Risk Factors for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Detection in Blood of Critically Ill Patients

Abstract

TO THE EDITOR—Data on detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral RNA in the blood of patients with coronavirus disease 2019 (COVID-19) are scarce [1–3]. The recently published article by Veyer et al [4] described plasma SARS-CoV-2 viral load in 58 non–critically and critically ill patients. In their multivariate analysis the authors showed that RNAaemia (ie, SARS-CoV-2 circulating in blood) was strongly associated with the clinical class, with higher level RNAaemia among critically ill patients. However, to date, risk factors for detectable SARS-CoV-2 RNAaemia in critically ill patients remain unknown. Therefore, we conducted a similar study using prospectively collected data at the Bichat University Hospital, France, in order to identify risk factors for SARS-CoV-2 detection in blood in critically ill, intubated patients. All included patients had a SARS-CoV-2–positive nasopharyngeal swab before intensive care unit (ICU) admission. During the ICU stay, all patients underwent regular monitoring of SARS-CoV-2 RNAaemia. All blood specimens were sent to the virology laboratory and used for RNA extraction, using the MagnaPure Large Volume Total NA kit (Roche), and amplification by real-time polymerase chain reaction (RT-PCR) techniques using RealStar SARS-CoV-2 RT-PCR RUO assay (Altona). Of note, this RT-PCR assay presents a low limit of detection at 625 copies/mL [5], probably slightly higher than the droplet PCR assay used by Veyer et al. In order to identify risk factors for SARS-CoV-2 detection in blood, we used univariable and multivariable mixed-effects logistic models for clustered data (PROC GLIMMIX; SAS) and we adjusted for the time between symptom onset and date of sampling. This model takes into account the clustering effect of multiple sampling per patient. From March to April 2020, 81 blood samples in 42 patients for SARS-CoV-2 detection were collected; 30 samples (37%) were positive. Thirty-four (81%) patients were male and the median age was 58 (interquartile range [IQR], 46–67) years; 22 (52%) had a cardiovascular comorbidity and 8 (19%) were immunosuppressed. Twenty-two (52%) and 18 (43%) patients received corticosteroids and ritonavir/lopinavir, respectively. The median time to negativity (ie, time between onset of symptoms and viral clearance process from blood) was 17 (IQR, 12–21) days. Using univariable mixed-effects models after adjusting for the time interval between onset of symptoms and date of sampling, we showed that immunosuppression (odds ratio [OR], 12.16; 95% confidence interval [CI], 1.74–84.93; P = .013) and chronic renal failure (OR, 5.98; 95% CI, 1.14–31.35; P = .035) increased the probability of SARS-CoV-2 detection in blood (Table 1). Interestingly, SARS-CoV-2 detection in blood was not associated with 6-week mortality. In the multivariable analysis, immunosuppression significantly increased the probability of SARS-CoV-2 detection in blood (OR, 8.95; 95% CI, 1.17–68.38; P = .035) (Table 1). Risk Factors for SARS-CoV-2 Detection in Blood in Critically Ill Patients Values were adjusted for the time interval between onset of symptoms and date of sampling. A sensitivity analysis forcing the variables “SOFA” and “chronic respiratory failure” in the multivariable analysis showed similar results (OR for immunosuppression, 8.35; 95% CI, .89–78.29; P = .063). Abbreviations: CI, confidence interval; ICU, intensive care unit; OR, odds ratio; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; SOFA, Sequential Organ Failure Assessment score. Veyer et al [4] showed that SARS-CoV-2 RNAaemia was strongly correlated with disease severity. With our data and using classical RT-PCR, we observed that especially immunosuppressed, critically ill patients tended to be viremic with SARS-CoV-2. In contrast to Veyer et al, we did not observe any association with mortality. Further larger multicenter cohorts are urgently needed to investigate risk factors for RNAemia using classical and ultrasensitive RT-PCR methods in patients with severe and nonsevere COVID-19. Financial support. N. B. currently receives a postdoc Mobility grant from the Swiss National Science Foundation (grant number P400PM_183865) and a grant from the Bangerter-Rhyner Foundation. Potential conflicts of interest. J.-F. T. received fees for lectures to Gilead, 3M, MSD, Pfizer, and BioMérieux; research grants from Astellas, 3M, MSD, and Pfizer; and participated on advisory boards of 3M, MSD, BioMérieux, Paratek, Medimmune, Gilead, Bayer Pharma, Nabriva, and Pfizer. B. V. reports grants and personal fees for lectures and travel accommodations from Qiagen; personal fees for lecture and travel accommodations from BioMérieux; personal fees for lectures from Hologic; grants, personal fees, and nonfinancial support from Qiagen; and personal fees from Gilead. All other authors report no potential conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

Topics & Concepts