Rxivist logo

Validation of reported risk factors for disease classification and prognosis in COVID-19: a descriptive and retrospective study

By Li Tan, Xia Kang, Xinran Ji, Qi Wang, Yongsheng Li, Qiongshu Wang, Hongming Miao

Posted 07 Apr 2020
medRxiv DOI: 10.1101/2020.04.05.20053769

Risk indicators viral load (ORF1ab Ct), lymphocyte percentage (LYM%), C-reactive protein (CRP), interleukin-6 (IL-6), procalcitonin (PCT) and lactic acid (LA) in COVID-19 patients have been proposed in recent studies. However, the predictive effects of those indicators on disease classification and prognosis remains largely unknown. We dynamically measured those reported indicators in 132 cases of COVID-19 patients including the moderate-cured (moderated and cured), severe-cured (severe and cured) and critically ill (died). Our data showed that CRP, PCT, IL-6, LYM%, lactic acid and viral load could predict prognosis and guide classification of COVID-19 patients in different degrees. CRP, IL-6 and LYM% were more effective than other three factors in predicting prognosis. For disease classification, CRP and LYM% were sensitive in identifying the types between critically ill and severe (or moderate). Notably, among the investigated factors, LYM% was the only one that could distinguish between the severe and moderate types. Collectively, we concluded that LYM% was the most sensitive and reliable predictor for disease typing and prognosis. During the COVID-19 pandemic, the precise classification and prognosis prediction are critical for saving the insufficient medical resources, stratified treatment and improving the survival rate of critically ill patients. We recommend that LYM% be used independently or in combination with other indicators in the management of COVID-19.

Download data

  • Downloaded 790 times
  • Download rankings, all-time:
    • Site-wide: 32,176
    • In infectious diseases: 2,327
  • Year to date:
    • Site-wide: 79,492
  • Since beginning of last month:
    • Site-wide: 64,765

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide