糖化血红蛋白及随机血糖对急性大血管闭塞性缺血性卒中患者血管内治疗预后的影响

Effects of glycated hemoglobin and random blood glucose on the prognosis of patients with acute large-vessel occlusive ischemic stroke treated with endovascular therapy

目的探究糖化血红蛋白(HbA1c)及随机血糖与急性大血管闭塞性缺血性卒中患者血管内治疗临床预后的关系。方法回顾性连续纳入2019年1月至2020年12月于首都医科大学宣武医院卒中中心接受血管内治疗的急性大血管闭塞卒中患者,收集患者基线资料(年龄和性别)、血管危险因素(包括吸烟、饮酒、高血压病、高脂血症、糖尿病、心房颤动、既往卒中和冠心病)、既往降糖药物使用、入院时血压和随机血糖、术后在院期间空腹HbA1c、入院时美国国立卫生研究院卒中量表(NIHSS)评分、闭塞血管部位、脑梗死分型、血管内治疗后再灌注情况,以及术前实验室检查包括血红蛋白、纤维蛋白原、中性粒细胞与淋巴细胞比值(NLR)。根据入院随机血糖7.8 mmol/L和术后在院期间空腹HbA1c 6.5%,将患者分为HbA1c及随机血糖正常组(HbA1c<6.5%且随机血糖<7.8 mmol/L)、HbA1c或随机血糖增高组(HbA1c≥6.5%、随机血糖<7.8 mmol/L或HbA1c<6.5%、随机血糖≥7.8 mmol/L)、HbA1c及随机血糖均高组(HbA1c≥6.5%且随机血糖≥7.8 mmol/L)3组,对3组患者的基本资料和治疗情况进行比较。根据术后90 d改良Rankin量表(mRS)评分将患者分为预后不良和预后良好组,mRS评分0~2分为预后良好,3~6分为预后不良,6分为死亡,并进行单因素分析和二元多因素Logistic回归分析,分析随机血糖和HbA1c对术后90 d神经功能预后的影响。对随机血糖和HbA1c联合指标进行受试者工作特征(ROC)曲线分析,得出其对术后90 d不良预后的预测效能。结果最终纳入246例患者,其中HbA1c及随机血糖正常组77例,HbA1c或随机血糖增高组97例,HbA1c及随机血糖均高组72例。3组患者年龄、高血压病、糖尿病、既往降糖药物使用、入院随机血糖、术后在院空腹HbA1c、NLR、纤维蛋白原水平差异均有统计学意义(均P<0.05)。组间两两比较结果显示,HbA1c及随机血糖均高组患者相较于其他两组高血压病、糖尿病及既往降糖药物使用患者比例均高(均P<0.05),入院随机血糖、术后在院空腹HbA1c及纤维蛋白原水平偏高(均P<0.05);HbA1c及随机血糖均高组患者相较于HbA1c及随机血糖正常组年龄偏大(P=0.018),HbA1c或随机血糖增高组相较于HbA1c及随机血糖正常组入院随机血糖及NLR偏高(均P<0.05)。术后90 d预后良好者102例(41.46%),预后不良者144例(58.54%),死亡51例(20.73%)。与预后良好组患者相比,预后不良组患者年龄偏大,既往患糖尿病、冠心病比例较高,入院NIHSS评分、入院收缩压、纤维蛋白原水平偏高,改良脑梗死溶栓(mTICI)分级2b~3级比例较低,症状性颅内出血比例较高,HbA1c及随机血糖正常患者比例较低,HbA1c及随机血糖均高患者比例较高(均P<0.05)。矫正潜在协变量后,与HbA1c及随机血糖正常组相比,HbA1c及随机血糖均高组术后90 d预后不良风险增加(OR=2.532,95%CI:1.148~5.586,P=0.021)。HbA1c联合随机血糖预测术后90 d预后不良的曲线下面积为0.649(95%CI:0.580~0.719,P<0.01)。结论糖化血红蛋白与随机血糖同时增高与急性大血管闭塞性缺血性卒中患者接受血管内治疗后的90 d预后不良相关。

