Browse the corpus

We appreciate the thoughtful commentary by Dr. Michele Carron1 on our article describing the potential for intraoperative electroencephalography (EEG) alpha power being independently associated to a patient’s risk of postoperative mortality and adverse outcomes.2 Dr. Carron describes how low intraoperative EEG alpha power may not solely reflect patient vulnerability, but also unintended deep anesthesia that cannot be deciphered without reporting the anesthetic dosing between survivors and nonsurvivors in our study cohort. We agree that the potential confounding introduced by unintended deep anesthesia is necessary to describe, and so we analyzed the differences in propofol and sevoflurane dosing, the two most common anesthetics in our study cohort, between patients who survived and died at 1 yr after surgery. Here we show histograms of the cumulative propofol dosing and sevoflurane dosing throughout the surgery for survivors and nonsurvivors (figs. 1 and 2). Distribution of cumulative propofol dose throughout the surgery for surviving and deceased patients at 1 yr after surgery. The height of each bar is the proportion of patients with that corresponding cumulative propofol dose in order to standardize the distributions between survivor and nonsurvivor groups. Distribution of average sevoflurane dose throughout the surgery for surviving and deceased patients at 1 yr after surgery. The height of each bar is the proportion of patients with that corresponding average sevoflurane dose in order to standardize the distributions between survivor and nonsurvivor groups.

Distribution of cumulative propofol dose throughout the surgery for surviving and deceased patients at 1 yr after surgery. The height of each bar is the proportion of patients with that corresponding cumulative propofol dose in order to standardize the distributions between survivor and nonsurvivor groups. Distribution of average sevoflurane dose throughout the surgery for surviving and deceased patients at 1 yr after surgery. The height of each bar is the proportion of patients with that corresponding average sevoflurane dose in order to standardize the distributions between survivor and nonsurvivor groups. We found that the median (interquartile range) for the cumulative propofol dose throughout the surgery was 300.00 mg (200.00 to 845.02 mg) in survivors and 183.97 mg (100.00 to 311.90 mg) in nonsurvivors, and for average sevoflurane, end-tidal concentration was 0.82% (0.0 to 1.25%) in survivors and 0.86% (0.18 to 1.13%) in nonsurvivors. Generally, we can see that the distributions are highly overlapping between the survivor and nonsurvivor groups. Wilcoxon rank sum tests found that the nonsurvivor cohort had a lower cumulative propofol dose than those who survived (P < 0.001), and that there was no significant difference in sevoflurane dose between groups (P = 0.65). Our original analysis also adjusted for the duration of intraoperative hypotension, which should account for the potential impact of unintended deep sedation on hypotension and subsequent risk of postoperative mortality.

0.001), and that there was no significant difference in sevoflurane dose between groups (P = 0.65). Our original analysis also adjusted for the duration of intraoperative hypotension, which should account for the potential impact of unintended deep sedation on hypotension and subsequent risk of postoperative mortality. In conclusion, we thank Dr. Carron for highlighting the important consideration of anesthetic dosing as a potential confounder in our original findings. Our additional analysis of propofol and sevoflurane administration between survivors and nonsurvivors reveals substantial overlap in dosing distributions, and importantly, no evidence that nonsurvivors received higher doses that would suggest deeper anesthesia. In fact, a lower cumulative propofol dose in nonsurvivors may reflect underlying patient vulnerability rather than anesthetic overdosing, further reinforcing our central hypothesis. While we recognize that observational studies have inherent limitations, we believe our findings strengthen the evidence that suppressed intraoperative EEG alpha power is a reliable marker of heightened postoperative risk, independent of anesthetic depth. Future prospective studies with detailed anesthetic monitoring protocols are warranted to build on these findings and improve patient outcomes through more personalized intraoperative care.

This work was funded in part by the U.S. National Institutes of Health (Bethesda, Maryland; grant Nos. R01AG056015, R42DA053075) to Dr. Purdon; U.S. National Institutes of Health (grant Nos. T32GM007753, T32GM144273) to R. V. Mather; U.S. National Institutes of Health (grant No. F32GM148114) to Dr. Liu; and a charitable gift from Jon Baker Sr. (Wellesley, Massachusetts) to Dr. Kunitake. The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Dr. Purdon is a cofounder of PASCALL Systems, Inc. (Boston, Massachusetts), a startup company developing closed-loop physiologic control systems for anesthesiology. The other authors declare no competing interests.