated with decreased bloodstream infections likely outweigh the cost of chlorhexidine bathing.

Bottom line: Chlorhexidine bathing is a well-tolerated, sustainable intervention that significantly reduces central-venous-catheter-associated bloodstream infections.

Citation: Montecalvo MA, McKenna D, Yarrish R, et al. Chlorhexidine bathing to reduce central venous catheter-associated bloodstream infection: impact and sustainability.Am J Med. 2012;125(5):505-511.

For more physician reviews of recent HM-relevant literature, visit our website.

Clinical question: What is the impact, and sustainability, of chlorhexidine bathing on central-venous-catheter-associated bloodstream infections?

Background: Chlorhexidine bathing has been associated with reductions in healthcare-associated bloodstream infections, including vancomycin-resistant enterococci and methicillin-resistant Staphylococcus aureus. No prospective studies have evaluated the impact and sustainability of chlorhexidine bathing.

Study design: Prospective, three-phase study.

Setting: Medical-surgical ICUs and respiratory-care units at five New York hospitals.

Synopsis: In the pre-intervention phase (six to nine months, 1,808 admissions), patients were bathed with soap and water or nonmedicated bathing cloths. In the intervention phase (eight months, 1,832 admissions), patients were bathed with 2% chlorhexidine cloths. In the post-intervention phase (12 months, 2,834 admissions), chlorhexidine bathing was continued without oversight by researchers.

During the intervention phase, there were significantly fewer central-venous-catheter-associated bloodstream infections (2.6/1,000 catheter days vs. 6.4/1,000 pre-intervention). The reductions in bloodstream infections were sustained during the post-intervention period (2.9/1,000 catheter days). Compliance with chlorhexidine bathing was 82% and 88% during the intervention and post-intervention phases, and was well tolerated by the patients.

Limitations of this study include lack of patient-specific data and severity of illness data, as well as lack of randomization and blinding. Although not evaluated in this study, the savings associated with decreased bloodstream infections likely outweigh the cost of chlorhexidine bathing.

Bottom line: Chlorhexidine bathing is a well-tolerated, sustainable intervention that significantly reduces central-venous-catheter-associated bloodstream infections.

Citation: Montecalvo MA, McKenna D, Yarrish R, et al. Chlorhexidine bathing to reduce central venous catheter-associated bloodstream infection: impact and sustainability.Am J Med. 2012;125(5):505-511.

For more physician reviews of recent HM-relevant literature, visit our website.

Clinical question: Which clinical decision rule—Wells rule, simplified Wells rule, revised Geneva score, or simplified revised Geneva score—is the best for evaluating a patient with a possible acute pulmonary embolism?

Background: The use of standardized clinical decision rules to determine the probability of an acute pulmonary embolism (PE) has significantly improved the diagnostic evaluation of patients with suspected PE. Several clinical decision rules are available and widely used, but they have not been previously directly compared.

Study design: Prospective cohort.

Setting: Seven hospitals in the Netherlands.

Synopsis: A total of 807 patients with suspected first episode of acute PE had a sequential workup with clinical probability assessment and D-dimer testing. When PE was considered unlikely according to all four clinical decision rules and a normal D-dimer result, PE was excluded. In the remaining patients, a CT scan was used to confirm or exclude the diagnosis.

The prevalence of PE was 23%. Combined with a normal D-dimer, the decision rules excluded PE in 22% to 24% of patients. Thirty percent of patients had discordant decision rule outcomes, but PE was not detected by CT in any of these patients when combined with a normal D-dimer.

This study has practical limitations because management was based on a combination of four decision rules and D-dimer testing rather than only one rule and D-dimer testing, which is the more realistic clinical approach.

Bottom line: When used correctly and in conjunction with a D-dimer result, the Wells rule, simplified Wells rule, revised Geneva score, and simplified revised Geneva score all perform similarly in the exclusion of acute PE.

Citation: Douma RA, Mos IC, Erkens PM, et al. Performance of 4 clinical decision rules in the diagnostic management of acute pulmonary embolism: a prospective cohort study. Ann Intern Med. 2011;154:709-718.

For more of physician reviews of HM-related literature, check out this month's"In the Literature".

Clinical question: Which clinical decision rule—Wells rule, simplified Wells rule, revised Geneva score, or simplified revised Geneva score—is the best for evaluating a patient with a possible acute pulmonary embolism?

Background: The use of standardized clinical decision rules to determine the probability of an acute pulmonary embolism (PE) has significantly improved the diagnostic evaluation of patients with suspected PE. Several clinical decision rules are available and widely used, but they have not been previously directly compared.

Study design: Prospective cohort.

Setting: Seven hospitals in the Netherlands.

Synopsis: A total of 807 patients with suspected first episode of acute PE had a sequential workup with clinical probability assessment and D-dimer testing. When PE was considered unlikely according to all four clinical decision rules and a normal D-dimer result, PE was excluded. In the remaining patients, a CT scan was used to confirm or exclude the diagnosis.

The prevalence of PE was 23%. Combined with a normal D-dimer, the decision rules excluded PE in 22% to 24% of patients. Thirty percent of patients had discordant decision rule outcomes, but PE was not detected by CT in any of these patients when combined with a normal D-dimer.

This study has practical limitations because management was based on a combination of four decision rules and D-dimer testing rather than only one rule and D-dimer testing, which is the more realistic clinical approach.

Bottom line: When used correctly and in conjunction with a D-dimer result, the Wells rule, simplified Wells rule, revised Geneva score, and simplified revised Geneva score all perform similarly in the exclusion of acute PE.

Citation: Douma RA, Mos IC, Erkens PM, et al. Performance of 4 clinical decision rules in the diagnostic management of acute pulmonary embolism: a prospective cohort study. Ann Intern Med. 2011;154:709-718.

For more of physician reviews of HM-related literature, check out this month's"In the Literature".

Clinical question: Which clinical decision rule—Wells rule, simplified Wells rule, revised Geneva score, or simplified revised Geneva score—is the best for evaluating a patient with a possible acute pulmonary embolism?

Background: The use of standardized clinical decision rules to determine the probability of an acute pulmonary embolism (PE) has significantly improved the diagnostic evaluation of patients with suspected PE. Several clinical decision rules are available and widely used, but they have not been previously directly compared.

Study design: Prospective cohort.

Setting: Seven hospitals in the Netherlands.

Synopsis: A total of 807 patients with suspected first episode of acute PE had a sequential workup with clinical probability assessment and D-dimer testing. When PE was considered unlikely according to all four clinical decision rules and a normal D-dimer result, PE was excluded. In the remaining patients, a CT scan was used to confirm or exclude the diagnosis.

The prevalence of PE was 23%. Combined with a normal D-dimer, the decision rules excluded PE in 22% to 24% of patients. Thirty percent of patients had discordant decision rule outcomes, but PE was not detected by CT in any of these patients when combined with a normal D-dimer.

This study has practical limitations because management was based on a combination of four decision rules and D-dimer testing rather than only one rule and D-dimer testing, which is the more realistic clinical approach.

Bottom line: When used correctly and in conjunction with a D-dimer result, the Wells rule, simplified Wells rule, revised Geneva score, and simplified revised Geneva score all perform similarly in the exclusion of acute PE.

Citation: Douma RA, Mos IC, Erkens PM, et al. Performance of 4 clinical decision rules in the diagnostic management of acute pulmonary embolism: a prospective cohort study. Ann Intern Med. 2011;154:709-718.

For more of physician reviews of HM-related literature, check out this month's"In the Literature".

Clinical question: When do venous thromboembolism (VTE) events occur after cancer surgery?

Background: Cancer is a known risk factor for VTE. Prophylaxis for VTE after cancer surgery is commonly stopped at the time of hospital discharge despite evidence for extended-duration treatment.

Study design: Retrospective cohort.

Setting: Patients reported to the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database.

Synopsis: The authors examined the records of 46,656 patients who underwent surgery for one of nine specified cancers. Overall VTE rate was 1.6% (1.0% deep venous thrombosis and 0.6% pulmonary embolism), with 33.4% of VTE events occurring after hospital discharge. VTE risk was highest after esophagogastric and hepatopancreaticobiliary surgery, followed by lung, rectum, ovary/uterus, colon, and prostate. Breast and thyroid/parathyroid surgeries had the lowest incidence of VTE. VTE was associated with increased 30-day mortality. Use of VTE prophylaxis during or after hospitalization was not recorded.

Bottom line: Elevated VTE risk persists following hospital discharge after cancer surgery and consideration should be given to extended-duration thromboprophylaxis. Optimal duration of prophylaxis and its risks and benefits remain poorly defined.

Citation: Merkow RP, Bilimoria KY, McCarter MD, et al. Post-discharge venous thromboembolism after cancer surgery: extending the case for extended prophylaxis. Ann Surg. 2011;254:131-137.

For more physician reviews of HM-related literature, visit our website.

Clinical question: When do venous thromboembolism (VTE) events occur after cancer surgery?

Background: Cancer is a known risk factor for VTE. Prophylaxis for VTE after cancer surgery is commonly stopped at the time of hospital discharge despite evidence for extended-duration treatment.

Study design: Retrospective cohort.

Setting: Patients reported to the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database.

Synopsis: The authors examined the records of 46,656 patients who underwent surgery for one of nine specified cancers. Overall VTE rate was 1.6% (1.0% deep venous thrombosis and 0.6% pulmonary embolism), with 33.4% of VTE events occurring after hospital discharge. VTE risk was highest after esophagogastric and hepatopancreaticobiliary surgery, followed by lung, rectum, ovary/uterus, colon, and prostate. Breast and thyroid/parathyroid surgeries had the lowest incidence of VTE. VTE was associated with increased 30-day mortality. Use of VTE prophylaxis during or after hospitalization was not recorded.

Bottom line: Elevated VTE risk persists following hospital discharge after cancer surgery and consideration should be given to extended-duration thromboprophylaxis. Optimal duration of prophylaxis and its risks and benefits remain poorly defined.

Citation: Merkow RP, Bilimoria KY, McCarter MD, et al. Post-discharge venous thromboembolism after cancer surgery: extending the case for extended prophylaxis. Ann Surg. 2011;254:131-137.

For more physician reviews of HM-related literature, visit our website.

Clinical question: When do venous thromboembolism (VTE) events occur after cancer surgery?

Background: Cancer is a known risk factor for VTE. Prophylaxis for VTE after cancer surgery is commonly stopped at the time of hospital discharge despite evidence for extended-duration treatment.

Study design: Retrospective cohort.

Setting: Patients reported to the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database.

Synopsis: The authors examined the records of 46,656 patients who underwent surgery for one of nine specified cancers. Overall VTE rate was 1.6% (1.0% deep venous thrombosis and 0.6% pulmonary embolism), with 33.4% of VTE events occurring after hospital discharge. VTE risk was highest after esophagogastric and hepatopancreaticobiliary surgery, followed by lung, rectum, ovary/uterus, colon, and prostate. Breast and thyroid/parathyroid surgeries had the lowest incidence of VTE. VTE was associated with increased 30-day mortality. Use of VTE prophylaxis during or after hospitalization was not recorded.

Bottom line: Elevated VTE risk persists following hospital discharge after cancer surgery and consideration should be given to extended-duration thromboprophylaxis. Optimal duration of prophylaxis and its risks and benefits remain poorly defined.

Citation: Merkow RP, Bilimoria KY, McCarter MD, et al. Post-discharge venous thromboembolism after cancer surgery: extending the case for extended prophylaxis. Ann Surg. 2011;254:131-137.

For more physician reviews of HM-related literature, visit our website.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Use of a Care-Transitions Intervention Reduces 30-Day Hospital Readmissions
2. Routine Chest Radiographs after Pneumonia to Rule Out Lung Cancer Have Low Diagnostic Yield
3. Hospitalist Care Shifts Costs to the Outpatient Environment
4. Stopping Smoking at Any Time before Surgery Is Safe
5. Hospitalization for Infection Increases Risk of Stroke
6. Longer Duration of Perioperative Antibiotics May Be Beneficial after Cardiac Surgery
7. Stroke Unit Care Combined with Early Supported Discharge Improves Outcomes
8. Criteria May Help Identify Patients at Risk for Infective Endocarditis

Use of a Care-Transitions Intervention Reduces 30-Day Hospital Readmissions

Clinical question: Does use of a specific care-transitions intervention (CTI) reduce 30-day hospital readmissions in a nonintegrated healthcare system?

