Abstract
BACKGROUND: Emergency Laparotomy (EL) is recognized as high-risk surgery with high mortality. Established surgical risk assessment tools (NELA Risk Prediction Calculator, P-POSSUM, ACS-NSQIP) are accurate predictors of morbidity and mortality. However, their multicomponent complexity limits their use in practice. Sarcopenia is associated with poorer surgical outcomes. This study tests for an association between a simple measure of radiological sarcopenia and mortality in EL patients in an Australian cohort. METHODS: A retrospective analysis was conducted of 500 patients admitted to four Australian hospitals who underwent EL during 2016-2017. All patients had a contemporaneous abdomino-pelvic CT scan. Radiological sarcopenia was measured as the ratio of total psoas muscle area (PM) to L3 vertebral body cross sectional area (PM:L3). Patients were followed up to 12 months. Primary outcomes were 30-, 90- and 365-day mortality. RESULTS: The mean 30-day mortality predictions for NELA, P-POSSUM and ACS-NSQIP were 11.36%, 17.28% and 11.30% respectively. PM:L3 ratio was associated with 30-, 90- and 365-day mortality (P < 0.001) and sex (P < 0.001) and negatively correlated with age (r = -0.4612; P < 0.001). Radiological sarcopenia had a weak negative correlation with NELA (r = -0.2737; P < 0.001), P-POSSUM (r = -0.1880; P < 0.001) and ACS-NSQIP (r = -0.2351; P < 0.001). The latter three metrics were significantly correlated (r > 0.5696; P < 0.001). CONCLUSION: Radiological sarcopenia (CT-assessed PM:L3) is a significant predictor of mortality in EL patients in Australia. The results of this study suggest that radiological sarcopenia is equivalent to established risk assessment tools. The more timely and easily accessible CT-assessed PM:L3 metric is potentially automatable and may have significant utility in clinical practice.