Myocardial Injury After Noncardiac Surgery—Slightly Heavier May Be Better

Ashish K. Khanna, MD, FCCP, FCCM; Tong J. Gan, MD, MHS, FRCA, MBA

Disclosures

Anesth Analg. 2021;132(4):957-959. 

Myocardial injury after noncardiac surgery (MINS) has been a focus of attention for perioperative medicine for nearly a decade.[1] An international cohort of 15,065 patients aged 45 years or older who underwent noncardiac surgery were used to develop diagnostic criteria for this entity in 2012. MINS was defined as an elevated troponin (peak troponin T [TnT] >0.03 ng/mL), judged to be due to a myocardial ischemia. While 8% of patients experienced this outcome, less than a fifth of those experienced any symptoms at all. However, and rather soberingly, there was a 4-fold increase in the risk of 30-day mortality in patients with MINS.[2] While this study used fourth-generation TnT, consistent outcomes have been seen in follow-up investigations using high-sensitivity troponin T (hsTnT) as well.[3] With ongoing inconsistent use of fourth-generation and hsTnT, a more universal definition of MINS would entail postoperative troponin levels >99th percentile of the upper reference limit within 30 days after surgery that result from myocardial ischemia without the need of an ischemic feature.[4] A substantial amount of asymptomatic myocardial injury occurs during the postoperative period. Much has been published about the well-established and ever-increasing association of perioperative hypotension with MINS.[1] This is probably a major contributor to the fact that a third of all postoperative deaths are due to a cardiac complication, and in general death in 30 days after surgery is one of the most common causes of mortality in the United States.[5] However, the adjusted risks for MINS and the association of hypotension are much less than the risk associated with a patients' baseline clinical factors.[6] Here, the role of obesity as a risk is of importance. Obesity is a rampant epidemic across the world.[7] While the connection of obesity with cardiovascular disease is well established, there is also a possible protective effect or an "obesity paradox" seen both with and without cardiovascular disease.[8]

In this month's issue of Anesthesia and Analgesia, Lee[9] present a 9-year (2010–2019) retrospective cohort of 35,269 patients with measured postoperative troponins, where they analyzed the association of 3 predefined body mass index (BMI) subgroups (low <18.5 kg/m2, normal 18.5–25 kg/m2, and high >25 kg/m2) and MINS diagnosed with a Troponin I (TnI) above the 99th percentile upper reference as >40 ng/L. The authors concluded that there was a higher incidence of postoperative myocardial infarctions; however, significantly lower 1-year mortality in the high BMI group compared to the normal and low groups, with adjusted hazards of 0.75 and 0.56, respectively, for each of the comparisons. Interestingly, this mortality difference appeared to only differentiate across groups at 1 year, whereas the 30-day mortality outcomes were not different. While this may point to a long-term protective effect of a higher BMI that was not evident in the 30 days postoperative period, this deserves more investigation. In this large, single-center registry study of consecutive patients undergoing noncardiac surgery, they used a TnI immunoassay instead of the hsTnT-based values that have been used by the Vascular events In non-cardiac Surgery patIents cohOrt evaluatioN study (VISION) investigators and are now in use in most parts of the world.[3] Diagnostic accuracy of the TnI assay had a sensitivity and specificity of 74% and 84%, respectively, on admission and 94% and 81% at follow-up (6–24 hours after admission), for the detection of myocardial ischemia at the 99th percentile (>40 ng/L). This along with an event rate for a cardiac event or death within 60 days at 24.1%, 55.1%, respectively, for >40–100 ng/L, or >100 ng/L for the TnI immunoassay, puts it in alignment with hsTnT.[10]

Anywhere from 2% to 40% of patients from 50 to 70 years of age will have elevated preoperative troponins.[11] Therefore, it is important to measure baseline troponin when using hsTnT. This is also because the presence of MINS is defined as when there is an increase by at least 5 ng/L from baseline in the postoperative period up to at least 20 ng/L, or irrespective of the preoperative baseline when the postoperative concentration is >65 ng/L.[3] Quite often cardiologists are called to evaluate isolated troponin elevations in an asymptomatic patient, and the lack of a baseline value can be a barrier to clinical intervention. Lee[9] makes an effort to include baseline troponin values in the data set analyzed, though the numbers of included patients are very low. The question on whom to perform postoperative surveillance monitoring of troponin has been widely debated as well. Most experts agree on this need in all patients >65 years of age, and those >45 years of age with at least 1 cardiovascular risk factor.[6] Here, the authors performed routine troponin measurements for moderate- to high-risk surgery or for patients with 1 cardiovascular risk factor. This is reflected by the 35,296 patients with measured troponin out of an initial cohort of 43,019 patients, though an argument could be made for a slightly higher risk selection here compared to recommendations. The authors did include some low-risk patients whose troponins were sampled at the discretion of the attending physician, likely because they were of a higher age or had ongoing recently suspected symptoms of ischemic disease. Despite all this, the troponin surveillance process did not strictly involve serial checks on the first 3 days after surgery, but rather included any value within 30 days after surgery.