Background: Previous interventions addressing improved communication between members of the healthcare team, patients, and families at time of discharge show promise for reducing hospital readmissions. Although these interventions revealed positive results, no research has been completed within a system where healthcare is integrated across settings.

Study design: Quasi-experimental prospective cohort study.

Setting: Six Rhode Island acute-care hospitals, including two community hospitals, three teaching hospitals, and a tertiary-care center and teaching hospital. Facilities ranged from 129 beds to 719 beds.

Synopsis: The CTI is a patient-centered intervention occurring across 30 days. The intervention includes a home visit by a coach within three days of hospital discharge, a telephone call within seven to 10 days of discharge, and a final telephone call no later than 30 days after admission. During these contacts, coaches encourage patient and family participation in care, and active communication with their primary-care provider regarding their disease state. A convenience sample of fee-for-service Medicare beneficiaries was identified by admission diagnoses of acute myocardial infarction, congestive heart failure, or specific pulmonary conditions. Overall, 74% participants completed the entire intervention. The odds of a hospital readmission were significantly lower in the intervention population compared with those who did not receive the intervention (OR 0.61; 95% CI, 0.42-0.88).

Study design: Study design was limited by ability to provide coaching (only 8% of total population was approached), and therefore may not be representative of a typical integrated healthcare setting. In addition, the sample consisted of a convenience sample, which may limit generalizability.

Bottom line: The CTI appears to decrease the rate of 30-day hospital readmissions in Medicare patients with certain cardiac and pulmonary diagnoses.

Citation: Voss R, Gardner R, Baier R, Butterfield K, Lehrman S, Gravenstein S. The care transitions intervention: translating from efficacy to effectiveness. Arch Intern Med. 2011;171:1232-1237.

Routine Chest Radiographs after Pneumonia to Rule Out Lung Cancer Have Low Diagnostic Yield

Clinical question: What is the incidence of new lung cancer detected on routine post-pneumonia chest radiographs?

Background: Routine chest radiographs have been recommended four to eight weeks after resolution of pneumonia to exclude underlying lung cancer. The diagnostic yield of this practice is uncertain.

Study design: Population-based cohort.

Setting: Seven emergency departments and six hospitals in Edmonton, Alberta, Canada.

Synopsis: Authors enrolled 3,398 patients with clinical and radiographic evidence of pneumonia. Of these, 59% were aged 50 and older, 52% were male, 17% were current smokers, 18% had COPD, and 49% were treated as inpatients. At 90-day follow-up, 1.1% of patients received a new diagnosis of lung cancer, with incidence steadily increasing to 2.2% at three-year follow-up. In multivariate analysis, age 50 and older, male sex, and current smoking were independent predictors of post-pneumonia new lung cancer diagnosis. Limiting follow-up chest radiographs to patients aged 50 and older would have detected 98% of new lung cancers and improved diagnostic yield to 2.8%.

Bottom line: Routine post-pneumonia chest radiographs for lung cancer screening have low diagnostic yield that is only marginally improved by selecting high-risk populations.

Citation: Tang KL, Eurich DT, Minhas-Sandhu JK, Marrie TJ, Majumdar SR. Incidence, correlates, and chest radiographic yield of new lung cancer diagnosis in 3398 patients with pneumonia. Arch Intern Med. 2011;171:1193-1198.

Hospitalist Care Shifts Costs to the Outpatient Environment

Clinical question: How does hospitalist care affect medical utilization costs after hospital discharge?

Background: The number of patients cared for by hospitalists is growing rapidly. Some studies have shown hospitalists to decrease length of stay and inpatient costs. The impact of shorter hospitalization on outpatient medical utilization and costs is not known.

Study design: Population-based national cohort.

Setting: Hospitalized Medicare patients.

Synopsis: In this study of 58,125 Medicare admissions at 454 hospitals, hospitalist care was associated with a 0.64-day shorter adjusted length of stay and $282 lower hospital charges compared with patients cared for by their primary-care physicians (PCPs). This was offset by $332 higher Medicare spending in the 30 days following hospitalization. Patients cared for by hospitalists were less likely to be discharged home (OR 0.82, 95% CI, 0.78-0.86), and were more likely to require emergency department visits (OR 1.18, 95% CI, 1.12-1.24) and readmissions (OR 1.08, 95% CI, 1.02-1.14). The authors postulate that shorter length of stay associated with hospitalist care is achieved at the expense of shifting costs to the outpatient environment. The discharged patients are sicker and, as a result, require more skilled care and repeat hospital visits.

Bottom line: Hospitalist care may be associated with higher overall costs and more medical utilization.

Citation: Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011;155:152-159.

Stopping Smoking at Any Time before Surgery Is Safe

Clinical question: Is smoking cessation within eight weeks of surgery safe?

Background: Smoking cessation before surgery can decrease the risk of surgical complications. However, several studies found increased risk for perioperative complications when smoking was stopped within eight weeks of surgery. These findings created uncertainty about general safety of tobacco cessation counseling before surgery.

Study design: Systematic review and meta-analysis.

Setting: Smokers undergoing any type of surgery.

Synopsis: The authors identified nine studies involving 889 patients that compared smokers who quit within eight weeks of surgery with those who continued to smoke. There was considerable heterogeneity in the studies but no overall difference in perioperative complications between those who quit smoking and those who continued to smoke (OR 0.78, 95% CI, 0.57-1.07). The subset of studies examining pulmonary complications also found no difference (OR 1.18, 95% CI, 0.95-1.46).

Bottom line: Smoking cessation at any time before surgery appears to be safe.

Citation: Myers K, Hajek P, Hinds C, McRobbie H. Stopping smoking shortly before surgery and postoperative complications: a systematic review and meta-analysis. Arch Intern Med. 2011;171:983-989.

Hospitalization for Infection Increases Risk of Stroke

Clinical question: Can infection act as a precipitant for acute ischemic stroke?

Background: Little is known about precipitants of acute ischemic stroke. Severe infections have been shown to promote hypercoagulability and platelet activation, and to induce endothelial dysfunction. Authors postulated that infections severe enough to warrant hospitalization might transiently increase the risk for stroke.

Study design: Case-crossover analysis of data from a multicenter prospective cohort (Cardiovascular Health Study).

Setting: Medicare patients in four communities.

Synopsis: During a median follow-up of 12.2 years, 669 strokes occurred in 5,639 study participants. Hospitalization for infection within 14 days was associated with increased risk of stroke (OR 8.0, 95% CI, 1.6-77.3), and the risk remained elevated for hospitalizations within 90 days (OR 3.4, 95% CI, 1.8-6.5). The findings remained significant after adjusting for comorbidities, including age, sex, race, smoking, and diabetes. The number of patients hospitalized for infection before stroke was small—eight within 14 days, and 29 within 90 days.

Bottom line: Infection severe enough to require hospitalization may act as a trigger for acute ischemic stroke.

Citation: Elkind MS, Carty CL, O’Meara ES, et al. Hospitalization for infection and risk of acute ischemic stroke: the Cardiovascular Health Study. Stroke. 2011;42:1851-1856.

Longer Duration of Perioperative Antibiotics May Be Beneficial after Cardiac Surgery

Clinical question: Is antibiotic prophylaxis for 24 or more hours better than shorter duration of treatment after cardiac surgery?

Background: Sternal surgical site infections are a serious complication of cardiac surgery. The optimal duration of perioperative antibiotic prophylaxis is not known, with recommendations ranging from a single dose to 72 hours. The Society of Thoracic Surgeons’ recommendation for 24 to 72 hours of prophylaxis is not based on a systematic review and meta-analysis.

Study design: Systematic review and meta-analysis.

Setting: Adult patients undergoing open-heart surgery who received perioperative antibiotic prophylaxis.

Synopsis: Authors identified 12 trials encompassing 7,893 patients. Compared with prophylaxis of ≥24 hours, prophylaxis of <24 hours was associated with a higher risk of sternal surgical site infections (RR 1.38, 95% CI, 1.13-1.69) and deep infections (RR 1.68, 95% CI, 1.12-2.53). There was no difference in mortality, other infections, or adverse events. Most studies had methodological limitations with a high risk for bias.

Bottom line: Perioperative antibiotic prophylaxis of ≥24 hours reduces sternal surgical infections.

Citation: Mertz D, Johnstone J, Loeb M. Does duration of perioperative antibiotic prophylaxis matter in cardiac surgery? A systematic review and meta-analysis. Ann Surg. 2011;254:48-54.

Stroke Unit Care Combined with Early Supported Discharge Improves Outcomes

Clinical question: Does early supported discharge (ESD) improve outcomes more than conventional follow-up in stroke patients?

Background: ESD is a mobile team that coordinates follow-up and rehabilitation. Previous studies have shown it to be beneficial in patients with mild to moderate disability at one year, but long-term effects of ESD are not known.

Study design: Randomized controlled trial.

Setting: Single center in Norway.

Synopsis: Stroke-unit patients were recruited and received standard care or ESD after discharge. All 320 patients received standard acute care. The proportion of patients with modified Rankin Score (mRS) of ≤2 was not significantly different in the two groups but identified a trend toward improvement in the intervention group (38% vs. 30%, P=0.106). More patients receiving conventional follow-up died or were institutionalized (P=0.032) but mortality rates at five years were similar (ESD 46% vs. 51%). Secondary outcomes (Scandinavian Stroke Scale, Barthel Index, Frenchay Activity Index, and Mini Mental Status Examination) were not statistically different. Predictors of good outcome in the ESD group included young age, low mRS, and living with others.

This study recruited patients from 1995 to 1997 and followed the patients for five years. Limitations to the applicability include advances in stroke rehabilitation in the last 10 years. The cost of a mobile multidisciplinary team consisting of a physiotherapist, occupational therapist, nurse, and part-time physician was not discussed and may limit the availability to many patients.

Bottom line: Early supported discharge may increase the proportion of patients living at home five years after stroke.

Citation: Fjaertoft H, Rohweder G, Indredavik B. Stroke unit care combined with early supported discharge improves 5-year outcome: a randomized controlled trial. Stroke. 2011;42:1707-1711.

Criteria May Help Identify Patients at Risk for Infective Endocarditis

Clinical question: Which patients with Staphylococcus aureus bacteremia benefit the most from transesophageal echocardiography?

Background: Infective endocarditis is a serious complication of S. aureus bacteremia (SAB), occurring in 5% to 17% of patients with documented SAB. It has been recommended to perform transesophageal echocardiography (TEE) in all patients with SAB. Large variation exists in rates of TEE, and identifying patients at low risk for endocarditis may help with more appropriate utilization of this test.

Study design: Retrospective cohort analysis.

Setting: Two university-based German tertiary hospitals (INSTINCT cohort) and one North American university-based hospital from October 1994 to December 2009 (SABG cohort).

Synopsis: A total of 736 cases of nosocomial SAB were analyzed. Age, source of infection, and 30-day and 90-day case fatality rates were similar between the two cohorts. Patients were followed during the index hospitalization and for three months after discharge.

Patients with infective endocarditis were more likely to have prolonged bacteremia; a permanent intracardiac device, such as a pacemaker or a heart valve; be recipients of hemodialysis; and have osteomyelitis. Of the 83 patients who did not fulfill any of the prediction criteria, no cases of infective endocarditis were found.

Bottom line: A set of simple criteria may help identify patients with nosocomial SAB who are at risk for infective endocarditis. The subset of patients who do not meet any of these criteria may not need diagnostic evaluation with TEE.

