Nearly all CDIs related to exposure in healthcare setting; CT angiography safe to determine ED discharge in potential ACS; predicting physiological deterioration through EMR use; accuracy of rapid influenza diagnostic tests; multifaceted QI intervention improves ACS therapy adherence; rivaroxaban vs. heparin plus vitamin K antagonist for PE treatment; coronary stent implantation in CAD; prolonged prophylaxis use in hip-replacement patients.
by Sanjay Galhotra, MD, Paul Martin, MD, MPH, Martina Stehlikova, MD, Claire Brown, MD, Valentyna Goloborodko, MD, Andrea Porovecchio, MD, Jeff Ceresnak, MD, William Southern, MD, MS, SFHM
A guide to this month’s studies
Clinical question: What is the contribution of non-hospital healthcare exposures to the burden of Clostridium difficile infection (CDI)?
Background: CDI is common among hospitalized patients and is associated with high rates of morbidity. Most CDI prevention programs are aimed at hospital-based risk factors. However, non-hospital healthcare exposures might contribute significantly to the burden of CDI, and hospital-based prevention programs might not address these risks.
Study design: Retrospective analysis of three population-based data sets.
Setting: Three population data sets tracking infections on a national and state level in the U.S.
Synopsis: This study analyzed two data sets to determine the contribution of non-hospital healthcare exposure to CDIs, and examined a third to determine whether a program addressing hospital-based exposures reduced the rate of CDI.Using data from the CDC’s Emerging Infections Program, 10,342 CDI cases from 111 hospitals were examined. Each case was classified as hospital onset (i.e. occurring three days after admission), nursing home onset, or community onset. A total of 94% of cases were associated with recent exposure to healthcare, 75% were classified as non-hospital onset, and 20% of hospital-onset cases were in patients who had recently been in a nursing home, suggesting that non-hospital-based healthcare exposures play a significant role. The second analysis used data from the National Healthcare Safety Network to examine 42,157 CDIs in 711 hospitals.
The authors found that 52% of CDIs were present on admission. The pooled rate of hospital-onset CDI was low (7.4 per 10,000 patient-days). The final analysis examined hospital-onset CDI rates across three hospital-based prevention programs over a period of nearly two years. The interventions were associated with a 20% decrease in CDI rates (to 7.5 from 9.3 per 10,000 patient days).
Bottom line: Nearly all CDIs are related to exposure in healthcare settings. Exposure to nursing home and ambulatory care seem to play a major role, and infection-control measures in addition to hospital-based strategies need to target these areas. Hospital-based interventions moderately decreased overall rates of CDI.
Citation: Centers for Disease Control and Prevention. Vital signs: preventing Clostridium difficile infections. MMWR Morb Mortal Wkly Rep. 2012;61:157-162.
Clinical question: Is a strategy utilizing CT angiography safe to allow discharge from the emergency department low- to intermediate-risk patients with possible acute coronary syndrome (ACS)?
Background: Admission rates for patients with possible acute coronary syndrome are high, but the majority of patients will not have cardiac causes of the symptoms. Coronary computed tomographic angiography (CCTA) has a high negative predictive value for the detection of coronary artery disease, but its utility in determining if a patient can be safely discharged from the emergency department has not been established.
Study design: Prospective, randomized, controlled, multicenter trial.
Setting: Five EDs in the U.S.; three included observation units.
Synopsis: The study enrolled 1,392 patients in a 2:1 ratio, with two-thirds undergoing CCTA and the other third treated with usual care (those admitted for possible ACS with a TIMI score of 0 to 2). Of 908 patients assigned to CCTA, 640 had a negative test, and none of them died or had a myocardial infarction within 30 days. Patients in the CCTA group were more likely than the group undergoing usual care to be discharged from the ED (49.6% vs. 22.7%), have a shorter length of stay (18 hours vs. 24.8 hours, P<.0001), and receive a coronary disease diagnosis.
Bottom line: In a randomized controlled trial, the use of CCTA to evaluate low- to intermediate-risk emergency department patients with possible ACS appears to be safe and effective, and leads to more discharges from the ED.
Citation: Litt HI, Gatsonis C, Snyder B, et al. CT angiography for safe discharge of patients with possible acute coronary syndromes. N Engl J Med. 2012;366:1393-1403.
Clinical question: Can impending physiological deterioration be predicted in medical-surgical ward patients using data from a comprehensive electronic health record (EHR)?
Background: Unplanned ICU transfer is associated with increased mortality and morbidity. Previous studies have examined physiological variables and track-and-trigger systems that might help in the identification of ward patients who are at risk for deterioration and transfer to a higher level of care. More hospitals across the U.S. are using EHRs, and eventually all hospitals are expected to be using one. However, it is unclear if an EHR-based model can accurately predict patients’ clinical deterioration.
Study design: Retrospective case-control study; the unit of analysis was the “patient shift,” each defined by a particular 12-hour period and patient. A complex model to predict patient deterioration and transfer to the ICU was developed and validated using EHR data that was available prior to the deterioration shift.
