This Edition Of The APSF Newsletter - Anesthesia Patient Safety

Inside: Performance Enhancing Drugs......................................................................................Page 5Ventilator Failure (Dear Sirs) ........................................................................................Page 6Cure for the Dopey Doc?..................................................................................................Page 9Managing Fatigue............................................................................................................0Britain & Ireland’s Approach to Fatigue ..................................................................Page 13Costs of Fatigue ..............................................................................................................Page 15Donors ................................................................................................................................Page 17Report on Long-Term Outcome Workshop................................................................Page 18ACS Collaborates With APSF ......................................................................................Page 21

NEWSLETTERVolume 20, No. 1, 1-24 Circulation 75,648 Spring 2005

www.apsf.org The Official Journal of the Anesthesia Patient Safety Foundation

Fatigue and SafetyFatigue has played a causal or contributory role

in some famous accidents.1 In 1986, the PresidentialCommission found that faulty decision-making bysleep-deprived managers contributed to the unto-ward launch of the space shuttle Challenger. Thenuclear accidents at Three Mile Island and Cher-nobyl both occurred during the early morninghours when our bodies are craving sleep. Thegrounding of the Exxon Valdez was a monumentalenvironmental catastrophe. The National Trans-portation Safety Board found that the probablecause of this accident was the fatigue of the personsailing the ship. The National Highway TrafficSafety Administrations estimates that over 100,000people are killed or injured each year in crashesattributed to drivers who fell asleep at the wheel orwere impaired by severe drowsiness. These exam-ples and many others reveal that fatigue is a prob-lem that extends beyond health care and is deeplyembedded within our society.

Studies have shown a correlation between theperformance effects of sleep deprivation andethanol intoxication.2 At 24 hours of continuouswakefulness, psychomotor function was equivalentto a blood alcohol concentration of 0.1%. This is ator above the legal limit for driving in most states.Think of the professional and personal liability ofcoming to work intoxicated!

Anesthesia providers, like all health careproviders, are required to care for patients whenthey present for care—anytime of the day or night.This is often in opposition to what our physiologydemands. An irrefutable fact is that fatigue andsleep deprivation negatively impact performanceand mood (see Table 1). In fact, the anesthesiolo-gist’s role of monitoring the patient in a vigilantmanner may be particularly vulnerable to the effectsof fatigue.3 Vigilance is defined as the act of beingalertly watchful, especially to avoid danger. Theword “vigilance” is at the center of the seal and isthe motto of the American Society of Anesthesiolo-gists. If we become disengaged from our environ-ment (such as the “microsleeps” that happen whenwe are sleep-deprived), all vigilance is lost.

®

Everyone has seen it. For example, watching a colleague fallasleep at a meeting or watching an intern struggle to remain alert whileholding a surgical retractor. Everyone has felt it. Eyelids get heavy and theenvironment starts to “grey out.” Ask yourself if you desire to be cared for byhealth care workers who look and feel this way. This clearly is a dangeroussituation for our patients. It is also unsafe for the practitioner when you consider the possibility of harm due to occupational injury (e.g.,needlesticks) and the increased risk of driving while sleepy.

This edition of the APSF Newsletter will focus on fatigue and the anesthesia careprovider. There is renewed interest in this topic, and we have gathered a cadre ofindividuals who will present important new information on this topic. Anesthesiol-ogy has been very forward-looking regarding many aspects of safety, and there is againan opportunity to be at the “cutting edge” in dealing with this pervasive problem. Wehope that the material in this issue will encourage others in our field to join with us tochange the manner in which we practice and care for patients.

See “Fatigue,” Page 3

Fatigue & the Practice of Anesthesiologyby Steve Howard, MD, Guest Editor

APSF NEWSLETTER Spring 2005 PAGE 2

The Anesthesia Patient Safety FoundationNewsletter is the official publication of the nonprofitAnesthesia Patient Safety Foundation and is publishedquarterly at Wilmington, Delaware. Annual member-ship: Individual - $100.00, Corporate - $500.00. This andany additional contributions to the Foundation are taxdeductible. © Copyright, Anesthesia Patient SafetyFoundation, 2005.

The opinions expressed in this Newsletter are notnecessarily those of the Anesthesia Patient SafetyFoundation or its members or board of directors. Validityof opinions presented, drug dosages, accuracy, andcompleteness of content are not guaranteed by the APSF.APSF Executive Committee:

Robert K. Stoelting, MD, President; Jeffrey B. Cooper,PhD, Executive Vice President; George A. Schapiro,Executive Vice President; David M. Gaba, MD, Secretary;Casey D. Blitt, MD, Treasurer; Sorin J. Brull, MD, RobertA. Caplan, MD, Nassib G. Chamoun, Robert C. Morell,MD, Michael A. Olympio, MD, Richard C. Prielipp, MD.Newsletter Editorial Board:

Robert C. Morell, MD, Editor; Sorin J. Brull, MD;Joan Christie, MD; Jan Ehrenwerth, MD; John H.Eichhorn, MD; Regina King; Lorri A. Lee, MD ; RodneyC. Lester, PhD, CRNA; Glenn S. Murphy, MD; DeniseO’Brien, BSN, RN; Karen Posner, PhD; Keith Ruskin,MD; Wilson Somerville, PhD; Jeffery Vender, MD.

Address all general, membership, and subscriptioncorrespondence to:Administrator, Deanna WalkerAnesthesia Patient Safety FoundationBuilding One, Suite Two8007 South Meridian StreetIndianapolis, IN 46217-2922e-mail address: [email protected]

Address Newsletter editorial comments, questions, letters, and suggestions to:Robert C. Morell, MDEditor, APSF Newsletterc/o Addie Larimore, Editorial AssistantDepartment of AnesthesiologyWake Forest University School of Medicine9th Floor CSBMedical Center BoulevardWinston-Salem, NC 27157-1009e-mail: [email protected]

The APSF continuesto accept and

appreciatecontributions.

Please make checks payable to the APSF

and mail donations to:

Anesthesia Patient Safety Foundation (APSF)520 N. Northwest HighwayPark Ridge, IL 60068-2573

Check out our newand improved

website atwww.apsf.org

• Grant Guidelines• Search Engine• Back Issues of Newsletter• Safety Links• Sponsor List• Foundation Information

User-Friendly & Informative!

George A. Schapiro, Chair

Abe Abramovich ....................Datascope Corporation

Michael Argentieri ................Dräger Medical, Inc.

Paul A. Baumgart ..................GE Healthcare

Casey D. Blitt, MD

Judson Boothe ........................Kimberly-Clark

Nassib G. Chamoun ..............Aspect Medical Systems

Stuart M. Cohen, MD ............Cerner Corporation

Philip Croxford ......................Arrow International

Joseph Davin ..........................Spacelabs Medical

William T. Denman, MD ......Tyco Healthcare

Cari Duncan ............................Organon

Robert Evans, MD ..................Docusys, Inc.

Michael S. Ferrara ..................Becton Dickinson

William Floyd ........................Aspect Medical Systems

Bruce Gingles ..........................Cook Critical Care

David Goodale, PhD, DDS....AstraZeneca Pharmaceuticals LP

Elizabeth Holland ..................Abbott Laboratories

Walter Huehn ........................Philips Medical Systems

Regina King ............................Baxter Anesthesia andCritical Care

Charles McLeskey, MD ........Abbott Laboratories

Steven C. Mendell ..................LMA of North America

Edward C. Mills......................Preferred PhysiciansMedical Risk Retention Group

Brian Mulnix ..........................Endo Pharmaceuticals

Lou Obendorf..........................Medical Education Technologies, Inc.

Patty Reilly ..............................Tyco Nellcor

Marcus Schabacker, MD........B. Braun Medical Inc.

Laurie Schechter ....................Smiths Medical

Maria Shepherd ......................Oridion

Robert K. Stoelting, MD

Robert A. Virag ......................Trifid Medical Group, LLC

Barry Wicker ..........................Vital Signs

A N E S T H E S I A P A T I E N T

S A F E T Y F O U N D A T I O N

CORPORATE ADVISORY COUNCIL

Contact Information

Anesthesia Patient Safety FoundationBuilding One, Suite Two8007 South Meridian StreetIndianapolis, IN 46217-2922Telephone: 317-885-6610Facsimile: 317-888-1482

Administrator:Deanna M. WalkerE-mail: [email protected]

President: Robert K. Stoelting, MDE-mail (President): [email protected]

Fatigue and HealthPeople who work irregular overnight shifts are

at increased health risks.4 Gastrointestinal and car-diovascular complaints increase, and there is someevidence to suggest that long, strenuous work isalso associated with obstetric complications (e.g.,preterm delivery, pregnancy-induced hyperten-sion). Immune function, carbohydrate metabolism,and endocrine function are impaired to somedegree. One study revealed an increased mortalitywith severe shortening of daily sleep—sleep of <4hours increased likelihood of death by a factor of2.8 in a 6-year follow-up period.5 The health effectsof chronic sleep deprivation are insidious and theirimpact may not manifest fully for years.

Physiology of FatigueSleep Homeostasis

Sleep homeostasis can be thought of as sleepbalance. Every adult has a genetically hard-wiredsleep requirement that does not change with ageand cannot be trained. The average sleep need foradults is over 8 hours per 24 hours, and most in oursociety do not achieve this requirement. In fact, sur-veys done by the National Sleep Foundation revealthat we are a society of chronic under sleepers byover an hour per night. Sleep is a physiologic drivestate similar to hunger or thirst. When sleeprequirements are not met a “sleep debt” ensues andsleepiness becomes manifest. The only way to payoff this sleep debt is by acquiring adequate sleep.Laboratory studies of chronic partial sleep depriva-tion provide important information to practicingclinicians.6 Research has shown that sleeping 6hours or less per night over 2 weeks results in cog-nitive performance deficits equivalent to 2 nights oftotal sleep deprivation. Data on subjective sleepi-


“Fatigue,” From Page 1 shift attempt to function when their clocks areprimed for sleep. When they attempt to sleep dur-ing the day, their clock is programmed for wakeful-ness. Opposing this normal rhythmic pattern ofday-awake and night-asleep is made more difficultbecause our society’s activities are strongly linkedto this pattern. Some errands and family responsi-bilities can only be performed during the day, mak-ing adaptation to shift work difficult.