Citation: Kaasch, AJ, Fowler VG Jr., Rieg S, et al. Use of a simple criteria set for guiding echocardiography in nosocomial Staphylococcus aureus bacteremia. Clin Infect Dis. 2011;53:1-9.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Use of a Care-Transitions Intervention Reduces 30-Day Hospital Readmissions
2. Routine Chest Radiographs after Pneumonia to Rule Out Lung Cancer Have Low Diagnostic Yield
3. Hospitalist Care Shifts Costs to the Outpatient Environment
4. Stopping Smoking at Any Time before Surgery Is Safe
5. Hospitalization for Infection Increases Risk of Stroke
6. Longer Duration of Perioperative Antibiotics May Be Beneficial after Cardiac Surgery
7. Stroke Unit Care Combined with Early Supported Discharge Improves Outcomes
8. Criteria May Help Identify Patients at Risk for Infective Endocarditis

Use of a Care-Transitions Intervention Reduces 30-Day Hospital Readmissions

Clinical question: Does use of a specific care-transitions intervention (CTI) reduce 30-day hospital readmissions in a nonintegrated healthcare system?

Background: Previous interventions addressing improved communication between members of the healthcare team, patients, and families at time of discharge show promise for reducing hospital readmissions. Although these interventions revealed positive results, no research has been completed within a system where healthcare is integrated across settings.

Study design: Quasi-experimental prospective cohort study.

Setting: Six Rhode Island acute-care hospitals, including two community hospitals, three teaching hospitals, and a tertiary-care center and teaching hospital. Facilities ranged from 129 beds to 719 beds.

Synopsis: The CTI is a patient-centered intervention occurring across 30 days. The intervention includes a home visit by a coach within three days of hospital discharge, a telephone call within seven to 10 days of discharge, and a final telephone call no later than 30 days after admission. During these contacts, coaches encourage patient and family participation in care, and active communication with their primary-care provider regarding their disease state. A convenience sample of fee-for-service Medicare beneficiaries was identified by admission diagnoses of acute myocardial infarction, congestive heart failure, or specific pulmonary conditions. Overall, 74% participants completed the entire intervention. The odds of a hospital readmission were significantly lower in the intervention population compared with those who did not receive the intervention (OR 0.61; 95% CI, 0.42-0.88).

Study design: Study design was limited by ability to provide coaching (only 8% of total population was approached), and therefore may not be representative of a typical integrated healthcare setting. In addition, the sample consisted of a convenience sample, which may limit generalizability.

Bottom line: The CTI appears to decrease the rate of 30-day hospital readmissions in Medicare patients with certain cardiac and pulmonary diagnoses.

Citation: Voss R, Gardner R, Baier R, Butterfield K, Lehrman S, Gravenstein S. The care transitions intervention: translating from efficacy to effectiveness. Arch Intern Med. 2011;171:1232-1237.

Routine Chest Radiographs after Pneumonia to Rule Out Lung Cancer Have Low Diagnostic Yield

Clinical question: What is the incidence of new lung cancer detected on routine post-pneumonia chest radiographs?

Background: Routine chest radiographs have been recommended four to eight weeks after resolution of pneumonia to exclude underlying lung cancer. The diagnostic yield of this practice is uncertain.

Study design: Population-based cohort.

Setting: Seven emergency departments and six hospitals in Edmonton, Alberta, Canada.

Synopsis: Authors enrolled 3,398 patients with clinical and radiographic evidence of pneumonia. Of these, 59% were aged 50 and older, 52% were male, 17% were current smokers, 18% had COPD, and 49% were treated as inpatients. At 90-day follow-up, 1.1% of patients received a new diagnosis of lung cancer, with incidence steadily increasing to 2.2% at three-year follow-up. In multivariate analysis, age 50 and older, male sex, and current smoking were independent predictors of post-pneumonia new lung cancer diagnosis. Limiting follow-up chest radiographs to patients aged 50 and older would have detected 98% of new lung cancers and improved diagnostic yield to 2.8%.

Bottom line: Routine post-pneumonia chest radiographs for lung cancer screening have low diagnostic yield that is only marginally improved by selecting high-risk populations.

Citation: Tang KL, Eurich DT, Minhas-Sandhu JK, Marrie TJ, Majumdar SR. Incidence, correlates, and chest radiographic yield of new lung cancer diagnosis in 3398 patients with pneumonia. Arch Intern Med. 2011;171:1193-1198.

Hospitalist Care Shifts Costs to the Outpatient Environment

Clinical question: How does hospitalist care affect medical utilization costs after hospital discharge?

Background: The number of patients cared for by hospitalists is growing rapidly. Some studies have shown hospitalists to decrease length of stay and inpatient costs. The impact of shorter hospitalization on outpatient medical utilization and costs is not known.

Study design: Population-based national cohort.

Setting: Hospitalized Medicare patients.

Synopsis: In this study of 58,125 Medicare admissions at 454 hospitals, hospitalist care was associated with a 0.64-day shorter adjusted length of stay and $282 lower hospital charges compared with patients cared for by their primary-care physicians (PCPs). This was offset by $332 higher Medicare spending in the 30 days following hospitalization. Patients cared for by hospitalists were less likely to be discharged home (OR 0.82, 95% CI, 0.78-0.86), and were more likely to require emergency department visits (OR 1.18, 95% CI, 1.12-1.24) and readmissions (OR 1.08, 95% CI, 1.02-1.14). The authors postulate that shorter length of stay associated with hospitalist care is achieved at the expense of shifting costs to the outpatient environment. The discharged patients are sicker and, as a result, require more skilled care and repeat hospital visits.

Bottom line: Hospitalist care may be associated with higher overall costs and more medical utilization.

Citation: Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011;155:152-159.

Stopping Smoking at Any Time before Surgery Is Safe

Clinical question: Is smoking cessation within eight weeks of surgery safe?

Background: Smoking cessation before surgery can decrease the risk of surgical complications. However, several studies found increased risk for perioperative complications when smoking was stopped within eight weeks of surgery. These findings created uncertainty about general safety of tobacco cessation counseling before surgery.

Study design: Systematic review and meta-analysis.

Setting: Smokers undergoing any type of surgery.

Synopsis: The authors identified nine studies involving 889 patients that compared smokers who quit within eight weeks of surgery with those who continued to smoke. There was considerable heterogeneity in the studies but no overall difference in perioperative complications between those who quit smoking and those who continued to smoke (OR 0.78, 95% CI, 0.57-1.07). The subset of studies examining pulmonary complications also found no difference (OR 1.18, 95% CI, 0.95-1.46).

Bottom line: Smoking cessation at any time before surgery appears to be safe.

Citation: Myers K, Hajek P, Hinds C, McRobbie H. Stopping smoking shortly before surgery and postoperative complications: a systematic review and meta-analysis. Arch Intern Med. 2011;171:983-989.

Hospitalization for Infection Increases Risk of Stroke

Clinical question: Can infection act as a precipitant for acute ischemic stroke?

Background: Little is known about precipitants of acute ischemic stroke. Severe infections have been shown to promote hypercoagulability and platelet activation, and to induce endothelial dysfunction. Authors postulated that infections severe enough to warrant hospitalization might transiently increase the risk for stroke.

Study design: Case-crossover analysis of data from a multicenter prospective cohort (Cardiovascular Health Study).

Setting: Medicare patients in four communities.

Synopsis: During a median follow-up of 12.2 years, 669 strokes occurred in 5,639 study participants. Hospitalization for infection within 14 days was associated with increased risk of stroke (OR 8.0, 95% CI, 1.6-77.3), and the risk remained elevated for hospitalizations within 90 days (OR 3.4, 95% CI, 1.8-6.5). The findings remained significant after adjusting for comorbidities, including age, sex, race, smoking, and diabetes. The number of patients hospitalized for infection before stroke was small—eight within 14 days, and 29 within 90 days.

Bottom line: Infection severe enough to require hospitalization may act as a trigger for acute ischemic stroke.

Citation: Elkind MS, Carty CL, O’Meara ES, et al. Hospitalization for infection and risk of acute ischemic stroke: the Cardiovascular Health Study. Stroke. 2011;42:1851-1856.

Longer Duration of Perioperative Antibiotics May Be Beneficial after Cardiac Surgery

Clinical question: Is antibiotic prophylaxis for 24 or more hours better than shorter duration of treatment after cardiac surgery?

Background: Sternal surgical site infections are a serious complication of cardiac surgery. The optimal duration of perioperative antibiotic prophylaxis is not known, with recommendations ranging from a single dose to 72 hours. The Society of Thoracic Surgeons’ recommendation for 24 to 72 hours of prophylaxis is not based on a systematic review and meta-analysis.

Study design: Systematic review and meta-analysis.

Setting: Adult patients undergoing open-heart surgery who received perioperative antibiotic prophylaxis.

Synopsis: Authors identified 12 trials encompassing 7,893 patients. Compared with prophylaxis of ≥24 hours, prophylaxis of <24 hours was associated with a higher risk of sternal surgical site infections (RR 1.38, 95% CI, 1.13-1.69) and deep infections (RR 1.68, 95% CI, 1.12-2.53). There was no difference in mortality, other infections, or adverse events. Most studies had methodological limitations with a high risk for bias.

Bottom line: Perioperative antibiotic prophylaxis of ≥24 hours reduces sternal surgical infections.

Citation: Mertz D, Johnstone J, Loeb M. Does duration of perioperative antibiotic prophylaxis matter in cardiac surgery? A systematic review and meta-analysis. Ann Surg. 2011;254:48-54.

Stroke Unit Care Combined with Early Supported Discharge Improves Outcomes

Clinical question: Does early supported discharge (ESD) improve outcomes more than conventional follow-up in stroke patients?

Background: ESD is a mobile team that coordinates follow-up and rehabilitation. Previous studies have shown it to be beneficial in patients with mild to moderate disability at one year, but long-term effects of ESD are not known.

Study design: Randomized controlled trial.

Setting: Single center in Norway.

Synopsis: Stroke-unit patients were recruited and received standard care or ESD after discharge. All 320 patients received standard acute care. The proportion of patients with modified Rankin Score (mRS) of ≤2 was not significantly different in the two groups but identified a trend toward improvement in the intervention group (38% vs. 30%, P=0.106). More patients receiving conventional follow-up died or were institutionalized (P=0.032) but mortality rates at five years were similar (ESD 46% vs. 51%). Secondary outcomes (Scandinavian Stroke Scale, Barthel Index, Frenchay Activity Index, and Mini Mental Status Examination) were not statistically different. Predictors of good outcome in the ESD group included young age, low mRS, and living with others.

This study recruited patients from 1995 to 1997 and followed the patients for five years. Limitations to the applicability include advances in stroke rehabilitation in the last 10 years. The cost of a mobile multidisciplinary team consisting of a physiotherapist, occupational therapist, nurse, and part-time physician was not discussed and may limit the availability to many patients.

Bottom line: Early supported discharge may increase the proportion of patients living at home five years after stroke.

Citation: Fjaertoft H, Rohweder G, Indredavik B. Stroke unit care combined with early supported discharge improves 5-year outcome: a randomized controlled trial. Stroke. 2011;42:1707-1711.

Criteria May Help Identify Patients at Risk for Infective Endocarditis

Clinical question: Which patients with Staphylococcus aureus bacteremia benefit the most from transesophageal echocardiography?

Background: Infective endocarditis is a serious complication of S. aureus bacteremia (SAB), occurring in 5% to 17% of patients with documented SAB. It has been recommended to perform transesophageal echocardiography (TEE) in all patients with SAB. Large variation exists in rates of TEE, and identifying patients at low risk for endocarditis may help with more appropriate utilization of this test.

Study design: Retrospective cohort analysis.

Setting: Two university-based German tertiary hospitals (INSTINCT cohort) and one North American university-based hospital from October 1994 to December 2009 (SABG cohort).

Synopsis: A total of 736 cases of nosocomial SAB were analyzed. Age, source of infection, and 30-day and 90-day case fatality rates were similar between the two cohorts. Patients were followed during the index hospitalization and for three months after discharge.

