key: cord-0716261-fpc6nyby
authors: Fraissé, Megan; Logre, Elsa; Mentec, Hervé; Cally, Radj; Plantefève, Gaëtan; Contou, Damien
title: Eosinophilia in critically ill COVID-19 patients: a French monocenter retrospective study
date: 2020-11-03
journal: Crit Care
DOI: 10.1186/s13054-020-03361-z
sha: a84843a75893f4ab2bd9ce89e189299086c09961
doc_id: 716261
cord_uid: fpc6nyby

nan

As reported in bacterial sepsis [1] , the early phase of SARS-CoV-2 infection seems to be accompanied by eosinopenia [2] [3] [4] . Conversely, our team noticed that several of our critically ill COVID-19 patients developed unexpected and unexplained eosinophilia during their ICU stay. Indeed, as white blood cells count is performed almost daily, monitoring and studying eosinophil course is simple in the ICU setting. To our knowledge, no study has focused on eosinophilia in COVID-19 although eosinophil recovery seven days after initial eosinopenia seems to be associated with a better outcome [3] .

We aimed to assess the incidence and to describe eosinophilia in critically ill COVID-19 patients, as well as to compare the outcome between patients developing or not eosinophilia during their ICU stay.

We retrospectively reviewed all daily white blood cells counts performed in adult COVID-19 patients (RT-PCR positive for SARS-CoV-2) admitted to our 40-bed COVID-19 ICU between March 6 and July 30, 2020. Eosinophilia was defined by an eosinophil count higher than 500/mm 3 and was considered as severe when exceeding 1500/mm 3 . Eosinopenia was defined by an eosinophil count lower than 40/mm 3 .

During the study period, 99 patients were admitted for acute respiratory failure related to SARS-CoV-2 pneumonia. After excluding 21 patients transferred to another ICU, 78 patients remained in the analysis. Among them, 69 (88%) had eosinopenia at ICU admission and 26 (33%, 95% confidence interval 23-45%) developed eosinophilia during ICU stay.

Among the 26 patients who developed eosinophilia (Table 1) , 22 (85%) had eosinopenia at ICU admission. Eosinophilia occurred 19 [13-28] days after ICU admission and lasted 5 [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] days. Median eosinophil count was 900 [678-1350]/mm 3 . Six patients (23%) developed severe eosinophilia. Seven patients (29%) had a biphasic eosinophilia. Ten (38%) patients were treated with a β-lactam antibiotic when eosinophilia occurred. Eleven (42%) patients had at least once a temperature > 38.3 °C and 2 (8%) had an erythematous skin rash during the eosinophilia period, respectively.

Comparison between patients with and without eosinophilia during ICU stay is detailed in Table 2 .

By using a Cox model with time-varying covariate and after adjustment for SAPSII, age and administration of glucocorticoids, eosinophilia was associated with a decreased ICU mortality (HR = 0.44, 95% CI 0.23-0.85, p = 0.014).

Despite an 88% rate of eosinopenia at ICU admission, we report that one-third of our critically ill COVID-19 patients developed an unexpected late-onset and prolonged ICU-acquired eosinophilia which was severe in Open Access *Correspondence: damien.contou@ch-argenteuil.fr Service de Réanimation Polyvalente, Centre Hospitalier Victor Dupouy, 69, rue du Lieutenant-Colonel Prud'hon, 95100 Argenteuil, France almost one quarter of them. Such a high rate of eosinophilia is uncommon in non-COVID-19 critically ill patients and has never been documented in other viral infection such as influenza. A clear explanation for eosinophilia was not retrieved in our patients even if a drug-induced eosinophilia could not be formally ruled out. Some patients with severe eosinophilia were even empirically treated with ivermectin for a hypothetic helminthiasis-related eosinophilia. Given that eosinophilia was a late-onset event in the course of ICU stay, its positive impact on survival is difficult to interpret, patients developing eosinophilia being exposed to a survival bias.

Considering the absence of a clear explanation for the high rate of eosinophilia observed in our patients, we can legitimately hypothesize that SARS-CoV-2 was directly or indirectly responsible for eosinophilia, as a consequence of infection or recovery. The late occurrence of eosinophilia is consistent with the prolonged SARS-CoV-2 RNAaemia reported in critically ill patients [5] . Whether eosinophilia is a marker of an excessive immune recovery or a dysregulated immune response during the cytokine storm [6] induced by the infection is unknown.

Even if our study suffers from several limitations, our findings emphasize the underestimated and understudied role of eosinophils in COVID-19. Further, larger studies are needed to overcome these limitations. 

Eosinopenia is a reliable marker of sepsis on admission to medical intensive care units

Eosinopenia and elevated C-reactive protein facilitate triage of COVID-19 patients in fever clinic: a retrospective case-control study

The prognostic value of eosinophil recovery in COVID-19: a multicentre, retrospective cohort study on patients hospitalised in Spanish hospitals

Clinical features of 85 fatal cases of COVID-19 from Wuhan. A retrospective observational study

SARS-CoV-2 detection in the lower respiratory tract of invasively ventilated ARDS patients

COVID-19: consider cytokine storm syndromes and immunosuppression

We warmly acknowledge Dr Marie Dubert for statistical analysis and Dr Olivier Pajot for careful data interpretation.

Authors' contributions MF and DC are responsible for the conception and design. All the authors were responsible for the data acquisition, analysis and interpretation. DC takes responsibility for the paper as a whole. All authors read and approved the final manuscript.

No funding.

The dataset used and analyzed for the current study is available from the corresponding author on reasonable request.

This study was conducted in accordance with the amended Declaration of Helsinki and was approved by the Institutional Review Board (CE 2020-67) of the French Intensive Care Society.

Not applicable.

The authors declare that they have no competing interests.Received: 19 October 2020 Accepted: 26 October 2020

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.