Intraoperative Hypotension and Myocardial Injury After Noncardiac Surgery in Adults With or Without Chronic Hypertension: A Retrospective Cohort Analysis

Document Type


Publication Date


Publication Title

Anesthesia and analgesia


BACKGROUND: The risk of myocardial injury progressively increases at intraoperative mean arterial pressures (MAPs) ≤65 mm Hg. Higher pressures might be required in chronically hypertensive patients. We aimed to test the hypothesis that the harm threshold is higher in patients with chronic hypertension than in normotensive patients.

METHODS: We conducted a single-center retrospective cohort analysis of adults >45 years old who had noncardiac surgery between 2010 and 2018 and scheduled, rather than symptom-driven, postoperative troponin measurements. The MAP thresholds under which risk started to increase were compared between patients with chronic hypertension (baseline MAP ≥110 mm Hg) and normotensive patients (baseline MAPHg). The primary outcome was a composite of in-hospital mortality and myocardial injury within 30 days, defined by any postoperative 4th-generation troponin T measurement ≥0.03 ng/mL apparently due to cardiac ischemia. Multivariable logistic regression and moving average smoothing methods were used to evaluate confounder-adjusted associations between the composite outcome and the lowest intraoperative MAP sustained for either 5 or 10 cumulative minutes, and whether the relationship depended on baseline pressure (normotensive versus hypertensive).

RESULTS: Among 4576 eligible surgeries, 2066 were assigned to the normotensive group with mean (standard deviation [SD]) baseline MAP of 100 (7) mm Hg, and 2510 were assigned to the hypertensive group with mean baseline MAP of 122 (10) mm Hg. The overall incidence of the composite outcome was 5.6% in normotensive and 6.0% in hypertensive patients (P = .55). The relationship between intraoperative hypotension and the composite outcome was not found to depend on baseline MAP in a multivariable mixed effects logistic regression model. Furthermore, no statistical change points were found for either baseline MAP group.

CONCLUSIONS: Baseline blood pressure of the hypertensive patients was only moderately increased on average, and the event rate was low. Nonetheless, we were not able to demonstrate a difference in the harm threshold between normotensive and chronically hypertensive patients. Our results do not support the theory that hypertensive patients should be kept at higher intraoperative pressures than normotensive patients.

PubMed ID



ePub ahead of print