Patients with infective endocarditis were more likely to have prolonged bacteremia; a permanent intracardiac device, such as a pacemaker or a heart valve; be recipients of hemodialysis; and have osteomyelitis. Of the 83 patients who did not fulfill any of the prediction criteria, no cases of infective endocarditis were found.

Bottom line: A set of simple criteria may help identify patients with nosocomial SAB who are at risk for infective endocarditis. The subset of patients who do not meet any of these criteria may not need diagnostic evaluation with TEE.

Citation: Kaasch, AJ, Fowler VG Jr., Rieg S, et al. Use of a simple criteria set for guiding echocardiography in nosocomial Staphylococcus aureus bacteremia. Clin Infect Dis. 2011;53:1-9.

In This Edition

Literature At A Glance

A guide to this month’s studies

1. Use of a Care-Transitions Intervention Reduces 30-Day Hospital Readmissions
2. Routine Chest Radiographs after Pneumonia to Rule Out Lung Cancer Have Low Diagnostic Yield
3. Hospitalist Care Shifts Costs to the Outpatient Environment
4. Stopping Smoking at Any Time before Surgery Is Safe
5. Hospitalization for Infection Increases Risk of Stroke
6. Longer Duration of Perioperative Antibiotics May Be Beneficial after Cardiac Surgery
7. Stroke Unit Care Combined with Early Supported Discharge Improves Outcomes
8. Criteria May Help Identify Patients at Risk for Infective Endocarditis

Use of a Care-Transitions Intervention Reduces 30-Day Hospital Readmissions

Clinical question: Does use of a specific care-transitions intervention (CTI) reduce 30-day hospital readmissions in a nonintegrated healthcare system?

Background: Previous interventions addressing improved communication between members of the healthcare team, patients, and families at time of discharge show promise for reducing hospital readmissions. Although these interventions revealed positive results, no research has been completed within a system where healthcare is integrated across settings.

Study design: Quasi-experimental prospective cohort study.

Setting: Six Rhode Island acute-care hospitals, including two community hospitals, three teaching hospitals, and a tertiary-care center and teaching hospital. Facilities ranged from 129 beds to 719 beds.

Synopsis: The CTI is a patient-centered intervention occurring across 30 days. The intervention includes a home visit by a coach within three days of hospital discharge, a telephone call within seven to 10 days of discharge, and a final telephone call no later than 30 days after admission. During these contacts, coaches encourage patient and family participation in care, and active communication with their primary-care provider regarding their disease state. A convenience sample of fee-for-service Medicare beneficiaries was identified by admission diagnoses of acute myocardial infarction, congestive heart failure, or specific pulmonary conditions. Overall, 74% participants completed the entire intervention. The odds of a hospital readmission were significantly lower in the intervention population compared with those who did not receive the intervention (OR 0.61; 95% CI, 0.42-0.88).

Study design: Study design was limited by ability to provide coaching (only 8% of total population was approached), and therefore may not be representative of a typical integrated healthcare setting. In addition, the sample consisted of a convenience sample, which may limit generalizability.

Bottom line: The CTI appears to decrease the rate of 30-day hospital readmissions in Medicare patients with certain cardiac and pulmonary diagnoses.

Citation: Voss R, Gardner R, Baier R, Butterfield K, Lehrman S, Gravenstein S. The care transitions intervention: translating from efficacy to effectiveness. Arch Intern Med. 2011;171:1232-1237.

Routine Chest Radiographs after Pneumonia to Rule Out Lung Cancer Have Low Diagnostic Yield

Clinical question: What is the incidence of new lung cancer detected on routine post-pneumonia chest radiographs?

Background: Routine chest radiographs have been recommended four to eight weeks after resolution of pneumonia to exclude underlying lung cancer. The diagnostic yield of this practice is uncertain.

Study design: Population-based cohort.

Setting: Seven emergency departments and six hospitals in Edmonton, Alberta, Canada.

Synopsis: Authors enrolled 3,398 patients with clinical and radiographic evidence of pneumonia. Of these, 59% were aged 50 and older, 52% were male, 17% were current smokers, 18% had COPD, and 49% were treated as inpatients. At 90-day follow-up, 1.1% of patients received a new diagnosis of lung cancer, with incidence steadily increasing to 2.2% at three-year follow-up. In multivariate analysis, age 50 and older, male sex, and current smoking were independent predictors of post-pneumonia new lung cancer diagnosis. Limiting follow-up chest radiographs to patients aged 50 and older would have detected 98% of new lung cancers and improved diagnostic yield to 2.8%.

Bottom line: Routine post-pneumonia chest radiographs for lung cancer screening have low diagnostic yield that is only marginally improved by selecting high-risk populations.

Citation: Tang KL, Eurich DT, Minhas-Sandhu JK, Marrie TJ, Majumdar SR. Incidence, correlates, and chest radiographic yield of new lung cancer diagnosis in 3398 patients with pneumonia. Arch Intern Med. 2011;171:1193-1198.

Hospitalist Care Shifts Costs to the Outpatient Environment

Clinical question: How does hospitalist care affect medical utilization costs after hospital discharge?

Background: The number of patients cared for by hospitalists is growing rapidly. Some studies have shown hospitalists to decrease length of stay and inpatient costs. The impact of shorter hospitalization on outpatient medical utilization and costs is not known.

Study design: Population-based national cohort.

Setting: Hospitalized Medicare patients.

Synopsis: In this study of 58,125 Medicare admissions at 454 hospitals, hospitalist care was associated with a 0.64-day shorter adjusted length of stay and $282 lower hospital charges compared with patients cared for by their primary-care physicians (PCPs). This was offset by $332 higher Medicare spending in the 30 days following hospitalization. Patients cared for by hospitalists were less likely to be discharged home (OR 0.82, 95% CI, 0.78-0.86), and were more likely to require emergency department visits (OR 1.18, 95% CI, 1.12-1.24) and readmissions (OR 1.08, 95% CI, 1.02-1.14). The authors postulate that shorter length of stay associated with hospitalist care is achieved at the expense of shifting costs to the outpatient environment. The discharged patients are sicker and, as a result, require more skilled care and repeat hospital visits.

Bottom line: Hospitalist care may be associated with higher overall costs and more medical utilization.

Citation: Kuo YF, Goodwin JS. Association of hospitalist care with medical utilization after discharge: evidence of cost shift from a cohort study. Ann Intern Med. 2011;155:152-159.

Stopping Smoking at Any Time before Surgery Is Safe

Clinical question: Is smoking cessation within eight weeks of surgery safe?

Background: Smoking cessation before surgery can decrease the risk of surgical complications. However, several studies found increased risk for perioperative complications when smoking was stopped within eight weeks of surgery. These findings created uncertainty about general safety of tobacco cessation counseling before surgery.

Study design: Systematic review and meta-analysis.

Setting: Smokers undergoing any type of surgery.

Synopsis: The authors identified nine studies involving 889 patients that compared smokers who quit within eight weeks of surgery with those who continued to smoke. There was considerable heterogeneity in the studies but no overall difference in perioperative complications between those who quit smoking and those who continued to smoke (OR 0.78, 95% CI, 0.57-1.07). The subset of studies examining pulmonary complications also found no difference (OR 1.18, 95% CI, 0.95-1.46).

Bottom line: Smoking cessation at any time before surgery appears to be safe.

Citation: Myers K, Hajek P, Hinds C, McRobbie H. Stopping smoking shortly before surgery and postoperative complications: a systematic review and meta-analysis. Arch Intern Med. 2011;171:983-989.

Hospitalization for Infection Increases Risk of Stroke

Clinical question: Can infection act as a precipitant for acute ischemic stroke?

Background: Little is known about precipitants of acute ischemic stroke. Severe infections have been shown to promote hypercoagulability and platelet activation, and to induce endothelial dysfunction. Authors postulated that infections severe enough to warrant hospitalization might transiently increase the risk for stroke.

Study design: Case-crossover analysis of data from a multicenter prospective cohort (Cardiovascular Health Study).

Setting: Medicare patients in four communities.

Synopsis: During a median follow-up of 12.2 years, 669 strokes occurred in 5,639 study participants. Hospitalization for infection within 14 days was associated with increased risk of stroke (OR 8.0, 95% CI, 1.6-77.3), and the risk remained elevated for hospitalizations within 90 days (OR 3.4, 95% CI, 1.8-6.5). The findings remained significant after adjusting for comorbidities, including age, sex, race, smoking, and diabetes. The number of patients hospitalized for infection before stroke was small—eight within 14 days, and 29 within 90 days.

Bottom line: Infection severe enough to require hospitalization may act as a trigger for acute ischemic stroke.

Citation: Elkind MS, Carty CL, O’Meara ES, et al. Hospitalization for infection and risk of acute ischemic stroke: the Cardiovascular Health Study. Stroke. 2011;42:1851-1856.

Longer Duration of Perioperative Antibiotics May Be Beneficial after Cardiac Surgery

Clinical question: Is antibiotic prophylaxis for 24 or more hours better than shorter duration of treatment after cardiac surgery?

Background: Sternal surgical site infections are a serious complication of cardiac surgery. The optimal duration of perioperative antibiotic prophylaxis is not known, with recommendations ranging from a single dose to 72 hours. The Society of Thoracic Surgeons’ recommendation for 24 to 72 hours of prophylaxis is not based on a systematic review and meta-analysis.

Study design: Systematic review and meta-analysis.

Setting: Adult patients undergoing open-heart surgery who received perioperative antibiotic prophylaxis.

Synopsis: Authors identified 12 trials encompassing 7,893 patients. Compared with prophylaxis of ≥24 hours, prophylaxis of <24 hours was associated with a higher risk of sternal surgical site infections (RR 1.38, 95% CI, 1.13-1.69) and deep infections (RR 1.68, 95% CI, 1.12-2.53). There was no difference in mortality, other infections, or adverse events. Most studies had methodological limitations with a high risk for bias.

Bottom line: Perioperative antibiotic prophylaxis of ≥24 hours reduces sternal surgical infections.

Citation: Mertz D, Johnstone J, Loeb M. Does duration of perioperative antibiotic prophylaxis matter in cardiac surgery? A systematic review and meta-analysis. Ann Surg. 2011;254:48-54.

Stroke Unit Care Combined with Early Supported Discharge Improves Outcomes

Clinical question: Does early supported discharge (ESD) improve outcomes more than conventional follow-up in stroke patients?

Background: ESD is a mobile team that coordinates follow-up and rehabilitation. Previous studies have shown it to be beneficial in patients with mild to moderate disability at one year, but long-term effects of ESD are not known.

Study design: Randomized controlled trial.

Setting: Single center in Norway.

Synopsis: Stroke-unit patients were recruited and received standard care or ESD after discharge. All 320 patients received standard acute care. The proportion of patients with modified Rankin Score (mRS) of ≤2 was not significantly different in the two groups but identified a trend toward improvement in the intervention group (38% vs. 30%, P=0.106). More patients receiving conventional follow-up died or were institutionalized (P=0.032) but mortality rates at five years were similar (ESD 46% vs. 51%). Secondary outcomes (Scandinavian Stroke Scale, Barthel Index, Frenchay Activity Index, and Mini Mental Status Examination) were not statistically different. Predictors of good outcome in the ESD group included young age, low mRS, and living with others.

This study recruited patients from 1995 to 1997 and followed the patients for five years. Limitations to the applicability include advances in stroke rehabilitation in the last 10 years. The cost of a mobile multidisciplinary team consisting of a physiotherapist, occupational therapist, nurse, and part-time physician was not discussed and may limit the availability to many patients.

Bottom line: Early supported discharge may increase the proportion of patients living at home five years after stroke.

Citation: Fjaertoft H, Rohweder G, Indredavik B. Stroke unit care combined with early supported discharge improves 5-year outcome: a randomized controlled trial. Stroke. 2011;42:1707-1711.

Criteria May Help Identify Patients at Risk for Infective Endocarditis

Clinical question: Which patients with Staphylococcus aureus bacteremia benefit the most from transesophageal echocardiography?

