High Sensitivity Troponin in Noncardiac Surgery: Pandora’s Box or Opportunity for Precision Perioperative Care?

Worldwide, an estimated 200 million non-cardiac surgeries are performed each year¹ As many as 10 million of these surgeries have overt cardiovascular complications that lead to worse clinical outcomes, including increased risk of death in short-term follow up.² A recent report from the Vascular Events in Noncardiac Surgery Patients Cohort Evaluation (VISION) Study investigators suggested that these overt cardiovascular complications are just the tip of the iceberg. Among 21,842 patients age ≥45 years undergoing noncardiac surgery at 23 centers worldwide, nearly 1 in 5 had evidence of myocardial injury as determined by elevated high sensitivity troponin T (hsTnT) in the first 3 days post-operatively, 93% without any ischemic symptoms to support a diagnosis of myocardial infarction.³ Further, there was a graded increase in 30-day mortality with higher hsTnT levels. However, despite the clear association of hsTnT elevation with 30-day mortality, whether routine screening for post-operative hsTnT elevation would lead to care decisions or interventions that improve clinical outcomes or, conversely, opens a Pandora’s box of additional testing and treatment with potentially unintended consequences is an unanswered question.

In this issue of Circulation, Mueller and colleagues offer additional insights into the frequency, clinical presentation, and risk associated with peri-operative myocardial injury (PMI) among higher risk patients undergoing non-cardiac surgery that suggest the importance of answering this question.⁴ They examined PMI among 2018 consecutive patients at increased risk for cardiovascular complications (age ≥65 years or age ≥45 years with a history of coronary artery disease, peripheral artery disease, or stroke) who underwent 2546 noncardiac surgeries at a single hospital. Importantly, hsTnT was systematically assessed both preoperatively and postoperatively as part of a standard-of-care protocol that included specific assessments and an automatic trigger for cardiology consultation for patients with PMI. Although most centers in VISION provided the hsTnT results to physicians, there was no specific protocol for response to elevated levels, and hsTnT was not obtained preoperatively in all study participants.³

PMI was common in the Mueller cohort, occurring in 16% of all surgeries. It may seem surprising that the rate of PMI was lower among these high risk patients than the “all-comer” population in VISION (18%). However, this likely reflects that in VISION, PMI was defined as an elevation of hsTnT above 14 ng/L (99^th percentile) without consideration of a preoperative level.³ In the Mueller study, PMI required an absolute hsTnT increase of ≥14 ng/L above the preoperative level within 48 hours post-operatively.⁴ In this high-risk population, PMI was associated with nearly 3-fold higher adjusted 30-day mortality risk and 1.6-fold higher 1-year mortality risk. Similar to VISION, most cases of PMI were clinically silent: only 6% had typical anginal chest pain, only 18% had any symptoms of ischemia, and only 29% met any criteria for spontaneous myocardial infarction. Importantly, 30-day mortality was similar among cases of PMI with and without symptoms and whether or not any criteria for spontaneous myocardial infarction were met. Thus, it appears that systematic screening would be necessary to identify the minority of noncardiac surgery patients with PMI at increased risk of mortality. Unfortunately, as with many other clinical scenarios in which myocardial injury is associated with increased mortality, it remains unclear whether the increased risk associated with PMI can be mitigated or simply reflects sicker patients undergoing essential procedures.

