Clinical Research Directory

Predictors of In-Hospital Mortality in Severe Abdominal Trauma

Severe abdominal trauma is a leading cause of early death from bleeding and of later death from sepsis and organ failure. Clinicians use scoring systems to estimate the risk of dying, but most of these scores were developed and calibrated in high-income trauma systems and may perform differently in high-volume referral hospitals elsewhere, where patients frequently arrive after a prehospital delay and already in established shock. This study will prospectively follow consecutive patients aged 16 years and older admitted to a single tertiary trauma centre with severe abdominal injury over a 12-month period. Severe abdominal trauma is defined as an Abbreviated Injury Scale score of 3 or more for the abdominal region, and/or an abdominal injury requiring emergency laparotomy, laparoscopy or angioembolisation, and/or admission to intensive care for management of an abdominal injury. A total of 430 evaluable patients is anticipated. Information generated as part of routine trauma care will be recorded prospectively from arrival until hospital discharge: age, sex and comorbidity; injury mechanism, time from injury to arrival and mode of arrival; arrival physiology (systolic blood pressure, heart rate, shock index, Glasgow Coma Scale, respiratory rate, oxygen saturation, temperature); admission biochemistry (arterial lactate, base deficit, pH, haemoglobin, platelet count, INR); imaging and operative findings with American Association for the Surgery of Trauma organ injury grades; transfusion and operative management; and in-hospital complications. The primary outcome is in-hospital all-cause mortality. Survivors and non-survivors will be compared, and a multivariable logistic regression model will be built to identify admission and early in-hospital variables that independently predict death. Model discrimination and calibration will be assessed and internally validated by bootstrap resampling, and performance will be compared with that of the Injury Severity Score, Revised Trauma Score, TRISS and the MGAP score. If the model is robust, a simplified integer bedside risk score will be derived and internally validated. The study is purely observational: it mandates no additional investigations, interventions or visits, and all clinical decisions remain with the attending trauma team according to institutional ATLS-based protocols. The findings are intended to support locally valid triage, transfusion and damage-control decision-making and to enable fair outcome benchmarking between comparable centres.