Fighting against ‘cancel culture’ in cardiac surgery

‘I’m sorry for the late notice, but the surgery will have to be rescheduled’. In healthcare systems with limited human and/or infrastructure limitations, how many times has this very phrase been said to a patient or their family? What follows is typically some combination of surprise, disappointment, anger and even fear of what the additional waiting time will mean for their health. Although clinicians may recognize the psycho-social stress imposed by last-minute case cancellation (albeit the impact of which may not be well understood), the effect on patient clinical outcomes has been less well described. Scheenstra et al. [1] attempt to characterize last-minute cardiac surgery cancellations through an innovative analysis covering an ∼2.5-year period at their centre. The authors found that last-minute cancellations of surgical procedures could be categorized on the basis of one of a few reasons (i.e. medically related in the case of underlying or comorbid illness or process related in the case of logistical difficulties encountered at the level of the institution) and that—depending on the category of cancellation—they are potentially associated with adverse outcomes. This analysis is noteworthy in that it provides insights into opportunities for quality improvement interventions, some of which are based on the authors’ own experiences. Apart from proof of concept for the proposed classification, the main conclusion of this article can be summarized in the statement that not all cancellations are alike. In particular, Scheenstra et al. found that case cancellations for newly identified (or perhaps underoptimized) medical reasons were associated with an increased rate of mortality at 1 year. This finding is consistent with the findings of the seminal work by Shannon et al., who used phase-of-care mortality analysis to identify perioperative root causes or sentinel events attributable to death [2]. In this and other reports incorporating the phase-of-care mortality analysis, systematic errors in the preoperative phase of care (such as under-recognition or lack of optimization of comorbidities) were found to contribute substantially to a cascade of subsequent events culminating in death, many of which may have been avoidable [2]. Furthermore, ‘failure to rescue’ has been advanced as a metric for analysing the probability of death resulting from major complications and is an important component of characterizing adverse events in cardiac surgery [3]. Due to the retrospective nature of this study, it is unclear how many last-minute medically related cancellations were secondary to issues that could have been diagnosed at the time of preoperative consultation/admission versus new issues arising on the day of the operation. In addition, it is not known what treatments were (or were not) initiated. It is difficult, therefore, to understand the degree to which unforeseen medical complications producing or resulting from cancellations could have been identified and managed to alter the patient’s trajectory. Nonetheless, these findings should reinforce the importance of ensuring optimal patient preparation prior to cardiac surgery [4]. Future researchers should seek to understand whether additional monitoring and optimization (e.g. prehabiliation) [5] during the preoperative waiting period can serve to build a patient’s resilience for major cardiac surgical procedures. Another aspect of this study that should be considered is the use of risk-adjusted data incorporating the EuroSCORE II. Because the EuroSCORE II risk model was primarily developed to estimate in-hospital mortality after cardiac surgery and not as a baseline measure of comorbid illness or health status, one needs to interpret the presented data with some caution [6]. The use of the EuroSCORE II as the sole matching criterion in patients whose operations were cancelled is potentially problematic because the various included covariates are weighted based on their estimated impact on postoperative mortality, which does not account for their effect on prognosis in the absence of intervention. Moreover, several procedural factors such as urgency and type of operation that are included in the calculation would not have been of relevance to the 1.4% (n = 2) of patients in the process-related cancellation group and the 28.2% (n = 22) of patients in the medically related cancellation group whose surgical procedures were never rescheduled. In other words, many of the included patients—and a sizeable proportion of the medically related cancellation group—were never at risk for the outcome that the EuroSCORE II aims to estimate (i.e. postoperative mortality) because they did not ultimately undergo surgery. It is therefore reasonable to propose that a more accurate estimation of the impact of last-minute cancellations could be obtained by adjusting for covariates that are known to impact survival in the absence of surgical intervention or by incorporating composite risk scores that do the same. Such an analysis would ideally incorporate disease-specific measures, for example symptom severity/classification in the case of aortic stenosis and coronary artery disease or perhaps frailty. The results of this study are particularly relevant within the current context of the coronavirus disease 2019 (COVID-19) pandemic. Cardiac surgeons and their patients have felt the impact of the pandemic firsthand as healthcare systems the world over have been forced to adapt to increased volumes of hospitalized individuals [7, 8], often resulting in the cancellation of non-urgent surgical cases [9]. During the ‘first wave’ of the pandemic, cardiac surgical volumes decreased in North America to approximately 45% of the prepandemic baseline [10]. These would presumably be considered as process-related cancellations under the proposed classification and therefore unassociated with mortality, provided they could be rescheduled in a timely manner. Of course, therein lies an important caveat because rescheduling COVID-19-related cancellations probably would not resemble the distribution of process-related cancellations in this article. Such analyses will no doubt be forthcoming with many efforts already underway to further characterize the impact of COVID-19 on cardiac surgical care. Altogether, we congratulate the authors on a thought-provoking article that has increased our understanding of last-minute cardiac surgical cancellations and will hopefully contribute to increased discourse surrounding preoperative patient optimization. Conflict of interest: RCA has unrestricted grant from Pfizer Canada Inc and has received honoraria from Abbott Nutrition, Edwards Lifesciences and AVIR Pharmaceuticals for work unrelated to this manuscript.