OBVIOUSLY, we all wish for our patients to do well. We learn about our patients’ comorbidities and develop anesthetic plans that impress colleagues and oral board examiners. We make efforts to influence our patients’ outcomes by implementing protocols for perioperative β-adrenergic blockade and postoperative analgesia. But are there more important factors readily within our control that we are not addressing? Consider the following scenarios:

  • Are you working alone? Are you hesitant to ask a colleague to help as you induce or awaken a patient with a potential difficult airway or a bad heart?

  • Are you too experienced (or arrogant) to use a checklist when putting your anesthesia machine in working order early in the morning? Do you think checklists are only for rookies and paranoid airline pilots?

  • Are you overseeing patients while working as part of an anesthetic team, but too tired to wander to the back of the operating suite to help awaken and extubate an elderly patient who should be fine? You have an experienced member of your team in that room, so why should it matter?

  • Have you avoided using a neuromuscular blockade monitor because it is not very reliable and seems to have a bad battery most of the time? Why worry—today’s muscle relaxants wear off quickly and rarely need to be reversed, right?

In this issue of Anesthesiology, Arbous et al.  1provide a jolting report on the positive impact that anesthesia providers can have on their patients. How? Simple anesthetic management principles seem to have a major effect on perioperative mortality. The routine use of an equipment checklist (odds ratio, 0.61), direct availability of an anesthesiologist to help lend a hand or troubleshoot when needed (odds ratio, 0.46), the use of full-time compared with part-time anesthesia team members (odds ratio, 0.41), the presence of two members of the anesthesia team at emergence (odds ratio, 0.69), and reversal of muscle relaxants at the end of anesthesia (odds ratio, 0.10) had dramatic, positive effects that were associated with reduced perioperative mortality within 48 h after surgery and anesthesia.

This report is remarkable in several ways. First, it is one of the few that have shown anesthetic management processes to dramatically reduce perioperative mortality. Second, it reports perioperative mortality rates matching a number of recent reports. Importantly, it supports the recent insightful article by Lagasse2about perioperative mortality and his suggestion that the US anesthesia community may have overestimated its impact on improving patient safety in the past two decades. Finally, the authors have used a unique and thoughtfully planned multiinstitutional survey and case–control methodology to evaluate this low (but not low enough)–frequency outcome.

It should not be surprising that perioperative anesthetic management processes can make a difference. The US Federal Aviation Administration has long required the use of pilot checklists for evaluating the airworthiness of aircraft and starting procedures, a requirement strongly supported by outcomes of real and simulated air flight. Why would our specialty, so often compared to piloting, be different? The Federal Aviation Administration also requires two pilots for most commercial aircraft operations, nicely matching the report’s finding that the presence of two anesthesia providers at emergence is associated with lower perioperative mortality. The positive impacts of immediate availability of an anesthesiologist when needed and continuity of anesthesia providers in the care of individual patients likewise make sense but, until this study, rarely have been shown to be associated with reduced perioperative mortality.

Have we really overestimated our positive impact on patient safety? Clearly, a number of recent studies suggest that our oft-quoted estimate of 1:200,000 or more patients who have an anesthetic-related death may be flawed.2The basis for this estimate is accurate but usually misinterpreted. Eichhorn et al.  3reported this low rate of anesthetic-related mortality in healthy  patients, an important distinction occasionally neglected in anesthesia patient safety statements. This current study, like others, suggests that the anesthetic-related mortality rate is still too high. The good news is that we have room for improvement and, now, data to support anesthetic management changes that may help.

The study of rare medical events is extremely difficult; it often is extraordinarily frustrating to obtain numerators large enough or denominators that are sufficiently robust to allow calculation of frequencies of the events and subsequent analyses for potential risk factors. Arbous et al.  1have used a multiinstitutional study technique common to clinical research in other medical specialties, notably cardiology, but infrequently attempted in anesthesiology and the study of perioperative mortality. This process has provided the authors with (unfortunately) a sufficient number of perioperative deaths to allow case–control analyses, a good way to seek associations between rare events and potential risk factors.

In general, efforts to seek associations between rare medical events and potential risk factors follow a progression. First, case reports and small case series describe unusual outcomes. If enough of these unusual outcomes can be gathered (typically at least 20 are needed, assuming valid controls can be assessed), a case–control methodology can be used to seek possible but not proven risk factors. Subsequently, potential risk factors identified by case–control studies must be evaluated prospectively in large populations to ascertain their validity. Finally, potential interventions to decrease the frequency of these rare events can be tested in randomized, prospective trials. The current study’s elegant methodology takes advantage of the large numbers of perioperative death reports that they collected in multiple institutions by creatively and prospectively seeking data from randomly selected controls within each of those institutions. This methodology is applicable to many rare perioperative events and should be a model often copied in the future.

Although conclusions from one study should not lead to wholesale changes in practice, the findings in this study support many plausible assumptions that improvements in anesthetic management processes can positively influence patient outcomes. The use of equipment checklists, immediate availability of anesthesiologists to help when needed, especially to provide extra assistance at emergence from anesthesia, and routine reversal of muscle relaxants are processes that should be seriously considered when seeking opportunities to improve the perioperative outcomes of anesthetized patients.

Mark A. Warner, M.D.

Mayo Clinic College of Medicine, Roches-ter, Minnesota. warner.mark@mayo.edu

Arbous MS, Meursing AEE, van Kleef JW, de Lange JJ, Spoormans HHAJM, Touw P, Werner FM, Grobbee DE: Impact of anesthesia management characteristics on severe morbidity and mortality. Anesthesiology 2005; 102:257–68
Lagasse RS: Anesthesia safety: Model or myth? A review of the published literature and analysis of current original data. Anesthesiology 2002; 97:1609–17
Eichhorn JH: Prevention of intraoperative anesthesia accidents and related severe injury through safety monitoring. Anesthesiology 1989; 70:572–7