Risk factors for mortality of nosocomial bacteraemia in intensive care units.

OBJECTIVE: The aim of this study was to follow critically ill patients prospectively in intensive care units (ICUs) to determine risk factors for mortality and outcome associated with nosocomial bacteraemia (NB). SUBJECTS AND METHODS: A case-control study of 176 patients was conducted to identify the risk factors for mortality of NB in ICU patients. The study was performed in emergency, surgical and general surgical ICUs with 23 beds during a 15-month period. A total of 1,450 patients were admitted to the ICUs during the study period. The USA Center for Disease Control and Prevention definitions were used to diagnose nosocomial infections. Nosocomial bacteraemia was defined as the isolation of one or more organisms from blood cultures taken at least 48 h after admission, which were not related to a problem present on admission. An assessment of whether the isolated organisms represented true bacteraemia rather than contamination was made by clinical or laboratory evidence of infection. RESULTS: A total of 214 bacteraemia episodes were found in the 176 patients (64 female, 112 male; 51.3 +/- 21.3 years old), 90 of whom died and 86 survived. The bacteraemia rate was 12.1%. The most common etiological agents of bacteraemia were Klebsiella pneumoniae: 46 (21.5%), methicillin-resistant Staphylococcus aureus: 46 (21.5%), Pseudomonas aeruginosa: 32 (14.9%), and Escherichia coli: 20 (9.3%). Multivariate analysis showed that the requirement of mechanical ventilation for more than 7 days (p < 0.001), total parenteral nutrition (p = 0.034), inotropic drug (p < 0.001), and increased creatinine level (p = 0.034) were independent risk factors for mortality of NB in ICUs. CONCLUSIONS: Nosocomial infections caused by Gram-negative bacteria continue to be one of the major sources of morbidity and mortality.

Extrapulmonary tuberculosis in immunocompetent adults.

Tuberculosis continues to be a significant cause of morbidity and mortality. Although tuberculosis usually attacks the lungs, other organs can also be affected, leading to extrapulmonary tuberculosis (EPT) or disseminated tuberculosis. This study retrospectively analysed the incidence, clinical sites and risk factors for EPT in 252 patients with EPT between 1 January 1991 and 30 June 2003. EPT was defined as clinical, laboratory, imaging, and/or histopathological evidence of mycobacterial infection in a site other than hilar lymph nodes or lung parenchyma. In our study group, tuberculous lymphadenitis (36.5%) was found to be the most common clinical presentation of EPT. 119 (47.2%) patients developed the severe form of EPT, according to the WHO report, and 133 (52.8%) patients developed the less severe form. A case history of pulmonary tuberculosis was found to be a risk factor for the development of EPT (p <0.05). The study showed that EPT is still a public health problem. These findings suggested that pulmonary tuberculosis may play a critical role in the development of EPT. 12-month therapy may be chosen in patients with EPT considering acceptable adverse effects without relapses.