Comparative study of five coma assessment scales in prognosis prediction of patients with severe stroke
10.3969/j.issn.1672-5921.2025.01.003
- VernacularTitle:五种昏迷评估量表对重症卒中患者预后预测效能的对比研究
- Author:
Dongyang HU
1
;
Xiaochen HAN
;
Sheng YAO
;
Jianguo LIU
;
Hairong QIAN
;
Jiatang ZHANG
Author Information
1. 100853 北京,解放军医学院
- Publication Type:Journal Article
- Keywords:
Stroke;
Forecasting;
Prognosis;
Severe stroke;
Coma scale
- From:
Chinese Journal of Cerebrovascular Diseases
2025;22(1):15-22,37
- CountryChina
- Language:Chinese
-
Abstract:
Objective To compare the predictive effectiveness of the Glasgow coma scale(GCS),GCS-pupils scale(GCS-P),Glasgow-Pittsburgh coma scale(GPCS),full outline of unresponsiveness scale(FOUR),and coma recovery scale-revised(CRS-R)in forecasting the prognosis of severe stroke patients.Methods A prospective,consecutive cohort of severe stroke patients admitted to the Department of Neurology,First Medical Center of Chinese PLA General Hospital from September 2021 to April 2024 was enrolled.Demographic and clinical data were collected,including age,sex,length of hospital stay,diagnosis(severe ischemic stroke,severe cerebral hemorrhage,aneurysmal subarachnoid hemorrhage),medical history(hypertension,diabetes,coronary artery disease),smoking and drinking habits,vital signs upon admission(temperature,pulse,respiration,blood pressure),neurological examination findings(including speech and brainstem reflexes)at admission,head imaging results(CT,MRI)within 24 h of admission to assess the presence of brain herniation,and whether intubation occurred within 24 h of admission.Patients underwent GCS,GCS-P,GPCS,FOUR,and CRS-R scoring within 8h of admission.Telephone follow-up was conducted at 6 months post-stroke to assess outcomes using the modified Rankin scale(mRS),with mRS scores of 0-2 classified as the good prognosis group and 3-6 as the poor prognosis group.The receiver operating characteristic(ROC)curve was used to assess the prognostic prediction value of the five scales for poor outcomes at 6 months.The area under the ROC curve(AUC)was calculated,and pairwise comparisons of AUC were performed using the Delong test.Results A total of 179 severe stroke patients were enrolled,including 116 males and 63 females.The group consisted of 132 patients with severe ischemic stroke,30 with severe intracerebral hemorrhage,and 17 with aneurysmal subarachnoid hemorrhage.At 6months,126patients had a poor prognosis and 53 had a good prognosis.(1)There were statistically significant differences in age,temperature,pulse,history of coronary artery disease,smoking and drinking habits,presence of speech impairment,abnormal brainstem reflexes,brain herniation,intubation within 24 h of admission,and GCS,GCS-P,GPCS,FOUR,and CRS-R scores between the poor and good prognosis groups(all P＜0.05).(2)ROC analysis revealed that the AUC(95％CI)for predicting poor outcomes at 6 months in severe stroke patients for GCS,GCS-P,GPCS,FOUR,and CRS-R were 0.808(0.742-0.863),0.815(0.750-0.869),0.828(0.765-0.880),0.841(0.780-0.892),and 0.831(0.768-0.883),respectively.Sensitivities were 76.98％,78.57％,82.54％,84.13％,and 82.54％,and specificities were 73.58％,73.58％,67.92％,71.70％,and 73.58％,respectively.The FOUR had the highest AUC,with an optimal cutoff value of 13.(3)Pairwise comparisons of AUC showed a statistically significant difference between the FOUR and GCS(the difference value of AUC is 0.034,95％CI 0.004-0.064,Z=2.194,P=0.028),but no significant differences were observed between other scales(all P＞0.05).Conclusion Compared to GCS,GCS-P,GPCS,and CRS-R,FOUR may provide more valuable prognostic information for severe stroke patients.
