Thiamine use and short-term prognosis in critical cerebrovascular disease: a propensity score matched retrospective analysis from the MIMIC-IV database

In the realm of critical care, cerebrovascular disease remains a major factor in patient morbidity and mortality, with limited treatment options outside of acute interventions. The importance of thiamine (vitamin B1) in energy metabolism is well-known, and it may provide neuroprotective benefits, though its role in serious cerebrovascular diseases is still uncertain.

Using data from the MIMIC-IV database, this retrospective cohort study explored the relationship between thiamine and short-term outcomes in patients with critical cerebrovascular conditions. The application of propensity score matching (PSM) helped to balance covariates between users and non-users of thiamine. The study's primary and secondary endpoints were the all-cause mortality at 7, 14, and 28 days. To estimate hazard ratios (HR) and 95% confidence interval (CI), Cox proportional hazards models were used, along with analyses for subgroups, dose-response, and duration-response.

Among 11,814 patients, 1,647 received thiamine. With 1,585 matched pairs post-PSM, thiamine administration showed a significant association with decreased mortality at 7 days (HR = 0.68, 95%CI: 0.53-0.89, P = 0.004), 14 days (HR = 0.78, 95%CI: 0.64-0.96, P = 0.016), and 28 days (HR = 0.82, 95%CI: 0.69-0.98, P = 0.028) compared with non-thiamine use after adjustment. Subgroup analyses revealed stronger benefits in patients aged ≥ 65 years, females, and those without mannitol or thrombolysis. Enhanced outcomes were most consistently observed with treatment durations longer than 5 days and daily doses lower than 100 mg.

Thiamine supplementation correlated with a reduction in short-term mortality for patients with critical cerebrovascular conditions, especially with prolonged treatment and reduced daily doses. While these observations propose thiamine as a possible supportive metabolic therapy, the evidence remains insufficient to guide clinical practice and should be considered hypothesis-generating pending confirmation in randomized trials.