医学
倾向得分匹配
脑积水
危险系数
蛛网膜下腔出血
外科
队列
比例危险模型
混淆
剪裁(形态学)
人口
队列研究
脑出血
内科学
置信区间
环境卫生
语言学
哲学
作者
Abdelsimar Tan Omar,Jose Danilo Bengzon Diestro,Julian Spears,Elisabetta Patorno
标识
DOI:10.1136/jnis-2024-021852
摘要
BACKGROUND: Hydrocephalus is a significant contributor to morbidity following aneurysmal subarachnoid hemorrhage (aSAH). We aimed to investigate the association between primary treatment modality and the incidence of hydrocephalus requiring CSF diversion, using a target trial approach for causal inference. METHODS: This cohort study used US administrative health claims data (Clinformatics Data Mart) and was conducted among aSAH patients undergoing primary treatment with either clipping or coiling, from January 1, 2004, to February 28, 2023. The primary outcome was hydrocephalus requiring CSF diversion surgery while the secondary outcome was mortality. Multivariable regression and 1:1 propensity score (PS) matching were used for confounder control. Crude and adjusted hazard ratios (HRs) with 95% CIs were calculated. RESULTS: A total of 5816 patients (mean age 59 years; 72% women) undergoing clipping (n=1794) or coiling (n=4022) were included in the primary cohort. The 1:1 PS matched cohort had 1794 participants per arm. Clipping demonstrated higher hazards of shunt dependent hydrocephalus compared with coiling in both the multivariable Fine-Gray model (HR 1.39, 95% CI 1.19 to 1.62) and the PS matched cohorts (HR 1.39, 95% CI 1.16 to 1.66). Mortality analysis favored clipping in the crude analysis (HR 0.78, 95% CI 0.69 to 0.88) but leaned toward coiling after confounder adjustment (HR 1.13, 95% CI 1.00 to 1.29 in the multivariable model; HR 1.11, 95% CI 0.95 to 1.29 in the PS matched cohort). CONCLUSION: These findings suggest that coiling is associated with reduced hazards of shunt dependent hydrocephalus following aSAH compared with clipping, and provide valuable insights for shared decision making among clinicians and patients, in the context of conflicting evidence from smaller observational studies.
科研通智能强力驱动
Strongly Powered by AbleSci AI