Frontiers in Microbiology (Dec 2022)

A practical scoring model to predict the occurrence of critical illness in hospitalized patients with SARS-CoV-2 omicron infection

  • Yao Zhang,
  • Jiajia Han,
  • Feng Sun,
  • Yue Guo,
  • Yifei Guo,
  • Haoxiang Zhu,
  • Feng Long,
  • Zhijie Xia,
  • Shanlin Mao,
  • Hui Zhao,
  • Zi Ge,
  • Jie Yu,
  • Yongmei Zhang,
  • Lunxiu Qin,
  • Ke Ma,
  • Richeng Mao,
  • Jiming Zhang,
  • Jiming Zhang,
  • Jiming Zhang

DOI
https://doi.org/10.3389/fmicb.2022.1031231
Journal volume & issue
Vol. 13

Abstract

Read online

BackgroundThe variants of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have emerged repeatedly, especially the Omicron strain which is extremely infectious, so early identification of patients who may develop critical illness will aid in delivering proper treatment and optimizing use of resources. We aimed to develop and validate a practical scoring model at hospital admission for predicting which patients with Omicron infection will develop critical illness.MethodsA total of 2,459 patients with Omicron infection were enrolled in this retrospective study. Univariate and multivariate logistic regression analysis were performed to evaluate predictors associated with critical illness. Moreover, the area under the receiver operating characteristic curve (AUROC), continuous net reclassification improvement, and integrated discrimination index were assessed.ResultsThe derivation cohort included 1721 patients and the validation cohort included 738 patients. A total of 98 patients developed critical illness. Thirteen variables were independent predictive factors and were included in the risk score: age > 65, C-reactive protein > 10 mg/L, lactate dehydrogenase > 250 U/L, lymphocyte < 0.8*10^9/L, white blood cell > 10*10^9/L, Oxygen saturation < 90%, malignancy, chronic kidney disease, chronic cardiac disease, chronic obstructive pulmonary disease, diabetes, cerebrovascular disease, and non-vaccination. AUROC in the derivation cohort and validation cohort were 0.926 (95% CI, 0.903–0.948) and 0.907 (95% CI, 0.860-0.955), respectively. Moreover, the critical illness risk scoring model had the highest AUROC compared with CURB-65, sequential organ failure assessment (SOFA) and 4C mortality scores, and always obtained more net benefit.ConclusionThe risk scoring model based on the characteristics of patients at the time of admission to the hospital may help medical practitioners to identify critically ill patients and take prompt measures.

Keywords