Extended-spectrum beta-lactamase-producing Enterobacteriaceae carriage upon hospital admission: prevalence and risk factors.

The aim of this study was to assess the value of surveillance cultures in identifying extended-spectrum beta-lactamase-producing Enterobacteriaceae (ESBL) carriers upon admission to hospital, and to identify risk factors for carriage. This prospective cross-sectional study included all hospital admissions over one week. Of 525 patients screened, 56 were positive for ESBLs. Half were only identified through screening. Four independent risk factors were identified: nursing home residency, hospitalization in the previous year, prior antibiotic treatment and prior ESBL carriage. Over 50% of the screened patients had at least one risk factor. By screening this targeted population, 87.5% of positive patients would have been identified.

The effect of empiric antibiotic therapy on mortality in debilitated patients with dementia.

The purpose of this investigation was to assess the effect of empirical antibiotic treatment on 30-day mortality among debilitated inpatients with dementia and Gram-negative bacteremia. A retrospective cohort study in the years 2005-2007 was undertaken. Data were collected through patient chart review. The association between individual variables and 30-day mortality was assessed through univariate analysis. Variables significantly associated with mortality (p < 0.05) were entered into a logistic regression analysis. Adjusted odds ratios (ORs) for mortality with 95% confidence intervals (CIs) are shown. Subgroup analysis of patients with and without decubitus ulcers was performed. In our cohort of 378 patients with dementia and Gram-negative bacteremia, the 30-day mortality was 39% overall and 61% in the subgroup of patients with decubitus ulcers. Inappropriate empirical therapy was associated with higher mortality, although this effect was not statistically significant (OR 1.41, 95% CI 0.86-2.29). Inappropriate empirical therapy did not affect mortality in the subgroup of patients with decubitus ulcers (OR 0.37, 95% CI 0.11-1.28). Other factors found to independently affect mortality included age, co-morbidities, source of infection, sepsis severity, and hospital-acquired infection. Appropriate empirical antibiotic therapy for patients with dementia and severe bacterial infection did not have a clear advantage, especially in the sickest group of patients with decubitus ulcers.

Nucleoside reverse transcriptase inhibitors in combination therapy for HIV patients: systematic review and meta-analysis.

Treatment guidelines recommend dual nucleoside reverse transcriptase inhibitors (NRTI ) as a part of combination antiretroviral therapy. The objective of this study was to assess the relative efficacy and toxicity of the dual NRTI part of the regimen in antiretroviral-naïve HIV-1-infected adults. A systematic review and meta-analysis of randomized controlled trials assessing highly active antiretroviral therapy (HAART) for treatment-naïve HIV-infected adults with a 48-week follow-up were done. We searched the PubMed, CENTRAL, and EMBASE electronic databases up to April 2009. Proceedings from conferences were reviewed. Data were extracted independently by two reviewers. Primary outcome was viral suppression at 48 weeks. The odds ratio (OR) is reported with its corresponding 95% confidence interval (CI). Twenty-two randomized controlled trials, including 8,184 HIV-treatment-naïve patients, were included. The combination didanosine + lamivudine/emtricitabine (four trials, 1,148 patients) was more effective (OR 0.53, 95% CI 0.41-0.68) for viral load (VL) >50 copies/ml and less toxic (OR 0.52, 95% CI 0.36-0.76) for discontinuation due to adverse events (AE) than its comparators. The combination tenofovir + lamivudine/emtricitabine was more effective and less toxic (OR 0.75, 95% CI 0.58-0.96) only in the 144-week follow-up data (two trials, 1,119 patients). Abacavir + lamivudine had similar efficacy to its comparators (OR 0.81, 95% CI 0.8-1.1), but more AIDS-defining events (OR 3.22, 95% CI 1.24, 8.40). The once-daily combination didanosine + lamivudine/emtricitabine was found to be effective and tolerable. This combination, soon to be generic, should be compared to the current standard of care in a large randomized trial. An effective, safe, and inexpensive alternative to current options is needed.

Low albumin gradient ascites complicating severe pseudomembranous colitis.

Pseudomembranous colitis (PMC) is a frequently severe, sometimes fatal iatrogenic disease that is antibiotic-associated in almost all cases. The most common clinical features of PMC include abdominal pain, watery diarrhea, fever, leukocytosis, hypoalbuminemia, and hypovolemia. Ascites, not considered a well-known feature of PMC, is fairly common, based on a review of the English language literature but has not been characterized fully. This case report describes 5 patients with PMC who presented with low serum-ascites albumin gradient (SAAG) and neutrocytic ascites, without evidence of infectious, malignant, or inflammatory peritoneal disease, which has not been reported previously. In 1 patient, massive low SAAG ascites was the presenting manifestation of PMC, a feature also not reported previously. Three of the 5 (60%) patients had acquired immunodeficiency syndrome. The characteristics of the fluid specimens in these 5 patients and the possible pathogenetic mechanisms are proposed. The findings suggest that PMC should be included in the differential diagnosis of low SAAG ascites, especially in patients with acquired immunodeficiency syndrome.

Contamination of blood cultures during venepuncture: fact or myth?

Contamination of blood cultures is believed to occur mainly during the venepuncture procedure. Consequently, meticulous preparation of the venepuncture site is widely recommended. To determine whether the contamination rate is indeed affected by the quality of the antiseptic procedure at the venepuncture site, 181 paired cultures were collected from 176 patients during a 6-month period after either strict antiseptic cleansing of skin with alcohol followed by povidone-iodine, or after brief disinfection with alcohol alone. The contamination rate was not influenced by the antiseptic procedure, and corresponded to the accepted percentage reported in most other studies. Eight false positive cultures (4.4%) were obtained after strict antisepsis of the skin and 6 (3.3%) after short simple cleansing with alcohol (P = 0.39). Our results suggest that contamination of blood cultures may not be related to the venepuncture procedure--regardless of the antiseptic technique used--but may be due to later stages of laboratory handling and processing of the specimens. Review of the literature has provided further indirect evidence to support this conclusion.