In This Edition
Literature at a Glance
A guide to this month’s studies
- Eplerenone and heart failure mortality
- Fidaxomicin for C. difficile diarrhea
- Guidelines for intensive insulin therapy
- Benefits of hospitalist comanagement
- Peritoneal dialysis versus hemodialysis
- Pneumococcal urinary antigen to guide CAP treatment
- Race and readmission rate
- Factors associated with readmission
- Unplanned transfers to the ICU
Eplerenone Improves Mortality in Patients with Systolic Heart Failure and Mild Symptoms
Clinical question: Does the selective mineralocorticoid antagonist eplerenone improve outcomes in patients with chronic heart failure and mild symptoms?
Background: In prior studies of miner alocorticoid antagonists in systolic heart failure, spironolactone reduced mortality in patients with moderate to severe heart failure symptoms, and eplerenone reduced mortality in patients with acute myocardial infarction complicated by left ventricular dysfunction. The use of eplerenone in patients with systolic heart failure and mild symptoms has not previously been examined.
Study design: Randomized, double-blind, multicenter, placebo-controlled trial.
Setting: Two hundred seventy-eight centers in 29 countries.
Synopsis: The study authors randomized 2,737 patients with New York Heart Association Class II heart failure and an ejection fraction of no more than 35% to either eplerenone (up to 50 mg daily) or placebo, in addition to recommended therapy. Patients with baseline potassium levels >5 mmol/L or estimated GFR <30 were excluded. The primary outcome was a composite of death from cardiovascular causes or hospitalization for heart failure.
The trial was stopped early, after a median follow-up period of 21 months, when an interim analysis showed significant benefit with eplerenone. The primary outcome occurred in 18.3% of patients in the eplerenone group and 25.9% in the placebo group (hazard ratio [HR], 0.63; 95% CI, 0.54 to 0.74; P<0.001). All-cause mortality was 12.5% in the eplerenone group and 15.5% in the placebo group (HR 0.76; 95% CI, 0.62 to 0.93; P=0.008). A serum potassium level exceeding 5.5 mmol/L occurred in 11.8% of patients in the eplerenone group and 7.2% of those in the placebo group (P<0.001).
Bottom line: Eplerenone reduces both the risk of death and the risk of hospitalization in patients with systolic heart failure and mild symptoms.
Citation: Zannad F, McMurray JJ, Krum H, et al. Eplerenone in patients with systolic heart failure and mild symptoms. N Engl J Med. 2011;364(1):11-21.
Fidaxomicin Noninferior to Vancomycin for C. Difficile Treatment
Clinical question: What is the safety and efficacy of fidaxomicin compared to vancomycin in the treatment of patients with C. difficile infection?
Background: Fidaxomicin, a new macrocyclic antibiotic, has shown high efficacy in vitro against C. diff, minimal systemic absorption, and a narrow-spectrum profile. In previously published Phase 2 trials of fidaxomicin for the treatment of C. diff, it has been associated with good clinical response and low recurrence rates.
Study design: Prospective, multicenter, double-blind, randomized trial.
Setting: Fifty-two sites in the United States and 15 in Canada.
Synopsis: The study included 629 adults with acute symptoms of C. diff and a positive stool toxin test. The patients were randomly assigned to 200-mg twice-daily fidaxomicin or 125-mg four-times-daily vancomycin for a course of 10 days. The primary endpoint was rate of clinical cure (resolution of diarrhea), and secondary endpoints were recurrence of C. diff and global cure (clinical cure and lack of relapse within four weeks of cessation of therapy).
The rate of clinical cure associated with fidaxomicin was noninferior to that associated with vancomycin (88.2% vs. 85.8%, respectively). Patients receiving fidaxomicin had a lower rate of relapse than those receiving vancomycin (15.4% vs. 25.3%, respectively, P=0.005) and a higher global cure rate (74.6 vs. 61.1%, P=0.006). In subgroup analysis, the lower rate of recurrence was seen in patients with non-North American pulsed-field Type 1 strain (NAP1/BI/027 strain), while in patients with the NAP1/BI/027 strain, the recurrence rate was similar for both drugs. There was no difference in adverse event rates.
Bottom line: Clinical cure rates of C. diff with fidaxomicin are noninferior to those with vancomycin; however, fidaxomicin is associated with a significantly lower rate of recurrence among those infected with the non-NAP1/BI/027 strain.
Citation: Louie TJ, Miller MA, Mullane KM, et al. Fidaxomicin versus vancomycin for Clostridium difficile infection. N Engl J Med. 2011;364(5):422-431.
