Performance and Impact on Antibiotic Prescriptions of a Multiplex PCR in a Real-Life Cohort of Critically Ill Patients with Suspected Ventilated Pneumonia: A Retrospective Monocentric Observational Study

Abstract

Pulmonary multiplex polymerase chain reaction (m-PCR) allows rapid pathogen detection. We aimed to assess its impact on initial antibiotic prescriptions in ventilated patients with suspected pneumonia. Between November 2020 and March 2022,ventilated patients with suspected pneumonia hospitalized in our ICU who benefited from respiratory sampling simultaneously tested using conventional microbiological methods and m-PCR were included. The proportion of appropriate changes in the initial antibiotic therapy following m-PCR results was assessed. We analyzed 104 clinical samples. Of the 47 negative m-PCR results, 16 (34%) led to an appropriate antibiotic strategy: 8 cessationsand 8 lack of initiation. Of the 57 positive m-PCR results, 51 (89%) resulted in an appropriate antibiotic strategy: 33 initiations, 2 optimizations, and 9 de-escalations. In the multivariate analysis, a positive m-PCR was associated with an appropriate antibiotic change (OR: 96.60; IC95% [9.72; 960.20], p < 0.001). A higher SAPS II score was negatively associated with an appropriate antibiotic change (OR: 0.96; IC95% [0.931; 0.997], p = 0.034). In our cohort, a positive m-PCR allowed for early initiation or adjustment of antibiotic therapy in almost 90% of cases. A negative m-PCR spared antibiotic use in onethird of cases. The impact of m-PCR results was reduced in the most severe patients.