User login
In This Edition
Literature at a Glance: A guide to this month’s studies
- PPI use with clopidrogrel in ACS.
- Chlorhexidine sponge use reduces line infections.
- Extended thienopyridine use does not benefit DES patients.
- CABG is revascularization choice for severe CAD.
- Pre-treated CVC use reduces bloodstream infections.
- Hospitalist use grows in U.S.
- Sepsis order set improves outcomes.
- Admission day predicts acute PE mortality.
PEDIATRIC HM LITerature
Early Transition to Oral Therapy in Acute Osteomyelitis in Children Associated with Fewer Complications I By Mark Shen, MD
Clinical question: Can acute osteomyelitis in children be managed with early transition to oral antimicrobial therapy?
Background: Traditional treatment of acute osteomyelitis in children has involved prolonged intravenous antimicrobial therapy, typically greater than three to four weeks in duration. Small studies suggest the feasibility of a transition to prolonged oral therapy after an initial response to intravenous antimicrobial agents.
Study design: Retrospective, cohort study.
Setting: 29 freestanding children’s hospitals.
Synopsis: Using the Pediatric Health Information System (PHIS) database, 1,969 children were identified via ICD-9-CM codes as having been hospitalized with acute osteomyelitis—and without comorbid conditions—between January 2000 and June 2005. Rehospitalizations were reviewed for complications of both treatment failure and the treatment itself. More than half (1,021) of the children underwent central venous catheterization for prolonged intravenous (IV) therapy; 948 were assigned to the oral therapy group. No significant differences in treatment failure existed between the two groups (5% in the IV group; 4% in the oral group).
Overall, the clinical characteristics of the two groups were indistinguishable. Propensity score analysis was used to handle possible patient-level confounders; a validation study was performed to address misclassifications in assignment. No significant confounding effects were found.
Secondary findings included marked variation across hospitals in the use of oral therapy and increased treatment complications in the IV group. Although unmeasured factors not present in an administrative database may affect results, the strikingly similar groups and outcomes suggest the data are relevant.
Bottom line: Early transition to oral therapy for acute osteomyelitis in children does not increase treatment failure and is associated with fewer treatment complications.
Reference: Zaoutis T, Localio AR, Leckerman K, Saddlemire S, Bertoch D, Keren R. Prolonged intravenous therapy versus early transition to oral antimicrobial therapy for acute osteomyelitis in children. Pediatrics. 2009;123:636-642.
PPI Use with Clopidogrel in Acute Coronary Syndrome Is Associated with Readmissions and Mortality
Clinical question: Does concomitant use of clopidogrel and a proton pump inhibitor (PPI) following hospitalization for acute coronary syndrome (ACS) lead to adverse outcomes?
Background: Prophylactic PPIs often are prescribed with clopidogrel to reduce the risk of gastrointestinal bleeding. Mechanistic studies have shown that omeprazole decreases the platelet-inhibitory effect of clopidogrel, raising concerns that PPIs might interfere with clopidogrel’s beneficial effects. The clinical significance of this finding is unknown.
Study design: Retrospective cohort study.
Setting: 127 VA hospitals.
Synopsis: Investigators used data from the Cardiac Care Follow-up Clinical Study and VA pharmacy records to examine 8,205 male veterans who were hospitalized for ACS and treated with clopidogrel. Patients who filled prescriptions for both clopidogrel and a PPI were at significantly higher risk for death or readmission with ACS compared with those who filled prescriptions for clopidogrel only (adjusted odds ratio, 1.25; 95% confidence interval, 1.11-1.41). Patients who filled prescriptions for PPIs alone had similar risk for adverse events as those who took neither medication.
Subanalyses found similarly increased risk among patients prescribed omeprazole and rabeprazole, but those taking lanzoprazole and pantoprazole were not examined due to the small sample size. Although causality cannot be inferred from this observational study, and the risk associated with combined clopidigrel and PPI use appeared small, alternatives for gastric acid reduction exist. Thus, it may be prudent to restrict PPI use to patients who have a clear indication for their use until more definitive clinical trials can be conducted.
Bottom line: Among patients who are treated with clopidogrel for ACS, PPIs should be reserved for patients with a clear indication for gastric acid reduction and who cannot use alternative therapies.
Citation: Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. JAMA. 2009;301(9):937-944.
Chlorhexidine-Impregnated Sponge Use Reduces Line Infections
Clinical question: Does dressing vascular catheters with chlorhexidine gluconate-impregnated sponges (CHGIS) reduce rates of catheter-related infections, and are dressing changes every seven days inferior to every three days?
Background: Process improvement strategies—including educating providers, strictly adhering to sterile technique, and promptly removing unnecessary catheters—greatly decrease catheter-related infections. It is unclear if CHGIS dressings offer additional benefit. Also uncertain is whether weekly dressing changes are as safe as changing dressings every three days.
Study design: A 2x2 factorial, assessor-blinded, randomized controlled trial.
Setting: ICUs in three university hospitals and two general hospitals in France.
Synopsis: 1,636 French adults expected to require arterial and central venous catheters for >48 hours were randomly assigned to one of four groups. Each group received either CHGIS dressings or standard dressings, and each group had dressing changes every three or seven days. Dressings were changed sooner if soiled or nonadherent. CHGIS dressings were associated with fewer catheter-related infections than standard dressings (0.6 vs. 1.4 infections per 1,000 catheter days; P=0.03). No significant difference in rates of catheter colonization existed between the three-day and seven-day dressing change strategies (10.4 vs. 11 events per 1,000 catheter days, P>0.05).
Although microbiology assessors were blinded to patients’ status, the ICU staff was not, potentially creating experimenter bias. Approximately 30% of the venous catheters and 40% of the arterial catheters were in a femoral site. Secondary analyses found higher rates of severe dermatitis among patients with CHGIS dressings but no difference in minimal bactericidal concentration (MBC) or colonizing organisms. Preliminary calculations suggested CHGIS dressings could be cost-effective.
Bottom line: Among critically ill adults, CHGIS catheter dressings may marginally reduce catheter-related infection rates, but further evaluation is needed before this technology can be adopted widely.
Citation: Timsit JF, Schwebel C, Bouadma L, et al. Chlorhexidine-impregnated sponges and less frequent dressing changes for prevention of catheter-related infections in critically-ill adults: a randomized controlled trial. JAMA. 2009;301(12):1231-1241.