The definition of obesity as used by Lee[9] also needs close examination. The Centers for Disease Control (CDC) define obesity as a BMI >30 kg/m2, while >25 kg/m2 is overweight, whereas the authors used the term high BMI for >25 kg/m2.[12] Therefore, one could argue that authors did not have truly obese patients in their cohort. Obesity needs to factor in ethnicity and Asian patients that constituted all of this cohort may actually be reclassified as obese by different threshold standards for this population. Also, an important physiological consideration is that Asians may have a distribution of fat that favors truncal obesity, which per se has a stronger association with cardiovascular complications even at lower BMI numbers. The World Health Organization (WHO) has also proposed a definition of overweight (BMI 23.0–24.9) and obesity (BMI ≥25.0) for this population.[13] Interestingly, at this modified Asian obesity cutoff BMI above 25.0 kg/m2, mortality risk was higher among Asians in comparison to the US population.[14] The authors did include in their analysis, a BMI of >30 kg/m2 as obese in accordance with the standard CDC definition. However, mortality of these truly obese patients (BMI >30 kg/m2) was only numerically lower though not statistically significant. It also seems difficult to interpret these classifications of BMI as done by Lee,[9] into 3 specific categories when translated into the clinical context. Does a difference of a single decimal point on the BMI scale suddenly escalate the same patient into a lower risk of MINS-related mortality? An ideal exposure would have used BMI as a continuous variable and allowed the clinician to interpret these results to see the progressively increasing protective effect of BMI on MINS. The authors do provide us smoothened hazard ratio plots of a progressively increasing BMI, but the number of patients drops off sharply at a BMI >30 kg/m2. It is therefore critically important that interested clinician scientists use these results to collaborate on larger patient cohorts in environments where a truly higher BMI patient population is more prevalent. For example, an elegant experiment would be to look at the North American population where there would be a substantial proportion of patients that qualify as truly obese and morbidly obese as well. It would be of great interest to the perioperative medicine community to see the entire spread of MINS and if this obesity paradox is observed along the entire BMI scale up to as high as >40 kg/m2. It is reasonable to assume that at some stage the cardiovascular harm associated with a very high BMI, coupled with the effect modification of a sedentary lifestyle, high underlying lipid subcategories and so forth, would make it difficult to strictly see this so-called protective effect persist at these very high BMI states.

There are potential mechanisms that explain the protective effective of a higher BMI in their findings. In general, cytokine is released to mediate protection from inflammation and maintenance of homeostasis, along with specifically for ischemic heart disease, a younger age, higher metabolic reserve, less cachexia, robust blood pressure with allowance for more aggressive cardiac medications are all likely benefit mechanisms for the observed obesity paradox.[15] As a single-center cohort, with the possibility of unidentified confounding factors, these results should be interpreted with caution. Troponins were measured in a slightly higher risk population and there is no clear description of intraoperative and postoperative blood pressure management, where it could well be that this higher risk group had more aggressive blood pressure control compared to the nonincluded group. This could have introduced a significant selection bias. Lee[9] deserves to be congratulated on the routine use of troponin monitoring with an effort to include baseline monitoring in a very real-world patient cohort that was appropriately adjusted across a wide variety of covariates. The protective effect of a high BMI is easy to label as an obesity paradox; however, this should not tempt the anesthesiologist to reassure the high BMI patient in the preoperative clinic. Until we fully explore the entire range of BMI with an adequate sample size, we are left at best to guess, and a likely speculation would be in fact a U-shaped curve of 1-year mortality and BMI. As of now, being slightly heavier may be better for MINS; however, this relationship may be more complicated and await confirmation from other investigations.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....