Risk Factors Analysis And Model Prediction Of Symptomatic Intracranial Hemorrhage After Interventional Therapy For Acute Occlusion Of Large Anterior Circulation Vessels

Objective: To investigate the related risk factors of symptomatic intracranial hemorrhage(s ICH)in patients with acute anterior circulation occlusion after endovascular treatment,and to obtain the effective prediction scale of s ICH after emergency interventional treatment.Methods: Retrospective analysis of our hospital and Xinqiao Hospital of the Third Military Medical University from January 1,2019 to February 1,2021,patients with anterior circulation acute vascular occlusion,who underwent emergency endovascular treatment and hospitalized,collected general information,treatment,imaging characteristics.All cases were randomly divided into derivation group(60%)and verification group(40%).Through univariate analysis,the clinical and imaging characteristics of patients with or without s ICH in derivation group were compared.In multivariate logistic regression analysis,multiple variables with statistical significance were combined to obtain the predictive score of s ICH in patients with acute anterior circulation occlusion undergoing emergency vascular intervention,And internal verification.Result:(1)A total of 186 cases were included in this study,32 cases(17.2%)underwent interventional therapy after thrombolysis,164 cases(88.2%)achieved vascular recanalization(blood flow grade 2b-3),113 cases(60.8%)had collateral circulation score(ASITN/SIR Score)of 2-3,40 cases had s ICH,accounting for21.5% of the total number of patients.(2)In the monofactor analysis of 186 cases,the NIHSS score of s ICH group was significantly higher than that of non s ICH Group [14.5(12.75-17.25)vs 12(10-15),P < 0.001],the blood platelet count was higher [175.5(148.8-222)vs 196(160.5-241),P=0.022],the ASPECT score was lower [5.5(2-8)vs 8(6-9),P < 0.001],and the collateral circulation was worse [14(35%)vs 99(67.8%),P< 0.001].(3)By multivariate logistic regression analysis,high NIHSS score was the independent risk factor of s ICH(OR = 1.132;95% CI,1.034-1.238),high ASPECT score(OR = 0.813;95% CI,0.698-0.946)and good collateral circulation(OR = 0.063,95% CI,0.159-0.827)were the protective factors of s ICH.(4)In the derivation group,the single factor analysis showed that NIHSS score was higher in s ICH group(15.5(13-17.8)vs12(10-15.3),P = 0.001],ASPECT score was lower [5.5(2.25-8)vs8(6-9),P=0.002],collateral circulation was worse [9(32.1%)vs63(75%),P < 0.001],more patients were treated with arterial thrombolysis(6(21.4%)VS2(2.4%),P = 0.001],and more patients had atrial fibrillation(19(69.9%)vs39(46.4%),P=0.049].(5)Multivariate logistic regression analysis of derivation group showed that high NIHSS score(OR = 1.141;95%CI,1.006-1.294)and arterial thrombolysis(OR= 12.641;95% CI,1.730-92.349)were independent risk factors for s ICH,while high ASPECT score(OR = 0.809;95%CI,0.660-0.991)and good collateral circulation(OR = 0.295;95%CI,0.093-0.935)were protective factors for s ICH.(6)After the scale derivation,the prediction model was finally incorporated into NIHSS score,ASPECT score and collateral circulation,and the collaborative circulation-ASPECTS-NIHSS(CAN)prediction scale was obtained to evaluate the occurrence of s ICH in patients with acute anterior circulation occlusion.The area under curve(AUC)of the derived group was0.797(95%CI,0.699-0.859).The AUC of the validation group was 0.736(95%CI,0.604-0.868).(7)Scale correction: the prediction model passed the Hosmer-Lemeshow test,and the goodness of fit between the derivation group and the validation group and the model was good(P = 0.772 in the derivation group and P = 0.242 in the validation group).There was no significant difference in the predicted value and observed value between the two groups.Pearson correlation analysis showed that there was a significant correlation between the predicted value of CAN scale and the actual s ICH in the derivation group and validation group(r =0.46,P < 0.001 in the derivation group;r = 0.251,P < 0.032 in the validation group).(8)The scores of the CAN scale were divided into low-risk group,medium risk group and high-risk group.In the derivation group,the risk of s ICH in the high-risk group(> 14 points)was 13.2 times higher than that in the low-risk group(< 8 points)(95% CI,3.777-46.131),and the risk of moderate risk group(8-14 points)was 8.308 times higher than that of low risk group(95% CI,2.388-28.906).Conclusion:(1)In this study,we found that high NIHSS score is an independent risk factor for s ICH in the endovascular treatment of anterior circulation acute occlusive stroke,and high ASPECT score and good collateral circulation are protective factors for postoperative s ICH.(2)Through the statistics of the derivation group and the verification group,the can prediction scale can be obtained,which can quickly and effectively evaluate before endovascular treatment,and predict the risk of s ICH in the endovascular treatment of anterior circulation large vessel acute occlusive stroke.For patients with score > 14,it is recommended to carefully choose intravascular intervention.