Thienopyridine Use Six Months after Sirolimus-Eluting Stent Implant-ation Offers No Benefit
Clinical question: What are the relative contributions of aspirin and thienopyridine on preventing stent thrombosis in patients with sirolimus-eluting stents?
Background: There are no randomized clinical trials addressing the optimal duration, or the risks associated with discontinuation, of dual-antiplatelet therapy after drug-eluting stent (DES) implantation. Nevertheless, many patients continue to be maintained on dual-antiplatelet therapy beyond one year of their index DES implantation.
Study design: Prospective multicenter observational study.
Setting: Hospitals in Japan.
Synopsis: This study observed 10,778 Japanese patients undergoing sirolimus-eluting stent implantation. Patients discontinuing both thienopyridine and aspirin had a significantly higher rate of stent thrombosis than those who continued both medications for up to 18 months. However, discontinuation of thienopyridine alone was not associated with an excess risk of stent thrombosis. Additionally, a landmark analysis of patients who were free of events at six months showed rates of death for myocardial infarction (MI) at 24 months were 4.1% for patients taking thienopyridine and 4.1% for patients not taking thienopyridine (P=0.99). Ticlodipine was the thienopyridine used by more than 95% of patients.
Hospitalists should be aware that the role thienopyridine therapy plays in reducing stent thrombosis beyond one month after implantation has not been well addressed.
Bottom line: Discontinuation of thienopyridine therapy after six months while maintaining aspirin therapy is not associated with increased risk of stent thrombosis in patients with sirolimus-eluting stents.
Citation: Kimura T, Morimoto T, Nakagawa Y, et al. Antiplatelet therapy and stent thrombosis after sirolimus-eluting stent implantation. Circulation. 2009;119(7):987-995.
Compared with PCI, CABG Results in Lower Rates of Major Adverse Events in Severe CAD Patients
Clinical question: What is the optimal revascularization strategy for previously untreated severe coronary artery disease (CAD)?
Background: Coronary artery bypass grafting (CABG) is the treatment of choice in three-vessel and left-main CAD. However, percutaneous coronary intervention (PCI) with drug-eluting stents often is utilized despite the lack of adequately powered randomized trials.
Study design: Prospective multicenter randomized clinical trial.
Setting: 85 hospitals in Europe and the U.S.
Synopsis: 1,800 patients with an average age of 65 and previously untreated three-vessel or left-main CAD amenable to therapy with both PCI and CABG were randomized to CABG or PCI. The primary combined endpoint was a major adverse cardiac or cerebrovascular event, defined as death, stroke, MI, or repeat revascularization. PCI was associated with a significantly higher rate of major adverse cardiac or cerebrovascular events, due mostly to a higher rate of repeat revascularization (13.5% vs. 5.9%, P<0.001). At 12 months, the two groups had similar rates of death from any cause or MI, and similar rates of the combined endpoint of death from any cause, stroke, or MI; however, the rate of stroke was 1.6% higher in the CABG group.
Hospitalists should continue to favor CABG over PCI but give consideration to the risks involved with such an intervention.
Bottom line: CABG remains the revascularization choice in patients with severe CAD.
Citation: Serruys PW, Morice MC, Kappetein AP, et al. Percuta-neous coronary intervention versus coronary artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-972.
Pre-Medicated Central Venous Catheters Reduce Risk of Catheter-Related Bloodstream Infections
Clinical question: Does pre-treating central venous catheters with anti-infective agents prevent catheter-related bloodstream infections?
Background: Use of central venous catheters (CVC) is associated with catheter-related bloodstream infection (CRBSI), with CRBSI-related mortality rates as high as 25%. Previous reviews have indicated that CVCs coated or impregnated with anti-infectives may reduce CRBSI incidence. This review integrates new trial data with information from prior reviews.
Study design: Meta-analysis of 27 randomized controlled trials.
Setting: Meta-analysis.
Synopsis: The authors report CVCs pre-treated with anti-infectives (AI-CVCs) are clinically effective in reducing the risk of CRBSI. The odds of having a CRBSI with a treated CVC versus an untreated CVC are 0.49 to 1 (95% CI, 0.37–0.64, 27 studies, fixed effects). The study also finds the use of AI-CVCs might provide a large cost savings in Great Britain. Because the findings are based on a meta-analysis, they are limited by the quality, context, and consistency of the original studies. The authors note that many of the studies had unsatisfactory descriptions of methodology. The current study is unable to separate the risk reduction attributable to AI-CVC versus that attributable to other infection control practices. Also, original data is insufficient to assess the benefits of AI-CVCs placed for longer than 12 days.
To summarize, AI-CVCs may present a means to reduce CRBSI, but more investigation of its role within infection control protocols is needed, as is investigation of longer duration of treatment.
Bottom line: Central venous catheters pre-treated with anti-infectives significantly reduce catheter-related bloodstream infections.
Citation: Hockenhull JC, Dwan KM, Smith GW, et al. The clinical effectiveness of central venous catheters treated with anti-infective agents in preventing catheter-related bloodstream infections: a systematic review. Crit Care Med. 2009;37(2):702-712.
Fivefold Increase in Hospitalists in the U.S. from 1995 to 2006
Clinical question: What is the growth rate of hospitalists and hospitalist-provided care?
Background: Survey data has shown a sharp increase in the number of hospitalists, but until now there have not been any national or population-based data on the growth of hospitalist care.
Study design: Descriptive analysis.
Setting: Medicare-enrolled patients.
Synopsis: The study is based on national Medicare data from 2.1 million admissions involving 990,785 patients in 5,800 hospitals and 120,226 general internists. It represents 5% of inpatient Medicare claims generated by general internists. The authors define “hospitalist” as a general internist who generates >90% of his or her claims from the care of hospitalized patients.
U.S. hospitals have seen substantial growth in hospitalists over the period examined. The nation saw a 500% increase in the number of general-internist hospitalists, and a 28% increase (to 37.1% in 2006 from 9.1% in 1995) in the number of Medicare patients who received care from a hospitalist. The odds that a hospitalized Medicare patient received care from a hospitalist increased 29.2% per year from 1997 to 2006. The percentage of hospitals with at least three hospitalists rose to 47.1% in 2006 from 11.6% in 1995.
