Portal de Periódicos da CAPES

Anesthetic Depth Is Not (Yet) a Predictor of Mortality!

2004; Lippincott Williams & Wilkins; Volume: 100; Issue: 1 Linguagem: Inglês

10.1213/01.ane.0000147507.23991.47

1526-7598

Neal H. Cohen,

Anesthesia and Neurotoxicity Research

The report by Monk et al. (1) in this issue of Anesthesia & Analgesia addresses the potential impact of anesthesia on morbidity and mortality. Currently the information available to us suggests that the morbidity and mortality associated with anesthesia is infrequent. The frequency of deaths attributable to anesthesia is reported to be <1 in 200,000 anesthetics (2,3). Whether this incidence reflects the true mortality associated with anesthesia is not clear for several reasons. First, the cause of perioperative death is not always clear and the relationship to anesthesia and intraoperative management is often hard to define. Second, very few patients die in the operating room. A common scenario is that the unstable patient is transferred to the critical care unit for further care and stabilization. If the patient then dies, the cause of death may not be thought to relate to anesthetic management or to intraoperative events under the control of the anesthesiologist. Finally, for the sickest patients and those undergoing complex surgical procedures, the relationship between anesthetic management and perioperative outcome is difficult to define. As a result, most deaths and many complications may be attributed to underlying pathology or physiology rather than to management strategies under the control of the anesthesiologist in the operating room. The impact of anesthesia on morbidity is even more difficult to assess, but it is important for anesthesiologists to understand if we are to improve outcomes associated with our care. Recently the influence of anesthetic management and intraoperative events on long-term outcome has generated intense interest (4). Many studies have demonstrated that preoperative and intraoperative management strategies impact postoperative outcome. The perioperative administration of adrenergic β-blockers has become standard as a result of studies that demonstrate improved outcome associated with their use (5,6). Also, intraoperative management strategies might influence outcome. For example, intraoperative glucose control reduces postoperative infections (7–9), improves wound healing (10,11) and, most importantly, decreases mortality (12–14). In the article by Monk et al. (1), the relationship between anesthesia itself and outcome is addressed more directly. They report on the findings of a study designed to evaluate mortality after major noncardiac surgery. They found, not surprisingly, that patient comorbidity and intraoperative hypotension are predictors of death, although in some cases mortality did not occur until months after the surgical procedure. These two factors probably represent markers of underlying physiologic instability for which increased morbidity and mortality might be expected. The study describes another factor: the cumulative deep hypnotic time, which the authors found is associated with increased mortality. Although underlying clinical conditions and intraoperative hypotension might be expected to be associated with poorer outcome, both initially and perhaps up to 1 yr after a major surgical procedure, the importance of anesthetic duration and depth is interesting and somewhat surprising. The findings of this study are important and emphasize the need to more carefully evaluate the impact of intraoperative management strategies on outcome. The most interesting aspect of this study is the finding related to the cumulative anesthetic time. The importance of anesthetic depth on intraoperative and postoperative outcomes has been emphasized for some time. For most patients, the primary concern related to anesthesia is the level of consciousness, potential inadequate anesthesia and recall of events in the operating room. Anesthesiologists have the same concerns and want to ensure adequate depth of anesthesia for both clinical and professional liability reasons. In addition, anesthetic depth is often used as a tool to provide better control of arterial blood pressure or other hemodynamic variables. Monk et al. suggest that anesthetic depth may have negative impact on patient outcome and should perhaps be minimized to prevent previously unanticipated complications and morbidity. The impact of this study, should the findings be validated, could result in major changes in how anesthesia is provided, what anesthetics are used, and how hemodynamic variables are controlled during surgery. One of the most significant aspects of this investigation is the description of a relatively new measure of anesthetic depth, the cumulative deep hypnotic time. It is not entirely clear how the authors were able to prospectively evaluate deep hypnotic time, but apparently the use of this measure of anesthetic depth is based on a study that demonstrated that patients whose anesthetic depth was titrated to maintain a Bispectral Index (BIS) value between 45 and 60 had faster emergence, less drug usage and more postoperative alertness than those who did not have their anesthetic depth titrated to a specific range (15). This measure of anesthetic depth may be an important variable of anesthetic management, but before we accept cumulative deep hypnotic time as an important measure, its relevance to other long-term morbidity and mortality requires much more detailed evaluation. Because the clinicians were blinded to the BIS data and titrated anesthetic depth based on other measures of anesthetic depth, including traditionally monitored physiologic changes, the validity of this measure is unclear. Its independent relationship to mortality warrants further validation. Perhaps as important as the description of the cumulative deep hypnotic time is the implication that a monitor of anesthetic depth is critical to patient management. This study, although identifying a potential relationship between BIS value and mortality, does not necessarily establish the BIS value as an independent predictor, as the authors did not perform