Effect of intravenous thiamine administration on critically ill patients: A systematic review and meta-analysis of randomized controlled trials
Nobuto Nakanishi,Yoshinobu Abe,Mizue Matsuo,Akihito Tampo,Kohei Yamada,Junji Hatakeyama,Minoru Yoshida,Ryo Yamamoto,Naoki Higashibeppu,Kensuke Nakamura,Joji Kotani
DOI: https://doi.org/10.1016/j.clnu.2024.09.002
Abstract:Background &aims: Thiamine is an essential micronutrient for energy metabolism. Thiamine deficiency is frequently observed in critically ill patients. However, the effect of thiamine administration is unclear in critically ill patients. Methods: We conducted a systematic review and meta-analysis. To identify randomized controlled trials on the effect of thiamine administration in critically ill patients, a literature search was conducted in MEDLINE, CENTRAL, and ICHUSHI databases from inception to April 2023. Pooled effect estimates were calculated about mortality as the primary outcome and shock duration, lactate level, Sequential Organ Failure Assessment (SOFA) score, delirium, length of mechanical ventilation, length of intensive care unit (ICU) stay, infection rate, all adverse events, and Short-Form Health Survey (SF-36) as the secondary outcomes. The certainty of evidence (CoE) was assessed using the Grading of Recommendations Assessment, Development, and Evaluation approach. Results: Overall, 35 studies (3494 patients) were included. Evidence suggested that thiamine administration resulted in little to no difference in mortality (risk ratio [RR], 0.89; 95% confidence interval [CI], 0.75 to 1.06; Low CoE); however, thiamine administration may reduce shock duration (mean difference [MD], -11.43 h; 95% CI, -20.16 to -2.69 h; Low CoE), lactate level (MD, -0.34 mmol/L; 95% CI, -0.63 to -0.05 mmol/L; Low CoE), and SOFA score (MD, -1.29; 95% CI, -1.91 to -0.66; Low CoE). Conversely, thiamine administration resulted in a slight increase in the length of ICU stay (MD, 0.40 days; 95% CI, 0.01-0.79 days; High CoE). Conclusions: Although thiamine administration may reduce shock state, it may not reduce mortality, and slightly increases the length of ICU stay.