This analysis might actually have underestimated HM’s growth. Analysis of Medicare claims does not identify pediatric hospitalists and hospitalists who work exclusively within HMOs. This analysis also did not include family practitioners or internal-medicine subspecialists who are hospitalists.
Bottom line: Medicare claims data confirm survey data findings: Hospitalists and hospitalist care has grown sharply over the last decade.
Citation: Kuo YF, Sharma G, Freeman JL, Goodwin JS. Growth in the care of older patients by hospitalists in the United States. N Engl J Med. 2009;360:1102-1012.
Standardized Order Set for Bacteremic Sepsis Improves LOS and Mortality
Clinical question: Does a standardized order set for bacteremic sepsis impact patient management and outcomes?
Background: Prompt cardiovascular resuscitation and appropriate antibiotics decrease morbidity and mortality in bacteremic sepsis. This study examined whether hospitalwide, standardized sepsis order set improved management and outcomes.
Study design: Retrospective, before-and-after study design.
Setting: 1,200-bed academic medical center.
Synopsis: Two hundred patients with bacteremic severe sepsis were randomly selected from 18 months before the order set was introduced, and 200 were selected from 18 months after the order set was introduced. Primary outcomes measured were quantity of fluid administered and appropriate initial antibiotics. Secondary outcomes measured were hospital mortality and length of stay. Patients in the “after” group received more intravenous fluid (1627±1862 ml vs. 2054±2237 ml, P=.04), more appropriate antibiotics (53.0% vs. 65.5%, P=.01), had shorter hospital stays (28.7±30.1 days vs. 22.4±20.9 days, P=.02), and decreased in-house mortality (55.0% vs. 39.5%, P =<0.01).
The retrospective design of the study limited its ability to determine causal relationship. Extensive education may have contributed to the change (Hawthorne effect). Management in the ICU and ED, not the hospital wards, was the primary reason for mortality difference.
Bottom line: A standardized order set for bacteremic sepsis was associated with increased compliance with evidence-based treatment and improved outcomes. Hospitalists should promptly treat bacteremic sepsis with appropriate fluid resuscitation and antibiotics.
Citation: Thiel SW, Asghar MF, Micek ST, Reichley RM, Doherty JA, Kollef MH. Hospital-wide impact of a standardized order set for the management of bacteremic severe sepsis. Crit Care Med. 2009;37(3):819-824.
Admission Day of the Week Predicts Mortality in Patients with Acute Pulmonary Embolus
Clinical question: Do weekend pulmonary embolus (PE) admissions have worse outcomes than weekday admissions?
Background: Studies of patients with acute cardiovascular diagnoses (e.g., stroke, cardiac arrest) have shown higher short-term mortality and longer length of stay (LOS) for weekend versus weekday admissions. PE diagnosis is complex, requiring timely testing and experienced staff who are sometimes unavailable on weekends. Optimal anticoagulation therapy also depends on provider skill.
Study design: Retrospective observational study.
Setting: 186 private Pennsylvania hospitals, January 2000 through November 2002.
Synopsis: Using the Pennsylvania Health Care Cost Containment Council database, the authors reviewed 15,531 records of patients with a primary or secondary PE diagnosis code. The primary outcome was all-cause mortality over 30 days; LOS was the secondary outcome.
Weekend admissions in the highest severity of illness risk class had higher 30-day mortality than weekday admissions. Weekend admissions were significantly more likely than weekday admissions to be clinically unstable and to have abnormal lab parameters. Adjusted for severity of illness risk class, overall mortality was 1.4% higher for weekend versus weekday admissions. All excess mortality came from the sickest group of patients. LOS did not differ.
Less-experienced caregivers or delayed diagnostic testing may play a role in poor outcomes. Patients admitted on weekends might receive delayed care from the first onset of symptoms. This is important because timely therapy has been shown to influence outcomes in acute PE. Reasons for these observed differences should be explored further to help provide more consistent PE management, regardless of admission day.
Bottom line: The sickest patients with PE admitted on weekends experienced small but significantly greater 30-day mortality compared with those admitted on weekdays.
Citation: Aujesky D, Jimenez D, Mor M, Geng M, Fine M, Ibrahim S. Weekend versus weekday admission and mortality after acute pulmonary embolism. Circulation. 2009;119:962-968. TH
In This Edition
Literature at a Glance: A guide to this month’s studies
- PPI use with clopidrogrel in ACS.
- Chlorhexidine sponge use reduces line infections.
- Extended thienopyridine use does not benefit DES patients.
- CABG is revascularization choice for severe CAD.
- Pre-treated CVC use reduces bloodstream infections.
- Hospitalist use grows in U.S.
- Sepsis order set improves outcomes.
- Admission day predicts acute PE mortality.
PEDIATRIC HM LITerature
Early Transition to Oral Therapy in Acute Osteomyelitis in Children Associated with Fewer Complications I By Mark Shen, MD
Clinical question: Can acute osteomyelitis in children be managed with early transition to oral antimicrobial therapy?
Background: Traditional treatment of acute osteomyelitis in children has involved prolonged intravenous antimicrobial therapy, typically greater than three to four weeks in duration. Small studies suggest the feasibility of a transition to prolonged oral therapy after an initial response to intravenous antimicrobial agents.
Study design: Retrospective, cohort study.
Setting: 29 freestanding children’s hospitals.
Synopsis: Using the Pediatric Health Information System (PHIS) database, 1,969 children were identified via ICD-9-CM codes as having been hospitalized with acute osteomyelitis—and without comorbid conditions—between January 2000 and June 2005. Rehospitalizations were reviewed for complications of both treatment failure and the treatment itself. More than half (1,021) of the children underwent central venous catheterization for prolonged intravenous (IV) therapy; 948 were assigned to the oral therapy group. No significant differences in treatment failure existed between the two groups (5% in the IV group; 4% in the oral group).
Overall, the clinical characteristics of the two groups were indistinguishable. Propensity score analysis was used to handle possible patient-level confounders; a validation study was performed to address misclassifications in assignment. No significant confounding effects were found.
Secondary findings included marked variation across hospitals in the use of oral therapy and increased treatment complications in the IV group. Although unmeasured factors not present in an administrative database may affect results, the strikingly similar groups and outcomes suggest the data are relevant.
Bottom line: Early transition to oral therapy for acute osteomyelitis in children does not increase treatment failure and is associated with fewer treatment complications.
Reference: Zaoutis T, Localio AR, Leckerman K, Saddlemire S, Bertoch D, Keren R. Prolonged intravenous therapy versus early transition to oral antimicrobial therapy for acute osteomyelitis in children. Pediatrics. 2009;123:636-642.
PPI Use with Clopidogrel in Acute Coronary Syndrome Is Associated with Readmissions and Mortality
Clinical question: Does concomitant use of clopidogrel and a proton pump inhibitor (PPI) following hospitalization for acute coronary syndrome (ACS) lead to adverse outcomes?
Background: Prophylactic PPIs often are prescribed with clopidogrel to reduce the risk of gastrointestinal bleeding. Mechanistic studies have shown that omeprazole decreases the platelet-inhibitory effect of clopidogrel, raising concerns that PPIs might interfere with clopidogrel’s beneficial effects. The clinical significance of this finding is unknown.
Study design: Retrospective cohort study.
Setting: 127 VA hospitals.
Synopsis: Investigators used data from the Cardiac Care Follow-up Clinical Study and VA pharmacy records to examine 8,205 male veterans who were hospitalized for ACS and treated with clopidogrel. Patients who filled prescriptions for both clopidogrel and a PPI were at significantly higher risk for death or readmission with ACS compared with those who filled prescriptions for clopidogrel only (adjusted odds ratio, 1.25; 95% confidence interval, 1.11-1.41). Patients who filled prescriptions for PPIs alone had similar risk for adverse events as those who took neither medication.
Subanalyses found similarly increased risk among patients prescribed omeprazole and rabeprazole, but those taking lanzoprazole and pantoprazole were not examined due to the small sample size. Although causality cannot be inferred from this observational study, and the risk associated with combined clopidigrel and PPI use appeared small, alternatives for gastric acid reduction exist. Thus, it may be prudent to restrict PPI use to patients who have a clear indication for their use until more definitive clinical trials can be conducted.
Bottom line: Among patients who are treated with clopidogrel for ACS, PPIs should be reserved for patients with a clear indication for gastric acid reduction and who cannot use alternative therapies.
Citation: Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. JAMA. 2009;301(9):937-944.
Chlorhexidine-Impregnated Sponge Use Reduces Line Infections
Clinical question: Does dressing vascular catheters with chlorhexidine gluconate-impregnated sponges (CHGIS) reduce rates of catheter-related infections, and are dressing changes every seven days inferior to every three days?
Background: Process improvement strategies—including educating providers, strictly adhering to sterile technique, and promptly removing unnecessary catheters—greatly decrease catheter-related infections. It is unclear if CHGIS dressings offer additional benefit. Also uncertain is whether weekly dressing changes are as safe as changing dressings every three days.
Study design: A 2x2 factorial, assessor-blinded, randomized controlled trial.
Setting: ICUs in three university hospitals and two general hospitals in France.
Synopsis: 1,636 French adults expected to require arterial and central venous catheters for >48 hours were randomly assigned to one of four groups. Each group received either CHGIS dressings or standard dressings, and each group had dressing changes every three or seven days. Dressings were changed sooner if soiled or nonadherent. CHGIS dressings were associated with fewer catheter-related infections than standard dressings (0.6 vs. 1.4 infections per 1,000 catheter days; P=0.03). No significant difference in rates of catheter colonization existed between the three-day and seven-day dressing change strategies (10.4 vs. 11 events per 1,000 catheter days, P>0.05).
Although microbiology assessors were blinded to patients’ status, the ICU staff was not, potentially creating experimenter bias. Approximately 30% of the venous catheters and 40% of the arterial catheters were in a femoral site. Secondary analyses found higher rates of severe dermatitis among patients with CHGIS dressings but no difference in minimal bactericidal concentration (MBC) or colonizing organisms. Preliminary calculations suggested CHGIS dressings could be cost-effective.
Bottom line: Among critically ill adults, CHGIS catheter dressings may marginally reduce catheter-related infection rates, but further evaluation is needed before this technology can be adopted widely.
Citation: Timsit JF, Schwebel C, Bouadma L, et al. Chlorhexidine-impregnated sponges and less frequent dressing changes for prevention of catheter-related infections in critically-ill adults: a randomized controlled trial. JAMA. 2009;301(12):1231-1241.
Thienopyridine Use Six Months after Sirolimus-Eluting Stent Implant-ation Offers No Benefit
Clinical question: What are the relative contributions of aspirin and thienopyridine on preventing stent thrombosis in patients with sirolimus-eluting stents?
Background: There are no randomized clinical trials addressing the optimal duration, or the risks associated with discontinuation, of dual-antiplatelet therapy after drug-eluting stent (DES) implantation. Nevertheless, many patients continue to be maintained on dual-antiplatelet therapy beyond one year of their index DES implantation.
Study design: Prospective multicenter observational study.
Setting: Hospitals in Japan.
Synopsis: This study observed 10,778 Japanese patients undergoing sirolimus-eluting stent implantation. Patients discontinuing both thienopyridine and aspirin had a significantly higher rate of stent thrombosis than those who continued both medications for up to 18 months. However, discontinuation of thienopyridine alone was not associated with an excess risk of stent thrombosis. Additionally, a landmark analysis of patients who were free of events at six months showed rates of death for myocardial infarction (MI) at 24 months were 4.1% for patients taking thienopyridine and 4.1% for patients not taking thienopyridine (P=0.99). Ticlodipine was the thienopyridine used by more than 95% of patients.
Hospitalists should be aware that the role thienopyridine therapy plays in reducing stent thrombosis beyond one month after implantation has not been well addressed.
Bottom line: Discontinuation of thienopyridine therapy after six months while maintaining aspirin therapy is not associated with increased risk of stent thrombosis in patients with sirolimus-eluting stents.
Citation: Kimura T, Morimoto T, Nakagawa Y, et al. Antiplatelet therapy and stent thrombosis after sirolimus-eluting stent implantation. Circulation. 2009;119(7):987-995.
Compared with PCI, CABG Results in Lower Rates of Major Adverse Events in Severe CAD Patients
Clinical question: What is the optimal revascularization strategy for previously untreated severe coronary artery disease (CAD)?
Background: Coronary artery bypass grafting (CABG) is the treatment of choice in three-vessel and left-main CAD. However, percutaneous coronary intervention (PCI) with drug-eluting stents often is utilized despite the lack of adequately powered randomized trials.