Overall, 86% of PMI cases were adjudicated as primarily cardiac in etiology, with 30-day mortality of 6.1%, of which 60% was cardiovascular.⁴ Considering this, the potential to improve outcomes simply through targeted cardiovascular medical interventions could be substantial. It is noteworthy, therefore, that only half of PMI cases adjudicated as primarily cardiac in the Mueller study had cardiology consultation, very few underwent diagnostic evaluation for coronary artery disease, and only 29% had any intensification of cardiovascular medications (addition and/or dose titration of aspirin, statins, angiotensin converting enzyme inhibitors, beta-blockers). This observation may reflect the broad spectrum of presumed etiologies other than plaque rupture that likely underlies at least half of cardiac PMI, and that leads to substantial clinical judgment in guiding work up and treatment. However, given high rates of prior coronary artery disease, myocardial infarction and heart failure, and high rates of cardiovascular risk factors such as hypertension and diabetes within the enriched population studied, such intensification of medical treatment in PMI patients may be reasonable. Supporting this hypothesis, in an autopsy series of 1841 consecutive cases, among 26 cases who died after undergoing noncardiac surgery, 12 had evidence of plaque rupture, and 9 of those had evident coronary thrombosis.⁵ Large randomized clinical trials would be needed to confirm clinical benefit of medical interventions based on routine hsTnT screening post-operatively, but seem warranted given the scope of the problem and the potential to impact outcome if treatment benefit were found. In the meantime, medical treatment is generally safe and would seem to have few downsides. A bigger challenge will be determining which patients diagnosed with PMI should undergo additional noninvasive or invasive cardiac testing. Indiscriminately adding stress testing or coronary angiography to the management of 1 in 5 of the 200 million patients undergoing noncardiac surgery worldwide would be resource intensive, and it is not clear that the potential benefits would outweigh risks or expense of testing.

Perhaps equally or more important than determining the appropriate treatment of PMI once it has occurred is determining what can be done pre-operatively to more effectively identify patients at risk for PMI and prevent or minimize it. At present, pre-operative risk stratification for clinically overt cardiovascular complications and proven prevention strategies for cardiovascular complications of noncardiac surgery are limited. The current American College of Cardiology/American Heart Association guidelines for peri-operative cardiovascular evaluation and management have no Class I indications for risk assessment beyond clinical/surgical risk assessment and determination of functional capacity, and no Class I indications for addition or intensification of pre-operative cardiovascular medical treatments.⁶ Mueller and colleagues provided a glimpse into the potential role of preoperative hsTnT measurement in risk stratification for PMI and its associated adverse consequences. They found that among their population of noncardiac surgery patients at increased risk of cardiovascular complications, 51% had an hsTnT value above 14 ng/L (99^th percentile) pre-operatively. Not only were these preoperative hsTnT elevations independently associated with increased 30-day and 1-year mortality, patients with pre-operative hsTnT elevation were at increased risk of adverse outcomes following non-cardiac surgery whether or not they also had PMI. This may suggest that patients with asymptomatic hsTnT elevations preoperatively are simply sicker patients and that occurrence of PMI and associated mortality may reflect this as much as surgical risk per se. Careful consideration of medical and surgical approaches to reduce pre-operative and operative cardiovascular risk⁷ and/or the necessity and timing of proceeding with the surgery itself is warranted in these patients and likely will require multi-disciplinary collaborations between cardiologists, anesthesiologists, and surgeons. However, because only 16% of patients had PMI despite 51% having preoperative hsTnT elevation, and 20% of patients who developed PMI did not have elevated preoperative levels, relying on preoperative hsTnT to aid in decision-making could have substantial downsides, including frequent use of expensive and low yield preoperative testing as part of evaluation of the elevated hsTnT, and potentially unnecessary delays to and/or cancellations of needed surgeries in many patients in whom PMI will not occur or cannot be mitigated.

In summary, there is currently limited information on strategies for prevention or treatment of PMI. Many of the clinical and economic implications of a strategy of systematic perioperative hsTnT screening of noncardiac surgery patients remain to be elucidated. The work of Mueller and colleagues raises the possibility that among higher risk patients undergoing noncardiac surgery, serial measurements of hsTnT to identify pre-operative elevations and PMI could be useful, even among patients without clinical signs or symptoms of ischemia, but definitive evidence is lacking. Thus, a potential Pandora’s Box has been opened. It is incumbent upon the clinical and scientific communities to provide the evidence that informs the role of hsTnT testing peri-operatively and the appropriate clinical response to elevated hsTnT levels to mitigate risk associated with noncardiac surgery if possible without undo delays to needed surgeries and unnecessary risks and expense. While PMI may be attributable to plaque rupture, supply-demand mismatch, or extra-cardiac causes, the large hazard conferred in all cases should prompt clinicians and researchers to pursue this evidence. Until such evidence is available, additional diagnostic testing and/or intensification of medication regimens may be warranted for patients with PMI prior to or after discharge, but will require clinical judgment, additional evidence, and in the meantime should be individualized to each patient’s circumstances.