ACP Guideline Discourages Use of Intensive Insulin Therapy in Hospitalized Patients
Clinical question: Does the use of intensive insulin therapy (IIT) to achieve tight glycemic control in hospitalized patients (whether in the SICU, MICU, or on the general medicine floor) improve important health outcomes?
Background: Hyperglycemia is a common finding in hospitalized patients and is associated with prolonged length of stay (LOS), death, and worsening health outcomes. Despite this, prospective studies have yet to provide consistent evidence that using IIT to achieve strict glycemic control (80 mg/dL-110 mg/dL) improves outcomes in hospitalized patients.
Study design: Systematic review of MEDLINE and the Cochrane Database of Systematic Reviews from 1950 to January 2010.
Setting: Trials included subjects with myocardial infarction, stroke, and brain injury, as well as those in perioperative settings and ICUs.
Synopsis: The review informing this guideline meta-analyzed 21 trials and found that IIT did not improve short-term mortality, long-term mortality, infection rates, LOS, or the need for renal replacement therapy. Furthermore, IIT was associated with a sixfold increase in risk for severe hypoglycemia in all hospital settings.
Based on these findings, the American College of Physicians (ACP) issued three recommendations:
- To not use IIT to strictly control blood glucose in non-SICU/non-MICU patients with or without diabetes (strong recommendation, moderate-quality evidence);
- To not use IIT to normalize blood glucose in SICU or MICU patients with or without diabetes (strong recommendation, high-quality evidence); and
- To consider a target blood glucose level of 140 mg to 200 mg if insulin therapy is used in SICU or MICU patients (weak recommendation, moderate-quality evidence).
Bottom line: The ACP recommends against using IIT to strictly control blood glucose (80 mg/dL-180 mg/dL) in hospitalized patients, whether in the SICU, MICU, or on the general medicine floor.
Citation: Qaseem A, Humphrey LL, Chou R, Snow V, Shekelle P. Use of intensive insulin therapy for the management of glycemic control in hospitalized patients: a clinical practice guideline from the American College of Physicians. Ann Intern Med. 2011;154(4):260-267.
Limited Benefits Seen with Hospitalist-Neurosurgeon Comanagement
Clinical question: Does hospitalist-neurosurgeon comanagement improve patient outcomes?
Background: The shared management of surgical patients between surgeons and hospitalists is increasingly common despite limited data supporting its effectiveness in reducing costs or improving patient outcomes.
Study design: Single-center, retrospective study.
Setting: Tertiary-care academic medical center.
Synopsis: Data were collected on the 7,596 patients who were admitted to the neurosurgical service of the University of California San Francisco Medical Center from June 1, 2005, to December 31, 2008. The study looked at 4,203 patients (55.3%) admitted before July 1, 2007, when hospitalist comanagement was implemented, and 3,393 patients (44.7%) after comanagement began. Of those admitted during the post-implementation period, 988 (29.1%) were comanaged.
After adjusting for patient characteristics and background trends, and accounting for clustering at the physician level, no differences were found in patient mortality rate, readmissions, or LOS after implementation of comanagement. No consistent improvements were seen in patient satisfaction.
However, physician and staff perceptions of safety and quality of care were significantly better after comanagement. There was a moderate decrease in adjusted hospital costs after implementation (adjusted cost ratio 0.94, range 0.88-1.00) equivalent to a cost savings of about $1,439 per hospitalization.
Bottom line: The implementation of a hospitalist-neurosurgery comanagement service did not improve patient outcomes or satisfaction, but it did appear to improve providers’ perception of care quality and reduce hospital costs.
Citation: Auerbach AD, Wachter RM, Cheng HQ, et al. Comanagement of surgical patients between neurosurgeons and hospitalists. Arch Intern Med. 2010;170(22):2004-2010.
Comparable Mortality Between Hemodialysis and Peritoneal Dialysis
Clinical question: What effect does the initial dialysis modality used have on mortality for patients with end-stage renal disease (ESRD)?
Background: Despite the substantially lower annual per-person costs of peritoneal dialysis (PD) as compared with hemodialysis (HD), only 7% of dialysis patients were treated with PD in 2008. It is unknown whether there are differences in mortality between those using PD and HD when examined in contemporary cohorts.
Study design: Retrospective cohort study.
Setting: National cohort.
