开发和验证一种评分系统以预测住院 SARS-CoV-2 奥密克戎感染者的死亡率：一项全国性多中心研究。

Development and validation of a scoring system to predict the mortality of hospitalized patients with SARS-CoV-2 Omicron: a nationwide, multicentre study.

机构信息

State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, The First Affiliated Hospital, National Medical Center for Infectious Diseases, Zhejiang University School of Medicine, Hangzhou, China.

Department of Infectious Diseases, Hospital Affiliated to Zhejiang Shuren University Shulan International Medical College, Shulan, Hangzhou, China.

出版信息

BMC Pulm Med. 2024 Jul 3;24(1):312. doi: 10.1186/s12890-024-03131-5.

DOI:10.1186/s12890-024-03131-5

PMID:38961438

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC11223413/

Abstract

BACKGROUND

The Omicron variant broke out in China at the end of 2022, causing a considerable number of severe cases and even deaths. The study aimed to identify risk factors for death in patients hospitalized with SARS-CoV-2 Omicron infection and to establish a scoring system for predicting mortality.

METHODS

1817 patients were enrolled at eight hospitals in China from December 2022 to May 2023, including 815 patients in the training group and 1002 patients in the validation group. Forty-six clinical and laboratory features were screened using LASSO regression and multivariable logistic regression.

RESULTS

In the training set, 730 patients were discharged and 85 patients died. In the validation set, 918 patients were discharged and 84 patients died. LASSO regression identified age, levels of interleukin (IL) -6, blood urea nitrogen (BUN), lactate dehydrogenase (LDH), and D-dimer; neutrophil count, neutrophil-to-lymphocyte ratio (NLR) as associated with mortality. Multivariable logistic regression analysis showed that older age, IL-6, BUN, LDH and D-dimer were significant independent risk factors. Based on these variables, a scoring system was developed with a sensitivity of 83.6% and a specificity of 83.5% in the training group, and a sensitivity of 79.8% and a sensitivity of 83.0% in the validation group.

CONCLUSIONS

A scoring system based on age, IL-6, BUN, LDH and D-dime can help clinicians identify patients with poor prognosis early.

摘要

背景

奥密克戎变异株于 2022 年底在中国爆发，导致相当数量的重症病例甚至死亡。本研究旨在确定住院的 SARS-CoV-2 奥密克戎感染患者死亡的危险因素，并建立预测死亡率的评分系统。

方法

2022 年 12 月至 2023 年 5 月，中国 8 家医院共纳入 1817 例患者，其中 815 例患者纳入训练组，1002 例患者纳入验证组。使用 LASSO 回归和多变量逻辑回归筛选 46 项临床和实验室特征。

结果

在训练集中，730 例患者出院，85 例患者死亡。在验证集中，918 例患者出院，84 例患者死亡。LASSO 回归确定了年龄、白细胞介素（IL）-6、血尿素氮（BUN）、乳酸脱氢酶（LDH）和 D-二聚体；中性粒细胞计数、中性粒细胞与淋巴细胞比值（NLR）与死亡率相关。多变量逻辑回归分析表明，年龄较大、IL-6、BUN、LDH 和 D-二聚体是显著的独立危险因素。基于这些变量，在训练组中开发了一个评分系统，其灵敏度为 83.6%，特异性为 83.5%，在验证组中灵敏度为 79.8%，特异性为 83.0%。