ObjectivesThe aim of this study was to evaluate the appropriateness of piperacillin/tazobactam (Tazocin®; Pfizer, New York, NY) usage in our hospital.Subjects and methodsThis retrospective study was designed to involve all patients admitted to Hamad General Hospital and prescribed piperacillin/tazobactam as an empiric therapy from January 1 to March 31, 2008. The medical records of such patients were retrospectively reviewed and studied.ResultsDuring this period, 610 prescriptions were ordered for 596 patients. The main indication for initiation of Tazocin was sepsis (207/610; 34%). The overall rate of appropriateness of empirical therapy was 348/610 (57%). Most of the inappropriate prescriptions were in cases of aspiration pneumonia and abdominal infections, with inappropriate prescriptions found mostly in surgical wards (86%) and the surgical intensive care unit (66.7%). Septic work-up results showed positive cultures in 57% (345/610) of cases. There were 198/254 prescriptions (78%) where antibiotics were changed according to the sensitivity data to narrow-spectrum antimicrobials. In 56/254 (22%) cases, pathogens were susceptible to narrow-spectrum antibiotics even though piperacillin/tazobactam was continued.ConclusionOur study showed that there was an injudicious use of piperacillin/tazobactam at our hospital, evidenced by the significant number of inappropriate empiric prescriptions and inappropriate drug modifications, based on the results of microbial cultures and antibiograms.

ObjectiveTo investigate the diagnostic utility of interferon-gamma (IFN-γ) and adenosine deaminase (ADA) in tuberculous pleural effusions by determining the best cutoff levels of these two markers for pleural tuberculosis, in the context of the local epidemiological settings in Qatar.MethodsWe prospectively studied IFN-γ and ADA levels in the pleural fluid of patients presenting to Hamad General Hospital between June 1, 2009 and May 31, 2010.ResultsWe studied 103 patients with pleural effusions, 72 (69.9%) with pleural tuberculosis, and 31 (30.1%) with nontuberculous etiologies. The mean IFN-γ concentration for the group with tuberculous effusions was significantly higher than that in the group with nontuberculous effusions (1.98 ± 81 vs 0.26 ± 10 pg/mL [P < 0.0001]). The mean ADA activity for the tuberculous effusions group was significantly higher than that in group with nontuberculous effusions (41.30 ± 20.09 vs 14.93 ± 14.87 U/L [P < 0.0001]). By analysis of receiver operating characteristic (ROC) curves, the best cutoff values for IFN-γ and ADA were 0.5 pg/mL and 16.65 U/L, respectively. The results for IFN-γ vs ADA were: for sensitivity, 100% vs 86%, respectively; for specificity, 100% vs 74%, respectively; for positive predictive value, 100% vs 88.5%, respectively; and for negative predictive value, 100% vs 69.7%, respectively.ConclusionIFN-γ and ADA could be used as valuable parameters for the differentiation of tuberculous from nontuberculous effusion, and IFN-γ was more sensitive and specific for tuberculous effusion than ADA.