Abstract

Objective: Candida species are often isolated from respiratory specimens of patients with a history of long duration of hospitalization, immunosuppression, diabetes, alcoholism, history of wide spectrum antibiotic or corticosteroid usage. The aim of this retrospective study is to investigate the effect of isolation of Candida species from respiratory specimens on prognosis.

Material and Method: Patients in the Department of Chest Disease in whom Candida species had been grown in ≥2 respiratory specimens obtained at the same or at different time periods within the last two years were enrolled in the study. Demographic features of the patients, risk factors for fungal infection, and the effect of this infection on prognosis were evaluated.

Results: Out of 47 cases (mean age 67.6±14.6 yrs, 33 males) enrolled in the study, 29 were diagnosed as pneumonia at admission. The most frequent species isolated was C. albicans (68.1%). Fifty bacterial agents were isolated from respiratory specimens of 29 patients (61.7%) initially or on follow-up along with Candida species. Immunosuppression was detected in 28 patients (59.6%). Mean duration of hospitalization was 22.3±15.2 days and 26 patients (55.3%) died. Radiologically, multiple-zone involvement, C. tropicalis isolation, high APACHE II score, co-existing bacterial infection, low PaO2/FiO2 ratio, and low albumin levels were found to be associated with mortality.

Conclusion: This study showed that Candida species could also be isolated in respiratory specimens of non-immunosuppressed patients and mortality was high. It was concluded that bacterial co-infection and C. tropicalis isolation affected the prognosis negatively. (Tur Toraks Der 2011; 12: 153-7)