Synopsis: Data for patients with incident ESRD over a nine-year period were obtained from the U.S. Renal Data Systems (USRDS), a national registry of all patients with ESRD. Initial dialysis modality was defined as the dialysis modality used 90 days after initiation of dialysis. Patients were divided into three three-year cohorts (1996-1998, 1999-2001, and 2002-2004) based on the date dialysis was initiated and followed for up to five years.
A substantial and consistent reduction in mortality was seen for PD patients across the three time periods. No such improvements were observed across the time periods for the HD patients. PD patients were, on average, younger, healthier, and more likely to be white. In an analysis of the most recent cohort adjusting for these factors, there was no significant difference in the risk of death between HD and PD patients. The median life expectancy of HD and PD patients was 38.4 and 36.6 months, respectively.
Limitations of the study include a lack of randomization and failure to consider switches from one dialysis modality to the other.
Bottom line: Patients beginning their renal replacement therapy with PD had similar mortality after five years compared to patients using in-center HD.
Citation: Mehrotra R, Chiu YW, Kalantar-Zadeh K, Bargman J, Vonesh E. Similar outcomes with hemodialysis and peritoneal dialysis in patients with end-stage renal disease. Arch Intern Med. 2011;171(2):110-118.
Pneumococcal Urinary Antigen Test Might Guide Community-Acquired Pneumonia Treatment
Clinical question: What is the diagnostic accuracy and clinical utility of pneumococcal urinary antigen testing in adult patients hospitalized with community-acquired pneumonia (CAP)?
Background: Although CAP is common, our ability to determine its etiology is limited, and empirical broad-spectrum antibiotic therapy is the norm. Pneumococcal urinary antigen testing could allow for the more frequent use of narrow-spectrum pathogen-focused antibiotic therapy.
Study design: Prospective cohort study.
Setting: University-affiliated hospital in Spain.
Synopsis: This study included consecutive adult patients hospitalized with CAP from February 2007 though January 2008. A total of 464 patients with 474 episodes of CAP were included. Pneumococcal urinary antigen testing was performed in 383 (80.8%) episodes of CAP. Streptococcus pneumoniae was felt to be the causative pathogen in 171 cases (36.1%). It was detected exclusively by urinary antigen test in 75 of those cases (43.8%).
For the urine antigen test, specificity was 96% (95% CI, 86.5 to 99.5), and the positive predictive value was 96.5% (95% CI, 87.9 to 99.5). The results of the test led clinicians to reduce the spectrum of antibiotics in 41 patients, and pneumonia was cured in all 41 of these patients. Treatment was not modified despite positive antigen test results in 89 patients.
Limitations of this study include a lack of complete microbiological data for all patients. The study also highlighted the difficulty in changing clinicians’ prescribing patterns, even when test results indicate the need for treatment modification.
Bottom line: A positive pneumococcal urinary antigen test result in adult patients hospitalized with CAP can help clinicians narrow antimicrobial therapy with good clinical outcomes.
Citation: Sordé R, Falcó V, Lowak M, et al. Current and potential usefulness of pneumococcal urinary antigen detection in hospitalized patients with community-acquired pneumonia to guide antimicrobial therapy. Arch Intern Med. 2011;171(2):166-172.
Racial Disparities Detected in Hospital Readmission Rates
Clinical question: Do black patients have higher odds of readmission than white patients, and, if so, are these disparities related to where black patients receive care?
Background: Racial disparities in healthcare are well documented. Understanding and eliminating those disparities remains a national priority. Reducing hospital readmissions also is a policy focus, as it represents an opportunity to improve quality while reducing costs. Whether there are racial disparities in hospital readmissions at the national level is unknown.
Study design: Retrospective cohort study.
Setting: Medicare fee-for-service beneficiaries from 2006 to 2008.
Synopsis: Medicare discharge data for more than 3 million Medicare fee-for-service beneficiaries aged 65 years or older discharged from January 1, 2006, to November 30, 2008, with the primary discharge diagnosis of acute myocardial infarction (MI), congestive heart failure, or pneumonia were used to calculate risk-adjusted odds of readmission within 30 days of discharge. Hospitals in the highest decile of proportion of black patients were categorized as minority-serving.
