Anesthetic Depth Is a Predictor of Mortality

Abstract

It's Time to Take the Next Step " The recognition of the existence of a problem is the first step in its solution. " —Martin H. Fischer I N 2005, a study by Monk et al. 1 was the first to suggest that cumulative deep hypnotic time (cumulative duration of low bispectral index ͓BIS͔) was an independent predictor of 1-yr mortality after major noncardiac surgery. Not surpris-ingly, the " validity of the mortality– hypnosis finding " was questioned, necessitating further investigation. 2 The report by Kertai et al. 3 in this issue of ANESTHESIOLOGY adds to the growing body of evidence confirming the association be-tween cumulative duration of low BIS and postoperative mortality. In this investigation, 17.8% of patients died in the first 3 yr after cardiac surgery, with the risk of death increas-ing by 29% for every cumulative hour for which the BIS was less than 45 during surgery. Other perioperative factors asso-ciated with increased mortality included the EuroSCORE, erythrocyte transfusion, intraoperative infusion of norepi-nephrine, and prolonged intensive care unit stay. As in the current study, previously published reports con-firming the mortality– hypnosis association were all derived from secondary analyses of data collected prospectively for other purposes. 3– 6 Lindholm et al. 4 examined the BIS data from a study originally designed to evaluate the effect of BIS monitoring on the incidence of intraoperative awareness and found that cumulative time at BIS less than 45 was associated with an increased risk of death for up to 2 yr after surgery. However, when preexisting malignancy was included in the analysis, the association between low BIS and mortality was found only in patients with malignancies associated with poorer life expectancies. The investigators in the B-Aware Trial recently reported their secondary analysis of long-term mortality and found that the absence of low BIS values (BIS Ͻ 40) was associated with improved survival and re-duced morbidity (myocardial infarction and stroke). 5 Fi-nally, a post hoc analysis of an observational study originally designed to investigate the use of BIS to monitor conscious-ness in mechanically ventilated, sedated adult intensive care unit patients confirmed this association in a nonsurgical pop-ulation. 6 The authors compared two groups of patients with similar demographic and severity of illness characteristics and found that 39% experienced burst suppression. Patients who experienced burst suppression had a significantly higher 6-month mortality rate compared with patients who did not have burst suppression when sedated (59 vs. 33%). These three studies along with the current report by Kertai et al. 3 indicate that the mortality– hypnosis association is valid and deserves further rigorous investigation. All the previous publications investigating the effect of low BIS on postoperative outcomes have identified preoper-ative comorbidity as an important independent risk factor for postoperative mortality, 1,4,5 leading to the hypothesis that these high-risk patients may have an increased suscepti-bility to anesthetic effects. 1 Kertai et al. 3 also found that increased preoperative comorbidity as defined by the EuroSCORE, a measure of cardiac operative risk, was asso-ciated with increased postoperative mortality. In this study, most patients (83%) with the longest duration of BIS less than 45 (Ͼ 4 h) had abnormal left ventricular ejection frac-tions. Similarly, 67% of these patients were on ␤-receptor blockers before surgery, suggesting a history of chronic hy-pertension. Hypertension and a history of heart disease are two of the most important factors associated with the pres-ence of cerebral white matter lesions and brain atrophy. 7,8