Sleep DisordersOver 80 sleep disorders have been described.

Common complaints include insomnia, excessivedaytime sleepiness, and abnormal movements orbreathing during sleep. Some of these disorders arerelatively common and have documented negativeeffects on wake performance. The most commonsleep disorders are obstructive sleep apnea (OSA),insomnia, and periodic limb movements. Shift worksleep disorder has gained attention recently and hasan approximate incidence of 10% in night and rotat-ing shift work populations.11

OSA is of special significance because it isprevalent in the population (at least 3-5%), but stillunderdiagnosed. It is associated with the risk fac-tors of obesity (BMI >27), habitual snoring, andlarge neck circumference, and is most commonlyfound in middle-aged males. Excessive daytimesleepiness is a common complaint of patients withOSA and occurs because of the cyclical awakeningscaused by airway obstruction during sleep. Peoplewith OSA have great difficulty getting restorativesleep because it is like having a pager go off 100times during each night. The incidence of automo-bile accidents increases in people with OSA, andexperiments reveal impairments that are equivalentto ethanol intoxication.

Other Factors That AffectWaking Function

Depressant and stimulant drugs, whether pre-scription drugs or substances of abuse, have obvi-ous effects on alertness and performance. In low tomoderate doses, caffeine can transiently improvealertness and performance. Social interaction, restbreaks, and exercise can all improve subjectivealertness and performance. These factors have noeffect on the underlying physiologic level of sleepi-ness. Other performance shaping factors includechronic or acute illness, boredom, noise, extremes oftemperature, and lack of environmental stimuli.

Literature ReviewThere have been many studies on the effect of

sleep deprivation and fatigue on the mood and per-formance of health care personnel. Extensivereviews on this topic have been published in thelast 5 years and the following highlight the impor-tant findings.12,13

ness (how sleepy the subjects felt) suggest that sub-jects are largely unaware of their level of impair-ment; this may explain why the impact of this levelof chronic sleep deprivation is assumed to be benign.

Many anesthesiologists and nurse anesthetistsare chronically sleep-deprived at baseline and peri-odically push the envelope further by performingcall duties, thus layering acute sleep deprivation ontop of a significant sleep debt. When the clinicianworks on the day after a busy on call night, a poten-tially catastrophic situation can exist.

Normal aging affects sleep.7 Disruptionsbecome more frequent, the amount of slow wave(restorative) sleep decreases, and sleep becomes lessconsolidated. Since sleep needs remain essentiallyunchanged, the result is daytime sleepiness. Thiscan be countered by daytime naps, but few workingpeople are able to take naps in our society.

Consecutive Hours ofWakefulness

As hours since the last sleep episode accrue, thephysiologic pressure to acquire sleep increases.Extending work shifts past 17 consecutive hoursincreases the likelihood of errors in laboratory6 andclinical settings.8 In a recent study reported in theNew England Journal of Medicine, interns rotating inthe intensive care unit were evaluated on two dif-ferent schedules.9 The traditional schedule allowedfor work of up to 30 consecutive hours, while theintervention schedule limited the number of consec-utive hours to <17. In this well designed study, resi-dents on the reduced schedule obtained more sleepand made 36% fewer errors. The intervention groupalso showed fewer “attentional failures,” whichwere actual electroencephalographic (EEG)episodes of sleepiness during patient care.8 Thesefindings match those of many laboratory studies:wakefulness for periods of over 16 hours predictsperformance lapses.6

Circadian FactorsHumans have a circadian timer that regulates

many body processes such as temperature and hor-mone secretion, but for this discussion the 24-hrsleep-wake cycle is of importance. The circadianclock is located in the suprachiasmatic nucleus ofthe hypothalamus. The daily light/dark variationentrains this clock to the 24-hour day. We are pro-grammed for 2 periods of decreased alertness every24 hours: between 3-7 AM and 1-4 PM. The lowestpoint in this cycle occurs during the early morningone, making it the period of greatest vulnerabilityto fatigue-related performance impairment.10

As we know from shift work and transmeridianair travel, the circadian pacemaker is resistant tochange. This is the primary reason for the inabilityof humans to readily adapt to shift work and whyjet lag occurs. For example, workers on the night

Rested Residents Make Fewer Errors

Table 1: Fatigue Effects on Human Functioning

■ Cognitive slowing

■ Impaired vigilance

■ Increased performance variability

■ Neglect of non-essential activities

■ Learning of new information decreases

■ Problem solving decays

■ Memory degrades

■ Motivation declines

■ Sleep intrudes into wakefulnessSee “Fatigue,” Next Page


compelling real-world evidence of how lack ofsleep leads to decreased alertness and impacts ourabilities to perform.

To summarize, we face many challenges in ourcurrent health care environment. As you will read inthe accompanying articles in this special issue of theAPSF Newsletter, finding solutions to the issues thatrevolve around fatigued health care personnel arenot simple and clear-cut. We owe it to our patientsand ourselves to come to work optimally prepared.We can lead the way for all of health care by helpingto solve the problem.

Dr. Howard is a Staff Physician at the VA Palo AltoHealth Care System and Director of the Patient SafetyCenter of Inquiry at VA Palo Alto. He is also an Associ-ate Professor of Anesthesia at the Stanford UniversitySchool of Medicine, Stanford, CA.

References

1. Mitler MM, Dement WC, Dinges DF. Sleep medicine,public policy, and public health. In: Kryger MH, et al.,eds. Principles and Practice of Sleep Medicine, 3rd ed.Philadelphia: W.B. Saunders Company, 2000: 580-8.

2. Dawson D, Reid K. Fatigue, alcohol and performanceimpairment [Scientific Correspondence]. Nature1997;388:235.

3. Weinger MB, Englund CE. Ergonomic and human fac-tors affecting anesthetic vigilance and monitoring per-formance in the operating room environment.Anesthesiology 1990; 73:995-1021.

4. Costa G. Shift work and occupational medicine: anoverview. Occup Med (Lond.) 2003;53:83-8.

5. Kripke DF, Simons RN, Garfinkel L, Hammond EC:Short and long sleep and sleeping pills. Is increased mor-tality associated? Arch Gen Psychiatry 1979;36:103-16.

6. Van Dongen HP, Maislin G, Mullington JM, Dinges DF.The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleepphysiology from chronic sleep restriction and total sleepdeprivation. Sleep 2003;26:117-26.

7. Bliwise DL. Normal aging. In: Kryger MH, et al., eds.Principles and Practice of Sleep Medicine, 3rd ed.Philadelphia: W.B. Saunders, Co., 2000: 26-42.

8. Lockley SW, Cronin JW, Evans EE, et al. Effect of reduc-ing interns' weekly work hours on sleep and attentionalfailures. N Engl J Med 2004;351:1829-37.

9. Landrigan CP, Rothschild JM, Cronin JW, et al. Effect ofreducing interns' work hours on serious medical errorsin intensive care units. N Engl J Med 2004;351:1838-48.

10. Van Dongen HP, Dinges DF. Circadian rhythms infatigue, alertness, and performance. In: Kryger MH, etal., eds. Principles and Practice of Sleep Medicine, 3rded. Philadelphia: W.B. Saunders, Co., 2000: 391-9.

11. Drake CL, Roehrs T, Richardson G, et al. Shift worksleep disorder: prevalence and consequences beyondthat of symptomatic day workers. Sleep 2004;27:1453-62.

12. Weinger MB, Ancoli-Israel S. Sleep deprivation andclinical performance. JAMA 2002;287:955-7.

13. Owens JA. Sleep loss and fatigue in medical training.Curr Opin Pulm Med 2001;7:411-8.

14. Pilcher JJ, Huffcutt AI. Effects of sleep deprivation onperformance: a meta-analysis. Sleep 1996;19:318-26.

15. Howard SK, Gaba DM, Smith BE, et al. Simulationstudy of rested versus sleep-deprived anesthesiologists.Anesthesiology 2003;98:1345-55; discussion 5A.

16. Barger LK, Cade BE, Ayas NT, et al. Extended workshifts and the risk of motor vehicle crashes amonginterns. N Engl J Med 2005;352:125-34.

Subjective ReportsAnesthesia personnel report working long

hours, often without taking breaks. In these sur-veys, over half the respondents reported havingcommitted an error in judgment due to fatigue andfeel that fatigue impairs patient safety.

MoodMood is found to be consistently worse as work

hours are extended and with work done at night.Levels of anger, hostility, tension, bewilderment,confusion, fatigue, anxiety, and depression increasewhile measures such as vigor and happinessdecrease. No studies have been done to evaluatehow these mood shifts impact patient care, but theylikely play a role in decreased job satisfaction andthe occurrence of burnout.

PerformanceMeta-analyses have been performed concerning

the effects of sleep loss on performance.14 Theseanalyses suggest that sleep deprivation impairs cog-nitive performance and mood with less effect onmotor performance. Known performance effectsinclude reduced vigilance, impaired memory, pro-longed reaction time, and poor communication. Per-formance also becomes more variable—one momentperformance is adequate, followed by perceptualdisengagement from the environment at sleep onset.Performance is zero upon drifting into sleep.

Sleep During Patient CareAmbulatory EEG and videotape of clinicians

have been used to quantify sleep in field and simu-lator studies. “Attentional failures” are greater ininterns working 30-hour duty periods when com-pared to shorter work shifts.8 Studies of anesthesi-ologists providing care to a simulated patient revealsimilar data.15 Based on detailed videotape analysisof “sleepy behaviors” (eyes closing, head nodding,and actual sleep), the most impaired anesthesiolo-gist in this study had these behaviors for over 30%of a 4-hour case! These studies support what iscommonly seen in the operating room, the class-room, and in conference rooms in hospitals aroundthe world. People manifest severe levels of sleepi-ness at critical time periods.