Background: Infective endocarditis is a serious complication of S. aureus bacteremia (SAB), occurring in 5% to 17% of patients with documented SAB. It has been recommended to perform transesophageal echocardiography (TEE) in all patients with SAB. Large variation exists in rates of TEE, and identifying patients at low risk for endocarditis may help with more appropriate utilization of this test.

Study design: Retrospective cohort analysis.

Setting: Two university-based German tertiary hospitals (INSTINCT cohort) and one North American university-based hospital from October 1994 to December 2009 (SABG cohort).

Synopsis: A total of 736 cases of nosocomial SAB were analyzed. Age, source of infection, and 30-day and 90-day case fatality rates were similar between the two cohorts. Patients were followed during the index hospitalization and for three months after discharge.

Patients with infective endocarditis were more likely to have prolonged bacteremia; a permanent intracardiac device, such as a pacemaker or a heart valve; be recipients of hemodialysis; and have osteomyelitis. Of the 83 patients who did not fulfill any of the prediction criteria, no cases of infective endocarditis were found.

Bottom line: A set of simple criteria may help identify patients with nosocomial SAB who are at risk for infective endocarditis. The subset of patients who do not meet any of these criteria may not need diagnostic evaluation with TEE.

Citation: Kaasch, AJ, Fowler VG Jr., Rieg S, et al. Use of a simple criteria set for guiding echocardiography in nosocomial Staphylococcus aureus bacteremia. Clin Infect Dis. 2011;53:1-9.

In This Edition

Literature at a Glance

A guide to this month’s studies

• PE and COPD exacerbations.
• Care bundles and readmission rates.
• Family history and VTE risk.
• Vasopressor choice and mortality in sepsis.
• Vitamin K use in overanticoagulation.
• Appropriate treatment of asymptomatic bacteriuria.
• Guideline adherence in thrombocytopenia.

Pulmonary Embolism Frequently Complicates COPD Exacerbations

Clinical question: What percentage of patients with acute chronic obstructive pulmonary disease (COPD) exacerbations has pulmonary emboli?

Background: As many as 30% of COPD exacerbations have no apparent precipitating event. Even in patients with evidence of a precipitating event, such as an upper-respiratory illness or increased environmental irritants, pulmonary emboli (PE) may coexist and warrant evaluation.

Study design: Literature review.

Setting: Multiple studies in Europe and the U.S.

Synopsis: This literature review included five studies to estimate the rate of PE in patients with a COPD exacerbation. Overall incidence of PE in COPD exacerbations was 19.9%, but of those patients requiring hospitalization, the incidence was as high as 25.5%. Incidence estimates varied based on interpretation of data that were missing or inconsistent between studies. Patients most commonly present with dyspnea, chest pain, hemoptysis, cough, and palpitations. Six percent of PE patients presented with syncope; no patients with an exacerbation without a PE presented with syncope.

Risk of mortality from PE is almost twice as high in patients with a COPD exacerbation compared with PE in other settings. A significant number of patients have PE without history or evidence of DVT, so in situ thrombosis is a significant factor. The interpretation of these results is limited by the heterogeneity of the study designs, and by the relatively low number of cases. Larger trials are necessary.

Bottom line: Pulmonary emboli are present in as many as 25% of all COPD exacerbations. Delay in diagnosis of PE in COPD patients affects morbidity and mortality. PE should be a consideration in many COPD exacerbations.

Citation: Rizkallah J, Man SF, Din DD. Prevalence of pulmonary embolism in acute exacerbations of COPD: a systematic review and metaanalysis. Chest. 2009;135(3):786-793.

Targeted-Care Bundle Can Reduce ED Visits and Readmission Rates in High-Risk Elderly Patients

Clinical question: Can a care coordination bundle reduce length of stay (LOS), ED visits, or readmissions within 30 days of a hospital admission?

Background: Hospital-based care coordination interventions have shown mixed results in affecting LOS, post-discharge ED visits, and readmission rates. Although there has been some success with particular interventions, no consistent benefit has been demonstrated. Most notably, a recent meta-analysis of several different interventions showed no improvement in mortality, LOS, or readmission rates.

Study design: A randomized, controlled trial of select high-risk elderly patients.

Setting: A large teaching hospital at Baylor University Medical Center.

Synopsis: A “targeted-care bundle” was implemented with high-risk elderly patients to try to reduce LOS, readmissions, and ED visits. High-risk patients were identified by age, diagnosis-related group (DRG), number of medications at admission, comorbid conditions, and need for assistance in activities of daily living. Subjects were randomized to usual care or to receive a targeted-care bundle. The targeted-care bundle included multiple interventions. A study care coordinator provided daily patient education, including condition-specific teaching, discharge teaching and planning, and a follow-up phone call at five to seven days after discharge. A clinical pharmacist intervened for medication reconciliation at admission and discharge, medication teaching, and a follow-up phone call at five to seven days after discharge. Structured documents, including a personal health record and supplemental discharge form, were implemented.

The study had low enrollment, largely due to the requirement to obtain informed consent from all participants. Therefore, the study was underpowered to detect such target endpoints as LOS. A significant decrease in 30-day readmission rates/ED visits was noticed, but there was no persistent effect at 60 days.

The intervention was designed to use existing hospital staff in order to be practical for broad utilization. Future studies need to focus on increased enrollment to demonstrate beneficial effect.

Bottom line: Targeted health interventions focusing on education and coordination of care might effect some significant outcomes, most notably readmissions or ED visits within 30 days, but the nature of the clinical problem makes rigorous testing of interventions a challenge.

Citation: Kohler BE, Richter KM, Youngblood L, et al. Reduction of 30-day postdischarge hospital readmission or emergency department (ED) visit rates in high-risk elderly medical patients through delivery of a targeted care bundle. J Hosp Med. 2009;4(4):211-218.

Family History Is a Risk Factor for Venous Thrombosis

Clinical question: Is family history of additional value in predicting an individual’s risk of venous thrombosis once a genetic risk factor is identified?

Background: A positive family history of venous thrombosis might suggest the presence of genetic risk factors in a given family. However, it is not known whether family history is of additional significance—once a risk factor is identified—in predicting an individual’s risk for venous thrombosis.

Study design: Population-based, case-control study.

Setting: Participants in the Multiple Environmental and Genetic Assessment (MEGA) of risk factors for venous thrombosis study.

Synopsis: Recruitment, data collection, and blood samples were obtained from individuals in the MEGA study. Participants completed a questionnaire about risk factors for venous thrombosis and family history. A positive family history more than doubled the risk of venous thrombosis, and when more than one family member was affected, the risk increased fourfold. The risk for venous thrombosis increased 64 times for individuals who had a family history, genetic risk factor, and environmental risk factor when compared with those with a negative family history and no known risk factors.

The underreporting or overestimation of the prevalence of a positive family history might limit this study.

Bottom line: Family history is a risk indictor for a first venous thrombosis, despite the presence of other risk factors.

Citation: Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen CJ. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med. 2009;169(6):610-615.

Vasopressor Choice Predicts Mortality in Septic Shock

Clinical question: Does vasopressor choice affect mortality in patients with community-acquired septic shock?

Background: Community-acquired septic shock is a common illness and, despite aggressive care, a leading cause of death. Randomized clinical control trials evaluating the efficacy and safety of different adrenergic supportive agents are lacking. Thus, both norepinephrine and dopamine are recommended as first-line agents in the treatment of septic shock by the Surviving Sepsis Campaign guidelines.

Study design: Multicenter, cohort observational study.

Setting: Seventeen intensive-care units in Portugal.

Synopsis: In adjusted analysis controlling for Simplified Acute Physiology Score (SAPS) II, use of norepinephrine in community-acquired septic shock was associated with higher hospital mortality and lower 28-day survival when compared with dopamine. Specifically, patients treated with norepinephrine had a statistically significant higher hospital mortality rate than those treated with dopamine (52% and 38.5%, respectively, P=0.002) and a lower 28-day survival (log rank=22.6; P<0.001). While this data is valuable, the nonrandomized, observational study design limits firm conclusions regarding vasopressor choice. Further results from three large trials comparing vasopressor use in septic shock should continue to shed light on this debate.

Bottom line: Norepinephrine administration is associated with higher hospital mortality and lower 28-day survival when compared with dopamine in patients with community-acquired septic shock.

Citation: Póvoa PR, Carneiro AH, Ribeiro OS, Pereira AC, Portuguese Community-Acquired Sepsis Study Group. Influence of vasopressor agent in septic shock mortality. Results from the Portuguese Community-Acquired Sepsis Study (SACiUCI study). Crit Care Med. 2009;37(2):410-416.

Oral Vitamin K Versus Placebo to Correct Excess Anticoagulation in Warfarin Patients

Clinical question: In nonbleeding patients with warfarin-associated coagulopathy, does oral vitamin K reduce bleeding events when compared to placebo?

Background: Warfarin is a common drug for primary and secondary prevention of thromboembolism, but it requires continued monitoring of the international normalized ratio (INR) value. INR values >4.0 are associated with an increase in bleeding complications, with specific concern for intracranial bleeding when INR values exceed 4.5. Small, randomized trials have shown that single, low-dose administration of oral vitamin K effectively reduces the INR in nonbleeding, overanticoagulated patients.

However, these studies have not shown if vitamin K reduces risk for bleeding without increasing the risk for thromboembolism.

Study design: Randomized, placebo-controlled trial.

Setting: Fourteen anticoagulation clinics in Canada, Italy, and the U.S.

Synopsis: Nonbleeding patients with supratherapeutic INR values between 4.5 and 10.0 were randomly assigned to receive 1.25 mg of oral vitamin K or placebo, then evaluated for all forms of bleeding for 90 days. Bleeding events were defined as “major bleeding,” “minor bleeding,” and “trivial bleeding.”

Though patients who received oral vitamin K had a significantly more rapid INR decrease, there were no differences between the two groups with regard to all bleeding events, thromboembolism, or death. The study was underpowered to detect differences in major bleeding.

Bottom line: Low-dose oral vitamin K leads to more rapid correction of the INR in overanticoagulated patients on warfarin therapy, but has little effect on clinical outcomes at 90 days.

Citation: Crowther MA, Ageno W, Garcia D, et al. Oral vitamin K versus placebo to correct excessive anticoagulation in patients receiving warfarin: a randomized trial. Ann Intern Med. 2009;150(5):293-300.

Inappropriate Treatment of Catheter-Associated Asymptomatic Bacteriuria

Clinical question: Are hospitalized patients with urinary catheters inappropriately treated with antibiotics for asymptomatic bacteriuria?

Background: Persons with catheters acquire bacteriuria at the rate of 3% to 10% per day, but in the majority of cases, no symptoms or secondary complications occur. Evidenced-based guidelines state that asymptomatic bacteriuria is not a clinically significant infection, and numerous studies have shown that treatment is unlikely to confer clinical benefit.

Study design: Retrospective cohort study.

Setting: A single-site Veterans Affairs hospital.

Synopsis: Using urine culture results over a three-month period from a single VA medical center, 280 cases were analyzed: 164 catheter-associated asymptomatic bacteriuria and 116 catheter-associated urinary tract infections (UTIs). A UTI was defined as having one or more of these symptoms: fever, urgency, frequency, dysuria, suprapubic tenderness, altered mental status, or hypotension in a patient without another recognized infection and a positive urine culture. Of the asymptomatic bacteriuria cases, 68% were managed appropriately with no antibiotic treatment; 32% were inappropriately treated with antibiotics.

In multivariate analysis, older patient age, predominance of gram-negative bacteria, and higher urine white blood cell count were significantly associated with inappropriate treatment.

This study highlights the fact that antibiotics continue to be used inappropriately in patients with catheters. Current guidelines do not distinguish well between asymptomatic bacteriuria and UTI, so there might be a knowledge gap. This study was based on urine culture data, not urinalysis of all patients with a catheter, so the symptomatic patients were likely over-represented.