Study design: Prospective multicenter randomized clinical trial.
Setting: 85 hospitals in Europe and the U.S.
Synopsis: 1,800 patients with an average age of 65 and previously untreated three-vessel or left-main CAD amenable to therapy with both PCI and CABG were randomized to CABG or PCI. The primary combined endpoint was a major adverse cardiac or cerebrovascular event, defined as death, stroke, MI, or repeat revascularization. PCI was associated with a significantly higher rate of major adverse cardiac or cerebrovascular events, due mostly to a higher rate of repeat revascularization (13.5% vs. 5.9%, P<0.001). At 12 months, the two groups had similar rates of death from any cause or MI, and similar rates of the combined endpoint of death from any cause, stroke, or MI; however, the rate of stroke was 1.6% higher in the CABG group.
Hospitalists should continue to favor CABG over PCI but give consideration to the risks involved with such an intervention.
Bottom line: CABG remains the revascularization choice in patients with severe CAD.
Citation: Serruys PW, Morice MC, Kappetein AP, et al. Percuta-neous coronary intervention versus coronary artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-972.
Pre-Medicated Central Venous Catheters Reduce Risk of Catheter-Related Bloodstream Infections
Clinical question: Does pre-treating central venous catheters with anti-infective agents prevent catheter-related bloodstream infections?
Background: Use of central venous catheters (CVC) is associated with catheter-related bloodstream infection (CRBSI), with CRBSI-related mortality rates as high as 25%. Previous reviews have indicated that CVCs coated or impregnated with anti-infectives may reduce CRBSI incidence. This review integrates new trial data with information from prior reviews.
Study design: Meta-analysis of 27 randomized controlled trials.
Setting: Meta-analysis.
Synopsis: The authors report CVCs pre-treated with anti-infectives (AI-CVCs) are clinically effective in reducing the risk of CRBSI. The odds of having a CRBSI with a treated CVC versus an untreated CVC are 0.49 to 1 (95% CI, 0.37–0.64, 27 studies, fixed effects). The study also finds the use of AI-CVCs might provide a large cost savings in Great Britain. Because the findings are based on a meta-analysis, they are limited by the quality, context, and consistency of the original studies. The authors note that many of the studies had unsatisfactory descriptions of methodology. The current study is unable to separate the risk reduction attributable to AI-CVC versus that attributable to other infection control practices. Also, original data is insufficient to assess the benefits of AI-CVCs placed for longer than 12 days.
To summarize, AI-CVCs may present a means to reduce CRBSI, but more investigation of its role within infection control protocols is needed, as is investigation of longer duration of treatment.
Bottom line: Central venous catheters pre-treated with anti-infectives significantly reduce catheter-related bloodstream infections.
Citation: Hockenhull JC, Dwan KM, Smith GW, et al. The clinical effectiveness of central venous catheters treated with anti-infective agents in preventing catheter-related bloodstream infections: a systematic review. Crit Care Med. 2009;37(2):702-712.
Fivefold Increase in Hospitalists in the U.S. from 1995 to 2006
Clinical question: What is the growth rate of hospitalists and hospitalist-provided care?
Background: Survey data has shown a sharp increase in the number of hospitalists, but until now there have not been any national or population-based data on the growth of hospitalist care.
Study design: Descriptive analysis.
Setting: Medicare-enrolled patients.
Synopsis: The study is based on national Medicare data from 2.1 million admissions involving 990,785 patients in 5,800 hospitals and 120,226 general internists. It represents 5% of inpatient Medicare claims generated by general internists. The authors define “hospitalist” as a general internist who generates >90% of his or her claims from the care of hospitalized patients.
U.S. hospitals have seen substantial growth in hospitalists over the period examined. The nation saw a 500% increase in the number of general-internist hospitalists, and a 28% increase (to 37.1% in 2006 from 9.1% in 1995) in the number of Medicare patients who received care from a hospitalist. The odds that a hospitalized Medicare patient received care from a hospitalist increased 29.2% per year from 1997 to 2006. The percentage of hospitals with at least three hospitalists rose to 47.1% in 2006 from 11.6% in 1995.
This analysis might actually have underestimated HM’s growth. Analysis of Medicare claims does not identify pediatric hospitalists and hospitalists who work exclusively within HMOs. This analysis also did not include family practitioners or internal-medicine subspecialists who are hospitalists.
Bottom line: Medicare claims data confirm survey data findings: Hospitalists and hospitalist care has grown sharply over the last decade.
Citation: Kuo YF, Sharma G, Freeman JL, Goodwin JS. Growth in the care of older patients by hospitalists in the United States. N Engl J Med. 2009;360:1102-1012.
Standardized Order Set for Bacteremic Sepsis Improves LOS and Mortality
Clinical question: Does a standardized order set for bacteremic sepsis impact patient management and outcomes?
Background: Prompt cardiovascular resuscitation and appropriate antibiotics decrease morbidity and mortality in bacteremic sepsis. This study examined whether hospitalwide, standardized sepsis order set improved management and outcomes.
Study design: Retrospective, before-and-after study design.
Setting: 1,200-bed academic medical center.
Synopsis: Two hundred patients with bacteremic severe sepsis were randomly selected from 18 months before the order set was introduced, and 200 were selected from 18 months after the order set was introduced. Primary outcomes measured were quantity of fluid administered and appropriate initial antibiotics. Secondary outcomes measured were hospital mortality and length of stay. Patients in the “after” group received more intravenous fluid (1627±1862 ml vs. 2054±2237 ml, P=.04), more appropriate antibiotics (53.0% vs. 65.5%, P=.01), had shorter hospital stays (28.7±30.1 days vs. 22.4±20.9 days, P=.02), and decreased in-house mortality (55.0% vs. 39.5%, P =<0.01).
The retrospective design of the study limited its ability to determine causal relationship. Extensive education may have contributed to the change (Hawthorne effect). Management in the ICU and ED, not the hospital wards, was the primary reason for mortality difference.
Bottom line: A standardized order set for bacteremic sepsis was associated with increased compliance with evidence-based treatment and improved outcomes. Hospitalists should promptly treat bacteremic sepsis with appropriate fluid resuscitation and antibiotics.