Driving StudiesA growing number of studies describe an

increased risk of drowsy driving in health careproviders. A recent study from the New EnglandJournal of Medicine found that interns were 2.3 timesas likely to report a crash after working anextended shift.16 They calculated that for everyextended work shift scheduled in a month, the riskof a motor vehicle crash increased by 9.1%. This is

Lack of Sleep Also Endangers Anesthesia Providers

Steve Howard, MD, Guest Editor

“Fatigue,” From Preceding Page

APSF GrantApplication

Information Available

on Our Website

�8 �Applications must

be submittedelectronically to the

website and received by

Monday, June 20, 2005


diagnosed with narcolepsy, obstructive sleep apneasyndrome, and shift work sleep disorder.Modafinil’s mechanisms of action and long-termeffects are unknown. Modafinil is distinctly differ-ent from stimulants such as methylphenidate,amphetamines, and cocaine. Elite athletes werequick to discover modafinil. In November 2004, theU.S. Olympic gold medal for the 1600 meter relaywas revoked and given to France because US ath-lete Calvin Harrison tested positive for modafinil.The International Associations of Athletics Federa-tions and WADA consider modafinil a perfor-mance-enhancing drug. Thus a third argumentagainst the use of performance-enhancing drugs isthe long-standing legal precedent of condemnationby the international athletic community.

Discovering molecules that enhance cognition(nootropics) comprises one of the most excitingareas in drug development. There is uniform sup-port for use of these drugs as therapeutic agents.Nootropic drugs will also be used by healthy indi-viduals as enhancing agents. Perceptions about theappropriateness of performance-enhancing druguse is contextual. The International Associations ofAthletics Federations and WADA consider druguse as cheating. In contrast, the Defense AdvancedResearch Projects Agency (DARPA) actively sup-ports performance-enhancing drugs for militaryuse. The DARPA website notes that, “Sleep depri-vation is a fact of modern combat. Current opera-tions depend on a warfighter’s ability to functionfor extended periods of time without adequate

sleep.” By changing only 2 words in the DARPAquote, one can describe a performance challengeoften encountered by health care providers.

Performance outcome also influences attitudesconcerning drug enhancement. The 2003 tragedy inwhich Canadian ground troops were killed by USpilots who had taken amphetamines has been inter-preted as a condemnation of performance-enhanc-ing drugs. For decades, US astronauts have usedamphetamines therapeutically to suppress nauseaand to enhance performance. Concerns over druguse have never sullied celebrations of a successfulspace mission. Likewise, if a clinical case goes well,and the anesthesia provider has taken a perfor-mance-enhancing drug, the probability of repercus-sions is low. One can easily envision a verydifferent response to the same scenario terminatingin a negative clinical outcome. Ultimately, perfor-mance-enhancing drug use will be decided not byeditorials or ethicists but by the drug-using public.For performance-enhancing drugs, presently thereis no rational code, but there is a clear coda: thegenie is out of the bottle.

Ralph Lydic, PhD, is the Bert La Du Professor andAssociate Chair for Research in the Department of Anes-thesiology at the University of Michigan, Ann Arbor,MI. He is supported by grants HL57120, HL40881, andHL65272 from the National Institutes of Health,Bethesda, MD, and by the Department of Anesthesiology.

Reference

1. Juliano LM, Griffiths RR. A critical review of caffeinewithdrawal: empirical validation of symptoms andsigns, incidence, severity, and associated features. Psy-chopharmacology (Berl) 2004;176:1-29.

by Ralph Lydic, PhD

I thank Steven K. Howard for the invitation tocontribute to this APSF Newsletter consideration offatigue as a potential detriment to anesthesia safety.The editorial assignment was to outline argumentsagainst use of performance-enhancing drugs as acounter-measure for excessive physician fatigue.Advances in neuroscience and economic impera-tives ensure the increasing relevance of perfor-mance-enhancing drugs. This is a topic presentlycharacterized by a paucity of objective data, unsci-entific argument, and hypocrisy. Some perfor-mance-enhancing drugs are willingly accepted, yetother drugs are condemned. Most clinical facilitiesnow maintain a “smoke-free” environment. Onecan easily envision the outrage that would be elicitedby proposals for a “caffeine-free” environment.Caffeine is the most widely consumed performance-enhancing drug. Recent reviews suggest that rapidcessation of caffeine intake causes a “clinically sig-nificant distress or impairment in functioning” thatwarrants “inclusion of caffeine withdrawal as anofficial diagnosis” by the World Health Organiza-tion (ICD-10) and the American Psychiatric Associ-ation (DSM-IV).1 This article highlights someconventional arguments against the healthy indi-viduals’ use of drugs that enhance performance(steroids), arousal (modafinil), and cognition(nootropics).

Anabolic steroids enhance performance buthave negative side effects that illustrate the counter-argument of excessive medical risks. The Interna-tional Olympic Committee relies on a WorldAnti-Doping Agency (WADA) for drug testing. Theneed for WADA is illustrated by German court con-victions of former East German government offi-cials who sponsored use of performance enhancingsteroids by their Olympic athletes. In March 2005,leaders of a laboratory co-operative are scheduledto stand trial for a number of charges including thealleged synthesis and sale to elite athletes of a pre-viously undetectable steroid called tetrahydro-gestrinone (THG). High profile athletes exert anenormous influence on younger, aspiring athletes.A second argument against performance-enhancingdrug use is that it encourages drug use by others.Dr. Linn Goldberg, professor of medicine at OregonHealth and Science University, reported on the 10December 2004 radio show Talk of the Nation thatabout one million teenagers in the US currently useanabolic steroids.

Modafinil is a schedule IV, wake-promotingdrug (www.provigil.com). Modafinil is approved totreat “unusually sleepy people” who have been

Performance-Enhancing DrugsPresent A Double-Edged Sword

Ralph Lydic, PhD


See “Dear Sirs,” Next Page

Anesthesia machines in general and the Divanventilator, in particular, require stable and uninter-rupted electrical power in order to function prop-erly. Power supply design is modestly complex;electrical efficiency, heat dissipation, weight, shape,size, and cost must all be considered as the designcriteria are set and achieved. The engineering, how-ever, is well understood and the power generatingindustry is stable and mature. Even so, manufactur-ers understand that component failures do happenduring routine use of complex equipment, andtherefore manufacturers attempt to incorporateappropriate design safeguards.

It has been truthfully said that the basic func-tions of the anesthesia provider are to put air in thelungs and water in the veins. With that in mind, itwould seem intuitive that the one function forwhich the anesthesia machine must have powerfulredundancy is ventilation. Dräger has clearly con-sidered the loss of external power in its Narkomed6000 machine. However, our episode suggests thatthe potential failure of internal components did notreceive the same level of attention. Specifically, webelieve it is a machine design fault to have com-bined the back-up battery system and AC powersupply in such a way that the failure of the ACpower supply subsystem does not initiate both bat-tery back-up and alarm systems in the same man-ner as does the loss of external AC power. Instead,the failure of the AC power supply caused a totalpower failure in the ventilator. This design seems tous to be “counterintuitive” and would suggest thatDräger engineers should readdress this design.

Location of the VentilatorOverride Button

One drawback, in our opinion, of the Divanventilator is that electrically controlled valves havethe potential to allow it to reach a state, following apower failure, where manual patient ventilation isnot possible. The Ventilator Override button isspecifically intended to alleviate this scenario anddid indeed do so in the present instance. The loca-tion of the button is a concern, however.

While Dräger has taken the trouble to build abright red light into the button, the button islocated low on the left side of the machine,beneath the power switch and behind both theflow sensor assembly and the inspiratory andexpiratory valves. In general use, this button mayor may not be visible to the operator, depending onhis height and position relative to the anesthesiamachine. In this instance, the button was not immedi-ately visible to the anesthesiologist because the lineof sight was obscured by the aforementioned items.

Dear SIRS:We would like to report an incident of anesthe-

sia machine failure with possible patient safetyimplications. After the induction and intubation ofan 8-year-old female and during the sterile skinpreparation for orthopedic surgery, a loud warningbuzzer was heard by the operating room staff,apparently emanating from the Dräger Narkomed6000 (Dräger Medical, Inc., Telford, PA) anesthesiamachine. A “walk-around” of the machine wasrequired in order to isolate the buzzing noise aris-ing from the vicinity of the red ventilator bypassbutton located below the machine’s power switch.Pressing the button caused the buzzing to stop andthe Ventilator Override button to illuminate. Thedisplay screen of the machine did not appear to beaffected by the problem, but during the walk-around the display was not visible to the anesthesi-ologist. Upon the anesthesiologist’s return to thehead of the patient, examination of the machine’sdisplay (absence of any ventilator information onthe Narkomed LCD display and a lack of end-tidalCO2 on a separate monitor) revealed that the venti-lator had stopped functioning. The case start wasdelayed, and the patient was hand ventilated untilthe machine was exchanged. The case subsequentlyproceeded without further incident. Later examina-tion determined that the large sealed power supply,located at the rear of the base of the machine, hadfailed. This incident exposes, in our opinion, severaldesign shortcomings in the Dräger Narkomed 6000machine, some of which may pose threats to patientsafety.

Power Supply FailureIt is disconcerting that the power supply failure

in the present instance led to a failure of the ventila-tor. Given that the Divan ventilator (Dräger Med-ical, Inc., Telford, PA.) in the Narkomed 6000 is anelectronically controlled, electrically powered pis-ton ventilator, the continuous availability of electri-cal power must be assured. We think it isreasonable to expect that some sort of “fail-over”back-up system, such as exists in the computer, air-line, and shipping industries, would be in place onthis ventilator. Such was not the case in the presentinstance. When the power supply failed, the patientwent without ventilation for somewhat less thanone minute. We think this episode highlights whatwe believe is a design shortcoming of theNarkomed 6000 machine: there exists a linkagebetween the power supply and the reserve batteriessuch that a failure of the power supply itself alsoeliminates the ability of the back-up batteries tofunction. In other words, the machine is protectedfrom external power loss, but not from internallyoriginating power interruptions.

Cause of Ventilator Failure is Unclear

Dear SIRS refers to the Safety Information

Response System. The purpose of this column is

to expeditiously communicate technology-related

safety concerns raised by our readers, with input

and responses from manufacturers and industry

representatives. This process was developed by

Drs. Michael Olympio, Chair of the Committee on

Technology, and Robert Morell, Editor of this

newsletter. Dr. Olympio is overseeing the column

and coordinating the readers’ inquiries and the

responses from industry. Dear SIRS made its

debut in the Spring 2003 issue.

