The genus of Nocardia is rapidly expanding and the species distribution varies with different geographical locations. We retrospectively reviewed the laboratory records of the bacteriology laboratory at National Taiwan University Hospital from January 1998 to June 2008 to identify patients with nocardiosis. During the study period, 164 isolates of Nocardia spp. were identified from 134 patients but only 113 patients had Nocardia infection. Nocardia brasiliensis (n = 54) was the most common pathogen, followed by N. asteroides (n = 36), N. farcinica (n = 7), N. flavorosea (n = 4), N. otitidiscaviarum (n = 3), N. nova (n = 3), N. beijingensis (n = 2) and one each of N. puris, N. jinanensis and N. takedensis. The major types of infection were cutaneous infection (56.6%), pulmonary infection (33.6%) and disseminated infection (7.1%). Eighty-eight patients received sulfonamide-containing antibiotic and eight of 100 patients with available data on outcomes died during the episode of nocardiosis. In conclusion, the clinical and microbiological manifestations of Nocardiosis vary with the different Nocardia species. Accurate identification of the species is crucial to make the diagnosis.

No abstract

We describe five patients with positive blood culture for Kocuria species. Three patients had catheter-related bacteraemia and one had infective endocarditis caused by Kocuria kristinae, and one had a K. marina isolate, which was considered to be a contaminant. Identification of the isolates was further confirmed by 16S rRNA gene sequence analysis. In conclusion, Kocuria species are an unusual cause of infection in immunocompromised patients. Accurate identification with molecular methods is imperative for the diagnosis of these unusual pathogens.

Patients presenting with pleural effusion of undetermined aetiology were prospectively enrolled, and an enzyme-linked immunospot (ELISPOT) assay on pleural fluid and peripheral blood was performed. Forty patients were studied, including 19 with culture- or biopsy-confirmed (n = 15) or clinically compatible (n = 4) tuberculous pleurisy, and 21 with pleural effusions due to non-tuberculous causes. The sensitivity, specificity and positive and negative predictive values of the assay were 94.7%, 85.7%, 85.7% and 94.7%, respectively, on pleural fluid, and 77.8%, 90.5%, 87.5% and 82.6%, respectively, on blood. Antigen-specific, interferon-gamma-secreting T-cells were concentrated eight to ten times in pleural fluid as compared with blood. Among the seven patients not suitable for pleural biopsy and three patients whose biopsy results were non-diagnostic, nine had positive ELISPOT result with pleural fluid. The ELISPOT assay for interferon-gamma can accurately diagnose tuberculous pleurisy and is helpful for patients not suitable for pleural biopsy and those whose biopsy results are non-diagnostic.

No abstract