Citation: Thiel SW, Asghar MF, Micek ST, Reichley RM, Doherty JA, Kollef MH. Hospital-wide impact of a standardized order set for the management of bacteremic severe sepsis. Crit Care Med. 2009;37(3):819-824.
Admission Day of the Week Predicts Mortality in Patients with Acute Pulmonary Embolus
Clinical question: Do weekend pulmonary embolus (PE) admissions have worse outcomes than weekday admissions?
Background: Studies of patients with acute cardiovascular diagnoses (e.g., stroke, cardiac arrest) have shown higher short-term mortality and longer length of stay (LOS) for weekend versus weekday admissions. PE diagnosis is complex, requiring timely testing and experienced staff who are sometimes unavailable on weekends. Optimal anticoagulation therapy also depends on provider skill.
Study design: Retrospective observational study.
Setting: 186 private Pennsylvania hospitals, January 2000 through November 2002.
Synopsis: Using the Pennsylvania Health Care Cost Containment Council database, the authors reviewed 15,531 records of patients with a primary or secondary PE diagnosis code. The primary outcome was all-cause mortality over 30 days; LOS was the secondary outcome.
Weekend admissions in the highest severity of illness risk class had higher 30-day mortality than weekday admissions. Weekend admissions were significantly more likely than weekday admissions to be clinically unstable and to have abnormal lab parameters. Adjusted for severity of illness risk class, overall mortality was 1.4% higher for weekend versus weekday admissions. All excess mortality came from the sickest group of patients. LOS did not differ.
Less-experienced caregivers or delayed diagnostic testing may play a role in poor outcomes. Patients admitted on weekends might receive delayed care from the first onset of symptoms. This is important because timely therapy has been shown to influence outcomes in acute PE. Reasons for these observed differences should be explored further to help provide more consistent PE management, regardless of admission day.
Bottom line: The sickest patients with PE admitted on weekends experienced small but significantly greater 30-day mortality compared with those admitted on weekdays.
Citation: Aujesky D, Jimenez D, Mor M, Geng M, Fine M, Ibrahim S. Weekend versus weekday admission and mortality after acute pulmonary embolism. Circulation. 2009;119:962-968. TH
In This Edition
Literature at a Glance: A guide to this month’s studies
- PPI use with clopidrogrel in ACS.
- Chlorhexidine sponge use reduces line infections.
- Extended thienopyridine use does not benefit DES patients.
- CABG is revascularization choice for severe CAD.
- Pre-treated CVC use reduces bloodstream infections.
- Hospitalist use grows in U.S.
- Sepsis order set improves outcomes.
- Admission day predicts acute PE mortality.
PEDIATRIC HM LITerature
Early Transition to Oral Therapy in Acute Osteomyelitis in Children Associated with Fewer Complications I By Mark Shen, MD
Clinical question: Can acute osteomyelitis in children be managed with early transition to oral antimicrobial therapy?
Background: Traditional treatment of acute osteomyelitis in children has involved prolonged intravenous antimicrobial therapy, typically greater than three to four weeks in duration. Small studies suggest the feasibility of a transition to prolonged oral therapy after an initial response to intravenous antimicrobial agents.
Study design: Retrospective, cohort study.
Setting: 29 freestanding children’s hospitals.
Synopsis: Using the Pediatric Health Information System (PHIS) database, 1,969 children were identified via ICD-9-CM codes as having been hospitalized with acute osteomyelitis—and without comorbid conditions—between January 2000 and June 2005. Rehospitalizations were reviewed for complications of both treatment failure and the treatment itself. More than half (1,021) of the children underwent central venous catheterization for prolonged intravenous (IV) therapy; 948 were assigned to the oral therapy group. No significant differences in treatment failure existed between the two groups (5% in the IV group; 4% in the oral group).
Overall, the clinical characteristics of the two groups were indistinguishable. Propensity score analysis was used to handle possible patient-level confounders; a validation study was performed to address misclassifications in assignment. No significant confounding effects were found.
Secondary findings included marked variation across hospitals in the use of oral therapy and increased treatment complications in the IV group. Although unmeasured factors not present in an administrative database may affect results, the strikingly similar groups and outcomes suggest the data are relevant.
Bottom line: Early transition to oral therapy for acute osteomyelitis in children does not increase treatment failure and is associated with fewer treatment complications.
Reference: Zaoutis T, Localio AR, Leckerman K, Saddlemire S, Bertoch D, Keren R. Prolonged intravenous therapy versus early transition to oral antimicrobial therapy for acute osteomyelitis in children. Pediatrics. 2009;123:636-642.
PPI Use with Clopidogrel in Acute Coronary Syndrome Is Associated with Readmissions and Mortality
Clinical question: Does concomitant use of clopidogrel and a proton pump inhibitor (PPI) following hospitalization for acute coronary syndrome (ACS) lead to adverse outcomes?
Background: Prophylactic PPIs often are prescribed with clopidogrel to reduce the risk of gastrointestinal bleeding. Mechanistic studies have shown that omeprazole decreases the platelet-inhibitory effect of clopidogrel, raising concerns that PPIs might interfere with clopidogrel’s beneficial effects. The clinical significance of this finding is unknown.
Study design: Retrospective cohort study.
Setting: 127 VA hospitals.
Synopsis: Investigators used data from the Cardiac Care Follow-up Clinical Study and VA pharmacy records to examine 8,205 male veterans who were hospitalized for ACS and treated with clopidogrel. Patients who filled prescriptions for both clopidogrel and a PPI were at significantly higher risk for death or readmission with ACS compared with those who filled prescriptions for clopidogrel only (adjusted odds ratio, 1.25; 95% confidence interval, 1.11-1.41). Patients who filled prescriptions for PPIs alone had similar risk for adverse events as those who took neither medication.
Subanalyses found similarly increased risk among patients prescribed omeprazole and rabeprazole, but those taking lanzoprazole and pantoprazole were not examined due to the small sample size. Although causality cannot be inferred from this observational study, and the risk associated with combined clopidigrel and PPI use appeared small, alternatives for gastric acid reduction exist. Thus, it may be prudent to restrict PPI use to patients who have a clear indication for their use until more definitive clinical trials can be conducted.
Bottom line: Among patients who are treated with clopidogrel for ACS, PPIs should be reserved for patients with a clear indication for gastric acid reduction and who cannot use alternative therapies.
Citation: Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. JAMA. 2009;301(9):937-944.