Michael Olympio, MD, Chair of the APSF Committee onTechnology and Co-Founder of the SIRS Initiative.

S AFETY

I NFORMATION

RESPONSE

S YSTEM

Dear SIRS


tributed to our delay in realizing that ventilator fail-ure had occurred. While there may (or may not) bean increase in reliability afforded by placing theventilator in a metal cabinet rather than a clear plas-tic column, the loss of visibility of the ventilator is,in our opinion, a high cost trade-off. We suggestthat Dräger reconsider this design and find a wayto combine reliable function, safe location and visi-bility in future machines.

On-Screen Ventilator FailureAlarm/Notification

The LCD display screen of the Narkomed 6000is one of the machine’s most attractive features. Theflexibility of such a display is unparalleled; it isintuitive and informative. One must wonder thenwhy there was not a very large, very noticeablealert on the display indicating that there had been amajor failure. Interestingly, exactly this type ofalarm does appear if an external power failureoccurs or the back-up battery system is engaged forany other reason.

The diagnostic information available to ourtechnical staff after the incident reveals that the sys-tem was aware that there had been a power supplyfailure. We propose making this vital informationaccessible to the clinician. We would like to suggestthat Dräger add an intrusive, informative, and pos-sibly even instructive on-screen alarm in case of anysignificant detectable internal or external powerfailure. This software upgrade should be installedin the field as soon as possible.

Power Supply Failure HistoryIn the publication of a power supply failure in

the same model anesthesia machine in September2003,1 Dräger was reported to have redesigned andreplaced the power supply in all affected machinesin the US and Canada. The earlier published failurewas tracked to a specific design flaw allowing con-tact between a capacitor can and a circuit boardtracing. Analysis of the power supply in the currentinstance is pending, but the circumstances of ourpower supply failure vary significantly from theearlier reported incident, in which there was a“pop” and smoke emanated from the power sup-ply. The machine serial number in the presentinstance is within the range of all the serial numbersof machines affected by the earlier power supplyrecall (Dräger correspondence), but this specificmachine was not affected by the recall.

Incident TrackingVentilator failure in an anesthesia machine is a

potentially devastating device mishap. Although itdoes not appear that FDA incident tracking is man-dated by this incident (there was no death orinjury), one could consider that such tracking mightgive Dräger the ability to cast a wider net as thecompany attempts to discover whether the powersupply/ventilator failure presented here is indeedan isolated incident or is the sentinel event in achain of potential adverse outcomes.

Albert R. Davis, MDBruce Kleinman, MDW. Scott Jellish, MD, PhDLoyola University Medical CenterMaywood, IL

Reference

1. Usher AG, Cave DA, Finegan BA. Critical incident withNarkomed 6000 anesthesia system. Anesthesiology2003;99:762.

The anesthesiologist had to walk completelyaround the machine and approach it from the leftside (as viewed by an operator in the usual posi-tion) before seeing the red light.

In sum, the location, appearance, and use of thebutton do not cause the user to have a high degreeof confidence or certainty in its function. It is nearlyhidden from sight and ambiguously named andlabeled, and the fact that it turns bright red from alight built into the switch makes it somewhat star-tling to use the first time. The button is, essentially,the last line of defense against electrical failure ofthe Divan ventilator. The Narkomed 6000 opera-tor’s manual states that the button “is provided foruse in the unlikely event of a fault which does notallow the clinician to ventilate a patient in normalManual/Spontaneous Mode or safe state.” Goodergonomic design suggests that a rarely used but-ton would be best placed away from the primarylines of sight and reach of a machine operator.However, in terms of patient safety, it is our opin-ion that a rarely used button intended as a last ditchresponse to a potentially life-threatening situationshould be highly visible, unambiguously labeled,and easily accessible. We believe that futureNarkomed anesthesia machines equipped with thecurrent iteration of the Divan ventilator should con-tinue to feature the Ventilator Override button, butthat button should be redesigned and relocated tosatisfy both ergonomic and safety considerations.

Ventilator VisibilityOne of the more controversial aspects of the

Narkomed 6000 machine is the placement of theDivan ventilator. The Divan ventilator is locatedbeneath a metal panel which serves as the “tabletop” for the anesthesia machine, and is thus com-pletely hidden from view. When the Narkomed6000 machine was first introduced, many experi-enced anesthesiologists were taken aback at theabsence of a visible bellows, particularly in light ofthe years of controversy over the “hanging vs.standing bellows” issue. Indeed, when this ventila-tor failed, no specific visible indication that theevent had occurred was noted except for the buzzersound and the red light in the ventilator overridebutton. An inexperienced operator, or one dis-tracted by other simultaneous problems, mighthave taken significantly longer to notice the failurethat occurred in this case. The placement of theDivan ventilator out of the operator’s sight con-

Invisible “Bellows” Requires Different Mindset“Dear Sirs,” From Preceding Page

Check out the Virtual Anesthesia Machine Website

and theAPSF Anesthesia

Machine Workbookat

www.anest.ufl.edu/vam


The ventilator override button was designed inthe event that an unforeseen condition occurs thatthe safety software does not recognize. Activatingthis button removes power from the ventilator forc-ing it into Safe State which allows for manual orspontaneous ventilation.

Power Supply FailureThe author’s concerns about the design of the

power supply are unfounded. The functionality andbehavior that the author is proposing is indeed howthe system works. The Narkomed 6000 is designedto primarily function on AC power. In the absenceof AC power, the anesthesia machine will functionon back-up battery power for a minimum of 30 min-utes with a fully charged battery. The anesthesiamachine will inform the user that AC power is nolonger present by a visual message and audiblealarm.

The author expresses concern about the depen-dence of the ventilator on electrical power. Thedependence of the NM6000 series anesthesiamachine on electrical power is no different than pre-vious anesthesia machines manufactured by DrägerMedical, which include the Narkomed product lineof anesthesia machines. The Narkomed productsalso require AC electrical power to function duringautomatic ventilation.

On-Screen Ventilator FailureAlarm/Notification

Current software released for the NM6000 anes-thesia machine does incorporate such features as dis-cussed into the alarm structure. In an AC powerfailure condition, the device posts an advisory alarmmessage, “AC Power Fail,” plus a single audibletone. A dialog box is also displayed on the screenwith a warning tone. Upon activation of back-up bat-tery power, the device will function for a minimumof 30 minutes from a full charge. Prior to depletion ofbattery power and machine shut down, a battery lowmessage will be posted indicating that approxi-mately 10 minutes of power are remaining. Prior touse, the power on self-test will post appropriate indi-cations to the user regarding the power status to thedevice. The system will post a message indicating thestatus of the device as functional, conditionally func-tional, or non-functional.

Power Supply Failure HistoryThe power supply returned from Loyola was

not within the suspect population for the recall thathas occurred. The power supply recall was issuedfor 151 machines. All facilities with machines thatwere affected by the recall were properly informedand had their power supplies replaced. The recallstarted November 01, 2002, and was completedFebruary 28, 2003.

Ventilator Override LocationCustomer input and design determined the cur-

rent location of the ventilator override switch to theventilator on the machine. The switch is clearlylabeled (see Figure 1), and the defined use is pro-vided to the user in the Operator’s Manual.

Ventilator VisibilityDräger Medical recognizes that the location of

the ventilator is a significant change for the userversus past designs. As such, Dräger Medical offersas an option, a top cover for the ventilator with atransparent window to allow the user to easily visu-alize movement of the piston ventilator.

Incident TrackingDräger Medical complies with post market sur-

veillance and the medical device reporting regula-tions.

We appreciate all feedback regarding the fea-tures and functionality of Dräger anesthesia work-stations. This feedback is important in helping toidentify potential performance issues and also helpsto incorporate new features that our customersrequire. Additionally, anesthesia machines are notall the same, and patient safety may be managedthrough various means and systems. As anesthesiamachines are required to perform more complexfunctions, machine features and usage are alwayschanging. As a result, it is recommended that allstaff undergo initial and periodic refresher trainingto ensure a high level of proficiency for all types ofanesthesia machines. Dräger Medical makes its per-sonnel available to help in the continuing educationof its customers.

Robert ClarkCareArea DirectorPerioperative CareDräger Medical, Inc.

Response:Dräger Medical is grateful for the opportunity to

respond to the concerns identified by the clinicalstaff at Loyola University Medical Center. DrägerMedical considers patient safety to be paramount,and Dräger Medical’s history of equipment innova-tion is evidence of this commitment to patient safety.

Reported ProblemDräger Medical was contacted regarding the

reported condition by a biomed from the facility.The biomed reported that during a case the Divanventilator was noted to lose power and the displayinformation went blank. An audible alarm was notedand the monitor screen for the anesthesia machinecontinued to function. The user depressed the venti-lator override button and manually ventilated thepatient on the machine until the machine wasreplaced. Upon replacing the anesthesia machine, thebiomed performance tested the device and wasunable to reproduce the reported occurrence. Thebiomed replaced the power supply to be cautious.

FindingsThe power supply was returned to Dräger Medical

for evaluation. Performance testing found thepower supply to function normally.

AssessmentThe fact that the rest of the system remained

active indicates that this was likely not a powersupply failure. This was confirmed when the powersupply was returned, and the failure analysis foundthe power supply to function normally.

It would appear from the report that a conditionmay have arisen in which an alarm was posted bythe Divan. The Divan ventilator does have an inter-nal audible speaker that is activated during anatypical hardware or software condition.

The Divan Ventilator incorporates safety soft-ware that monitors the function of the ventilatorduring the power-on self-tests, as well as duringnormal use. In the event that the safety softwaremonitors a fault condition, the Divan may post anerror message on the Divan Control Panel displayand an audible alarm. Upon posting the message,the Divan ventilator may configure a warm start orconfigure itself into Safe State. Safe State automati-cally configures the Divan for manual or sponta-neous ventilation and the next power-on self-testcannot be aborted. A warm start by the Divanwould cause the ventilator to momentarily reset,power off, and then on. The reset will take approxi-mately five to ten seconds and then the Divan venti-lator will resume ventilation at the user setparameters.

