357 Leukocytes, Platelets, and Positive SARS-CoV-2 Results in Admitted Emergency Department Patients

Abstract

Since the first few cases of Coronavirus disease 2019 (COVID-19) were reported in December 2019, the pandemic has affected over four million patients worldwide with over 250,000 deaths as of early May, 2020. As testing for COVID-19 is a limited resource, increasing its efficacy will play a critical role in managing this pandemic. Early reports have indicated an association between blood cell count and COVID-19. The objective of this study is to determine whether or not white blood cell and platelet count at presentation can guide clinical decisionmaking regarding management of persons under investigation (PUIs) Retrospective electronic medical record review (EMR) over one month from 3/15/2020 to 4/15/2020. Data was collected from three community emergency departments (ED), all sites of a single emergency medicine residency program. Included were all ED patients tested in the ED for SARS_CoV-2 and admitted to the hospital. Excluded were patients less than 18, cardiac arrests, and missing data. Abstracted data include demographics, admission diagnosis, ED vital signs, white blood cell (WBC) count, platelet (PLT) count, and result of COVID-19 testing. With power set at 0.80 and significance set at 0.05, a sample size of 340 patients would detect at least 15% differences among the variables related to positive SARS_CoV-2. ROC analysis was used to define cut off points for prediction of Covid result based on WBC and Platelet counts. Sensitivity, specificity, positive and negative likelihood ratio, all with 95% CI are reported. 484 cases met inclusion/exclusion criteria. The mean age was 67.9 years (SD: 16.6) with 39.9% females. SAR-CoV-2 virus was detected in 245 patients (50.6%). COVID-19 positive patients had significantly lower WBC (7.2 versus 11.2; p<.001) and platelet counts (211 versus 239; p<.001). ROC analysis of both WBC and PLT were significant with area under the curve of 75% and 64.5% (p<001). At a white count of 7.0 or less, sensitivity was .624, specificity was .766, positive likelihood ratio 2.65 (95% CI: 2.06 to 3.40), and negative likelihood ratio 0.50 (95% CI: 0.42 to 0.59). At a platelet count of 200 or less, sensitivity was .543, specificity was .716, positive likelihood ratio 1.91 (95% CI: 1.51 to 2.41), and negative likelihood ratio 0.64 (95% CI: 0.55 to 0.75). When combined, lower WBC plus lower platelets gave a sensitivity of .437, specificity of .849 with positive likelihood of 2.90 (95% CI: 2.08 to 4.04) and negative likelihood of .66 (95% CI: .59 to .75). In this multi-center community study of ED patients admitted over one month with suspected SARS-CoV-2 infection, both the initial white cell count and platelet counts were significantly lower in SARS-CoV-2 positive patients. Patients with both initial WBC less than 7.0 and Platelets less than 200 had increased odds of positive SARS-CoV-2 by a factor of three.