Chlorhexidine-Impregnated Sponge Use Reduces Line Infections
Clinical question: Does dressing vascular catheters with chlorhexidine gluconate-impregnated sponges (CHGIS) reduce rates of catheter-related infections, and are dressing changes every seven days inferior to every three days?
Background: Process improvement strategies—including educating providers, strictly adhering to sterile technique, and promptly removing unnecessary catheters—greatly decrease catheter-related infections. It is unclear if CHGIS dressings offer additional benefit. Also uncertain is whether weekly dressing changes are as safe as changing dressings every three days.
Study design: A 2x2 factorial, assessor-blinded, randomized controlled trial.
Setting: ICUs in three university hospitals and two general hospitals in France.
Synopsis: 1,636 French adults expected to require arterial and central venous catheters for >48 hours were randomly assigned to one of four groups. Each group received either CHGIS dressings or standard dressings, and each group had dressing changes every three or seven days. Dressings were changed sooner if soiled or nonadherent. CHGIS dressings were associated with fewer catheter-related infections than standard dressings (0.6 vs. 1.4 infections per 1,000 catheter days; P=0.03). No significant difference in rates of catheter colonization existed between the three-day and seven-day dressing change strategies (10.4 vs. 11 events per 1,000 catheter days, P>0.05).
Although microbiology assessors were blinded to patients’ status, the ICU staff was not, potentially creating experimenter bias. Approximately 30% of the venous catheters and 40% of the arterial catheters were in a femoral site. Secondary analyses found higher rates of severe dermatitis among patients with CHGIS dressings but no difference in minimal bactericidal concentration (MBC) or colonizing organisms. Preliminary calculations suggested CHGIS dressings could be cost-effective.
Bottom line: Among critically ill adults, CHGIS catheter dressings may marginally reduce catheter-related infection rates, but further evaluation is needed before this technology can be adopted widely.
Citation: Timsit JF, Schwebel C, Bouadma L, et al. Chlorhexidine-impregnated sponges and less frequent dressing changes for prevention of catheter-related infections in critically-ill adults: a randomized controlled trial. JAMA. 2009;301(12):1231-1241.
Thienopyridine Use Six Months after Sirolimus-Eluting Stent Implant-ation Offers No Benefit
Clinical question: What are the relative contributions of aspirin and thienopyridine on preventing stent thrombosis in patients with sirolimus-eluting stents?
Background: There are no randomized clinical trials addressing the optimal duration, or the risks associated with discontinuation, of dual-antiplatelet therapy after drug-eluting stent (DES) implantation. Nevertheless, many patients continue to be maintained on dual-antiplatelet therapy beyond one year of their index DES implantation.
Study design: Prospective multicenter observational study.
Setting: Hospitals in Japan.
Synopsis: This study observed 10,778 Japanese patients undergoing sirolimus-eluting stent implantation. Patients discontinuing both thienopyridine and aspirin had a significantly higher rate of stent thrombosis than those who continued both medications for up to 18 months. However, discontinuation of thienopyridine alone was not associated with an excess risk of stent thrombosis. Additionally, a landmark analysis of patients who were free of events at six months showed rates of death for myocardial infarction (MI) at 24 months were 4.1% for patients taking thienopyridine and 4.1% for patients not taking thienopyridine (P=0.99). Ticlodipine was the thienopyridine used by more than 95% of patients.
Hospitalists should be aware that the role thienopyridine therapy plays in reducing stent thrombosis beyond one month after implantation has not been well addressed.
Bottom line: Discontinuation of thienopyridine therapy after six months while maintaining aspirin therapy is not associated with increased risk of stent thrombosis in patients with sirolimus-eluting stents.
Citation: Kimura T, Morimoto T, Nakagawa Y, et al. Antiplatelet therapy and stent thrombosis after sirolimus-eluting stent implantation. Circulation. 2009;119(7):987-995.
Compared with PCI, CABG Results in Lower Rates of Major Adverse Events in Severe CAD Patients
Clinical question: What is the optimal revascularization strategy for previously untreated severe coronary artery disease (CAD)?
Background: Coronary artery bypass grafting (CABG) is the treatment of choice in three-vessel and left-main CAD. However, percutaneous coronary intervention (PCI) with drug-eluting stents often is utilized despite the lack of adequately powered randomized trials.
Study design: Prospective multicenter randomized clinical trial.
Setting: 85 hospitals in Europe and the U.S.
Synopsis: 1,800 patients with an average age of 65 and previously untreated three-vessel or left-main CAD amenable to therapy with both PCI and CABG were randomized to CABG or PCI. The primary combined endpoint was a major adverse cardiac or cerebrovascular event, defined as death, stroke, MI, or repeat revascularization. PCI was associated with a significantly higher rate of major adverse cardiac or cerebrovascular events, due mostly to a higher rate of repeat revascularization (13.5% vs. 5.9%, P<0.001). At 12 months, the two groups had similar rates of death from any cause or MI, and similar rates of the combined endpoint of death from any cause, stroke, or MI; however, the rate of stroke was 1.6% higher in the CABG group.
Hospitalists should continue to favor CABG over PCI but give consideration to the risks involved with such an intervention.
Bottom line: CABG remains the revascularization choice in patients with severe CAD.
Citation: Serruys PW, Morice MC, Kappetein AP, et al. Percuta-neous coronary intervention versus coronary artery bypass grafting for severe coronary artery disease. N Engl J Med. 2009;360(10):961-972.
Pre-Medicated Central Venous Catheters Reduce Risk of Catheter-Related Bloodstream Infections
Clinical question: Does pre-treating central venous catheters with anti-infective agents prevent catheter-related bloodstream infections?
Background: Use of central venous catheters (CVC) is associated with catheter-related bloodstream infection (CRBSI), with CRBSI-related mortality rates as high as 25%. Previous reviews have indicated that CVCs coated or impregnated with anti-infectives may reduce CRBSI incidence. This review integrates new trial data with information from prior reviews.
Study design: Meta-analysis of 27 randomized controlled trials.
Setting: Meta-analysis.