Dear Sirs Response

Manufacturer Analyzes Incident, Provides Feedback

Figure 1: General Location of Ventilator OverrideSwitch


for the practitioner. Research shows that the sleep-deprived anesthesiologist drives home at least asimpaired as one legally intoxicated with ethanol.

One could argue that the work hours of criticalhealth care providers should be regulated; societyshould make the investment to train and payenough personnel to provide the same level ofexpertise to everyone, whenever their need for carearises. However, that is not the case. Sleepy peopleare keeping people asleep. Before it was confirmedscientifically, it was widely known empirically.Anesthesiologists and nurse anesthetists in busyclinical practices are excessively sleepy. Fatigueaffects performance. While we know it empirically,this is a bit more tedious to establish scientifically,but progress is being made. Morbidity and mortal-ity are at stake. In 1999, the Institute of Medicinereport described a previously underappreciatednumber of deaths that were attributable to medicalerror. I have no doubt that these incidents are morecommon when the practitioner is fatigued, and myearly morning follies today indicate that I was ill-prepared to deal with simple tasks, much less thecomplex, rapidly evolving catastrophes that mysurgeons and patients occasionally conjure up.Unlike professional athletes, whose primary goalmay be to win personal fame and fortune, our goalis to maintain vigilance against disaster for ourpatients. We have an ethical duty to maximize ourpotential and to carefully and pragmatically con-sider any reasonable aid that presents itself.

Cognition is complex, and delivering anesthesiais not the same as flying a plane. Exactly howmodafinil improves performance is not known, andits effects on memory, problem solving, mood,motivation, interpersonal interactions, and a varietyof other factors important to the practice of anesthe-sia have not been fully investigated. Perhapsmodafinil will prove to diminish management ofcomplex diagnostic decision-making, or worsenmulti-task management. However, we have themeans of testing performance in realistic simula-tors, so we could begin to answer these questions.College students are taking modafinil to cram forfinals. Journalists are using it to combat jet-lag onoverseas assignments. Residents are using the drugto maintain alertness while on call. It seems reason-able to study this new frontier of patient (and prac-titioner) safety, to potentially arm ourselves withboth well crafted shifts, good sleep habits, strategicnapping, and a safe pharmaceutical to be usedwhen absolutely necessary, to help us maintain ourmotto: “vigilance.”

Dr. Smith is an attending anesthesiologist at theWashington Township Hospital in Fremont, CA, andserves as an Adjunct Assistant Professor in the Depart-ment of Anesthesiology at the Stanford UniversitySchool of Medicine, Stanford, CA.

by Brian E. Smith, MD

I was on call last night, and I was lucky.Although I worked moderately hard for 19 hours,the final 5 were spent dozing off and on in a noisy,cold “sleep” room. Departing for a 7:00 am depart-ment meeting, I promptly locked my lab coat andmy keys in the call room. A quick change of scrubsmade me feel a little fresher, until I spilled a dribbleof my morning double espresso on them. I foundmyself searching for the right word a few times dur-ing the meeting, not unusual as I age, but a notice-ably more common problem today. I was luckyagain after the meeting; while most of my colleagueswent to face a challenging day in the OR, one feweroutpatient rooms on the schedule spared me fromplacing a patient’s life in my hands. My group onceeschewed working post-call; with declining reim-bursement from profitable insurance conglomerates,it is now considered an economic necessity to main-tain the status quo. I have to finish writing this arti-cle and get some sleep; I have 2 cardiac casestomorrow, which means a 4:45 am wake-up time.

The concept of performance-enhancing drugshas gotten significant press lately, and the buzz isoverwhelmingly negative. Professional athletessearching for a competitive edge are vilified ascheats in the rare instances that they have beencaught using anabolic steroids. However, like manyother effective tools, performance-enhancing med-ications used correctly could decrease suffering andimprove life: lost in the hype about steroids arepotential benefits to society of a wide range of med-ications unrelated to steroids, and in my view, notat all analogous. Let’s focus on one.

Modafinil is a pill that improves mental alert-ness through a novel mechanism, probably involv-ing dopamine reuptake. It does not have theadrenergic side effects of amphetamines; in fact,most subjects cannot readily perceive the differencebetween active drug and placebo. It does notdecrease the user’s ability to recover sleep when anopportunity arises, and does not deliver a reboundeffect. It does improve performance on a host oflaboratory tasks, including tests of vigilance, mem-ory, mood, and attention, as well as real-worldtasks such as flying combat aircraft. The militaryhas a compelling interest in performance andfatigue because combat operations are typically24/7, and the stakes are extremely high for the par-ticipants. Hence, there is serious, ongoing researchon the use of modafinil by sleep-deprived aircrews,and the results are encouraging. Modafinil worksnearly as well as dexedrine and better than caffeinein maintaining performance during severe sleepdeprivation. It does have occasional side effectsincluding headaches and nausea, and undoubtedlymore will be found. However, sleepiness is not onlydetrimental to our patients; it is not entirely benign

Cure for the Dopey Doc?

Pleasesupport

your APSF

Individualdonations aremost welcome.

Yourcontributiondoes make adifference.

Contributions may be sent to:

Anesthesia PatientSafety Foundation

(APSF)520 N. Northwest Highway

Park Ridge, IL 60068-2573

(Make checks payable to the APSF)


providers suffer from acute sleep loss, cumulativesleep debt, prolonged periods of continuous wake-fulness, and circadian disruption. This physiologi-cal disruption is associated with decreased alertnessand performance and an increase in errors andrisks, including provider car crashes.1-3 No doubt,ongoing research will further define these risks andelaborate the additional outcomes of fatigue inmedical practice. However, it is critical to nowmove beyond just documenting that fatigue is pre-sent in 24/7 health care activities and has negativeeffects on performance and safety. It is time to testand implement effective strategies and approachesthat will manage the known risks associated withfatigue in health care.

Classically, the most widely used approach toaddress fatigue in operational settings is through“hours-of-service” (HOS) regulations. Typically,these regulations establish specific off-duty require-ments intended to provide rest opportunities andduty time limitations to prevent extended periodsof work and wakefulness. An HOS approach alsomight address total work hours in a certain numberof days or an identified number of days off forrecovery. In some HOS structures, the work timewithin a duty period also has specific limits, forexample, drive time for commercial truck operatorsand flight time for commercial pilots. Recently, theAccreditation Council for Graduate Medical Educa-tion (ACGME) utilized this same historicalapproach to establish intern and resident work hourlimitations. There is no question that “HOS” arenecessary but insufficient to effectively managefatigue-related risks. This is true because allowingpeople to work without limitations will clearly

engender fatigue, and providing minimal off-dutyperiods to prevent acute sleep loss and recoveryperiods to “zero” cumulative sleep debt are basicsof managing fatigue. However, given the complexi-ties previously identified, a general HOS structurecannot be expected to cover every circumstance orindividual in health care.

Though an HOS approach is necessary, 2 exam-ples illustrate its limitations. While an off-dutyperiod maybe mandated, this does not guaranteethat an individual will obtain required sleep duringthis time. Whether an individual chooses to remainawake and active, has sleep disturbed by externalfactors (e.g., noisy environment, sick child), isafflicted with one of the almost 90 identified sleepdisorders or a multitude of other possibilities, thatindividual may not obtain sufficient sleep. Acutesleep loss, regardless of etiology, will degrade sub-sequent alertness and performance.

Another example involves night work thatrequires a health care provider to be awake whenbiologically programmed for sleep by the circadianclock and then attempt to sleep during the daywhen programmed for wakefulness. Therefore,anyone working through the night is at risk fordegraded alertness and performance and the associ-ated errors related to the circadian time of day.Generally, people do not physiologically adjust tonight work,4 and successive nights of work increasesafety risks.5 While light administration has beenused successfully in the laboratory to adjust circa-dian phase,6 the effective, practical and economi-cally feasible application of this strategy to diversework settings has yet to be realized. Therefore, thecircadian clock will continue to create a physiologi-cal fatigue risk for individuals working at night.

by Mark R. Rosekind, PhD

If only it were that simple: just tell people to gettheir 8 hours of sleep and develop perfect workschedules, or invent a “magic bullet” that elimi-nates fatigue, manages the safety risks, and is effec-tive for everyone. Managing the fatigue-relatedrisks in health care is much more complicated.Health care tasks can be very different, so there arediverse operational requirements. The people them-selves are very different, so individual differencesrelated to age, gender, experience, sleep need, andmany other factors vary widely. The sleep and cir-cadian physiology that underlies fatigue, sleepi-ness, and performance decrements are complex.Historical precedents are difficult to change, andlongstanding practices can be resistant to progress.All of health care is contending with the economicsassociated with the modern practice of medicine,and addressing any significant patient safety issuewill involve cost/benefit scrutiny.

These real-world challenges clearly show thateffectively managing fatigue in any 24/7 opera-tional setting is a complex endeavor, with no singleor simple solution. Given our human physiologicaldesign that includes a requirement for sleep and thepowerful control and effects of the circadian clock,it is impractical to believe that fatigue can be elimi-nated from around-the-clock activities. However,there is no question that the safety-related risksassociated with fatigue and impaired alertness canbe managed better than current practice.

An extensive scientific literature exists thatdemonstrates the risks and costs associated withfatigue in health care.1,2 It is clear that health care

Managing Fatigue 24/7 in Health Care:Opportunities to Improve Safety

See “Managing Fatigue,” Next Page

Mark R. Rosekind, PhD


There is no HOS structure in existence that fullyaddresses this circadian component; in fact, mostpolicies do not incorporate specific circadian ele-ments because they are so complex.

Alertness Management ProgramA comprehensive alertness management

approach offers the greatest opportunity to addressthese issues and involves a multi-component pro-gram that includes education, alertness strategies,scheduling practices, healthy sleep, and scientificand policy support.7 An overview of this approachand each component is provided to illustrate theopportunity that a comprehensive alertness man-agement approach offers.

Component 1: EducationEducation is the foundation for any effort to

address fatigue. At a minimum, the educationalcomponent should provide information about therisks associated with fatigue, the physiologicalmechanisms that underlie fatigue, and availableapproaches to reduce risks and optimize perfor-mance and alertness during operations. Considerthat the basis for any significant attitude and behav-ior change begins with education. Health care pro-vides many such examples where changes in diet,exercise, and smoking begin with educational pro-grams that include information about risks andbehavior change. This same approach is a criticalfirst element for any fatigue or alertness manage-ment program.