Objective To explore the relationship between glycated hemoglobin(HbA1c),random blood glucose and clinical prognosis of patients with acute large-vessel occlusive ischemic stroke treated with endovascular therapy.Methods Acute large-vessel occlusive ischemic stroke patients who received endovascular therapy in Stroke Center of Xuanwu Hospital,Capital Medical University from January 2019 to December 2020 were enrolled retrospectively.Baseline data(age and gender),vascular risk factors(including smoking,drinking,hypertension,hyperlipidemia,diabetes,atrial fibrillation,previous stroke and coronary heart disease),hypoglycemic medication histories,blood pressure and random blood glucose on admission,fasting HbA1c during hospitalization after operation,the National Institutes of Health Stroke Scale(NIHSS)score on admission,site of occlusive vessels,classification of cerebral infarction,reperfusion after endovascular therapy and preoperative laboratory examination including hemoglobin,fibrinogen,neutrophil/lymphocyte ratio(NLR)were collected.According to the random blood glucose on admission 7.8 mmol/L and fasting HbA1c during hospitalization after operation 6.5%,the patients were divided into three groups:normal HbA1c and normal random blood glucose(HbA1c<6.5%and random blood glucose<7.8 mmol/L),high HbA1c or high random blood glucose(HbA1c≥6.5%and random blood glucose<7.8 mmol/L or HbA1c<6.5%and random blood glucose≥7.8 mmol/L),and high HbA1c and high random blood glucose(HbA1c≥6.5%and random blood glucose≥7.8 mmol/L).The basic data and treatment of the three groups were compared.According to the modified Rankin scale(mRS)score at 90 d after surgery,the patients were divided into two groups:poor prognosis group and good prognosis group.The mRS score 0-2 was classified as good prognosis,3-6 as poor prognosis and 6 as death.Univariate analysis and binary multivariate Logistic regression analysis were performed to analyze the effects of random blood glucose and HbA1c on the prognosis of neurological function at 90 d after surgery.The sensitivity and specificity of the combined indexes of random blood glucose and HbA1c in predicting poor prognosis at 90 d after surgery were obtained by receiver operating characteristic(ROC)curve analysis.Results All 246 patients were enrolled finally,including normal HbA1c and normal random blood glucose group(n=77),high HbA1c or high random blood glucose group(n=97),high HbA1c and high random blood glucose group(n=72).There were significant differences in age,hypertension,diabetes,hypoglycemic medication histories,random blood glucose on admission,fasting HbA1c during hospitalization after operation,NLR and fibrinogen among the three groups(all P<0.05).The results of pairwise comparison between two groups showed that the proportion of patients with hypertension,diabetes and hypoglycemic medication histories in the high HbA1c and high random blood glucose group was higher than that in the other two groups(all P<0.05),and the random blood glucose on admission,fasting HbA1c during hospitalization after operation and fibrinogen levels in the high HbA1c and high random blood glucose group was higher than that in the other two groups(all P<0.05).The patients in high HbA1c and high random blood glucose group were older than those in normal HbA1c and normal random blood glucose group(P=0.018),and the random blood glucose on admission and NLR in high HbA1c or high random blood glucose group were higher than those in normal HbA1c and normal random blood glucose group(all P<0.05).At 90 d after surgery,102 cases(41.46%)had a good prognosis and 144 cases(58.54%)had a poor prognosis,of which 51 cases(20.73%)died.Compared with the patients with good prognosis,the patients with poor prognosis were older and had a higher proportion of diabetes and coronary heart disease,higher NIHSS score on admission,higher systolic blood pressure and fibrinogen level,lower proportion of modified thrombolysis in cerebral infarction(mTICI)2b-3 grade,higher proportion of symptomatic intracranial hemorrhage(sICH),lower proportion of patients with normal HbA1c and normal random blood glucose and higher proportion of patients with high HbA1c and high random blood glucose(all P<0.05).After adjusting the potential covariates,compared with the normal HbA1c and normal random blood glucose group,the high HbA1c and high random blood glucose group had an increased risk of poor prognosis(OR,452.532,95%CI 1.148-5.586,P=0.021)at 90 d.The area under the curve of HbA1c combined with random blood glucose to predict poor prognosis at 90 d after surgery was 0.649(95%CI 0.580-0.719,P<0.01).Conclusion The simultaneous increase of HbA1c and random blood glucose is associated with poor prognosis at 90 d after endovascular therapy in patients with acute large-vessel occlusive stroke.