Home Site Map Contact Us Join ASA Members Only
 
ASA NEWSLETTER
 
 
February 2001
Volume 65
Number 2
   
The Anesthesiologist and Fatigue

Steven K. Howard, M.D.


For a variety of complex reasons, health care personnel have traditionally worked long hours. Fatigue among resident physicians has been studied for many years, but the results of these studies have been mixed, especially regarding the effect of fatigue on performance. No studies have been performed on different age groups of anesthesiologists after residency training, and to summarize in detail the results of the many published articles is beyond the scope of this article. In general, as the hours of sustained wakefulness increase, the studies have shown major detrimental effects on mood, subjective sleepiness (how sleepy you feel) and subtle detrimental effects on psychomotor performance. This has obvious operational implications, as patients rely on us to provide safe care both day and night.

Determinants of Sleepiness

The major determinants of sleepiness are sleep quantity and sleep quality, circadian influences and the use of certain medications. If sleep is restricted below an individual's average normal requirement (which for most adults is greater than eight hours of sleep per 24 hours), then daytime sleepiness will result. This resultant sleep debt builds and continues to pressure the brain physiologically to fall asleep. Superimposed on this is the effect of the biphasic circadian clock that produces a state of increased sleep tendency and decreased performance capacity during two periods in the 24-hour day from about 2 a.m. to 6 a.m. and from about 2 p.m. to 6 p.m. Central nervous system (CNS) depressants (e.g., alcohol) can exacerbate latent sleepiness while CNS stimulants (e.g., caffeine) can mask it. However, the only way to actually pay back the sleep debt is by acquiring adequate sleep.

That sleepiness can have real-world operational importance is supported by recent evidence from other domains. Many studies have shown that automobile accidents oscillate with circadian-induced sleepiness. Studies from the National Aeronautics and Space Administration reveal pilots having microsleeps in the cockpit during critical portions of flight. Similar analysis has rarely been performed for health care accidents, but there is no reason to think that health care workers will be immune to the impairing effects of fatigue.

My laboratory has performed two studies of fatigue in resident anesthesiologists. The investigations include 1) a study of daytime sleepiness and 2) a study of performance using a realistic patient simulator. What follows is a brief description of the two experiments.

Measurement of Daytime Sleepiness

Daytime sleepiness was measured in a group of anesthesiology residents. Daytime sleepiness is routinely measured in sleep laboratories to evaluate patients for sleep disorders such as obstructive sleep apnea and narcolepsy. Daytime sleepiness is measured by a test called the multiple sleep latency test (MSLT). The MSLT measures the ease of falling asleep over the course of a day and is based on the firmly established principle that the sleepier you are, the faster you fall asleep. The sleep latency (time to fall asleep) is determined using specific electroencephalography (EEG) criteria. The MSLT score is the average latency across five attempts to fall asleep over the course of a day. Pathologic MSLT scores are < 5 minutes and are seen commonly in patients with sleep disorders and in healthy people who have been sleep-deprived for 24 hours. Normal MSLT scores are > 10 minutes.

In this study, each subject was observed in each of three conditions. In the baseline condition, the subjects were on typical operating room rotations and had not had an on-call period for at least 48 hours. In the post-call condition, the subjects were on a demanding clinical rotation and were studied immediately after a 24-hour, in-hospital on-call period. In the sleep-extended condition, the subjects were required to extend and maximize their sleep for four consecutive days prior to being tested.

The residents’ average total sleep time (hours ± standard deviation) for the four nights prior to the study period were: baseline 7.1 ± 1.5; post-call 6.3 ± 1.9; and sleep-extended 8.8 ± 1. In this study, subjects were found to have daytime sleepiness at or near pathologic levels in both the post-call condition (4.9 ± 4.7 minutes) and the baseline (6.7 ± 5.3 minutes) condition. The figure shows these data graphically over the course of the day. This documents a previously unknown level of chronic sleep debt in our subjects in the baseline condition even though subjects were averaging over seven hours of sleep per night. It is important to note that this level of daytime sleepiness was reversible, as shown for the sleep-extended condition. When the same subjects were allowed to obtain more sleep, the MSLT scores were normal (12.0 ± 6.4 minutes).

This study also revealed that subjects had poor ability to subjectively perceive their actual level of physiologic sleepiness as measured by the MSLT. They also had trouble perceiving that they had actually fallen asleep even when sleep was recorded unambiguously by the EEG. The study's findings suggest that there may be situations during our clinical work where we become vulnerable to falling asleep, are unable to perceive that this is occurring and do not even know that we have nodded off. This can only increase the risk to our patients or indeed to ourselves as we drive ourselves home.

Figure 1

Simulator Study

The next logical experiment was to test the effects of fatigue on performance using a patient simulator in a realistic environment. We asked the question: How does sleepiness and fatigue affect our ability to perform our job as anesthesiologists? Simulation is a useful technique for studying this question since it poses no risk to patients and allows for extensive manipulation of both the subject and the environment. Simulation also allows for the experiment to be controlled and reproduced between subjects, which is impossible to do with real patients.

