Pneumocystosis and quantitative PCR

Abstract

ObjectivePneumocystis pneumonia (PCP) is now predominantly observed in immunosuppressed non-HIV-infected patients. The sensitivity of the PCR is here higher than direct examination (DE) of respiratory secretions because the infection is caused by a lower inoculum of Pneumocystis jirovecii (P. jirovecii). The objective of our retrospective study was to assess the contribution of quantitative PCR (qPCR) in the diagnosis of PCP. Patients and methodsAll patients hospitalized for PCP suspicion with a positive qPCR were included. Irrespective of the qPCR value, patients were initially classified into two groups (infection and colonization [PCP ruled out]) based on clinical, radiological, and microbiological data. Both groups were then compared based on the qPCR value. ResultsBetween 2013 and 2016, 150 patients were included; 75% of them were not infected with HIV. The diagnosis of PCP was retained for 129 patients and rejected for 21 patients. The DE was negative in 60% of PCP cases. The median value of qPCR was 76,650copies/mL among infected patients and 3220copies/mL among colonized patients. The threshold corresponding to a specificity of 100% was 56,000copies/mL. The optimal value to distinguish an infection from a colonization was 10,100copies/mL. ConclusionOur study confirms the diagnostic value of the qPCR in immunosuppressed patients, especially when the DE is negative. When the qPCR is˂56,000copies/mL, the result should be interpreted based on the clinical context and paraclinical examinations.