Jianbo Chen, Qiong Wu, Xiangzhi Xiao, Sufen Chen, Jue Hu, Zhen Wang, Wengao Zeng
Documented inpatient aspirin use was not associated with lower in-hospital mortality after adjustment for measured covariates. Interpretation is limited by unavailable treatment timing and severity measures and by the absence of functional and post-discharge outcomes. These findings represent observational associations.
BACKGROUND: Aspirin has antiplatelet and anti-inflammatory effects that may be relevant to cerebrovascular complications of tuberculous meningitis (TBM), but its association with survival in routine clinical practice remains uncertain. We examined the association between documented inpatient aspirin use and all-cause in-hospital mortality in a large non-HIV TBM cohort.
METHODS: We conducted a single-center retrospective cohort study of 1,574 non-HIV patients hospitalized with TBM from October 2013 through January 2024. Exposure was defined as any documented aspirin administration during the index hospitalization. The primary outcome was all-cause in-hospital death. Missing covariates were handled using multiple imputation by chained equations (20 datasets, 20 iterations). Stabilized inverse probability of treatment weights based on 25 prespecified covariates standardized the comparison to the overall cohort. Sensitivity analyses included 1st-99th percentile weight truncation, five-fold cross-fitted augmented inverse probability weighting (AIPW), 1:1 propensity-score matching, and restriction to adults.
RESULTS: Among 1,574 patients, 844 (53.6%) had documented aspirin use and 187 (11.9%) died in hospital. Mortality occurred in 107/844 patients with documented aspirin use (12.7%) and 80/730 patients without documented use (11.0%). The maximum absolute standardized mean difference decreased from 0.202 before weighting to 0.020 after weighting, and the effective sample size remained approximately 1,519. The primary weighted analysis yielded a risk difference of 0.79 percentage points (95% CI,-2.49-4.06), a risk ratio of 1.07 (95% CI, 0.81-1.41), and an odds ratio of 1.08 (95% CI, 0.79-1.48). All 95% confidence intervals crossed the null, and sensitivity analyses produced similar estimates.
CONCLUSIONS: Documented inpatient aspirin use was not associated with lower in-hospital mortality after adjustment for measured covariates. Interpretation is limited by unavailable treatment timing and severity measures and by the absence of functional and post-discharge outcomes. These findings represent observational associations.