Validation of the COAST score for predicting acute traumatic coagulopathy: A retrospective single-centre cohort study

Abstract

Purpose Early identification of trauma patients at risk of developing acute traumatic coagulopathy is important in initiating appropriate, coagulopathy-focused treatment. A clinical acute traumatic coagulopathy prediction tool is a quick, simple method to evaluate risk. The COAST score was developed in Australia and we hypothesised that it could predict coagulopathy and bleeding-related adverse outcomes in other advanced trauma systems. We validated COAST on a single-centre cohort of trauma patients from a trauma centre in Belgium. Methods The COAST score was modified to suit available data; we used entrapment, blood pressure, temperature, major chest injury and abdominal injury to calculate the score. Acute traumatic coagulopathy was defined as international normalised ratio >1.5 or activated partial thromboplastin time >60 s upon arrival of the patient to the hospital. Data were extracted from the local trauma registry on patients that presented between 1 January and 31 December 2015. Results In all, 133 patients met the inclusion criteria (>16 years old, available COAST and outcome data) for analysis. The COAST score had an area under the receiver operating characteristics curve of 0.941 (95% CI: 0.884–0.999) and at COAST ≥3, it had 80% sensitivity and 96% specificity. The score also identified patients with higher rates of mortality, blood transfusion and emergent surgery. Conclusion This retrospective cohort study demonstrated the utility of the COAST score in identifying trauma patients who are likely to have bleeding-related poor outcomes. The early identification of these patients will facilitate timely, appropriate treatment for acute traumatic coagulopathy and minimise the risk of over-treatment. It can also be used to select patients with acute traumatic coagulopathy for trials involving therapeutic agents targeted at acute traumatic coagulopathy.