Presentation and Outcome of Severe Staphylococcal Pneumonia Depends Strongly on Age and the Presence of Panton-Valentine Leucocidin: A Prospective Multicenter Cohort Study

Abstract

Background: The first characterisation of Panton-Valentine Leucocidin (PVL)-associated staphylococcal necrotizing pneumonia (SNP) was published in 2002. A major controversy regarding the role played by PVL in the severity of community-acquired pneumonia (CAP) ensued. To address this issue, all French intensive care units (ICU) were asked to prospectively include their cases of staphylococcal CAP between January 2011 and December 2016. Method: Patients with CAP who fulfilled the following criteria were included: clinical and radiological signs of lung infection, at least one assessment of S. aureus involvement (positive culture from blood/pleural fluid, respiratory sample), and admission to ICU. At the time of data collection, clinicians were not aware of PVL status. Patterns in patient age distribution were detected using a clustering procedure. Risk factors for mortality were identified using multivariate Cox regression. Findings: One-hundred and sixty-three patients, aged from one month to 87 years, were included; 52.1% were PVL-positive. Unexpectedly, PVL-positive cases distributed into two independent clusters, with a cut-off of around 3 years of age, whereas PVL-negative cases were only observed in adolescents and adults. Features of the <3 year-old patients matched with the historical description of staphylococcal pleuropneumonia (SPP) in toddlers with round-shaped infiltrates and mild severity. In older patients, PVL was associated with an enhanced severity and a higher case-fatality rate despite its occurrence in more favourable underlying conditions. Interpretation: This study demonstrates the association of PVL with two distinct facets of staphylococcal CAP with marked differences between toddlers and adolescents/adults regarding clinical presentation and outcome. The strong link between SPP and PVL was found and the association of PVL with severity of SNP in adults is confirmed. This dichotomy between SPP and SNP elucidates the conflicting results of previous studies assessing the role of PVL in staphylococcal pneumonia. Trial Registration: The study is registered with ClinicalTrials.gov, number NCT02798497. Funding Statement: Funded by the French Ministry of Health, PHRC 2010-A01132-37. Declaration of Interests: FV have received a grant from the French Ministry of Research for the submitted work, has received grants from bioMerieux, and has been a consultant for Pfizer and Acccelerate Diagnosis. YG has been a consultant for Pfizer vaccine, GSK Vaccines and Sanofi-Pasteur MSD vaccines. AT has received a grant from bioMerieux and R-biopharm. JPR has received a grant from bioMerieux; and honoraria for lecturing from Pfizer, MSD and Procter&Gamble. CB has received a grant from bioMerieux and Pathoquest. GL has received a grant from Pfizer, Novartis, Basilea, Diadenode and Pierre Fabre, and has been consultant for Eumedica, GenMark and Beckton Dickinson. OD has received a grant from the French Ministry of Research and from Pfizer. CD has received honoraria for lecturing from Correvi. PV has received personnal fees for expertise and consulting from GSK, Astellas, Pfizer and bioMerieux, and received research grant from Sanofi, MSD and ANIOS. The other authors declare no competing interests. Ethics Approval Statement: The regional ethics committee (Comite de Protection des Personnes Sud-Est IV) approved the study (number: A11-39). Written informed consent was obtained from all patients, and patient records were anonymised prior to analysis.