Intra-operative heart rate and postoperative outcomes – rowing against the tide?

This Invited Commentary accompanies the following original article: Ruetzler K, Yilmaz HO, Turan A et al. Intra-operative tachycardia is not associated with a composite of myocardial injury and mortality after noncardiac surgery. A retrospective cohort analysis. Eur J Anaesthesiol 2019; 36:105–113. Worldwide, 300 million adults undergo noncardiac surgery every year.1 Associated mortality remains substantial with 0.5 to 4% of patients dying within the first 30 postoperative days.2–5 Peri-operative cardiac ischaemia or injury (as defined by elevated high-sensitivity troponin) occurs in at least 20% of noncardiac surgical patients6 and is associated with nearly 8% mortality within 30 days after surgery.7,8 In the absence of ischaemic signs and/or symptoms, a large proportion of postoperative cardiac ischaemic events do not appear to result from acute coronary occlusion and myocardial necrosis,9–13 but are postulated to occur following insufficient oxygen supply to meet cardiac demands (type 2 ischaemia). Peri-operative tachycardia may, therefore, be a likely driver of myocardial injury for a host of basic physiological reasons.14 The tachycardia hypothesis has been supported by multiple observational studies showing an association between different thresholds of pre, intra and postoperative heart rates and postoperative ischaemic events and mortality.14 However, attempts to mitigate the postulated negative effects of peri-operative tachycardia using beta-blockers or alpha-2 agonists to modulate sympathetic nervous system activity have resulted in a very mixed picture emerging from a number of high profile trials. Indeed, peri-operative beta-blockers significantly decreased intra and postoperative myocardial ischaemia15 and myocardial infarction15,16 in a heart rate dependent manner. However, two large pragmatic multicentre trials with metoprolol17 or clonidine18 revealed that these interventions were strongly associated with peri-operative hypotension and, at least for beta-blockers, it meant a clear signal towards increased postoperative mortality.16 Therefore, in recent years, much of the literature in the field has been focusing on the relationship between peri-operative hypotension and myocardial ischemia. In this issue of the European Journal of Anaesthesiology, the publication of a single-centre retrospective study from the Department of Outcomes Research at the Cleveland Clinic re-opens the debate on the role of heart rate in peri-operative cardiac ischaemia (K. Ruetzler, H.O. Yilmaz, A. Turan, et al., pp. 105–113). The authors reviewed the electronic anaesthesia records (EARs) of 2652 adult patients undergoing surgery between 2010 and 2015, who had routine scheduled troponin measurements. The main aim of the study was to evaluate the association between intra-operative ‘area above’ a heart rate more than 90 beats per minute (bpm) and a composite of Myocardial Injury after Noncardiac Surgery (MINS) and in-hospital mortality after surgery. The ‘area above’ a given heart rate threshold was defined as ‘the sum of each time interval between two adjacent heart rate measurements (where at least one measurement was above the threshold) multiplied by the average of the corresponding two heart rates’. The measure has units of hours x bpm and has the advantage of capturing both severity and duration of tachycardia, although it may be that it is not intuitively calculated. For example, an exposure of intra-operative area above heart rate more than 90 bpm of 6 h x bpm would result from a heart rate of 93 (i.e. 3 bpm above the threshold of 90 bpm) over two intra-operative hours. MINS, that is the elevation of postoperative troponin judged from myocardial ischaemia, occurred in 124 (4.6%) patients and six (0.2%) died while in hospital. The major finding of this study was the lack of an association between intra-operative area above heart rate more than 90 bpm and the composite outcome of MINS and in-hospital mortality, after adjustment for confounders [odds ratio (OR) 0.99; 95% confidence interval (95% CI) 0.97 to 1.01] per 1 h x bpm increase in area above heart rate more than 90 bpm. Area above heart rate more than 80 bpm or heart rate more than 100 bpm were also not associated with the composite outcome. The lack of an association between several measures of elevated intra-operative heart rate and myocardial injury/mortality challenges most of the previous literature, including a recent study from the VISION cohort with over 14 000 patients (> five times more than the present cohort).19 In that study, intra-operative heart rate more than 100 bpm was independently associated