Association between time to transfusion and outcomes in patients with traumatic haemorrhagic shock: an analysis of a nationwide trauma registry in Japan.

PURPOSE: To investigate the association of in-hospital time to transfusion on outcomes of patients with traumatic haemorrhagic shock. METHODS: Using data from the Japan Trauma Data Bank (January 2019-December 2023), we identified adult patients (aged ≥ 18 years) with haemorrhagic shock following trauma (systolic blood pressure < 90 mmHg; shock index ≥ 1.0 upon hospital arrival) and received ≥ 10 units red blood cells within 24 h of injury. We excluded patients transferred from or to other hospitals after emergency treatment, with burn injuries or cardiac arrest, without transfusion time records, and with first transfusion > 60 min after hospital arrival. Mixed-effect multivariable logistic regression estimated the association between time to transfusion for in-hospital mortality, adjusting for patient characteristics, comorbidities, and injury severity. Heterogeneity in the association between transfusion timing and mortality was assessed through nine subgroup analyses. RESULTS: Overall, 698 patients (median age 59 years, 66.2% male) from 132 hospitals were analysed, of whom 265 (38.0%) died during hospitalisation. Mean time to transfusion varied from 4 to 58 min across hospitals. A positive association was observed between time to transfusion and in-hospital mortality. Each 1-min delay in transfusion initiation was associated with higher in-hospital mortality (adjusted odds ratio 1.015, 95% confidence interval 1.002-1.028). Subgroup analyses did not show statistically significant evidence of interaction between transfusion timing and subgroup variables. CONCLUSION: Delayed transfusion initiation was significantly associated with higher in-hospital mortality in patients with traumatic haemorrhagic shock. Reducing in-hospital delays to transfusion may represent an important target for trauma care quality improvement.