Overall, black patients had 13% higher odds of all-cause 30-day readmission than white patients (24.8% vs. 22.6%, OR 1.13, 95% CI, 1.11-1.14), and patients discharged from minority-serving hospitals had 23% higher odds of readmission than patients from non-minority-serving hospitals (25.5% vs. 22.0%, OR 1.23, 95% CI, 1.20-1.27). Among those with acute MI, black patients had 13% higher odds of readmission (OR 1.13, 95% CI, 1.10-1.16), irrespective of the site of care, while patients from minority-serving hospitals had 22% higher odds of readmissions (OR 1.22, 95% CI, 1.17-1.27), even after adjusting for patient race. Similar disparities were seen for CHF and pneumonia. Results were unchanged after adjusting for hospital characteristics, including markers of caring for poor patients.
Bottom line: Compared with white patients, elderly black Medicare patients have a higher 30-day hospital readmission rate for MI, CHF, and pneumonia that is not fully explained by the higher readmission rates seen among hospitals that disproportionately care for black patients.
Citation: Joynt KE, Orav EJ, Jha AK. Thirty-day readmission rates for Medicare beneficiaries by race and site of care. JAMA. 2011;305(7):675-681.
Easily Identifiable Clinical and Demographic Factors Associated with Hospital Readmission
Clinical question: Which clinical, operational, or demographic factors are associated with 30-day readmission for general medicine patients?
Background: While a few clinical risk factors for hospital readmission have been well defined in subgroups of inpatients, there are still limited data regarding readmission risk that might be associated with a broad range of operational, demographic, and clinical factors in a heterogeneous population of general medicine patients.
Study design: Retrospective observational study.
Setting: Single academic medical center.
Synopsis: The study examined more than 10,300 consecutive admissions (6,805 patients) discharged over a two-year period from 2006 to 2008 from the general medicine service of an urban academic medical center. The 30-day readmission rate was 17.0%.
In multivariate analysis, factors associated with readmission included black race (OR 1.43, 95% CI, 1.24-1.65), inpatient use of narcotics (OR 1.33, 95% CI, 1.16-1.53) and corticosteroids (OR 1.24, 95% CI, 1.09-1.42), and the disease states of cancer (with metastasis 1.61, 95% CI, 1.33-1.95; without metastasis 1.95, 95% CI 1.54-2.47), renal failure (OR 1.19, 95% CI 1.05-1.36), congestive heart failure (OR 1.30, 95% CI, 1.09-1.56), and weight loss (OR 1.26, 95% CI, 1.09-1.47). Medicaid payor status (OR 1.15, 95% CI, 0.97-1.36) had a trend toward readmission. None of the operational factors were significantly associated with readmission, including discharge to skilled nursing facility or weekend discharge.
A major limitation of the study was its inability to capture readmissions to hospitals other than the study hospital, which, based on prior studies, could have accounted for nearly a quarter of readmissions.
Bottom line: Readmission of general medicine patients within 30 days is common and associated with several easily identifiable clinical and nonclinical factors.
Citation: Allaudeen N, Vidyarthi A, Maselli J, Auerbach A. Redefining readmission risk factors for general medicine patients. J Hosp Med. 2011;6(2):54-60.
Unplanned Medical ICU Transfers Tied to Preventable Errors
Clinical question: What fraction of unplanned medical ICU (MICU) transfers result from errors in care and why do they occur?
Background: Prior studies have suggested that 14% to 28% of patients admitted to the MICU are unplanned transfers. It is not known what fraction of these transfers result from errors in care, and whether these transfers could be prevented.
Study design: Retrospective cohort study.
Setting: University-affiliated academic medical center.
Synopsis: All unplanned transfers to the MICU from June 1, 2005, to May 30, 2006, were included in the study. Three independent observers, all hospitalists for more than three years, reviewed patient records to determine the cause of unplanned transfers according to a taxonomy the researchers developed for classifying the transfers. They also determined whether the transfer could have been prevented.
Of the 4,468 general medicine admissions during the study period, 152 met inclusion criteria for an unplanned MICU transfer. Errors in care were judged to account for 19% (n=29) of unplanned transfers, 15 of which were due to incorrect triage at admission and 14 to iatrogenic errors, such as opiate overdose during pain treatment or delayed treatment. All 15 triage errors were considered preventable. Of the iatrogenic errors, eight were considered preventable through an earlier intervention. Overall, 23 errors (15%) were thought to be preventable. Observer agreement was moderate to almost perfect (κ0.55-0.90).
Bottom line: Nearly 1 in 7 unplanned transfers to the medical ICU are associated with preventable errors in care, with the most common error being inappropriate admission triage.
Citation: Bapoje SR, Gaudiani JL, Narayanan V, Albert RK. Unplanned transfers to a medical intensive care unit: causes and relationship to preventable errors in care. J Hosp Med. 2011;6(2):68-72. TH