BACKGROUNDThe diagnosis of Pneumocystis pneumonia (PCP) in immunocompromised patients is still challenging today due to the absence of an in vitro culture system and the low diagnostic accuracy of microscopic examinations. Herein, we performed a meta-analysis to evaluate the accuracy of real-time polymerase chain reaction (PCR) in the diagnosis of PCP.

METHODSWe searched Web of Knowledge and Medline from 1990 to May 2010 for studies reporting diagnostic accuracy data regarding the use of real-time PCR in the diagnosis of PCP in immunocompromised patients.

RESULTSTen individual studies were included. Overall, the sensitivity of real-time PCR was 97% (95%CI: 93% - 99%); the specificity was 94% (95%CI: 90% - 96%). The area under the HSROC curve (95%CI) for real-time PCR was 0.99 (0.97 - 0.99). In a subgroup analysis regarding studies involving HIV patients among the study population, the sensitivity and specificity were 97% (95%CI: 93% - 99%) and 93% (95%CI: 89% - 96%), respectively. Regarding studies using Bronchoalveolar lavage (BAL) samples only: sensitivity = 98% (95%CI: 94% - 99%); specificity = 93% (95%CI: 89% - 96%), respectively. Regarding studies using microscopy as a reference standard: sensitivity = 97% (95%CI: 92% - 99%); specificity = 93% (95%CI: 88% - 96%). However, high between-study statistical heterogeneity was observed in all analyses.

CONCLUSIONSReal-time PCR has a good diagnostic accuracy and may provide a useful adjunctive tool for the diagnosis of PCP in immunocompromised patients. Further studies are needed in order to identify any differences in the diagnostic performance of real-time PCR in HIV and non-HIV immunocompromised patients.