ABSTRACT
Introduction: COVID-19 is characterized by a maladaptive cytokine release, requiring hospitalization in intensive care unit (ICU) for the more severe forms, or in conventional wards for less aggressive forms. However, in a recent meta-analysis, the prevalence of complications during hospitalization in patients with COVID-19 was as high as 36%, requiring in some cases transfer into ICU, particularly in patients with comorbidities such as diabetes, hypertension or obesity. Cell Population Data (@CPD), provided as part of CBC-Diff analysis by the DxH 800 analyzers (Beckman Coulter) were previously shown to be affected by SARS-CoV-2 infection. Thus, the aim of this study was to analyze whether it was possible, from conventional biological tests collected on admission, and comorbidities, to identify patients with a moderate initial form but at high risk of worsening, who will require subsequent ICU admission. Method(s): The study included 357 consecutive hospitalised COVID-19 patients, from which 40 patients required ICU after staying in the ward. CBC-Diff results including @ CPD from DxH 800, C-reactive protein (CRP) and clinical information (hypertension, diabetes, Body Mass Index (BMI)) were used for the statistical analysis with MedCalc software, version 20.106 (MedCalc Software Ltd, Ostend, Belgium). Result(s): Using logistic regression we obtained ICU factor, which allowed to identify those patients who will require transfer to ICU. ICU factor, based on several @CPD, platelet count, BMI and diabetes, demonstrated AUC 0.843, sensitivity 40% and specificity 97.5% at cut-off 0.39. This approach would allow us correct identification of 16 out of 40 (40%) patients who require ICU hospitalization, with only 8 (2.5%) incorrectly classified patients from hospitalization only group. Adding CRP did not improve the performance of the model. Conclusion(s): ICU factor can constitute an easy and inexpensive tool for screening patients with a moderate form of COVID-19, but at high risk of worsening. This also makes it possible to anticipate the management of intensive care beds, a limiting factor when patients arrive in large numbers at the hospital in the event of a new aggressive variant.
ABSTRACT
Background-aim: Coronavirus disease (COVID-19) caused by SARS-CoV-2 is characterized by high contagiousness requiring isolation measures. Currently, diagnosis is based on the RT-PCR and/or chest computed tomography (CT) scan, but these methodologies are time-consuming and may delay the diagnosis. CBC-Diff analysis is the first step in patient assessment and may contribute to the diagnosis of COVID-19. Morphological changes of the immune cells can be identified by electro-optical analysis on the hematology analyzer DxH 800 (Beckman Coulter, Inc., Brea, CA). We studied whether the analysis of cellular population data (CPD), provided as part of CBC-Diff analysis by the DxH 800 analyzers can help to identify SARS-CoV-2 infection. Methods: The study included 322 consecutive patients from the emergency unit with positive RT-PCR (Allplex 2019 nCoV Assay, Eurobio, Les Ulis, France) and 285 consecutive patients with clinical suspicion for COVID-19, but who had negative RT-PCR and CT-scan not suggestive of SARS-CoV-2 infection. We also included 137 subjects with a normal CBC-Diff, referred to our institution without evidence of infection and when prevalence of SARS-CoV-2 was very low in France. Blood was collected in EDTA-K3 tubes and analyzed within 6 h after collection. Results: The majority of CPD was different between the 3 groups;CPD of patients (with or without COVID-19) were significantly different of CPD of controls. Four, six and nine CPD for NE, LY, and MO, respectively, were significantly different between COVID-19+ and COVID-19- patients. Using ROC analysis, we identified parameters, which were able to discriminate COVID-19+ patients from COVID-19- patients. The best parameter was SD-V-Mo (Standard Deviation of Monocyte Volume), with AUC 0.819, sensitivity of 91.59% and specificity of 63.03% at the cut-off>21.71. MN-V-Mo (Mean Monocyte Volume) demonstrated AUC 0.742 with sensitivity of 76.64%, specificity of 65.85% at cut-off>180. SD-AL2-MO (Standard Deviation of Axial Light Loss for Monocytes) provided AUC 0.722 with sensitivity of 85.67%, specificity of 52.11% at cut-off>17.51. Currently CPD are research use only;their clinical utility has not been established. Conclusions: Consideration of CPD could constitute a first step and potentially aid in the early diagnosis of COVID-19.