Hypotension and Postoperative Delirium: Reply

In Reply: We thank Pruc and Szarpak1 for their interest in our study on the association between intraoperative hypotension and postoperative delirium.2 Their thoughtful engagement and valuable insights highlight important considerations and the potential for future research. We agree that recent data suggest that cerebral oxygenation may play a significant role in the development of postoperative delirium. Although blood pressure is one of the determinants of cerebral perfusion and oxygenation, it is neither the sole determinant nor the most significant one.3,4 Previous studies have demonstrated that most of intraoperative changes in cerebral oxygenation are determined by hemoglobin, body temperature, and arterial partial pressure of carbon dioxide rather than by blood pressure.3 Our study focused on mean arterial pressure as a measurable indicator of intraoperative hypotension, because blood pressure management represents a potentially modifiable factor affecting postoperative delirium.5 We recognize that future studies incorporating cerebral oxygenation monitoring, such as near-infrared spectroscopy, could provide a more nuanced understanding of the relationship between intraoperative hemodynamics and postoperative delirium. Importantly, the use of near-infrared spectroscopy is not common in most noncardiac surgeries performed in older adults. Given the real-life nature of our data set, such information is not available. Therefore, investigating the association between cerebral oxygenation and postoperative delirium is beyond the scope of our study. The observations regarding the differential effects of vasopressors such as phenylephrine and ephedrine are highly relevant.6 These agents’ distinct mechanisms of action and their implications on cerebral perfusion warrant further investigation, because they may influence clinical decision-making during surgery. However, as we mentioned in the “Discussion” section, we did not adjust our models to intra- and postoperative variables, even those potentially associated with delirium, because such variables occur at the same time period or after the relevant exposure in the study’s primary analysis (intraoperative hypotension) and as such cannot confound the association of interest.7 As Pruc and Szarpak1 noted, we chose not to elaborate on the potential role of vasopressors or other postoperative variables, such as postoperative hypotension, hypoxia, and the use of sedatives and analgesics, because, as intra- or postoperative variables, they cannot act as relevant confounders of the association between intraoperative hypotension and postoperative delirium. In fact, some of them (e.g., vasopressors used to treat intraoperative hypotension) might serve as mediators of the association of interest. Adjusting for mediators should be avoided, as it might eliminate or diminish true associations.7 Finally, we appreciate the authors’ comment regarding the potential role of blood pressure variability in the relationship between hypotension and postoperative delirium, which had been discussed thoroughly in our article.8,9 Their comment highlights the importance of our advocacy to determine which of the factors, hypotension or blood pressure fluctuations, contributes more significantly to the development of postoperative delirium. Incorporating detailed analyses of blood pressure variability, duration, and intensity of hypotension into future research could enhance our understanding of the thresholds and patterns most closely associated with postoperative delirium. Pruc and Szarpak1 also suggest that various hypotension definitions should be considered because patients might differ on harm thresholds. Of note, evidence is still scarce, and some actually suggest that a universal intraoperative threshold of 65 mmHg is reasonable in patients with chronic hypertension as it is for normotensive patients.10 Future research will hopefully shed light on this controversy, regarding postoperative delirium as well as many other postoperative complications associated with intraoperative hypotension. Again, we thank Pruc and Szarpak1 for their insights regarding our clinically focused study and their letter to the editor. We agree that this topic is of utmost importance and recognize the need for further investigations in this field, which can contribute to a more comprehensive understanding of the association between intraoperative hypotension and postoperative delirium. Competing Interests The authors declare no competing interests.