While it is critical to provide an appropriate off-duty sleep opportunity, there is no guarantee thatan individual will use the time for sleep. Perhapsthe only acceptable approach to address this issueinvolves providing effective education that portraysthe risks associated with not obtaining requiredsleep before work. It is unlikely that “bed checks”or technological monitoring of individuals to vali-date sleep will be readily accepted or adopted soon.Similarly, the circadian clock will continue to createphysiological risks during the nighttime window ofcircadian low. It is critical that night workers beinformed about this period of reduced alertness andperformance, and utilize strategies to manage it.

This education should be provided to all seg-ments within a health care organization. Often,these educational efforts will focus on trainees andphysicians. However, nurses, technicians, pharma-cists, maintenance personnel, and anyone workingirregular schedules are potentially at risk andshould be offered, at least, basic fatigue education.Also, a one-time “session” is less likely to createbehavior change than information provided in mul-tiple forums and formats and continually over time.Integrating fatigue education into an ongoing pro-gram creates the opportunity for this information tobecome a part of the corporate culture.

Component 2: Alertness Strategies

A second component of a comprehensive alert-ness management program involves alertnessstrategies. Beyond just information, the alertnessstrategies component should include two elements:1) specific guidance on the effective use of strate-gies, and 2) establishing appropriate policies andsupport for implementation of strategies. For exam-ple, two well-studied and effective fatigue counter-measures involve planned naps and strategic use ofcaffeine.8,9 However, it is critical that individualsunderstand the basics of their effective use. Howlong should naps be? What is the best timing? Whatshould be done to avoid sleep inertia? These areimportant considerations to obtain optimal resultsfrom naps. In a similar way, information about thestrategic and effective use of caffeine will increasethe benefits obtained. For example, what dose isneeded to enhance alertness and mental function-ing, how much caffeine is contained in differentdrinks and food, and what will be the best timingfor ingesting caffeine as an alertness strategy?Examples of some answers to these questions andother strategy implementation information is avail-able in other sources.2,7,10

Another important element of this informationshould be to assist individuals in discriminatingwhich strategies have a proven scientific basis fortheir effectiveness and use. When confronted withpotential recommendations on “how to stay awakeon the job” from a variety of sources (e.g., friends,co-workers, media), what criteria might an individ-ual use to determine their value, efficacy, and safety.

The implementation of alertness strategies oftenrequires policies and corporate support. An explicitpolicy on the use of planned naps is useful in a vari-ety of ways. First, the policy provides an explicitsanction for the use of planned naps in the context ofthe work environment. Second, it can delineate spe-cific procedures, timing, and circumstances for itsuse. Third, it is a mechanism to address any potentialconcerns about misuse. Beyond policies, corporatesupport might include the creation of specific naprooms or some space conducive to obtaining aplanned rest period. Also, while some may focus onthe use of naps during low workload periods onshift, they can be an invaluable alertness strategyprior to a drive home after a work period. In relationto caffeine, if information about its effective strategicuse is provided, then it is critical to have mechanismsin place to make it available when needed.

Similar to the education component, alertnessstrategies information and policies should be dis-tributed through multiple forums and formats andon an ongoing basis. One example of an integratededucation and alertness strategies tool was a videodeveloped on behalf of the American Society ofAnesthesiologists.11 This short video includes somebasics of fatigue risks and the underlying physio-

logical mechanisms, and concludes with guidanceand modeling of how to apply alertness strategiesin a health care setting.

Component 3: Scheduling Policies andPractices

The third component of a comprehensive alert-ness management program involves schedulingpolicies and practices. At a minimum, this compo-nent should focus on 2 core issues. First, a basic“fatigue” analysis can be conducted to determinefrom a system’s perspective how alertness and per-formance could be affected by current schedulingpolicies and practices. This might involve an exami-nation of minimum off-duty times, shift lengths,policies on multiple shifts, consecutive duty periods,recovery opportunities, and other factors. There areat least 14 potential scheduling-related factors thatcan be evaluated as they relate to alertness and per-formance.7 This analysis can identify both strengthsand vulnerabilities in current scheduling. Second,based on the system analysis, the schedulingstrengths should be reinforced and even extendedwhenever possible. The vulnerabilities identifiedpresent an opportunity to adjust scheduling policiesand practices to address fatigue-related risks andenhance alertness and performance.

It should be noted explicitly that schedulingissues are the most complex and contentious ele-ments of any comprehensive alertness managementprogram. Scheduling significantly affects both indi-vidual health care providers and the organization.From an individual perspective, scheduling policiesand practices will affect income, family, and socialactivities, and off-duty mood and performance.

Component 4: Healthy Sleep

The fourth component of a comprehensive alert-ness management program involves healthy sleep.At a minimum, this should involve providing infor-mation about the existence of the almost 90 sleepdisorders that can be diagnosed and treated. Someof the sleep disorders known to affect health andsafety, such as sleep apnea or shift work sleep dis-order, should be emphasized.12-14 Information canoutline classic signs and symptoms, as well as, thepotential health and safety risks that may be associ-ated. In this context, information should includeresources that identify accredited sleep disordersclinics available for referral, and possibly the extentof insurance coverage for these services. Beyondinformation and referral sources, it is possible tohave a work-based diagnostic and treatment pro-gram that identifies and treats individuals withsleep apnea.

Component 5: Scientific and Policy Support

The fifth component of a comprehensive alert-ness management program involves scientific andpolicy support. A guiding principle for the program

Comprehensive Approach Can Improve Safety“Managing Fatigue,” From Preceding Page

See “Managing Fatigue,” Next Page

3. Barger LK, Cade BE, Ayas NT, et al. Extended workshifts and the risk of motor vehicle crashes amonginterns. N Engl J Med 2005;352:125-34.

4. Gander PH, Gregory KB, Connell LJ, et al. Flight crewfatigue IV: overnight cargo operations. Aviat Space Envi-ron Med 1998;69(9 Suppl):B26-36.

5. Folkard S, Tucker P. Shift work, safety and productivity.Occup Med (Lond) 2003;53:95-101.

6. Horowitz TS, Cade BE, Wolfe JM, et al. Efficacy ofbright light and sleep/darkness scheduling in alleviat-ing circadian maladaptation to night work. Am J PhysiolEndocrinol Metab 2001;281:E384-91.

7. Rosekind MR. Managing work schedules: an alertnessand safety perspective. In: Kryger MH, Roth T, DementWC, eds. Principles and Practice of Sleep Medicine.Philadelphia: Elsevier Science, (in press).

8. Committee on Military Nutrition (Institute of Medi-cine). Caffeine for the sustainment of mental task per-formance: formulations for military operations.Washington, DC: National Academy of Sciences, 2001.

9. Rosekind MR, Graeber RC, Dinges DF, et al. Crew fac-tors in flight operations IX: effects of planned cockpitrest on crew performance and alertness in long-hauloperations. NASA Technical Memorandum #108839.Moffett Field, CA: NASA, 1994.

10. Rosekind MR, Gander PH, Connell LJ, et al. Crew fac-tors in flight operations X: Alertness management inflight operations. NASA Technical Memorandum#1999-208780. Moffett Field, CA: NASA, 1999.

11. Howard SK, Gaba DM, Rosekind MR. Fatigue: Implica-tions for the Anesthesiologist. American Society ofAnesthesiologists Patient Safety Videotape Series, Pro-gram 29. Park Ridge, IL: American Society of Anesthesi-ologists, Inc., 1998. Available on the web at:http://www.apsf.org/resource_center/educational_tools/video_library.mspx. (Accessed March 8, 2005)

12. Drake CL, Roehrs T, Richardson G, et al. Shift worksleep disorder: prevalence and consequences beyondthat of symptomatic day workers. Sleep 2004;27:1453-62.

13. Teran-Santos J, Jimenez-Gomez A, Cordero-Guevara J.The association between sleep apnea and the risk oftraffic accidents. N Engl J Med 1999;340:847-51.

14. The International Classification of Sleep Disorders(Revised): Diagnostic and Coding Manual. Westchester,IL: American Academy of Sleep Medicine, 2001.

should be the scientific foundation of all alertnessmanagement activities. Whenever possible andavailable, scientific data should be used to guideand create a foundation for each program compo-nent. Whether it is the core material provided in theeducational activities or examination of schedulingpractices, fatigue-related scientific knowledge cancreate a basis for guiding action.

A central ingredient of a comprehensive alert-ness management program involves the integrationof activities and multiple approaches to addressfatigue. The use of these various components pro-vides a more extensive intervention that acknowl-edges the complex factors that can createfatigue-related risks. Some issues may be mostamenable to educational activities, while others mayrequire explicit changes in scheduling practices orovert sanctioning of certain alertness strategies (e.g.,naps). An integrated program will ensure that infor-mation in the educational component is consistentlyrepresented in the strategies and scheduling activi-ties and that the alertness strategies component issupported by appropriate policy. This integrationcan be a critical element to the success of a compre-hensive alertness management program.

The scientific data are clear: fatigue createssafety and health risks in 24/7 operational settings,including health care. Given the complex factorsthat underlie fatigue, no single or simple solutioncan be expected to eliminate fatigue-related risks inhealth care. There are historical approaches, such asHOS policies, that are necessary, but insufficient tofully address the complexity. However, a compre-hensive alertness management approach offers anopportunity to address the complex nature offatigue and reduce the known risks. As more datademonstrating the safety and health risks continueto emerge, it is critical that health care settingsmove forward with interventions that addressfatigue and promote alertness, performance, andsafety. If action is not taken, then health care itselfmay be viewed as hypocritical when attempting toaddress these same issues with patients or in otheroccupational settings. However, the current statusof efforts in health care regarding fatigue also repre-sents an important and unique opportunity. Byacknowledging and addressing fatigue relatedrisks, health care can be a model for other around-the-clock environments and provide leadership oneffective interventions that can improve the alert-ness, safety, and health of our 24/7 society.

Dr. Rosekind is President and Chief Scientist ofAlertness Solutions, Cupertino, CA.

