Pseudomonas aeruginosa bacteraemia in patients with hematologic malignancies: risk factors, treatment and outcome.

Our aims were to identify factors associated with Pseudomonas aeruginosa (PA) bloodstream infection (BSI) in patients with hematological malignancies and evaluate the outcome of the affected patients. Consecutive patients with hematological malignancies who developed PA BSI were identified. Subsequently, two case-control studies were performed to evaluate the risk factors (i) for PA BSI and (ii) for carbapenem resistant (CR) PA BSI. Patients' outcome was evaluated at 28 days after the onset of bacteraemia. A total of 64 patients with PA BSI (45 caused by CS and 19 by CR organisms) and 128 without PA BSI were enrolled. Patients with rapidly fatal disease, steroid use, neutropenia or prior surgery were more likely to develop PA BSI, whereas patients with previous hospitalization and prior use of fluoroquinolones were more likely to develop CR PA BSI. The 28-day mortality rate was 35.9%. Severity of sepsis was the only independent predictor of adverse outcome.

Valuation and preliminary validation of the Greek 15D in a sample of patients with coronary artery disease.

OBJECTIVE: To replicate, using the Greek version of the 15D instrument, the three-stage valuation procedure, using a sample of coronary artery disease patients; to assess on a preliminary basis the psychometric properties of the instrument in the Greek health-care environment; to evaluate the health-related quality of life of patients with coronary artery disease. METHODS: The generic instrument 15D was translated and culturally adapted into the Greek language and setting. It was then administered during October 2005 to May 2006 to 420 coronary artery disease patients. The three-stage valuation procedure was employed and, with the use of elicited preference weights, a single health-related quality of life index score for the patients was derived. Scores were also calculated using the original Finnish valuation system and compared with the previously derived utilities. Sensitivity, reliability, and validity were assessed by examining response distributions, floor and ceiling effects, item-scale correlations, Cronbach's alpha coefficients, and hypothesized relationships between sociodemographic variables and health-related quality of life. RESULTS: The Greek valuation systematically generated higher 15D utilities than the Finnish one. Nevertheless, the utilities derived with the original Finnish valuation algorithm differed significantly with respect to sex, age, and education, just as they did using the Greek valuation system. In most cases, the full range of possible responses has been used satisfactorily and floor and ceiling effects were generally moderate. In general, internal consistency reliability was also satisfactory. CONCLUSION: The valuation system generated results demonstrating satisfactory psychometric properties. Further research should validate the 15D in the general Greek population.