with MINS (OR 1.27; 95% CI 1.07 to 1.50), myocardial infarction (OR 1.34; 95% CI 1.05 to 1.70) and 30-day mortality after surgery (OR 2.65; 95% CI 2.06 to 3.41), after adjustment for possible confounders.19 Secondary analyses suggest that the relation between high heart rate and MINS is likely duration-dependent (being more significant with > 30 min of exposure) and the risk is higher in the presence of hypotension (SBP <100 mmHg).19 So, what could explain this apparent contradiction? The increasing capacity to analyse large data sets in anaesthesia (minute-by-minute heart rate measurements were used in this study) has been proving a useful tool in our understanding of peri-operative morbidity and mortality.20 In the absence of universally accepted standards or definitions of pre, intra and postoperative heart rate, the authors are commended by their choice of heart rate exposure that encompasses both duration and severity, unlike previous studies that focused solely on the crossing of a certain heart rate threshold. Although the observational nature of the study is naturally at risk of confounding, it is noteworthy that the results were adjusted for some covariates known to be associated with postoperative myocardial ischaemia, namely intra-operative hypotension. Several aspects of all these studies are worth considering. Variable event rates of MINS limit generalisability, with clear inter-institutional differences (4.6% compared with 8% in the whole VISION cohort using the same troponin assay).7 The fragility of the current study is significant, given the low event rate and a post hoc sensitivity analysis wherein a borderline association was evident between the maximum heart rate sustained for at least 10 min and the all-or-none composite outcome of myocardial injury and in-hospital all-cause mortality (P = 0.052). In this study, patients may have benefited from a routine standard of care wherein episodes of tachycardia were treated promptly, thereby diminishing the likely impact of this single-centre observational study. Another possibility for the few episodes of tachycardia and low rate of MINS is the population's relatively low baseline risk for cardiovascular complications (only 19% of patients had coronary heart disease, 2% heart failure and 5% renal failure). Noteworthy, only a minority of patients were submitted to emergency surgery (1.1%), which is a known risk factor for MINS.7 Without access to pre-operative troponin on all patients, the authors were forced to exclude from the analysis a significant number of sick patients with comorbidities know to be associated not only with elevated baseline troponin but also higher cardiovascular risk, such as chronic kidney disease. Despite several limitations, the study by Ruetzler et al. (pp. 105–113) does challenge the preceding literature and broadly accepted aspects of current practice. Most fundamentally, these data challenge the concept that tachycardia – however that may be defined – is a direct mediator of peri-operative myocardial injury. Interestingly, a recent retrospective analysis of 41 140 patients found that a heart rate at least 90 bpm was associated with an increased risk of myocardial injury (adjusted OR 1.22; 95% CI 1.06 to 1.39; P = 0.005),21 mirroring closely the findings from the analogous VISION study.22 Detailed prospective physiological examination of similar higher-risk surgical patients pre-operatively found that a resting heart rate at least 87 bpm was strongly associated with distinct cardiovascular endotypes that may explain the unintended, adverse consequences of nonpersonalised treatments aimed at reducing intra-operative tachycardia in isolation from other haemodynamic factors.23 Taken together, the ‘perioperative tachycardia’ body of literature now brings into sharp focus whether the assumption remains valid that tachycardia per se drives peri-operative myocardial injury. New heart rate metrics (e.g. area above a certain threshold) may be valuable, but certainly require validation in other cohorts. Very little is known about the impact of postoperative tachycardia on myocardial ischaemia or death. This is particularly important, as we know that other postoperative events such as hypotension are related with mortality.24 In the meantime, anaesthesiologists and peri-operative physicians should continue to regard tachycardia as a clinical measurement indicative of peri-operative adversity. Acknowledgements relating to this article Assistance with the Invited Commentary: none. Financial support and sponsorship: none. Conflicts of interest: none. Comment from the Editor: this article was checked and accepted by the Editors, but was not sent for external peer-review.