We evaluated simulator performance of resident anesthesiologists in each of two conditions. In the sleep-deprived condition, the subjects were kept awake for 24 hours. This replicates a common clinical practice for some experienced anesthesiologists who provide anesthesia care for patients during the day after a busy on-call period. In the sleep-extended condition, the subjects were required to extend their sleep for four consecutive days prior to being tested in the simulator.

The simulations consisted of matched laparoscopic cases of four-hour duration beginning at 9 a.m. and ending at 1:30 p.m. Twenty-seven probes of vigilance were introduced into the scenario at randomly distributed times as were two subtle clinical events per case. All simulations were videotaped for retrospective analysis.

In this complex study, subjects had slightly slower average response times to vigilance probes in the fatigued condition, and certain subjects completely missed some probes because they were clearly asleep. An extensive analysis of 96 hours of the study’s videotapes was performed to quantify profoundly sleepy behaviors (i.e., head nodding or obvious sleep episodes). We found a far greater number and duration of the sleepy episodes in the sleep-deprived group. Two subjects during the experiment had profound sleepiness for a total of more than 25 percent of the four-hour scenario (essentially asleep for one hour!). While all subjects often cycled rapidly between awake and sleepy behaviors, these subjects had several continuous episodes of sleep lasting more than five minutes.

Sleepiness and Fatigue in Society

Throughout the industrialized world and in virtually all transport industries where impaired performance translates into risk of death or injuries, work hours are controlled by regulation to mitigate the effects of sleep loss, fatigue and circadian disruption. In these industries, accidents related to fatigue are not considered isolated occurrences but rather as the expected result of errors and incidents induced by human physiological limitations. The National Transportation Safety Board (NTSB) examines the sleep/wake history of any crew member involved in an accident, and fatigue has been formally identified as a contributing factor and probable cause in major transportation accidents.

In contrast, within health care in the United States, regulation of work hours, schedules and rest requirements are either nonexistent or are markedly less than those considered appropriate by the rest of our society. Great Britain and Australia have addressed the work hours of junior doctors, which has had a significant impact on how physicians are trained in those countries. In many cases, the trainees’ mentors now do the after hours work previously done by trainees. So now who is being sleep deprived?

Unlike pilots, nuclear power plant operators, truck drivers or train engineers, there is no U.S. federal regulatory body that has oversight responsibility for the work/rest practices of physicians. The incentives and disincentives to work while sleepy change with the practice setting. No doubt, trainees look at working the day after-call differently than does an experienced practitioner who is getting paid an incremental fee to care for a patient. This is a complicated issue, and answers to this problem have not and will not come easily.

Generally, the burden of proof has been placed on clinicians and investigators to prove that working long shifts (24-36 hours) and irregular on-call schedules is not safe for patients or clinicians. The case examples in which sleep loss and physician fatigue have been identified as important contributing factors to medical error have been treated as isolated individual lapses rather than as the tip of the iceberg of a widespread and serious risk. Perhaps the time has come for us to reverse the burden of proof in the interests of safety. The life we save might be our own.

Bibliography:

Howard SK, Gaba DM. Human performance and patient safety. In: Morrell RC, Eichhorn JH, eds. A Manual of Patient Safety in Anesthesia. New York: Churchill Livingstone; 1997.

Howard SK, Gaba DM, Rosekind MR, Zarcone VP. Excessive daytime sleepiness in resident physicians: Risks, intervention, and implications. Sleep. 2000 (submitted).

Howard SK, Smith BE, Gaba DM, Rosekind MR. Performance of well-rested vs. highly-fatigued residents: A simulator study. Anesthesiology. 1997; A-943.

Howard SK, Keshavacharya S, Smith BE, et al. Behavioral evidence of fatigue during a simulator experiment. Anesthesiology. 1998; A-1236.

Horne J, Reyner L. Sleep-related vehicle accidents. Br Med J. 1995; 310:565-567.

Pack AI, Pack AM, Rodgman E, et al. Characteristics of crashes attributed to the driver having fallen asleep. Accid Anal Prev. 1995; 27:769-775.

Rosekind M, Graeber R, Dinges D, et al. Crew factors in flight operations: IX. Effects of preplanned cockpit rest on crew performance and alertness in long-haul operations. Vol. NASA Technical Memorandum 108839. Moffett Field, CA: NASA Ames Research Center; 1995.

Mitler M, Dement D, Dement W. Sleep medicine, public policy, and public health. In: Kryger M, Roth T, Dement W, eds. Principles and Practice of Sleep Medicine. 3rd ed. Philadelphia, PA: W.B. Saunders Co; 2000:580-588.



    Steven K. Howard, M.D., is Staff Physician, V.A. Palo Alto Health Care System and Associate Professor, Stanford University School of Medicine, Palo Alto, California


return to top

Home >Newsletters >February 2001Home >Test

 


FEATURES

Data Management:
Pulling It All Together

ARTICLES


DEPARTMENTS


The views expressed herein are those of the authors and do not necessarily represent or reflect the views, policies or actions of the American Society of Anesthesiologists.

NL Archives

Information for Authors