Department of Respiratory Medicine, Japanese Red Cross Medical Center, 4-1-22 Hiroo, Shibuya-ku, Tokyo 150-8953, Japan.
Department of Infectious Diseases, Japanese Red Cross Medical Center, 4-1-22 Hiroo, Shibuya-ku, Tokyo 150-8953, Japan.
Respir Investig. 2021 Nov;59(6):804-809. doi: 10.1016/j.resinv.2021.08.001. Epub 2021 Sep 11.
This study aimed to examine risk factors associated with critical coronavirus disease 19 (COVID-19) and to establish a risk predictive model for Japanese patients.
We retrospectively assessed adult Japanese patients diagnosed with COVID-19 at the Japanese Red Cross Medical Center, Tokyo, Japan between February 1, 2020 and March 10, 2021. The patients were divided into critical and non-critical groups based on their condition during the clinical courses. Univariate and multivariate logistic regression analyses were performed to investigate the relationship between clinical characteristics and critical illness. Based on the results, we established a predictive model for the development of critical COVID-19.
In total, 300 patients were enrolled in this study. Among them, 86 were included in the critical group. Analyses revealed that age ≥65 y, hemodialysis, need for O supplementation upon diagnosis, and an initial serum C-reactive protein level of ≥6.5 mg/dL were independently associated with the development of critical COVID-19. Next, a predictive model for the development of critical COVID-19 was created, and this included the following variables: age ≥65 y, male sex, diabetes, hemodialysis, need for O supplementation upon diagnosis, and an initial serum C-reactive protein level of ≥6.5 mg/dL. The area under the receiver operating characteristic curve of the model was 0.86 (95% confidence interval, 0.81-0.90). Using a cutoff score of 12, the positive and negative predictive values of 74.0% and 80.4% were obtained, respectively.
Upon diagnosis, the predictive model can be used to identify adult Japanese patients with COVID-19 who will require intensive treatment.
本研究旨在探讨与危重型 2019 冠状病毒病（COVID-19）相关的危险因素，并为日本患者建立风险预测模型。
我们回顾性评估了 2020 年 2 月 1 日至 2021 年 3 月 10 日期间在日本红十字会医疗中心东京院区被诊断为 COVID-19 的成年日本患者。根据临床病程将患者分为危重型和非危重型。进行单因素和多因素逻辑回归分析，以研究临床特征与重症疾病之间的关系。根据分析结果，我们建立了预测 COVID-19 患者发展为危重型的模型。
本研究共纳入 300 例患者，其中 86 例纳入危重型组。分析表明，年龄≥65 岁、血液透析、诊断时需要补充氧气以及初始血清 C 反应蛋白水平≥6.5mg/dL 与危重型 COVID-19 的发生独立相关。随后，建立了预测 COVID-19 患者发展为危重型的模型，包括年龄≥65 岁、男性、糖尿病、血液透析、诊断时需要补充氧气以及初始血清 C 反应蛋白水平≥6.5mg/dL。该模型的受试者工作特征曲线下面积为 0.86（95%置信区间，0.81-0.90）。使用截断值为 12，模型的阳性预测值和阴性预测值分别为 74.0%和 80.4%。
该预测模型可用于识别需要强化治疗的成年日本 COVID-19 患者。
