幕上高血压性脑出血微创颅内血肿抽吸引流术后早期神经功能恶化危险因素分析 Risk Factors of Early Neurological Deterioration of Patients after Minimally Invasive Surgery in Supratentorial Hypertensive Intracerebral Hemorrhage

目的 分析幕上高血压性脑出血微创颅内血肿抽吸引流术后发生早期神经功能恶化（early neurological deterioration，END）的危险因素。 方法 本研究为回顾性病例研究，连续纳入2018年10月—2022年12月首都医科大学附属北京天坛医院急诊神经内科收治的微创颅内血肿抽吸引流术治疗的幕上高血压性脑出血患者，记录患者人口学特征和临床信息，以及血肿体积、部位、周围灌注等影像学信息。END定义为手术后24 h内NIHSS评分增加≥4分或GCS评分下降≥2分。将单因素分析中差异有统计学意义的变量纳入多因素logistic回归模型（后退法），分析影响患者END的独立危险因素。绘制独立危险因素预测END的ROC曲线并计算AUC，采取De Long检验比较不同独立危险因素的预测能力。 结果 共入组157例患者，平均年龄为（57.1±13.1）岁。20例（12.7%）患者出现END。多因素logistic回归分析显示，术前血肿体积（OR 1.024，95%CI 1.001～1.047，P＝0.043）、术后血肿扩大（OR 41.605，95%CI 7.405～233.765，P＜0.001）、术前低灌注体积（OR 1.011，95%CI 1.002～1.020，P=0.012）等3个因素可独立预测END的发生。其中术前低灌注体积ROC的AUC为0.921，敏感度为0.824，特异度为0.891，截断值为119.0 mL。De Long检验显示，术前低灌注体积对END的预测效力优于术前血肿体积及术后血肿扩大。 结论 术前低灌注体积与幕上高血压性脑出血患者微创颅内血肿抽吸引流术后发生END独立相关，低灌注体积越大，术后END发生的风险越高。 Abstract: Objective To analyze the risk factors of early neurological deterioration (END) after minimally invasive surgery of patients with supratentorial hypertensive intracerebral hemorrhage (ICH). Methods This study was a retrospective case study. Supratentorial hypertensive ICH patients who had undergone minimally invasive surgery in the Emergency Neurology Department, Beijing Tiantan Hospital, Capital Medical University from October 2018 to December 2022 were continuously included. Patients’ demographic characteristics and clinical information, as well as imaging information like hematoma volume, location, and peri-hematoma perfusion, were recorded. The evaluation criteria of END was that the NIHSS score increased by≥4 points or the GCS score decreased by≥2 points within 24 hours after surgery. The variables with statistically significant differences selected by univariate analysis were incorporated into the binary logistic regression model (regression method) to analyze the independent risk factors affecting the END of patients. At the same time, the ROC curve of independent risk factors was plotted and the AUC was calculated. De Long test was used to compare the prediction ability of different independent risk factors. Results A total of 157 patients were enrolled, with an average age of (57.1±13.1) years. END occurred in 20 patients (12.7%). Multivariate logistic regression analysis showed that preoperative hematoma volume (OR 1.024, 95%CI 1.001-1.047, P=0.043) and postoperative hematoma expansion (OR 41.605, 95%CI 7.405-233.765, P＜0.001) and preoperative hypoperfusion volume (OR 1.011, 95%CI 1.002-1.020, P=0.012) could independently predict the occurrence of END. The AUC, sensitivity, and specificity of ROC for preoperative hypoperfusion volume were 0.921, 0.824, and 0.891. The cut-off value was 119.0 mL. De Long test indicated that the prediction efficacy of preoperative hypoperfusion volume was better than that of preoperative hematoma volume and postoperative hematoma expansion (P