Key result
HAS model achieves superior 0.97 AUC over NELA score for predicting 30-day emergency laparotomy mortality.
Why the study?
The NELA mortality risk score is routinely used in the UK, but how its predictive performance compares to the novel HAS model for estimating mortality risk after emergency laparotomy was unknown.
Does the HAS model improve mortality risk prediction compared to the NELA score in adult patients undergoing emergency laparotomy?
Cohort (n=818)
No
Does the HAS model improve mortality risk prediction compared to the NELA score in adult patients undergoing emergency laparotomy?
Effect estimate: AUC 0.97 vs 0.86
Absolute Event Rate: 0.97% vs 0.86%
p-value: p=<0.0001
The HAS model demonstrated superior discrimination and calibration compared to the NELA score for predicting 30-day, in-hospital, and 90-day mortality after emergency laparotomy.
Supports HAS model for mortality risk stratification after emergency laparotomy; leaves open prospective validation before clinical adoption.
Background The National Emergency Laparotomy Audit (NELA) mortality risk score is currently used in the UK to estimate mortality risk after emergency laparotomy. The HAS (Hajibandeh Index, American Society of Anesthesiologists status, and sarcopenia) is a novel model with excellent accuracy in predicting the risk of mortality after emergency laparotomy. This study aimed to compare the predictive performance of the HAS model and NELA score in estimating mortality risk following emergency laparotomy. Methodology A retrospective cohort study was conducted including consecutive adult patients who underwent emergency laparotomy between January 2019 and January 2022. Thirty-day mortality was the primary outcome. In-hospital mortality and 90-day mortality were the secondary outcomes. The predictive tools were compared in terms of discrimination via receiver operating characteristic curve analysis, calibration via the Hosmer-Lemeshow test, and classification via classification table. Results Analysis of 818 patients showed that the area under the curve of HAS was superior to NELA for 30-day mortality (0.97 vs. 0.86, p < 0.0001), in-hospital mortality (0.90 vs. 0.83, p = 0.0004), and 90-day mortality (0.90 vs. 0.83, p = 0.0004). HAS demonstrated good calibration for 30-day mortality (p = 0.286), in-hospital mortality (p = 0.48), and 90-day mortality (p = 0.48) while NELA score showed poor calibration for 30-day mortality (p = 0.001), in-hospital mortality (p = 0.001), and 90-day mortality (p = 0.001). Conclusions The HAS model was superior to the NELA score in predicting mortality after emergency laparotomy. The HAS model may be worth paying attention to for external validation.
No takes yet. Share an insight, caveat, or question.
Linganathan et al. (2023) conducted a cohort in Emergency laparotomy for non-traumatic abdominal pathology (n=818). HAS model vs. NELA score was evaluated on 30-day postoperative mortality discrimination (AUC) (AUC 0.97 vs 0.86, p=<0.0001). The HAS model demonstrated superior discrimination compared to the NELA score in predicting 30-day mortality after emergency laparotomy (AUC 0.97 vs. 0.86, p < 0.0001).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: