Non Invasive Diagnosis of Pneumocystis Pneumonia (DANIPOP)

Performance of Non-targeted and/or Non-invasive Respiratory Samples for the Rapid Diagnosis of Pneumocystis Pneumonia Using the BDMAX TM Molecular Biology Platform (Becton Dickinson)

Incidence and morbi-mortality of Pneumocystis pneumonia (PCP) are increasing. Early and fast diagnosis and treatment improve PCP prognosis. Biological diagnosis is based on the detection of Pneumocystis jirovecii, mainly by PCR, in broncho-alveolar lavage (BAL) obtained from bronchial fibroscopy. However this invasive exam is not always possible in emergency in suspected patient and others non invasive (sputa) and/or non-targeted (bronchial aspiration) are sent to the laboratory (25% of cases, data from the Grenoble University Hospital). Diagnosis performances of these non invasive/non-targeted samples are not clearly established. In this study, the investigators aimed to establish the diagnosis value of non-invasive and/or non-targeted respiratory samples (oral fluids, sputa and bronchial aspiration) for the PCP diagnosis, compared to the gold-standard (Pneumocystis PCR on BAL, beta-D-glucans testing on serum and radio-clinical records).

Sampling of oral fluids, sputa, bronchial aspiration in addition to BAL for the molecular diagnosis of PCP

Sensitivity of Pneumocystis PCR on non-invasive and/or non-targeted respiratory samples compared to the gold-standard

Specificity of Pneumocystis PCR on non-invasive and/or non-targeted respiratory samples compared to the gold-standard

AUCs of Pneumocystis PCR on non-invasive and/or non-targeted respiratory samples compared to the gold-standard

Estimation of the predictive values of Pneumocystis PCR on non-invasive and/or non-targeted respiratory samples compared to the gold-standard

Estimation of the negative predictive values of Pneumocystis PCR on non-invasive and/or non-targeted respiratory samples compared to the gold-standard

Time-saving (in hours) of PCP diagnosis on non-invasive and/or non-targeted respiratory samples compared to the PCP diagnosis on BAL, taking into account the time needed for bronchial fibroscopy

Optimal cut-off values for interpretation of Pneumocystis fungal load on non-invasive and/or non-targeted respiratory samples

Impact of PCP diagnosis on non-invasive and/or non-targeted respiratory samples on the patient management

Estimation of the number of days of presumptive anti-PCP treatment that would have been avoided based on a PCP diagnosis on non-invasive and/or non-targeted respiratory samples

Impact of PCP diagnosis on non-invasive and/or non-targeted respiratory samples on the patient management

Estimation of the number of patients who would have received an earlier appropriate anti-PCP treatment based on a PCP diagnosis on non-invasive and/or non-targeted respiratory samples

Impact of PCP diagnosis on non-invasive and/or non-targeted respiratory samples on the patient management

Inclusion Criteria: Immunocompromised patient with (i) clinical and/or radiological suspicion of PCP, and (ii) bronchial fibroscopy with contributive BAL No immediate life-threatening conditions (estimated life expectancy >12h) No PCP treatment or PCP treatment < 48h Patient hospitalized in the Grenoble Alpes University Hospital with medical insurance Informed and written consent of the patient or its related Exclusion Criteria: Pregnancy, breastfeeding Exclusion period of another clinical trial Deprivation of liberty