An associated editorial observes that the study extrapolates data from studies that involved patients with uncomplicated UTIs and, therefore, might reach erroneous conclusions. Further, viewing catheter-associated symptomatic UTIs and catheter-associated asymptomatic bactiuria as dichotomous and warranting inherently different management fails to encompass a number of clinical factors, including co-infection, and further fails to acknowledge that removal of the catheter is the first step in treatment. However, the finding that antibiotics continue to be used inappropriately is useful.

Bottom line: A clinical determination of whether a patient with a catheter really has a symptomatic UTI/urosepsis or only has asymptomatic bacteriuria should precede starting antibiotics in hospitalized patients.

Citations: Cope M, Cevallos ME, Cadle RM, Darouiche RO, Musher DM, Trautner BW. Inappropriate treatment of catheter-associated asymptomatic bateriuria in a tertiary care hospital. Clin Infect Dis. 2009;48(9):1182-1188.

Kunin CM. Catheter-associated urinary tract infections: a syllogism compounded by a questionable dichotomy. Clin Infect Dis. 2009;48:1189-1190.

Current Practices in the Evaluation and Management of Thrombocytopenia in Heparin Patients

Clinical question: Are the current American College of Chest Physicians (ACCP) guidelines for the recognition, treatment, and prevention of heparin-induced thrombocytopenia (HIT) being followed?

Background: Heparin-based anticoagulation is frequently given to hospitalized patients, and approximately 1% to 5% of these patients develop HIT. In 2004, the ACCP published a consensus statement on the evaluation, management, and prevention of HIT.

Study design: Prospective, observational study.

Setting: Forty-eight U.S. hospitals in the Complications After Thrombocytopenia Caused by Heparin (CATCH) registry.

Synopsis: The CATCH trial enrolled patients receiving any form of heparin for >96 hours (n=2,420), cardiac-care-unit patients treated with heparin (n=1,090), and patients who had an HIT antibody assay performed (n=449), for a total of 3,536 total patients. The study included patients on unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH). Thrombocytopenia was defined at a platelet count <150,000, or a decrease of 50% when compared with admission.

In the prolonged heparin group, 36.4% of patients developed thrombocytopenia; however, HIT was suspected in only 19.8% of these high-risk patients. While physicians were more likely to consider HIT in the cardiac-care patients (37.6%), the diagnosis was considered>24 hours after the thrombocytopenia developed. Physicians often waited until after a thromboembolic complication occurred before evaluating for HIT. More often than not, preventive measures were missed (e.g., failing to check for HIT antibodies, continuing heparin after HIT was suspected).

Bottom line: Thrombocytopenia is a common occurrence in patients receiving heparin and, despite the risk of devastating complications from HIT, treatment infrequently conforms to the established guidelines.

Citation: Crespo EM, Oliveira GBF, Honeycutt EF, et al. Evaluation and management of thrombocytopenia and suspected heparin-induced thrombocytopenia in hospitalized patients: The Complications After Thrombocytopenia Caused by Heparin (CATCH) registry. Am Heart J. 2009;157(4):651-657. TH

PEDIATRIC HM LITerature

Epinephrine and Dexamethasone Alone Do Not Reduce Hospital Admissions in Infants with Bronchiolitis

By Mark Shen, MD

Clinical question: Does nebulized epinephrine, oral dexamethasone, or a combination of the two result in a decrease in hospital admissions when given in the ED to infants with bronchiolitis?

Background: Bronchiolitis is the most common lower-respiratory-tract infection of infancy. Rates of admission have climbed in the past two decades. Bronchodilators and corticosteroids are not routinely recommended, and the evidence surrounding epinephrine and dexamethasone has shown varying results.

Study design: Randomized, double-blind, placebo-controlled trial.

Setting: Eight Canadian pediatric EDs.

Synopsis: Eight hundred infants ages six weeks to 12 months with a first episode of bronchiolitis were randomized to nebulized epinephrine and oral dexamethasone, epinephrine and oral placebo, dexamethasone and nebulized placebo, or oral and nebulized placebo. Epinephrine was delivered as two nebulizations 30 minutes apart, and dexamethasone was given as a 1 mg/kg oral dose, followed by five once-daily doses of 0.6 mg/kg. In unadjusted analysis, only the infants who received epinephrine and dexamethasone were less likely to be admitted to the hospital by day seven.

Limitations of this study include a small effect size and a non-statistically-significant difference between the epinephrine/dexamethasone group and placebo after statistical adjustment for multiple comparisons. Given that admission rates did not decrease in the epinephrine/placebo group and the dexamethasone/placebo group, this study supports national guidelines that do not recommend the routine use of these agents in bronchiolitis. The authors suggest that synergy between epinephrine and dexamethasone in bronchiolitis be further studied.

Bottom line: When given alone, epinephrine and dexamethasone do not reduce hospital admissions in infants with bronchiolitis.

Citation: Plint AC, Johnson DW, Patel H, et al. Epinephrine and dexamethasone in children with bronchiolitis. N Engl J Med. 2009;360(20):2079-2089.

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

In This Edition

Literature at a Glance

A guide to this month’s studies

• PE and COPD exacerbations.
• Care bundles and readmission rates.
• Family history and VTE risk.
• Vasopressor choice and mortality in sepsis.
• Vitamin K use in overanticoagulation.
• Appropriate treatment of asymptomatic bacteriuria.
• Guideline adherence in thrombocytopenia.

Pulmonary Embolism Frequently Complicates COPD Exacerbations

Clinical question: What percentage of patients with acute chronic obstructive pulmonary disease (COPD) exacerbations has pulmonary emboli?

Background: As many as 30% of COPD exacerbations have no apparent precipitating event. Even in patients with evidence of a precipitating event, such as an upper-respiratory illness or increased environmental irritants, pulmonary emboli (PE) may coexist and warrant evaluation.

Study design: Literature review.

Setting: Multiple studies in Europe and the U.S.

Synopsis: This literature review included five studies to estimate the rate of PE in patients with a COPD exacerbation. Overall incidence of PE in COPD exacerbations was 19.9%, but of those patients requiring hospitalization, the incidence was as high as 25.5%. Incidence estimates varied based on interpretation of data that were missing or inconsistent between studies. Patients most commonly present with dyspnea, chest pain, hemoptysis, cough, and palpitations. Six percent of PE patients presented with syncope; no patients with an exacerbation without a PE presented with syncope.

Risk of mortality from PE is almost twice as high in patients with a COPD exacerbation compared with PE in other settings. A significant number of patients have PE without history or evidence of DVT, so in situ thrombosis is a significant factor. The interpretation of these results is limited by the heterogeneity of the study designs, and by the relatively low number of cases. Larger trials are necessary.

Bottom line: Pulmonary emboli are present in as many as 25% of all COPD exacerbations. Delay in diagnosis of PE in COPD patients affects morbidity and mortality. PE should be a consideration in many COPD exacerbations.

Citation: Rizkallah J, Man SF, Din DD. Prevalence of pulmonary embolism in acute exacerbations of COPD: a systematic review and metaanalysis. Chest. 2009;135(3):786-793.

Targeted-Care Bundle Can Reduce ED Visits and Readmission Rates in High-Risk Elderly Patients

Clinical question: Can a care coordination bundle reduce length of stay (LOS), ED visits, or readmissions within 30 days of a hospital admission?

Background: Hospital-based care coordination interventions have shown mixed results in affecting LOS, post-discharge ED visits, and readmission rates. Although there has been some success with particular interventions, no consistent benefit has been demonstrated. Most notably, a recent meta-analysis of several different interventions showed no improvement in mortality, LOS, or readmission rates.

Study design: A randomized, controlled trial of select high-risk elderly patients.

Setting: A large teaching hospital at Baylor University Medical Center.

Synopsis: A “targeted-care bundle” was implemented with high-risk elderly patients to try to reduce LOS, readmissions, and ED visits. High-risk patients were identified by age, diagnosis-related group (DRG), number of medications at admission, comorbid conditions, and need for assistance in activities of daily living. Subjects were randomized to usual care or to receive a targeted-care bundle. The targeted-care bundle included multiple interventions. A study care coordinator provided daily patient education, including condition-specific teaching, discharge teaching and planning, and a follow-up phone call at five to seven days after discharge. A clinical pharmacist intervened for medication reconciliation at admission and discharge, medication teaching, and a follow-up phone call at five to seven days after discharge. Structured documents, including a personal health record and supplemental discharge form, were implemented.

The study had low enrollment, largely due to the requirement to obtain informed consent from all participants. Therefore, the study was underpowered to detect such target endpoints as LOS. A significant decrease in 30-day readmission rates/ED visits was noticed, but there was no persistent effect at 60 days.

The intervention was designed to use existing hospital staff in order to be practical for broad utilization. Future studies need to focus on increased enrollment to demonstrate beneficial effect.

Bottom line: Targeted health interventions focusing on education and coordination of care might effect some significant outcomes, most notably readmissions or ED visits within 30 days, but the nature of the clinical problem makes rigorous testing of interventions a challenge.

Citation: Kohler BE, Richter KM, Youngblood L, et al. Reduction of 30-day postdischarge hospital readmission or emergency department (ED) visit rates in high-risk elderly medical patients through delivery of a targeted care bundle. J Hosp Med. 2009;4(4):211-218.

Family History Is a Risk Factor for Venous Thrombosis

Clinical question: Is family history of additional value in predicting an individual’s risk of venous thrombosis once a genetic risk factor is identified?

Background: A positive family history of venous thrombosis might suggest the presence of genetic risk factors in a given family. However, it is not known whether family history is of additional significance—once a risk factor is identified—in predicting an individual’s risk for venous thrombosis.

Study design: Population-based, case-control study.

Setting: Participants in the Multiple Environmental and Genetic Assessment (MEGA) of risk factors for venous thrombosis study.

Synopsis: Recruitment, data collection, and blood samples were obtained from individuals in the MEGA study. Participants completed a questionnaire about risk factors for venous thrombosis and family history. A positive family history more than doubled the risk of venous thrombosis, and when more than one family member was affected, the risk increased fourfold. The risk for venous thrombosis increased 64 times for individuals who had a family history, genetic risk factor, and environmental risk factor when compared with those with a negative family history and no known risk factors.

The underreporting or overestimation of the prevalence of a positive family history might limit this study.

Bottom line: Family history is a risk indictor for a first venous thrombosis, despite the presence of other risk factors.

Citation: Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen CJ. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med. 2009;169(6):610-615.

Vasopressor Choice Predicts Mortality in Septic Shock

Clinical question: Does vasopressor choice affect mortality in patients with community-acquired septic shock?

Background: Community-acquired septic shock is a common illness and, despite aggressive care, a leading cause of death. Randomized clinical control trials evaluating the efficacy and safety of different adrenergic supportive agents are lacking. Thus, both norepinephrine and dopamine are recommended as first-line agents in the treatment of septic shock by the Surviving Sepsis Campaign guidelines.

Study design: Multicenter, cohort observational study.

Setting: Seventeen intensive-care units in Portugal.

Synopsis: In adjusted analysis controlling for Simplified Acute Physiology Score (SAPS) II, use of norepinephrine in community-acquired septic shock was associated with higher hospital mortality and lower 28-day survival when compared with dopamine. Specifically, patients treated with norepinephrine had a statistically significant higher hospital mortality rate than those treated with dopamine (52% and 38.5%, respectively, P=0.002) and a lower 28-day survival (log rank=22.6; P<0.001). While this data is valuable, the nonrandomized, observational study design limits firm conclusions regarding vasopressor choice. Further results from three large trials comparing vasopressor use in septic shock should continue to shed light on this debate.

Bottom line: Norepinephrine administration is associated with higher hospital mortality and lower 28-day survival when compared with dopamine in patients with community-acquired septic shock.

Citation: Póvoa PR, Carneiro AH, Ribeiro OS, Pereira AC, Portuguese Community-Acquired Sepsis Study Group. Influence of vasopressor agent in septic shock mortality. Results from the Portuguese Community-Acquired Sepsis Study (SACiUCI study). Crit Care Med. 2009;37(2):410-416.