References

1. Gaba DM, Howard SK. Fatigue among clinicians andthe safety of patients. N Engl J Med 2002;347:1249-55.

2. Howard SK, Rosekind MR, Katz JD, et al. Fatigue inanesthesia: implications and strategies for patient andprovider safety. Anesthesiology 2002;97:1281-94.


Letter to the Editor

Surgeon ShouldGive Feedback to His Health Care Team FirstTo the Editor:

I am amazed by the response to the anonymoussurgeon/patient's perioperative experiencerecounted in the 2004-5 winter issue of the APSFNewsletter. I have been in practice for 20 years, inboth academic and private practice settings. Cur-rently, I practice in a tertiary care referral centerwith an anesthesia residency. The overwhelmingmajority of “anesthesia providers” are dedicated toexcellent patient care and very professional in theirapproach to the patient. I am sorry that this sur-geon/patient did not have quite the perioperativeexperience he was expecting. However, he needs todirect his criticism in a constructive way toward theanesthesia team responsible for his care, and nottoward the entire specialty. Interestingly, if I wroteto a surgical newsletter to complain about my sur-geon, the letter would probably not be published. Iwould be told, “Gee, honey, go discuss this withyour surgeon.” As a specialty we seem to have acollective inferiority complex! But, why? Advancesin surgery would not be possible without advancesin the field of anesthesiology. Anesthesiologistshave pioneered fields like intensive care medicineand pain medicine. The ABA is committed to thecontinuing education of its members. The residentsthat I train are as intelligent, dedicated, and hardworking as any surgery resident, if not more so.Coming to work is a pleasure, and I consider myselfblessed to work with such competent individuals. Ihave a great suggestion. Let's invite Dr. PhilMcGraw to the next ASA meeting. Maybe he canfigure out the basis for this collective inferioritycomplex, because I can't! I can, however, recom-mend one helpful exercise for those who might stillsuffer from such a complex. Before you leave forwork in the morning, look at yourself in the mirrorand tell yourself, “Damn I'm good!”

Elizabeth T. Young, MDNew Orleans, LA

“Managing Fatigue,” From Preceding Page

Leaders Should Acknowledge Fatigue Support Your APSFYour contribution• Supports research• Promotes initiatives• Supports the Newsletter• Enhances safety

Thank You!


In early 2003 Dr. Ed Charlton (an ex-member ofAAGBI Council) wrote to the AAGBI President todraw his attention to an editorial and review articlepublished in Anesthesiology on the subject of fatiguein anaesthesia.1 This was a timely occurrence aschanges in work patterns, due to the impendingimplementation of the European Working TimeDirectives and the anticipated New ConsultantContract, were expected to have a profound effecton the total weekly working hours and lifestyles ofboth career grade and non-career grade anaes-thetists, as well as other hospital doctors in the UKand Ireland. A paper with a similar message hadbeen published in June 2000 and reported thatfatigue was listed as a contributory factor in 152anesthetic incidents, or 2.7% of all reports.2

The President raised the matter at the next bi-monthly Advisory Meeting of the Council where Iwas tasked with setting up a working party toexamine the evidence and produce a document toexplore the problems of fatigue in anaesthesia andmake proposals that could reduce the risks for bothpatient and practitioner. A scoping meeting ofselected Council members met in March 2003 to dis-cuss the proposal, and it was agreed that the projectshould proceed with other invitees who could addknowledge and experience to the deliberations.

The first full meeting of the new Fatigue andAnaesthetists Working Party took place in May2003 and, in addition to members of Council andthe Association’s Trainees section, involved EdCharlton and representatives of the British MedicalAssociation, an Occupational Health Physician, andthe Royal College of Anaesthetists.

The working party met on a total of 6 occasionsand produced 8 drafts of the document. Much ofthe draft revision was done by email iteration andamendment with the formal meetings beingreserved for discussion of policy and drawing outthe recommendations from the text. When theworking party met in January 2004 to review draft5, it was over 8,000 words long and was fully refer-enced with 56 citations. It was clearly too long for atraditional glossy and a number of options wereconsidered. It could be published as it was in full, itcould be severely abridged, or it could be abridgedwith the full document and all its supporting refer-ences and material available on the AAGBI website.

The last proposal was accepted, and an abridgeddraft (4,000 words and 21 references) was submittedto Council for final approval in early March 2004.All Council members were able to feedback com-ments to the April Council meeting, and correctionsor amendments adopted there were included in thefinal document launched in July 2004.

Every Association document when first publishedhas a date assigned for a review, when Council decideswhether it should be withdrawn, is still a relevant doc-ument, or should be completely revised and repub-lished. These documents are known as “Glossies”because of the glossy cover assigned to them.

What are glossies for? First, they should educatemembers on topics of relevance to professionalpractice, but they are not meant as mini textbooksof anaesthesia. Second, they should be useful to themembership as references when discussing profes-sional activities with other specialty groups or man-agers. It is hoped that, with the support of a glossy,an anaesthetic division may be able to ensureproper provision of support, be it clinical, equip-ment, secretarial, or administrative to allow safepractice. Some glossies become very powerful tools,such as the “Recommendations for Standards ofMonitoring During Anaesthesia and Recovery,”which has set the standard, and is broken at theperil of the practitioner. Others such as “Manage-ment of Anaesthesia for Jehovah’s Witnesses” pro-vide handy pocket-sized reference to less commonproblems faced by the anaesthetist. The glossy,“Fatigue and the Anaesthetist,” can be used toguide members and the management of hospitalsinto following safe practices with a scientific back-ground in respect to the duration of working hoursand recovery from arduous duties. Guidance is alsooffered on better practices to be followed whereturnovers from one anaesthetist or team to anotherare necessary due to the onset of fatigue or therequirement to relieve a colleague who has

by Michael E. Ward, MB BS, MRCS LRCP, FRCA

The membership of the Association of Anaes-thetists of Great Britain and Ireland (AAGBI) istotally medical, and currently 95% of the anaes-thetists in the UK are members. It has in excess of9,000 members, including 4,500 full members (con-sultants in UK terminology is equivalent to attend-ings in the US), 1,000 members retired from activepractice, and the remainder who are staff in train-ing grades. Most reside in the British Isles, but 400currently work or live overseas (68 are currently inthe US).

Annual membership costs are £180 ($338). Theroles of the Association are

• to represent the interest of its members

• to provide advice and information to both mem-bers and public about anesthetic matters

• to organize regular (twice yearly) national meet-ings and 40 small seminar meetings annually atits London headquarters to which members havesignificant discounts

• to publish its monthly scientific journal, Anaes-thesia, sent free to all members together with itsnewsletter, Anaesthesia News

• to maintain a professional website

• to provide a significant level of automatic injuryand life insurance coverage for members duringpatient transfer

• to provide representation at Westminster and theDepartment of Health

• to work with the Royal College of Anaesthetiststo provide full professional support and mainte-nance of standards of practice and training

• to publish regular guidelines on professional andethical matters which form the basis for goodpractice, known as “Glossies.”

Fatigue and Anaesthetists:The Association of Anaesthetists of Great Britain and Ireland’s Approach

The Association of Anaesthetistsof Great Britain and Ireland

See “Britain and Ireland,” Next Page Michael E. Ward, MB BS, MRCS LRCP, FRCA


exceeded safe working hours. It makes a number ofsignificant proposals aimed to protect both thepatient and the practitioner.

Principal Recommendations:

• Every anaesthetist should be aware of the prob-lem of fatigue and carry a personal obligation toprovide safe and effective service.

• Departments must have a plan to manage staffat all grades who have suffered an onerous dutyperiod and consider themselves unfit to con-tinue work. For example, when a night on call isarduous there must be a mechanism/protocol toallow for replacement/relief of an overtiredanaesthetist.

• Job plans which are likely to lead to predictablefatigue should be avoided so that busy nightson call should neither follow nor precede a fullworking day in the operating room, intensivecare unit, or similar duty.

• Job plans of career grade staff should bedesigned to include flexibly worked fixed the-atre sessions without operating sessions allo-cated to a particular surgeon or service in orderto provide regular relief for colleagues.

• Routine rest breaks should be implemented. A“Handover Protocol” should be used beforeeven short rest breaks. A standardised checklistto be used when relieving a colleague should beintroduced to avoid omitting vital informationabout the patient, procedure, anaesthetic tech-nique, and equipment.

• All hospitals should ensure the availability of“on call” rooms for those doctors working nightshifts, to enable rest breaks. With the recentintroduction of shift working, many hospitalshave removed the “duty” or “on-call” rooms asa cost saving measure. The Association believesthat such facilities should be reintroduced orprotected in order that anaesthetists who get theopportunity to take a rest break during theirshift can do so in a suitable environment.

• Management should provide designated roomsadjacent to the operating room for napping and“post-call” sleeping facilities.

• Good quality accommodation should be avail-able for resident on-call staff.

• All staff should have access to good qualityrefreshments at all times. It is often not possiblefor anaesthetists to visit the dining hall/cafete-ria at the fixed hours of opening due to theunusual requirements of their workload. Alter-native refreshment facilities must be provided atthese “odd” hours.

• On-call responsibilities should be reviewed foranaesthetists over 45 years of age in conjunctionwith advice from an accredited specialist inoccupational medicine.

• Private practitioners should ensure that a combi-nation of National Health Service and Privatework does not lead them to practice when com-promised by fatigue.

While the proposals in the document are clearlyaimed at anaesthetic working practices, the princi-ples of the findings and recommendations will bejust as applicable to surgery or other branches ofhospital medical practice.

The “Glossy” has been sent to all members ofthe AAGBI and interested groups and posted onthe AABGI’s website.3 In addition, an expandedversion, which goes into some aspects in muchgreater detail, has also been posted on the website.4

AAGBI Has Important Recommendations

Reminder!APSF GrantApplications

DueMonday

June 20, 2005• See website for

guidelines andinstructions

• Applications must be submittedelectronically viawebsite

“Britain and Ireland,” From Preceding Page

ACGME RecognizesDangers of Fatigue

The Accreditation Council forGraduate Medical Education(ACGME) has recently instituted arequirement for “Sleep Loss andFatigue Management Training” inteaching specifications for all resi-dency programs.

