Acquired loss of cardiac vagal activity is associated with myocardial injury in patients undergoing non-cardiac surgery: prospective observational mechanistic cohort study

Abstract

Background: Myocardial injury is more frequent after non-cardiac surgery in patients with preoperative cardiac vagal dysfunction as quantified by delayed heart rate recovery after cessation of cardiopulmonary exercise testing. Here, we hypothesised that serial and dynamic measures of perioperative cardiac vagal activity should also be associated with myocardial injury after non-cardiac surgery. Methods: Serial measures in cardiac vagal activity were quantified preoperatively and daily using heart rate variability and a standardised orthostatic challenge in patients undergoing elective non-cardiac surgery. The primary outcome was myocardial injury (high-sensitivity troponin (hsTnT) ≥15ng.L -1 ) within 48h of surgery. Clinicians, patients and investigators were blinded to hsTnT. The exposure of interest was cardiac vagal activity (high-frequency power spectral analysis [HFlog]) and heart rate recovery after a standardised orthostatic challenge. Results: hsTnT≥15ng.L -1 occurred in 48/189 [25%] patients, of whom 41/48 [85%] had a revised cardiac risk index score <2. Patients with a post surgery troponin HsTnT ≥15ng.L -1 were associated with an early loss (within 24h) of cardiac vagal activity (HFLog) post surgery compared to day of surgery (4.19 [95%CI:3.62-4.75] vs 5.22 [95%CI:4.64-5.81]; p<0.001).Heart rate recovery after a standardised orthostatic challenge after surgery was slower in patients with hsTnT≥15ng.L -1 (5 beats minute -1 (95% CI: 3 -7), compared to heart rate recovery in patients who remained free of myocardial injury (10 beats minute -1 (95%CI:7 to 12]; p = 0.02). Conclusions:Real-time, serial heart rate measures indicating loss of cardiac vagal activity are associated with perioperative myocardial injury in lower-risk patients undergoing non-cardiac surgery.