To the Editor:
We read with interest the recent article by Lee et al.1 assessing the association between withholding angiotensin receptor blockers (ARBs) in the early postoperative period and 30-day mortality after noncardiac surgery. They showed that postoperative delay in resuming ARBs was strongly associated with increased 30-day postoperative mortality. Many things of this retrospective analysis are well done, and we can learn from their example. Strengths of this study include a large sample of patients from the Medical Statistical Analysis System and Corporate Data Warehouse files in the Veterans Affairs Informatics and Computing Infrastructure and use of multivariable and propensity score–matched Cox proportional hazards models to determine the independent effect of failure to resume ARBs by postoperative day 2 on the primary outcomes. Furthermore, the authors openly discuss the limitations of their work. However, the two important issues in this study seemed not to be well addressed.
First, the intraoperative risk factors affecting postoperative short-term mortality were not included in the multivariable and propensity score–matched Cox proportional hazards models. Actually, short durations of an intraoperative mean arterial pressure less than 55 mmHg can result in postoperative myocardial and kidney injury, with an independent graded relationship between duration of intraoperative hypotension and postoperative myocardial injury and kidney injury.2 Furthermore, intraoperative massive blood loss, hypotension, blood transfusion, tachycardia, and hypertension have been associated independently with short-term morbidity and mortality after noncardiac surgery.2–5 Typically, based on the estimated blood loss, lowest heart rate, and lowest mean arterial pressure during surgery, the Surgical Apgar score that is originally developed in patients undergoing general and vascular surgery has been shown to be a good predictor of major complications and 30-day postoperative mortality.6 In addition, in comparison with the surgical risk stratification based on only the preoperative factors, the inclusion of intraoperative factors can improve the ability to predict a perioperative mortality and adverse cardiac events.7 In this study, not taking intraoperative risk factors into account would have tampered with the inferences of multivariable analysis and propensity score–matching model for adjusted hazard ratios of 30-day postoperative mortality.
Second, postoperative complications were observed in this study. We noted that in the full cohort, most complications occurred at higher rates in the patients with failure to resume ARBs in the early postoperative period. Even in the propensity score–matched cohort, there are still more noncardiac complications in the patients with postoperative delay in resuming ARBs. Other than early postoperative hypotension, acute kidney injury and myocardial injury, it was unclear why other postoperative complications are not included in the Cox proportional hazards model for adjusted mortality. In fact, postoperative cardiac and noncardiac complications are strongly linked in patients undergoing noncardiac surgery, i.e., patients who experience one type of complication are at an increased risk of developing the other type of complication.8 The available literature provides compelling evidence that pulmonary complications are frequent after noncardiac surgery, especially elderly patients with comorbidities, and postoperative pulmonary complications are associated with increased short-term mortality.9 Furthermore, postoperative heart failure,10 arrhythmias,11 acute ischemic stroke,12,13 delirium,14 and deep vein thrombosis15 have also been shown to be significant independent predictors of decreased short-term survival after noncardiac surgery. Given that these postoperative factors are not included in adjusted potential confounders, we argue that the results of statistical adjustment for postoperative 30-day mortality in this study would have been biased in favor of patients with failure to resume ARBs in the early postoperative period.
The authors declare no competing interests.