769    

Background: The incidence of Pneumocystis pneumonia (PCP) has decreased dramatically during the HAART era, but it still represents an important cause of death among HIV-infected people. Colonized animals, both immunocompromised and immunocompetent, have been shown to be able to infect immunocompromised animals that then develop PCP. Preliminary studies indicate that humans may also be carriers of Pneumocystis. However, the prevalence of colonization and its role in the pathogenesis of PCP remain unclear. We studied the prevalence of Pneumocystis colonization using oropharyngeal washing.

Methods:  In this cohort study, oropharyngeal washing specimens were obtained from HIV⁺ patients at San Francisco General Hospital at the time sputum induction or bronchoscopy  was performed to rule out PCP. Subjects provided a daily oropharyngeal washing specimen until the diagnostic evaluation for PCP was complete. The specimens were stored at -70°C and processed at the Centers for Disease Control and Prevention, where DNA was extracted, PCR amplified and sequenced at the mitochondrial large subunit (Mtlsu) rRNA. Patients were defined as colonized if they had at least one oropharyngeal wash that was PCR-positive in the absence of a microscopic or clinical diagnosis of PCP.

Results: Of the 56 patients enrolled from August 2002 to June 2003, 18 had no clinical or microbiological diagnosis of PCP. These 18 subjects provided a total of 30 samples (range 1 to 5). If the 18 subjects, 13 (72%) without PCP had a positive-PCR oropharyngeal washing specimen and were considered to be colonized. Of the 5 non-colonized subjects, 4 provided only 1 specimen.The median CD4 count was 83 cells/m/L (range 9 to 702) in colonized subjects and 203 cells/mL (range 15 to 234) in those not colonized. Among the colonized subjects, 5 (38%) had a history of PCP, while none of the non-colonized subjects reported this history. Four colonized patients and 1 of those not colonized received sulpha-prophylaxis for PCP in the 3 months before the hospitalization. Serial oropharyngeal washing coming from the same patients were not constantly positive, indicating a possible effect of the modality of specimen collection or empirical therapy for PCP on the PCR results. All patients were alive after a minimum of 6 weeks follow-up except 1 (colonized) patient who died from ARDS and sepsis.

Conclusions: The prevalence of Pneumocystis colonization is high among HIV⁺ patients. These subjects could play a critical role in the epidemiology and transmission of PCP in immunocompromised populations.

Keywords: Pneumocystis; Colonization; Mtlsu