Prognostic dynamic nomogram integrated with metabolic acidosis for in-hospital mortality and organ malperfusion in acute type B aortic dissection patients undergoing thoracic endovascular aortic repair

Abstract

Background: Organ malperfusion is a lethal complication in acute type B aortic dissection (ATBAD). The aim of present study is to develop a nomogram integrated with metabolic acidosis to predict in-hospital mortality and organ malperfusion in patients with ATBAD undergoing thoracic endovascular aortic repair (TEVAR). Methods: The nomogram was derived from a retrospectively study of 286 ATBAD patients who underwent TEVAR from 2010 to 2017 at a single medical center. Model performance was evaluated from discrimination and calibration capacities, as well as clinical effectiveness. The results were validated using a prospective study on 77 patients from 2018 to 2019 at the same center. Results: In the multivariate analysis of the derivation cohort, the independent predictors of in-hospital mortality and organ malperfusion identified were base excess, maximum aortic diameter ≥ 5.5 cm, renal dysfunction, D-dimer level ≥ 5.44 μg/mL and albumin amount ≤ 30 g/L. The penalized model was internally validated by bootstrapping and showed excellent discriminatory (bias-corrected c-statistic, 0.85) and calibration capacities (Hosmer–Lemeshow P value, 0.471; Brier Score, 0.072; Calibration intercept, − 0.02; Slope, 0.98). After being applied to the external validation cohort, the model yielded a c-statistic of 0.86 and Brier Score of 0.097. The model had high negative predictive values (0.93–0.94) and moderate positive predictive values (0.60–0.71) for in-hospital mortality and organ malperfusion in both cohorts. Conclusions: A predictive nomogram combined with base excess has been established that can be used to identify high risk ATBAD patients of developing in-hospital mortality or organ malperfusion when undergoing TEVAR.