Use of pre-operative blood products in abdominal trauma: a planned secondary analysis of the GOAL-trauma study

<h2>Summary</h2><h3>Background</h3> Haemorrhage remains the leading cause of death in patients with abdominal trauma. However, variation remains globally in the standards of transfusion practice. The aim of this study was to characterise global pre-operative transfusion practices for patients undergoing trauma laparotomy and identify any associations between transfusion strategies and mortality. <h3>Methods</h3> This was a planned secondary analysis of the Global Outcomes After Laparotomy for Trauma (GOAL-Trauma) study, an international multicentre prospective observational study conducted at 187 hospital centres across 51 countries. Patients of any age with a blunt or penetrating traumatic injury who underwent a laparotomy within 5 days of presentation were eligible. Patients were excluded if they were undergoing a repeat laparotomy at the recruiting centre within 30 days of the index procedure. Eligible participants were recruited between April 1, 2024, and Dec 31, 2024 during select 30-day periods and were followed up until discharge, death, or 30 days post-operatively (if still hospitalised), whichever came first. Pre-operative blood product usage was recorded from time of injury until index procedure. Countries were stratified by Human Development Index (HDI) and the primary outcome was post-operative in-hospital 30-day mortality. Comparative regression analyses between blood component groups were performed, comparing high versus low transfusion ratios of fresh frozen plasma (FFP) to packed red blood cells (PRBC). The GOAL-Trauma study was registered with ClinicalTrials.gov, NCT06180668. <h3>Findings</h3> Overall, 721 (40.8%) of 1768 patients received any type of pre-operative blood product. Those in the upper HDI tertile received the highest proportions of blood components, across PRBC, FFP, and platelets (p < 0.0001). Whole blood usage was nearly double in the lower HDI tertile compared to middle and upper HDI tertiles (p < 0.0001). Tranexamic acid use was low across all HDI tertiles (529 of 1768 patients, 29.9%). No difference in overall 30-day post-operative mortality risk was observed between those in the high-ratio and low-ratio FFP:PRBC cohorts (OR = 1.52, CI: 0.89–2.64). <h3>Interpretation</h3> Our findings show significant disparity in the usage of pre-operative blood products for trauma patients globally. With ongoing equipoise regarding the optimum balance of blood products for pre-operative resuscitation in trauma, this work informs future research to support the development of global guidelines for blood transfusion practices in trauma and highlights the need for reciprocal learning across income settings. <h3>Funding</h3> Royal College of Surgeons Ratanji Dalal Research Fellowship and the Engineering and Physical Sciences Research Council.