Oral Vitamin K Versus Placebo to Correct Excess Anticoagulation in Warfarin Patients

Clinical question: In nonbleeding patients with warfarin-associated coagulopathy, does oral vitamin K reduce bleeding events when compared to placebo?

Background: Warfarin is a common drug for primary and secondary prevention of thromboembolism, but it requires continued monitoring of the international normalized ratio (INR) value. INR values >4.0 are associated with an increase in bleeding complications, with specific concern for intracranial bleeding when INR values exceed 4.5. Small, randomized trials have shown that single, low-dose administration of oral vitamin K effectively reduces the INR in nonbleeding, overanticoagulated patients.

However, these studies have not shown if vitamin K reduces risk for bleeding without increasing the risk for thromboembolism.

Study design: Randomized, placebo-controlled trial.

Setting: Fourteen anticoagulation clinics in Canada, Italy, and the U.S.

Synopsis: Nonbleeding patients with supratherapeutic INR values between 4.5 and 10.0 were randomly assigned to receive 1.25 mg of oral vitamin K or placebo, then evaluated for all forms of bleeding for 90 days. Bleeding events were defined as “major bleeding,” “minor bleeding,” and “trivial bleeding.”

Though patients who received oral vitamin K had a significantly more rapid INR decrease, there were no differences between the two groups with regard to all bleeding events, thromboembolism, or death. The study was underpowered to detect differences in major bleeding.

Bottom line: Low-dose oral vitamin K leads to more rapid correction of the INR in overanticoagulated patients on warfarin therapy, but has little effect on clinical outcomes at 90 days.

Citation: Crowther MA, Ageno W, Garcia D, et al. Oral vitamin K versus placebo to correct excessive anticoagulation in patients receiving warfarin: a randomized trial. Ann Intern Med. 2009;150(5):293-300.

Inappropriate Treatment of Catheter-Associated Asymptomatic Bacteriuria

Clinical question: Are hospitalized patients with urinary catheters inappropriately treated with antibiotics for asymptomatic bacteriuria?

Background: Persons with catheters acquire bacteriuria at the rate of 3% to 10% per day, but in the majority of cases, no symptoms or secondary complications occur. Evidenced-based guidelines state that asymptomatic bacteriuria is not a clinically significant infection, and numerous studies have shown that treatment is unlikely to confer clinical benefit.

Study design: Retrospective cohort study.

Setting: A single-site Veterans Affairs hospital.

Synopsis: Using urine culture results over a three-month period from a single VA medical center, 280 cases were analyzed: 164 catheter-associated asymptomatic bacteriuria and 116 catheter-associated urinary tract infections (UTIs). A UTI was defined as having one or more of these symptoms: fever, urgency, frequency, dysuria, suprapubic tenderness, altered mental status, or hypotension in a patient without another recognized infection and a positive urine culture. Of the asymptomatic bacteriuria cases, 68% were managed appropriately with no antibiotic treatment; 32% were inappropriately treated with antibiotics.

In multivariate analysis, older patient age, predominance of gram-negative bacteria, and higher urine white blood cell count were significantly associated with inappropriate treatment.

This study highlights the fact that antibiotics continue to be used inappropriately in patients with catheters. Current guidelines do not distinguish well between asymptomatic bacteriuria and UTI, so there might be a knowledge gap. This study was based on urine culture data, not urinalysis of all patients with a catheter, so the symptomatic patients were likely over-represented.

An associated editorial observes that the study extrapolates data from studies that involved patients with uncomplicated UTIs and, therefore, might reach erroneous conclusions. Further, viewing catheter-associated symptomatic UTIs and catheter-associated asymptomatic bactiuria as dichotomous and warranting inherently different management fails to encompass a number of clinical factors, including co-infection, and further fails to acknowledge that removal of the catheter is the first step in treatment. However, the finding that antibiotics continue to be used inappropriately is useful.

Bottom line: A clinical determination of whether a patient with a catheter really has a symptomatic UTI/urosepsis or only has asymptomatic bacteriuria should precede starting antibiotics in hospitalized patients.

Citations: Cope M, Cevallos ME, Cadle RM, Darouiche RO, Musher DM, Trautner BW. Inappropriate treatment of catheter-associated asymptomatic bateriuria in a tertiary care hospital. Clin Infect Dis. 2009;48(9):1182-1188.

Kunin CM. Catheter-associated urinary tract infections: a syllogism compounded by a questionable dichotomy. Clin Infect Dis. 2009;48:1189-1190.

Current Practices in the Evaluation and Management of Thrombocytopenia in Heparin Patients

Clinical question: Are the current American College of Chest Physicians (ACCP) guidelines for the recognition, treatment, and prevention of heparin-induced thrombocytopenia (HIT) being followed?

Background: Heparin-based anticoagulation is frequently given to hospitalized patients, and approximately 1% to 5% of these patients develop HIT. In 2004, the ACCP published a consensus statement on the evaluation, management, and prevention of HIT.

Study design: Prospective, observational study.

Setting: Forty-eight U.S. hospitals in the Complications After Thrombocytopenia Caused by Heparin (CATCH) registry.

Synopsis: The CATCH trial enrolled patients receiving any form of heparin for >96 hours (n=2,420), cardiac-care-unit patients treated with heparin (n=1,090), and patients who had an HIT antibody assay performed (n=449), for a total of 3,536 total patients. The study included patients on unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH). Thrombocytopenia was defined at a platelet count <150,000, or a decrease of 50% when compared with admission.

In the prolonged heparin group, 36.4% of patients developed thrombocytopenia; however, HIT was suspected in only 19.8% of these high-risk patients. While physicians were more likely to consider HIT in the cardiac-care patients (37.6%), the diagnosis was considered>24 hours after the thrombocytopenia developed. Physicians often waited until after a thromboembolic complication occurred before evaluating for HIT. More often than not, preventive measures were missed (e.g., failing to check for HIT antibodies, continuing heparin after HIT was suspected).

Bottom line: Thrombocytopenia is a common occurrence in patients receiving heparin and, despite the risk of devastating complications from HIT, treatment infrequently conforms to the established guidelines.

Citation: Crespo EM, Oliveira GBF, Honeycutt EF, et al. Evaluation and management of thrombocytopenia and suspected heparin-induced thrombocytopenia in hospitalized patients: The Complications After Thrombocytopenia Caused by Heparin (CATCH) registry. Am Heart J. 2009;157(4):651-657. TH

PEDIATRIC HM LITerature

Epinephrine and Dexamethasone Alone Do Not Reduce Hospital Admissions in Infants with Bronchiolitis

By Mark Shen, MD

Clinical question: Does nebulized epinephrine, oral dexamethasone, or a combination of the two result in a decrease in hospital admissions when given in the ED to infants with bronchiolitis?

Background: Bronchiolitis is the most common lower-respiratory-tract infection of infancy. Rates of admission have climbed in the past two decades. Bronchodilators and corticosteroids are not routinely recommended, and the evidence surrounding epinephrine and dexamethasone has shown varying results.

Study design: Randomized, double-blind, placebo-controlled trial.

Setting: Eight Canadian pediatric EDs.

Synopsis: Eight hundred infants ages six weeks to 12 months with a first episode of bronchiolitis were randomized to nebulized epinephrine and oral dexamethasone, epinephrine and oral placebo, dexamethasone and nebulized placebo, or oral and nebulized placebo. Epinephrine was delivered as two nebulizations 30 minutes apart, and dexamethasone was given as a 1 mg/kg oral dose, followed by five once-daily doses of 0.6 mg/kg. In unadjusted analysis, only the infants who received epinephrine and dexamethasone were less likely to be admitted to the hospital by day seven.

Limitations of this study include a small effect size and a non-statistically-significant difference between the epinephrine/dexamethasone group and placebo after statistical adjustment for multiple comparisons. Given that admission rates did not decrease in the epinephrine/placebo group and the dexamethasone/placebo group, this study supports national guidelines that do not recommend the routine use of these agents in bronchiolitis. The authors suggest that synergy between epinephrine and dexamethasone in bronchiolitis be further studied.

Bottom line: When given alone, epinephrine and dexamethasone do not reduce hospital admissions in infants with bronchiolitis.

Citation: Plint AC, Johnson DW, Patel H, et al. Epinephrine and dexamethasone in children with bronchiolitis. N Engl J Med. 2009;360(20):2079-2089.

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.

In This Edition

Literature at a Glance

A guide to this month’s studies

• PE and COPD exacerbations.
• Care bundles and readmission rates.
• Family history and VTE risk.
• Vasopressor choice and mortality in sepsis.
• Vitamin K use in overanticoagulation.
• Appropriate treatment of asymptomatic bacteriuria.
• Guideline adherence in thrombocytopenia.

Pulmonary Embolism Frequently Complicates COPD Exacerbations

Clinical question: What percentage of patients with acute chronic obstructive pulmonary disease (COPD) exacerbations has pulmonary emboli?

Background: As many as 30% of COPD exacerbations have no apparent precipitating event. Even in patients with evidence of a precipitating event, such as an upper-respiratory illness or increased environmental irritants, pulmonary emboli (PE) may coexist and warrant evaluation.

Study design: Literature review.

Setting: Multiple studies in Europe and the U.S.

Synopsis: This literature review included five studies to estimate the rate of PE in patients with a COPD exacerbation. Overall incidence of PE in COPD exacerbations was 19.9%, but of those patients requiring hospitalization, the incidence was as high as 25.5%. Incidence estimates varied based on interpretation of data that were missing or inconsistent between studies. Patients most commonly present with dyspnea, chest pain, hemoptysis, cough, and palpitations. Six percent of PE patients presented with syncope; no patients with an exacerbation without a PE presented with syncope.

Risk of mortality from PE is almost twice as high in patients with a COPD exacerbation compared with PE in other settings. A significant number of patients have PE without history or evidence of DVT, so in situ thrombosis is a significant factor. The interpretation of these results is limited by the heterogeneity of the study designs, and by the relatively low number of cases. Larger trials are necessary.

Bottom line: Pulmonary emboli are present in as many as 25% of all COPD exacerbations. Delay in diagnosis of PE in COPD patients affects morbidity and mortality. PE should be a consideration in many COPD exacerbations.

Citation: Rizkallah J, Man SF, Din DD. Prevalence of pulmonary embolism in acute exacerbations of COPD: a systematic review and metaanalysis. Chest. 2009;135(3):786-793.

Targeted-Care Bundle Can Reduce ED Visits and Readmission Rates in High-Risk Elderly Patients

Clinical question: Can a care coordination bundle reduce length of stay (LOS), ED visits, or readmissions within 30 days of a hospital admission?

Background: Hospital-based care coordination interventions have shown mixed results in affecting LOS, post-discharge ED visits, and readmission rates. Although there has been some success with particular interventions, no consistent benefit has been demonstrated. Most notably, a recent meta-analysis of several different interventions showed no improvement in mortality, LOS, or readmission rates.

Study design: A randomized, controlled trial of select high-risk elderly patients.

Setting: A large teaching hospital at Baylor University Medical Center.

Synopsis: A “targeted-care bundle” was implemented with high-risk elderly patients to try to reduce LOS, readmissions, and ED visits. High-risk patients were identified by age, diagnosis-related group (DRG), number of medications at admission, comorbid conditions, and need for assistance in activities of daily living. Subjects were randomized to usual care or to receive a targeted-care bundle. The targeted-care bundle included multiple interventions. A study care coordinator provided daily patient education, including condition-specific teaching, discharge teaching and planning, and a follow-up phone call at five to seven days after discharge. A clinical pharmacist intervened for medication reconciliation at admission and discharge, medication teaching, and a follow-up phone call at five to seven days after discharge. Structured documents, including a personal health record and supplemental discharge form, were implemented.

The study had low enrollment, largely due to the requirement to obtain informed consent from all participants. Therefore, the study was underpowered to detect such target endpoints as LOS. A significant decrease in 30-day readmission rates/ED visits was noticed, but there was no persistent effect at 60 days.

