Automated preoperative assessment of endothelial dysfunction and risk stratification for perioperative myocardial injury in patients undergoing non-cardiac surgery.

Abstract

Myocardial injury after non-cardiac surgery (MINS) is a common complication with associated serious morbidity and mortality. Endothelial dysfunction might play an important role in MINS, and its rapid assessment could provide a novel method of risk stratification before surgery. We studied 238 subjects scheduled to undergo intermediate or high-risk surgery in a two-centre prospective study to determine whether preoperative endothelial dysfunction identified by a reactive hyperaemia-peripheral arterial tonometry (RH-PAT) index could provide effective risk stratification for MINS, defined as serum troponin ≥0.04 μg litre(-1), within 3 postoperative days. The primary outcome occurred in 35 subjects (14.7%). Endothelial dysfunction was defined as an RH-PAT index of ≤1.22. Adjusted for age, Lee index and a composite measure of the extent of surgery, endothelial dysfunction was associated with MINS [odds ratio 10.1, 95% confidence interval (CI) 3.3-30.9, P=0.001] and increased time to discharge from hospital after surgery (hazard ratio 0.39, 95% CI 0.23-0.65, P=0.001). Endothelial dysfunction identified MINS with a sensitivity of 31%, a specificity of 96%, and a positive diagnostic likelihood ratio of 8.0. Risk classification for MINS was improved by the addition of RH-PAT-defined endothelial dysfunction to the Lee index (c-statistic increased from 0.69 to 0.77; integrated discrimination improvement 0.11, P=0.003). However, prognostic utility varied widely between sites. For patients undergoing non-cardiac surgery, non-invasive assessment of endothelial function might enhance preoperative risk stratification for perioperative myocardial injury. However, unexplained large inter-site variation in prognostic utility could limit widespread application and needs to be further understood.