Setting: Fourteen hospitals of the Northern California Kaiser Permanente Medical Care Program (an integrated healthcare delivery system) that have used an operational EHR for at least three months.
Synopsis: The study identified 4,036 patient-shifts during which a transfer to the ICU occurred, and compared these with 39,782 patient-shifts during which no transfer occurred. Variables in the model included patient demographic data, diagnoses, comorbidities, lab results, and vital signs. The EHR-based predictive model performed well at predicting clinical deterioration and transfer to the ICU (derivation c-statistic 0.84, validation c-statistic 0.77). The model performed best among patients with gastrointestinal diseases (c-statistic 0.84) and worst in patients with acute myocardial infarction (c-statistic 0.57).
Bottom line: Impending physiologic deterioration can be predicted using EHR-based models.
Citation: Escobar GJ, Laguardia JC, Turk BJ, et al. Early detection of impending physiologic deterioration among patients who are not in intensive care: Development of predictive models using data from an automated electronic medical record. J Hosp Med. 2012:doi:10.1002/jhm.1929 [Epub ahead of print].
Clinical question: How accurate are rapid influenza diagnostic tests, and can they be used to guide further management?
Background: Three million to 5 million people a year develop influenza, which has a high mortality rate and potential global implications. The gold-standard diagnostic tests, PCR and viral culture, have a long turnaround time and are expensive. There have been limited systematic reviews addressing the accuracy of various rapid influenza diagnostic tests (RIDTs), especially in adults.
Study design: Systematic review and meta-analysis.
Synopsis: Studies that compared commercial rapid influenza tests versus reference standards of PCR or viral culture were included. The 159 rapid tests were evaluated in 119 studies representing 26 different brands of commercial tests. Sixty-two of 119 studies included both adults and children, and 41 studies included children only. Overall, rapid influenza tests had a sensitivity of 62.3% and a specificity of 98.2%. There was a higher sensitivity for children than adults (67% vs. 54%) and a higher sensitivity for influenza A than influenza B (65% vs. 52%). The brand of rapid test made little difference. Most articles reviewed did not include a direct comparison between individual brands of RIDTs.
Bottom line: Rapid influenza diagnostic tests have low sensitivity and high specificity. A positive rapid test indicates likely influenza, but a negative test does not rule it out.
Citation: Chartrand C, Leeflang MM, Minion J, Brewer T, Pai M. Accuracy of rapid influenza diagnostic tests: a meta-analysis. Ann Intern Med. 2012;156:500-511.
Clinical question: Does a multipronged approach to QI interventions improve the adherence to evidence-based therapies in acute coronary syndromes?
Background: Despite evidence-based recommendations for the use of antiplatelet medication, anticoagulants, beta-blockers, and statins in the setting of an acute coronary syndrome, adherence to these recommendations is poor, especially in low- to middle-income patients.
Study design: Cluster-randomized, prospective, blinded trial.
Setting: Thirty-six urban, public hospitals in Brazil.
Synopsis: Hospitals were randomly allocated to a multifaceted QI strategy versus routine practice. At each hospital, patients were enrolled in a consecutive manner when they presented to the emergency department with an acute coronary syndrome. The QI intervention included multiple modalities, including checklists in the ED, reminders, case management (by a trained RN) to ensure compliance, and educational materials to the staff (pocket algorithms and interactive web programs).
The primary endpoint was adherence to all available evidence-based therapies in the first 24 hours of management. Patients were more likely to receive all evidence-based therapies in hospitals that received the intervention (67.9% vs. 49.5%, P=0.01). The secondary endpoint (adherence at 24 hours and at discharge) was also significantly in favor of the intervention group (50.9% vs. 31.9%, P=0.03).
The findings might not be applicable to institutions with high adherence rates to evidence-based therapy.
Bottom line: A multifaceted QI intervention improves adherence to evidence-based therapies in urban hospitals in medium-income countries.
Citation: Berwanger O, Guimaraes HP, Lanranjeira LN, et al. Effect of a multifaceted intervention on use of evidence-based therapies in patients with acute coronary syndromes in Brazil: the BRIDGE-ACS randomized trial. JAMA. 2012 [Epub ahead of print].
Clinical question: How does the efficacy and safety of rivaroxaban compare with low-molecular-weight heparin (LMWH) plus vitamin K antagonist (VKA) for treatment of acute pulmonary embolism (PE)?
Background: Data support the use of rivaroxaban for stroke prevention in atrial fibrillation, for VTE prophylaxis after orthopedic surgery, and for the treatment of acute coronary syndromes and acute deep vein thromboses. The EINSTEIN-PE investigators now study rivaroxaban for the treatment of acute symptomatic PE.
Study design: Randomized, open-label, noninferiority trial.
Setting: Two hundred sixty-three sites in 38 countries.
Synopsis: The investigators enrolled 4,832 patients with acute symptomatic PE (with or without DVT). Rivaroxaban alone was compared with standard therapy (enoxaparin followed by VKA to an international normalized ratio [INR] of 2 to 3) for three, six, or 12 months. The mean study duration was nine months. INR was in the therapeutic range 62.7% of the time.
The combined primary endpoint of fatal or nonfatal PE, or DVT, occurred in 2.1% in the rivaroxaban group and 1.8% in the standard-therapy group, meeting the pre-specified criteria for non-inferiority (HR 1.12; 95% CI: 0.75-1.68). There were no significant differences in the primary safety outcome of bleeding (HR, 0.90; 95% CI: 0.76-1.07). However, major bleeding events favored rivaroxaban, with events in 1.1% vs. 2.2% of those on VKA (HR 0.49, CI 0.31-0.79, P=0.003). Efficacy and safety of fixed-dose rivaroxaban did not vary by age, sex, weight, glomerular filtration rate, or extent of initial PE.
Bottom line: In patients with symptomatic acute PE, rivaroxaban is noninferior to standard treatment with LMWH plus VKA and might have a safety benefit of less major bleeding. Rivaroxaban and other new oral anticoagulants are likely to be used more frequently as more data is accumulated, but at this time, the guidelines continue to recommend VKA and LMWH as first-line agents.
Citation: The EINSTEIN-PE Investigators, Büller HR, Prins MH, et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N Engl J Med. 2012;366:1287-1297.
Clinical question: Does stent implantation lead to better outcomes when compared to medical therapy in patients with stable coronary artery disease (CAD)?
Background: It is unclear whether stent implantation improves outcomes any more than medical therapy in patients with stable CAD. This is a meta-analysis of trials comparing outcomes of coronary stent implantation with optimal medical therapy.
Study design: Meta-analysis of eight prospective randomized nonblinded trials comparing stent implantation plus medical therapy to medical therapy alone in patients with stable CAD. Outcomes of death, nonfatal myocardial infarction, unplanned revascularization, and persistent angina were compared between the two groups.
Synopsis: Eight trials with 7,229 patients were included with an average follow-up of 4.3 years. In a pooled analysis, coronary stenting versus medical therapy was associated with similar risk of death (8.9% vs. 9.1% P=0.83); nonfatal myocardial infarction (8.9% vs. 8.1%, P=0.22); and persistent angina (29% vs. 33%, P=0.10). Coronary stenting was associated with decreased risk of unplanned revascularization, but the difference was not significant (21.4% vs. 30.7%, P=0.11).
Bottom line: In stable CAD, coronary stent implantation provides no additional benefit when compared to medical therapy.
Citation: Stergiopoulos K, Brown DL. Initial coronary stent implantation with medical therapy vs. medical therapy alone for stable coronary artery disease: meta-analysis of randomized controlled trials. Arch Intern Med. 2012;172:312-319.
Clinical question: Should patients with hip or knee surgery receive venous thromboprophylaxis (VTE) prophylaxis for seven to 10 days post-operatively or for 21 days or more?
Background: Venous thromboembolism is the most frequent medical complication of knee and hip surgeries. Routine prophylaxis is recommended for minimum of seven to 10 days, but a longer duration might be more beneficial.
Study design: Systemic review and meta-analysis of eight randomized, controlled trials comparing different duration of VTE prophylaxis after hip fracture surgery and total hip or knee replacement.
Synopsis: Eight trials with 2,917 patients met inclusion criteria. Pharmacological agents used for VTE prophylaxis in these trials were coumadin, enoxaparin, dalteparin, and fondaparinux. In patients with hip replacement, undergoing VTE prophylaxis for ≥21 days (compared with seven to 10 days) was associated with a decreased risk of pulmonary embolism (odds ratio [OR] 0.14, 95% CI 0.04-0.47), symptomatic DVTs (OR 0.36, 95% CI 0.16-0.81), asymptomatic DVTs (OR 0.48, 95% CI 0.31-0.75), and proximal DVTs (RR 0.29, CI 0.16-0.52). Prophylaxis for ≥21 days was also associated with an increased risk of minor bleeding (OR 7.55, 95% CI 1.51-37.64).
Data were not sufficient for evaluation of prolonged prophylaxis for total knee replacement and hip fracture surgery.
Bottom line: Prolonged prophylaxis is associated with fewer VTE events in patients with hip replacement but also carries increased risk of minor bleeding.
Citation: Sobieraj DM, Lee S, Coleman CI, et al. Prolonged versus standard-duration venous thromboprophylaxis in major orthopedic surgery: a systematic review. Ann Intern Med. 2012 May 15 [Epub ahead of print].
The Hospitalist newsmagazine reports on issues and trends in hospital medicine. The Hospitalist reaches more than 25,000 hospitalists, physician assistants, nurse practitioners, residents, and medical administrators interested in the practice and business of hospital medicine.