空腹血糖与TG/HDL-C 对非糖尿病患者并发COVID-19 的预测价值研究

目的　探究空腹血糖与三酰甘油/ 高密度脂蛋白胆固醇(triglyceride/high density lipoprotein cholesterol, TG/HDL-C) 比值在对糖尿病患者并发新型冠状病毒肺炎(coronavirus disease 2019, COVID-19) 的预测价值。方法　连续收集2021 年12 月～ 2022 年1 月西安市第四医院收治的非糖尿病并发COVID-19 患者39 例，通过倾向性评分匹配同期西安交通大学第二附属医院健康体检者34 例( 健康对照组)，比较两组间临床特征与实验室检验指标，Logistic 回归及受试者工作特征(receiver operating characteristic, ROC) 曲线分析空腹血糖与TG/HDL-C 对非糖尿病患者并发COVID-19 的预测价值。结果　COVID-19 患者均为轻型(30 例) 或普通型(9 例)，整体症状较轻，中位年龄29.0 (20.0, 49.0) 岁，男性24 例(61.5%)，临床预后较好。COVID-19 患者入院时空腹血糖(4.30±0.47 mmol/L) 与HDL-C1.07(0.86, 1.30) mmol/L]较健康对照组5.15±0.70 mmol/L, 2.24(1.77, 3.05) mmol/L] 显著降低，差异具有统计学意义(t=6.277, P ＜ 0.001；Z =-6.026,P＜0.001)，而低密度脂蛋白胆固醇(low density lipoprotein cholesterol, LDL-C) 2.40(1.81, 2.91) mmol/L] 与TG/HDL-C 0.91(0.54, 1.52)] 较健康对照组1.11 (0.99, 1.30) mmol/L, 0.54 (0.33, 0.90)] 显著升高，差异具有统计学意义(Z=-6.271, -2.801,均P ＜ 0.005)。Logistic 回归分析显示入院时空腹血糖可能是非糖尿病患者感染COVID-19 的独立保护因素(OR :0.020,95% CI: 0.003 ～ 0.150)，TG/HDL-C 升高可能是COVID-19 发病的独立危险因素(OR :4.802, 95% CI: 1.249 ～ 18.460)。ROC 曲线结果显示，空腹血糖与TG/HDL-C 能较好地预测COVID-19 发病风险，曲线下面积(area under curve, AUC) 分别为0.871 和0.708，二者联合诊断的AUC 为0.895。结论　空腹血糖降低与TG/HDL-C 升高可能是非糖尿病人群并发COVID-19 的危险因素，对COVID-19 发病具有较好的预测价值。

Predictive Value of Fasting Blood Glucose and TG/HDL-C Ratio in Non-Diabetic Patients with COVID-19

Objective　To investigate the predictive values of fasting blood glucose and triglyceride/high-density lipoprotein cholesterol ratio (TG/HDL-C) in non-diabetic patients with COVID-19. Methods　A total of 39 non-diabetic patients with COVID-19 admitted to the Fourth Hospital of Xi’an from December 2021 to January 2022 were included. And 34 health examination subjects from the Second Affiliated Hospital of Xi’an Jiaotong University were matched as health control according to their propensity score. The clinical characteristics and laboratory test results between groups were compared, and the predictive value of fasting glucose and TG/HDL-C in non-diabetic COVID-19 patients was analyzed by logistic regression and receiver operating curve (ROC). Results　COVID-19 patients were either mild (30 cases) or common type (9 cases) with mild symptoms and good clinical prognosis. The median age was 29.0 (20.0, 49.0) years, 24 (61.5%) were males. Fasting blood glucose (4.30±0.47 mmol/L) and HDL-C 1.07 (0.86, 1.30) mmol/L] levels in COVID-19 patients were significantly lower than healthy controls 5.15±0.70 mmol/L, 2.24 (1.77, 3.05) mmol/L], the differences were statistically significant (t=6.277, P ＜ 0.001; Z=6.026, P ＜ 0.001). However, low density lipoprotein cholesterol (LDL-C) 2.40 (1.81, 2.91) mmol/L] and TG/HDL-C 0.91 (0.54, 1.52)] in COVID-19 patients were significantly increased compared to healthy controls 1.11 (0.99, 1.30) mmol/L, 0.54 (0.33, 0.90)], and the differences were statistically significant (Z=-6.271, -2.801, all P ＜ 0.005). Logistic regression analysis showed that fasting blood glucose on admission could be an independent protective factor (OR:0.020, 95% CI: 0.003 ～ 0.150) and elevated TG/HDL-C be a risk factor (OR:4.802, 95% CI: 1.249 ～ 18.460) for COVID-19 infection among non-diabetic populations. The ROC curve showed that fasting blood glucose and TG/HDL-C were good at predicting the risk of COVID-19, and the area under the curve (AUC) were 0.871 and 0.708, respectively, and was 0.895 when combined. Conclusion　Decreased fasting blood glucose and elevated TG/HDL-C would be risk factors for COVID-19 infection in the non-diabetic population, and both have good predictive value for the incidence of COVID-19.