The intervention was designed to use existing hospital staff in order to be practical for broad utilization. Future studies need to focus on increased enrollment to demonstrate beneficial effect.

Bottom line: Targeted health interventions focusing on education and coordination of care might effect some significant outcomes, most notably readmissions or ED visits within 30 days, but the nature of the clinical problem makes rigorous testing of interventions a challenge.

Citation: Kohler BE, Richter KM, Youngblood L, et al. Reduction of 30-day postdischarge hospital readmission or emergency department (ED) visit rates in high-risk elderly medical patients through delivery of a targeted care bundle. J Hosp Med. 2009;4(4):211-218.

Family History Is a Risk Factor for Venous Thrombosis

Clinical question: Is family history of additional value in predicting an individual’s risk of venous thrombosis once a genetic risk factor is identified?

Background: A positive family history of venous thrombosis might suggest the presence of genetic risk factors in a given family. However, it is not known whether family history is of additional significance—once a risk factor is identified—in predicting an individual’s risk for venous thrombosis.

Study design: Population-based, case-control study.

Setting: Participants in the Multiple Environmental and Genetic Assessment (MEGA) of risk factors for venous thrombosis study.

Synopsis: Recruitment, data collection, and blood samples were obtained from individuals in the MEGA study. Participants completed a questionnaire about risk factors for venous thrombosis and family history. A positive family history more than doubled the risk of venous thrombosis, and when more than one family member was affected, the risk increased fourfold. The risk for venous thrombosis increased 64 times for individuals who had a family history, genetic risk factor, and environmental risk factor when compared with those with a negative family history and no known risk factors.

The underreporting or overestimation of the prevalence of a positive family history might limit this study.

Bottom line: Family history is a risk indictor for a first venous thrombosis, despite the presence of other risk factors.

Citation: Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen CJ. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med. 2009;169(6):610-615.

Vasopressor Choice Predicts Mortality in Septic Shock

Clinical question: Does vasopressor choice affect mortality in patients with community-acquired septic shock?

Background: Community-acquired septic shock is a common illness and, despite aggressive care, a leading cause of death. Randomized clinical control trials evaluating the efficacy and safety of different adrenergic supportive agents are lacking. Thus, both norepinephrine and dopamine are recommended as first-line agents in the treatment of septic shock by the Surviving Sepsis Campaign guidelines.

Study design: Multicenter, cohort observational study.

Setting: Seventeen intensive-care units in Portugal.

Synopsis: In adjusted analysis controlling for Simplified Acute Physiology Score (SAPS) II, use of norepinephrine in community-acquired septic shock was associated with higher hospital mortality and lower 28-day survival when compared with dopamine. Specifically, patients treated with norepinephrine had a statistically significant higher hospital mortality rate than those treated with dopamine (52% and 38.5%, respectively, P=0.002) and a lower 28-day survival (log rank=22.6; P<0.001). While this data is valuable, the nonrandomized, observational study design limits firm conclusions regarding vasopressor choice. Further results from three large trials comparing vasopressor use in septic shock should continue to shed light on this debate.

Bottom line: Norepinephrine administration is associated with higher hospital mortality and lower 28-day survival when compared with dopamine in patients with community-acquired septic shock.

Citation: Póvoa PR, Carneiro AH, Ribeiro OS, Pereira AC, Portuguese Community-Acquired Sepsis Study Group. Influence of vasopressor agent in septic shock mortality. Results from the Portuguese Community-Acquired Sepsis Study (SACiUCI study). Crit Care Med. 2009;37(2):410-416.

Oral Vitamin K Versus Placebo to Correct Excess Anticoagulation in Warfarin Patients

Clinical question: In nonbleeding patients with warfarin-associated coagulopathy, does oral vitamin K reduce bleeding events when compared to placebo?

Background: Warfarin is a common drug for primary and secondary prevention of thromboembolism, but it requires continued monitoring of the international normalized ratio (INR) value. INR values >4.0 are associated with an increase in bleeding complications, with specific concern for intracranial bleeding when INR values exceed 4.5. Small, randomized trials have shown that single, low-dose administration of oral vitamin K effectively reduces the INR in nonbleeding, overanticoagulated patients.

However, these studies have not shown if vitamin K reduces risk for bleeding without increasing the risk for thromboembolism.

Study design: Randomized, placebo-controlled trial.

Setting: Fourteen anticoagulation clinics in Canada, Italy, and the U.S.

Synopsis: Nonbleeding patients with supratherapeutic INR values between 4.5 and 10.0 were randomly assigned to receive 1.25 mg of oral vitamin K or placebo, then evaluated for all forms of bleeding for 90 days. Bleeding events were defined as “major bleeding,” “minor bleeding,” and “trivial bleeding.”

Though patients who received oral vitamin K had a significantly more rapid INR decrease, there were no differences between the two groups with regard to all bleeding events, thromboembolism, or death. The study was underpowered to detect differences in major bleeding.

Bottom line: Low-dose oral vitamin K leads to more rapid correction of the INR in overanticoagulated patients on warfarin therapy, but has little effect on clinical outcomes at 90 days.

Citation: Crowther MA, Ageno W, Garcia D, et al. Oral vitamin K versus placebo to correct excessive anticoagulation in patients receiving warfarin: a randomized trial. Ann Intern Med. 2009;150(5):293-300.

Inappropriate Treatment of Catheter-Associated Asymptomatic Bacteriuria

Clinical question: Are hospitalized patients with urinary catheters inappropriately treated with antibiotics for asymptomatic bacteriuria?

Background: Persons with catheters acquire bacteriuria at the rate of 3% to 10% per day, but in the majority of cases, no symptoms or secondary complications occur. Evidenced-based guidelines state that asymptomatic bacteriuria is not a clinically significant infection, and numerous studies have shown that treatment is unlikely to confer clinical benefit.

Study design: Retrospective cohort study.

Setting: A single-site Veterans Affairs hospital.

Synopsis: Using urine culture results over a three-month period from a single VA medical center, 280 cases were analyzed: 164 catheter-associated asymptomatic bacteriuria and 116 catheter-associated urinary tract infections (UTIs). A UTI was defined as having one or more of these symptoms: fever, urgency, frequency, dysuria, suprapubic tenderness, altered mental status, or hypotension in a patient without another recognized infection and a positive urine culture. Of the asymptomatic bacteriuria cases, 68% were managed appropriately with no antibiotic treatment; 32% were inappropriately treated with antibiotics.

In multivariate analysis, older patient age, predominance of gram-negative bacteria, and higher urine white blood cell count were significantly associated with inappropriate treatment.

This study highlights the fact that antibiotics continue to be used inappropriately in patients with catheters. Current guidelines do not distinguish well between asymptomatic bacteriuria and UTI, so there might be a knowledge gap. This study was based on urine culture data, not urinalysis of all patients with a catheter, so the symptomatic patients were likely over-represented.

An associated editorial observes that the study extrapolates data from studies that involved patients with uncomplicated UTIs and, therefore, might reach erroneous conclusions. Further, viewing catheter-associated symptomatic UTIs and catheter-associated asymptomatic bactiuria as dichotomous and warranting inherently different management fails to encompass a number of clinical factors, including co-infection, and further fails to acknowledge that removal of the catheter is the first step in treatment. However, the finding that antibiotics continue to be used inappropriately is useful.

Bottom line: A clinical determination of whether a patient with a catheter really has a symptomatic UTI/urosepsis or only has asymptomatic bacteriuria should precede starting antibiotics in hospitalized patients.

Citations: Cope M, Cevallos ME, Cadle RM, Darouiche RO, Musher DM, Trautner BW. Inappropriate treatment of catheter-associated asymptomatic bateriuria in a tertiary care hospital. Clin Infect Dis. 2009;48(9):1182-1188.

Kunin CM. Catheter-associated urinary tract infections: a syllogism compounded by a questionable dichotomy. Clin Infect Dis. 2009;48:1189-1190.

Current Practices in the Evaluation and Management of Thrombocytopenia in Heparin Patients

Clinical question: Are the current American College of Chest Physicians (ACCP) guidelines for the recognition, treatment, and prevention of heparin-induced thrombocytopenia (HIT) being followed?

Background: Heparin-based anticoagulation is frequently given to hospitalized patients, and approximately 1% to 5% of these patients develop HIT. In 2004, the ACCP published a consensus statement on the evaluation, management, and prevention of HIT.

Study design: Prospective, observational study.

Setting: Forty-eight U.S. hospitals in the Complications After Thrombocytopenia Caused by Heparin (CATCH) registry.

Synopsis: The CATCH trial enrolled patients receiving any form of heparin for >96 hours (n=2,420), cardiac-care-unit patients treated with heparin (n=1,090), and patients who had an HIT antibody assay performed (n=449), for a total of 3,536 total patients. The study included patients on unfractionated heparin (UFH) or low-molecular-weight heparin (LMWH). Thrombocytopenia was defined at a platelet count <150,000, or a decrease of 50% when compared with admission.

In the prolonged heparin group, 36.4% of patients developed thrombocytopenia; however, HIT was suspected in only 19.8% of these high-risk patients. While physicians were more likely to consider HIT in the cardiac-care patients (37.6%), the diagnosis was considered>24 hours after the thrombocytopenia developed. Physicians often waited until after a thromboembolic complication occurred before evaluating for HIT. More often than not, preventive measures were missed (e.g., failing to check for HIT antibodies, continuing heparin after HIT was suspected).

Bottom line: Thrombocytopenia is a common occurrence in patients receiving heparin and, despite the risk of devastating complications from HIT, treatment infrequently conforms to the established guidelines.

Citation: Crespo EM, Oliveira GBF, Honeycutt EF, et al. Evaluation and management of thrombocytopenia and suspected heparin-induced thrombocytopenia in hospitalized patients: The Complications After Thrombocytopenia Caused by Heparin (CATCH) registry. Am Heart J. 2009;157(4):651-657. TH

PEDIATRIC HM LITerature

Epinephrine and Dexamethasone Alone Do Not Reduce Hospital Admissions in Infants with Bronchiolitis

By Mark Shen, MD

Clinical question: Does nebulized epinephrine, oral dexamethasone, or a combination of the two result in a decrease in hospital admissions when given in the ED to infants with bronchiolitis?

Background: Bronchiolitis is the most common lower-respiratory-tract infection of infancy. Rates of admission have climbed in the past two decades. Bronchodilators and corticosteroids are not routinely recommended, and the evidence surrounding epinephrine and dexamethasone has shown varying results.

Study design: Randomized, double-blind, placebo-controlled trial.

Setting: Eight Canadian pediatric EDs.

Synopsis: Eight hundred infants ages six weeks to 12 months with a first episode of bronchiolitis were randomized to nebulized epinephrine and oral dexamethasone, epinephrine and oral placebo, dexamethasone and nebulized placebo, or oral and nebulized placebo. Epinephrine was delivered as two nebulizations 30 minutes apart, and dexamethasone was given as a 1 mg/kg oral dose, followed by five once-daily doses of 0.6 mg/kg. In unadjusted analysis, only the infants who received epinephrine and dexamethasone were less likely to be admitted to the hospital by day seven.

Limitations of this study include a small effect size and a non-statistically-significant difference between the epinephrine/dexamethasone group and placebo after statistical adjustment for multiple comparisons. Given that admission rates did not decrease in the epinephrine/placebo group and the dexamethasone/placebo group, this study supports national guidelines that do not recommend the routine use of these agents in bronchiolitis. The authors suggest that synergy between epinephrine and dexamethasone in bronchiolitis be further studied.

Bottom line: When given alone, epinephrine and dexamethasone do not reduce hospital admissions in infants with bronchiolitis.

Citation: Plint AC, Johnson DW, Patel H, et al. Epinephrine and dexamethasone in children with bronchiolitis. N Engl J Med. 2009;360(20):2079-2089.

Reviewed by Pediatric Editor Mark Shen, MD, medical director of hospital medicine at Dell Children’s Medical Center, Austin, Texas.