Intraoperative Hypotension and Myocardial Infarction Development Among High-Risk Patients Undergoing Noncardiac Surgery

A Nested Case-Control Study

Linn Hallqvist, MD, PhD Student, DESA; Fredrik Granath, PhD; Michael Fored, MD, PhD; Max Bell, MD, PhD


Anesth Analg. 2021;133(1):6-15. 

In This Article


The source population is collected from large registries and databases, with possible reporting bias, coding errors, and risk of misclassification. Furthermore, it cannot be ensured that all physicians across Sweden used the universal definition when the MI diagnosis was made in the source population, which may result in that MI cases are missed. Another major limitation is the lack of cardiac biomarkers in all patients in the source population; cardiac troponins measured routinely would have captured a larger proportion of MI developed in the perioperative period (including myocardial injuries), since many of these incidents are clinically silent.[14] Troponins may more readily be analyzed in elderly/high-risk patients, possibly leading to an overrepresentation of more severe and frail patients among cases. Reverse causality must be considered; we are unable to exclude the possibility that the hypotensive event is a consequence of a major MI occurring on the operating table. However, from a clinical perspective, MI following a fall in BP is more probable. Moreover, all cases with a major hypotensive episode, leading to cardiac biomarker-analysis postsurgery despite the absence of other clinical signs and ischemic symptoms were excluded to minimize the risk of reversed causation. Additionally, an observed episode of IOH may increase the likelihood of MI diagnosis, leading to overestimation of risk. Intraoperative data were extracted manually from anesthetic charts, with inherent risk of errors. Mean arterial blood pressure (MAP) data are lacking, since this information is inadequately registered, and the SBP-based IOH definition may affect the ability to contextualize the results. The MI data were not available in the registry data; controls were sampled among patients alive and MI-free at day 30. This cumulative density sampling will overestimate relative risk; however, since MI is a rare event, the overestimation is small.[24] Furthermore, this sampling scheme precludes estimation of 30-day mortality related to MI. A previous study of our source population showed perioperative MI increasing 30-day mortality 5-fold. In this study, MI cases had a 27% 30-day mortality, compared to 26% in the source population.[18] The majority of cases are patients with elevated risk factor burden, and our ability to estimate IOH-associated MI risk among patients with low underlying risk is limited. The sensitivity analysis suggests lower relative impact of IOH in low-risk patients and higher impact among high-risk patients. Absolute excess among high-risk patients may thus be under-estimated and, correspondingly, in lower-risk patients, risks may be overestimated.

Our results are in line with previous studies,[2,4,5,7,9,25,26] but with a more pronounced effect of hypotension. The nested case-control design and the use of a well-defined population of high-risk surgical patients give reliable estimates of associations even in rare outcomes, reducing risk of residual confounding. Further plausible reasons for the strong association are our outcome—and exposure—definitions. Only symptomatic MIs, fulfilling the universal definition, are included, myocardial injuries are not. Regarding exposure, since we had access to pre- and intraoperative BP values, we could compare different definitions, relative to baseline (mm Hg), relative to baseline (%), and absolute intraoperative thresholds. All resulted in similar risk estimates with a gradual elevation of MI risk in relation to an increasing fall in BP. Statistically, relative drop in mm Hg from individual baseline was favored. Little is known about optimal BP thresholds perioperatively. A review of IOH identified 140 definitions in 130 studies.[27] Previous investigations are limited by use of specific systolic- or mean BP and may underestimate IOH as a risk factor. Many studies use binary cut-offs, MAP <55 mm Hg or systolic BP <80 mm Hg, showing associations with organ damage and mortality.[2–4] Individual IOH definitions being beneficial was strengthened by a randomized trial evaluating BP in septic shock, where outcomes were improved by high BP targets only in patients with hypertension.[28] In patients with preexisting hypertension, the autoregulatory capacity in the kidney and brain, an essential mechanism to preserve optimal blood perfusion when systemic BP fluctuates, is affected.[29,30] However, there are studies showing that absolute and relative thresholds are comparable in their ability to discriminate patients with myocardial injury from those without.[9] A randomized study showed that targeting an individualized SBP, compared with standard management, reduced postoperative organ dysfunction.[26]

Clinical Significance

MI in the perioperative period has a significant impact on postoperative morbidity and mortality. Thirty-day mortality is increased 5-fold,[18] and the risk increase remains; nonfatal perioperative MI patients have a doubled risk of death at 1 year postsurgery. Perioperative MI is an overall rare condition explaining why these striking findings have not been identified previously. Patients developing MI postsurgery are at increased risk of other complications, such as respiratory failure, pneumonia, wound infection, deep venous thrombosis, and confusion. They also have a prolonged postoperative length of stay and more commonly need treatment at the intensive care unit.[6,14,31–33] Our study identified IOH as a potential major contributor to MI, irrespective of MI type. IOH was equally common among patients with fatal and nonfatal MI, suggesting that IOH is merely a trigger and that the mortality is a result of other risk factors. Notably, IOH was significantly more frequent in lower-risk than in higher-risk groups, implying more vigilant anesthesia in comorbid and fragile patients. The reduction in mm Hg from individual baseline is a clinically appealing definition, the lowest acceptable threshold could be easily determined in the OR, before the anesthetic induction. Importantly, perioperative hemodynamic instability can be prevented in most clinical situations. Adequate intravascular volume and organ perfusion can be maintained using vasoactive drugs and protocolized hemodynamic algorithms to guide delivery of intravenous fluids and maximize stroke volume. An increasing population of elderly patients, with cardiovascular risk factors, are undergoing extensive surgery. Avoiding IOH, by an attentive and meticulous anesthetic treatment during and after surgery, could lower the risk of perioperative MI, as well as other postoperative complications, improving quality of life for these patients and reducing costs for the society.