Synopsis: The authors report CVCs pre-treated with anti-infectives (AI-CVCs) are clinically effective in reducing the risk of CRBSI. The odds of having a CRBSI with a treated CVC versus an untreated CVC are 0.49 to 1 (95% CI, 0.37–0.64, 27 studies, fixed effects). The study also finds the use of AI-CVCs might provide a large cost savings in Great Britain. Because the findings are based on a meta-analysis, they are limited by the quality, context, and consistency of the original studies. The authors note that many of the studies had unsatisfactory descriptions of methodology. The current study is unable to separate the risk reduction attributable to AI-CVC versus that attributable to other infection control practices. Also, original data is insufficient to assess the benefits of AI-CVCs placed for longer than 12 days.
To summarize, AI-CVCs may present a means to reduce CRBSI, but more investigation of its role within infection control protocols is needed, as is investigation of longer duration of treatment.
Bottom line: Central venous catheters pre-treated with anti-infectives significantly reduce catheter-related bloodstream infections.
Citation: Hockenhull JC, Dwan KM, Smith GW, et al. The clinical effectiveness of central venous catheters treated with anti-infective agents in preventing catheter-related bloodstream infections: a systematic review. Crit Care Med. 2009;37(2):702-712.
Fivefold Increase in Hospitalists in the U.S. from 1995 to 2006
Clinical question: What is the growth rate of hospitalists and hospitalist-provided care?
Background: Survey data has shown a sharp increase in the number of hospitalists, but until now there have not been any national or population-based data on the growth of hospitalist care.
Study design: Descriptive analysis.
Setting: Medicare-enrolled patients.
Synopsis: The study is based on national Medicare data from 2.1 million admissions involving 990,785 patients in 5,800 hospitals and 120,226 general internists. It represents 5% of inpatient Medicare claims generated by general internists. The authors define “hospitalist” as a general internist who generates >90% of his or her claims from the care of hospitalized patients.
U.S. hospitals have seen substantial growth in hospitalists over the period examined. The nation saw a 500% increase in the number of general-internist hospitalists, and a 28% increase (to 37.1% in 2006 from 9.1% in 1995) in the number of Medicare patients who received care from a hospitalist. The odds that a hospitalized Medicare patient received care from a hospitalist increased 29.2% per year from 1997 to 2006. The percentage of hospitals with at least three hospitalists rose to 47.1% in 2006 from 11.6% in 1995.
This analysis might actually have underestimated HM’s growth. Analysis of Medicare claims does not identify pediatric hospitalists and hospitalists who work exclusively within HMOs. This analysis also did not include family practitioners or internal-medicine subspecialists who are hospitalists.
Bottom line: Medicare claims data confirm survey data findings: Hospitalists and hospitalist care has grown sharply over the last decade.
Citation: Kuo YF, Sharma G, Freeman JL, Goodwin JS. Growth in the care of older patients by hospitalists in the United States. N Engl J Med. 2009;360:1102-1012.
Standardized Order Set for Bacteremic Sepsis Improves LOS and Mortality
Clinical question: Does a standardized order set for bacteremic sepsis impact patient management and outcomes?
Background: Prompt cardiovascular resuscitation and appropriate antibiotics decrease morbidity and mortality in bacteremic sepsis. This study examined whether hospitalwide, standardized sepsis order set improved management and outcomes.
Study design: Retrospective, before-and-after study design.
Setting: 1,200-bed academic medical center.
Synopsis: Two hundred patients with bacteremic severe sepsis were randomly selected from 18 months before the order set was introduced, and 200 were selected from 18 months after the order set was introduced. Primary outcomes measured were quantity of fluid administered and appropriate initial antibiotics. Secondary outcomes measured were hospital mortality and length of stay. Patients in the “after” group received more intravenous fluid (1627±1862 ml vs. 2054±2237 ml, P=.04), more appropriate antibiotics (53.0% vs. 65.5%, P=.01), had shorter hospital stays (28.7±30.1 days vs. 22.4±20.9 days, P=.02), and decreased in-house mortality (55.0% vs. 39.5%, P =<0.01).
The retrospective design of the study limited its ability to determine causal relationship. Extensive education may have contributed to the change (Hawthorne effect). Management in the ICU and ED, not the hospital wards, was the primary reason for mortality difference.
Bottom line: A standardized order set for bacteremic sepsis was associated with increased compliance with evidence-based treatment and improved outcomes. Hospitalists should promptly treat bacteremic sepsis with appropriate fluid resuscitation and antibiotics.
Citation: Thiel SW, Asghar MF, Micek ST, Reichley RM, Doherty JA, Kollef MH. Hospital-wide impact of a standardized order set for the management of bacteremic severe sepsis. Crit Care Med. 2009;37(3):819-824.
Admission Day of the Week Predicts Mortality in Patients with Acute Pulmonary Embolus
Clinical question: Do weekend pulmonary embolus (PE) admissions have worse outcomes than weekday admissions?
Background: Studies of patients with acute cardiovascular diagnoses (e.g., stroke, cardiac arrest) have shown higher short-term mortality and longer length of stay (LOS) for weekend versus weekday admissions. PE diagnosis is complex, requiring timely testing and experienced staff who are sometimes unavailable on weekends. Optimal anticoagulation therapy also depends on provider skill.
Study design: Retrospective observational study.
Setting: 186 private Pennsylvania hospitals, January 2000 through November 2002.
Synopsis: Using the Pennsylvania Health Care Cost Containment Council database, the authors reviewed 15,531 records of patients with a primary or secondary PE diagnosis code. The primary outcome was all-cause mortality over 30 days; LOS was the secondary outcome.
Weekend admissions in the highest severity of illness risk class had higher 30-day mortality than weekday admissions. Weekend admissions were significantly more likely than weekday admissions to be clinically unstable and to have abnormal lab parameters. Adjusted for severity of illness risk class, overall mortality was 1.4% higher for weekend versus weekday admissions. All excess mortality came from the sickest group of patients. LOS did not differ.
Less-experienced caregivers or delayed diagnostic testing may play a role in poor outcomes. Patients admitted on weekends might receive delayed care from the first onset of symptoms. This is important because timely therapy has been shown to influence outcomes in acute PE. Reasons for these observed differences should be explored further to help provide more consistent PE management, regardless of admission day.
Bottom line: The sickest patients with PE admitted on weekends experienced small but significantly greater 30-day mortality compared with those admitted on weekdays.
Citation: Aujesky D, Jimenez D, Mor M, Geng M, Fine M, Ibrahim S. Weekend versus weekday admission and mortality after acute pulmonary embolism. Circulation. 2009;119:962-968. TH