Epidemiology of post-influenza bacterial pneumonia due to Panton–Valentine leucocidin positive Staphylococcus aureus in intensive care units: a retrospective nationwide study

Abstract

Dear Editor, During influenza, Staphylococcus aureus (SA) superinfection is one of the major causes of death [1]. Very little is known regarding SA producing Panton–Valentine leucocidin (SAPVL) post-influenza pneumonia and specifically its epidemiology [2–5]. The main goal of this study was to assess, in a retrospective French multicentre study, the in-intensive care unit (ICU) mortality and prevalence of SAPVL postinfluenza pneumonia. This retrospective observational multicentre French study was conducted in 25 ICUs. Adult patients were included if they were admitted to the ICU between January 2009 and December 2017 for influenza pneumonia with a proven superinfection with SAPVL. The complete methodology has been provided in the online supplemental material. Among the 2053 patients identified with influenza pneumonia, 22 [1.1%, CI 95% (0.6–1.5)] presented an influenza superinfected by SAPVL. The included population was 44 years old [36–57]. At admission to ICU, diagnosis of influenza was systematically achieved in the first 24 h. Methicillin-susceptible SA was found in 17 patients (77.3%), while methicillin-resistant SA (MRSA) was found in 5 patients (22.7%). Duration between the onset of symptoms and admission to hospital was 3 days [3–4]. All patients were transferred to ICU within 24 h following hospitalisation. Eighteen patients (81.8%) had a diagnosis of acute respiratory distress syndrome with a median lowest PaO2/ FiO2 of 100 [61–121] mmHg. Among the latter, 16 patients (73%) required a rescue extracorporeal membrane oxygenation (ECMO) support. Severity at admission and complications are described in Tables 1, S1 and S2. The prevalence of influenza superinfected by SAPVL was 1.1% with an associated all-cause in-ICU mortality of 54.5%, while the global in-ICU mortality of patients admitted for severe influenza without SAPVL was 20.5%. Despite a high degree of failure at admission, no major prior comorbidities or risk factors was found for complicated influenza or MRSA infections. The relatively low rate of prone positioning could be in part explained by the clinical presentation of the studied patients with high SOFA scores. Thus, the intensivist might have been reluctant to turn such severe patients to a prone position. Furthermore, 6/22 patients were included before the PROSEVA study which had been published in 2013 [6]. This study presents several limitations due to its retrospective design. The estimated prevalence is probably underestimated, due to the non-systematic screening of Panton–Valentine leucocidin production in all patients superinfected with SA. Over the ten past *Correspondence: b.levy@chru‐nancy.fr 1 Medical Intensive Care Unit, University Hospital of Nancy, Brabois, Rue du Morvan, 54500 Vandœuvre‐Lès‐Nancy, France Full author information is available at the end of the article