any power analysis or other determination to define the range, mean score, or the significance of the duration of the BIS value at specific levels. Would one patient whose BIS value never decreased to <59 have the same outcome as a patient whose BIS value remained 45 for 2 h of the surgical procedure? Are there other reasons why the clinicians felt the need to deepen the anesthetic—related to the surgical procedure, to physiologic variables, or to other requirements—that result in an association between BIS value and anesthetic depth but that do not reflect the independent nature of the BIS score alone? If increasing concentrations of the volatile anesthetic are really related to mortality, and should anesthetic be used to control blood pressure or other physiologic variables, would alternative approaches that achieve the same hemodynamic outcomes be preferable? What alternatives are available to the clinician caring for this patient population? In addition, the authors must define how they would reduce the cumulative deep hypnotic time for a patient undergoing a lengthy surgical procedure for which deep anesthesia may be required. Furthermore, the term “cumulative deep hypnotic time” is entirely determined by the BIS number. Is that appropriate? Despite the reservations about reliance on the deep hypnotic time to minimize perioperative morbidity and mortality, the findings in this study are indirectly supported by other recent studies. Lennmarken et al. (16) provided a preliminary report that demonstrated that low intraoperative BIS levels were associated with increased postoperative mortality, as has been demonstrated in this study. In addition a recent study conducted by Liu (17) assessed the value of the BIS monitor during ambulatory surgery. In that study, Liu demonstrated a modest reduction in anesthetic consumption when the BIS monitor was used to assess anesthetic depth. The reduction in anesthetic consumption is most likely associated with reduced anesthetic depth. Although mortality was not evaluated in the study, it did demonstrate a reduction in postoperative nausea and vomiting and shorter recovery time. Mortality was not evaluated and would be expected to be infrequent for ambulatory procedures. However, could the reduced consumption and depth of anesthesia minimize other perioperative complications? Additional studies of deep hypnotic time in other patient populations may be able to answer that question. Another important consideration in evaluating the findings of this study is the fact that, as noted by the authors, the study had a number of different facets. In fact, the preoperative evaluation included extensive assessment of preoperative cognitive function and neurologic status. Because the authors did not report any of the findings related to cognitive function as part of this study, the actual purpose for which the study was completed is not clear. The extent to which the findings related to the deep hypnotic time might be an interesting association without apparent causal relationship will require further investigation. In addition to the concerns about the study design and its impact on interpretation of the significance of the deep hypnotic time, the exclusion criteria might have affected the findings. Because the investigators planned to assess impact of anesthetic management on cognitive function in their study design, they excluded patients with cognitive impairment and all procedures that might affect postoperative cognitive function. As a result, the “major surgical procedures” that are included in the study might not be representative of a large enough patient population to make the conclusions meaningful. Had the study included a more comprehensive group of surgical patients, the findings might have had greater impact. In addition, as many high-risk procedures were not included in the study population, the 1-yr mortality for all patients and those older than 65 years of age is frequent. The impact of this smaller subset of patients on the findings warrants further investigation. In defining potential independent variables that might be predictive of mortality, the authors collected only those variables that they thought might have a potential effect on mortality. Their list is interesting and includes some obvious potential predictors of mortality based on previously reported data. They evaluated some predictors, however, that might be a better reflection of underlying clinical condition or other factors unrelated to the one being evaluated. Few investigators would have selected cumulative deep hypnotic time as a potential predictor of mortality, particularly because its meaning is unknown and the range is relatively arbitrary. Finally, as noted in the manuscript, one of the authors of this manuscript is employed by Aspect, the manufacturer of the BIS monitor. Dr. Sigl and Aspect did not participate in the design or implementation of the study, but they assisted with the statistical analysis of the data and with the preparation of the manuscript. With Dr. Sigl’s assistance the authors were able to define the cumulative deep hypnotic index—a somewhat arbitrary measure of anesthetic depth, as noted. Whether this measure of anesthetic depth has validity and whether the measure has the same impact if it is achieved with volatile anesthetics or other drugs is unknown. Despite questions about the value of the cumulative deep hypnotic time as a predictor of 1-year mortality, the authors have identified a potentially interesting factor associated with anesthetic management that might have impact on outcome. The findings in this study are interesting and warrant our consideration. Further validation of the cumulative deep hypnotic time and the important contributors to it will be essential if the measure is to have predictive value or impact on anesthetic management of patients undergoing major surgical procedures. This study may help to stimulate more objective and thorough analyses of this important issue so that anesthesiologists have better information from which to make clinical decisions about anesthetic depth and the management of hemodynamic variables during surgery.