Specifically, it is stated that “faculty and residents must be educated to recognize the signs offatigue and adopt and apply policies to prevent and counteractits potential effects.”

Dr. Ward serves as Chair of the Association of Anaes-thetists Working Party on “Fatigue and Anaesthetists”and is Former Vice President of the Association of Anaes-thetists of Great Britain and Ireland. He is also a Consul-tant Anaesthetist and Senior Clinical Lecturer in theNuffield Department of Anaesthetics, Oxford, England.

References

1. Howard SK, Rosekind MR, Katz JD, Berry AJ. Fatigue inanesthesia: implications and strategies for patient andprovider safety. Anesthesiology 2002;97:1281-94.

2. Morris GP, Morris RW. Anaesthesia and fatigue: ananalysis of the first 10 years of the Australian IncidentMonitoring Study 1987-1997. Anaesth Intensive Care2000;28:300-4.

3. Fatigue and Anaesthetists. London: The Association ofAnaesthetists of Great Britain and Ireland, 2004. Availableon the web at: http://www.aagbi.org/pdf/Fatigue.pdf.

4. Fatigue and Anaesthetists (Extended Web Version).London: The Association of Anaesthetists of GreatBritain and Ireland, 2004. Available on the web at:http://www.aagbi.org/pdf/Web_Version_Fatigue.doc


See “Costs,” Next Page

productivity is not typically ascribed to fatigue ofpersonnel in the health care system but is a “real”cost that must also be considered.

Malpractice litigation has surprisingly notbecome a driving force regarding the safety andeconomic ramifications of fatigue. Sleep deprivationof clinicians involved in litigation has only rarelybeen a significant issue even though high levels offatigue are widespread throughout health care.Over many decades, a rise in such litigation hasbeen expected each time the issue is highlighted inthe media (e.g., after the Libby Zion case) or withincreasing unionization of residents, but so far thishas not happened. If it should ever happen, it willcertainly amplify the actual costs of fatigue-inducederrors or negative outcomes.

Finally, one of the most uniform findings offatigue studies of health care personnel is that tiredclinicians have negatively altered mood, whichsurely cannot contribute to caring and humaneinteractions with patients.3 It is hard to put a cost oncallous or insensitive patient care, but the publicoften seems to place a higher premium on personalattention and caring than on technical excellence.Thus, there are clear and major opportunities toreduce costs and improve care by amelioratingsleepy clinicians. Measuring these costs, or theextent to which they are mitigated by countermea-sures, will be very difficult. It is hard to measurethe benefits of the “accident that did not occur”(especially when the accidents are uncommon), orgratified patients who otherwise would have beenupset with their care. The signals of “safety” areinherently weaker than the signals of“production.”4,5

Having said that, let us look at the diverse coun-termeasures that have been suggested, as well assome of the direct and indirect costs they wouldimpose on the health care system. Although thesecosts have never been measured, and cannot bedetailed exhaustively in this article, many of themcan be articulated for consideration in any futurecomprehensive financial modeling of potentialfatigue intervention policies.

EducationA seemingly easy countermeasure is to educate

clinicians about the science of sleep and the ways inwhich they can act to protect their health andimprove their performance. Educational modalities(videos, lectures, and seminar series) require mod-erate investment to produce and disseminate. Forclinicians to attend such training may pull them outof revenue-producing work, and has to be tradedoff against other sorts of training that they arerequired to take. And while the costs may be onlymoderate, the efficacy and effectiveness remain tobe tested.

Sleep HygieneOne typical target of education is to exhort clini-

cians to make sleep a higher priority in their lives.On the one hand, requiring health care profession-als to come to work well-rested is no different fromforcing them to come to work sober and drug-free.On the other hand, to some degree this policy shiftsthe cost “burden” onto individual clinicians whoare being asked to give up an hour or two of theirother activities–whether moonlighting or spendingtime with their families–in favor of sleep. Whilethey will also recoup some direct benefits to theirhealth, mood, and possibly traffic safety, in partthey are being asked to bear the costs of the healthcare system’s 24/7 demands, rather than having thesystem bear those costs collectively by hiring morestaff.

Fitness for Duty Tests/StandardsThe logical conclusion of a variety of counter-

measures that attempt to change the pre-shift phys-iological state of clinicians would be to establishmetrics and standards for “fitness for duty.” In thiscase, clinicians who are impaired would not beallowed to work that day or that shift. The costs forimplementing such standards could be high.Replacement clinicians would need to be immedi-ately available to substitute for those who are not fitfor duty. Measures would have to be taken toensure that the system is not abused by those want-ing extra time off. And, while some causes of sleepdeprivation can be avoided by proper sleephygiene, others (such as the sick child at home) can-not reasonably be prevented even by the most dedi-cated clinician. In addition, the health care systemalready copes with absences due to illness, soadding “fatigue impairment” to other more com-mon illnesses might be a tractable marginal cost.

by David M. Gaba, MD

In the course of studying and writing about theproblem of fatigue in health care personnel, ourgroup and others have suggested a multitude ofpossible countermeasures. In this Special Issue, aworld expert on managing fatigue concerns in avariety of high-hazard industries discusses a num-ber of possible approaches. Many countermeasuresseem simple to implement at little cost. Otherswould require substantial reorganization of theclinical work environment and the clinical work-force. These are thought to have substantial costsassociated with them, and many of them have notbeen studied very thoroughly for efficacy (do theywork in the lab?), effectiveness (do they work inday-to-day practice?), let alone for their costs. Inthis article I will raise a number of questions con-cerning various costs of countermeasures. Few ofthe questions can be answered at this point, andnone can be answered definitively. Attempting toanswer them will be an appropriate component of along-term strategy to come to grips with thedemands of a “24/7” health care industry.

First, the growing literature on the adverseeffects of fatigue on clinician performance impliesthat there are substantial, sometimes hidden, coststo individuals and society already imposed by sub-optimal or inefficient patient care and outrighterrors caused or exacerbated by clinician fatigue.Such costs are difficult to compute. Recent datashow unequivocally that error rates were higher bytrainees working highly fatiguing schedules vs.those working less fatiguing schedules.1 Extrapo-lating the excess errors in the more fatigued groupfrom these studies to the wider clinician populationis appropriate, but imprecise. On the one hand, out-right errors are only the tip of the iceberg. Whatabout orders that are written a little late, justenough to push discharge from an ICU out anotherday? What about the clinic or OR schedule that getsbehind because people are working just a littleslower. On the other hand, not all the costs appar-ently due to fatigue-induced errors can necessarilybe recouped using any of the existing countermea-sures. Thus, in the absence of very complex andlong-term trials that capture direct and indirectcosts due to excess fatigue of clinicians, we cannotbe sure of how much money could be saved by lim-iting clinician sleepiness.

The current situation imposes other societalcosts related to the health care workers themselves.There is growing evidence that fatigued trainees areat risk for traffic accidents, particularly when dri-ving home after long and/or overnight duty peri-ods.2 There is also strong evidence among truckdrivers and others that occupational-related ill-nesses are related to sleep disturbances due to nightwork. The cost of such accidents, illnesses, and lost

The Costs of Clinician Fatigue and Its Prevention

David M. Gaba, MD


shift workers during only 8 or 12 hour shifts. Whatwould it take to provide a guaranteed nap for suchworkers? In some cases it might be feasible forexisting personnel to cross cover sufficiently toallow scheduled naps. In other cases it might benecessary to add one or more individuals on thenight shift to provide rotating coverage for the nap-ping individuals. Modeling the costs of on-shiftnapping will be difficult and will depend heavilyon the exact staffing and work demands of eachclinical setting. Union rules may play a role in whatcan and cannot be done with existing personnelversus with additional clinicians.

Post-Shift NappingNaps taken after the shift is completed might

help with the safety of the drive home and mightalso reduce accumulated sleep debt. If they are vol-untary, the only costs are to provide the facilities fordeparting workers to nap. Experience shows thatmost workers would prefer to depart tired in orderto get home earlier. Those with very long driveshome, or those likely to face heavy traffic, may bethe most likely to choose to nap rather than departimmediately. Thus, unless the preferences of work-ers are changed, or the nap is a mandatory use ofunpaid time, the post-shift nap is unlikely to be arobust strategy for the workforce as a whole, eventhough its costs are modest.

Drug TherapyMany clinicians already use a drug—caffeine—

on a regular basis to promote alertness during thenight. But caffeine is an imperfect drµg and is gen-erally not used very wisely. The new alertnessenhancing drug modafinil (approved for the use inpatients with narcolepsy, daytime sleep complaints

Work Schedule Changes andWork Hours Limitations

A common countermeasure being imposedaround the world, at least for trainees, is limitingtotal work hours and on duty-periods. The Euro-pean Union is currently at a 56-hour weekly limit,with other limits on shift duration and rotation.Eighteen months ago, the ACGME imposed regula-tions on all residency programs in the U.S. thatrequired substantial changes in work schedules.The imposed limit of 80 hours per week is far morelenient than European limitations. Recent studiespublished in the New England Journal of Medicinedemonstrate that altering work schedules candecrease error rates.1 But, the altered schedulesrequire hiring more net clinical employees, or shift-ing work from trainees to others. Both are relativelyexpensive solutions. Calculating the costs of chang-ing the schedules is not easy. There are also indirectcosts to the system. Medical educators are strug-gling with how to provide the same training in thefewer hours available. Some solutions to this prob-lem will themselves have costs.

At-Work NappingSome studies in other industries (and some in

health care) show a benefit to at-work napping dur-ing long or overnight duty periods.6 Even shortnaps of 45 minutes have been shown to be benefi-cial in some settings, improving alertness and per-formance.7 Naps are common for trainees working24-hour (or longer) shifts, but they are availablehaphazardly, and during many shifts little if anysleep can be accrued. Scheduled guaranteed napsmight be useful for such personnel or even for night

“Costs,” From Preceding Page in patients with obstructive sleep apnea, and shiftwork sleep disorder) may provide an alternative.Whether shift workers should use modafinil on aregular basis is one question (see Dr. Lydic’s andDr. Smith’s articles in this issue). But if they should,or if they do, who will pay the costs of the drug?Who will pay the costs of any side effects generatedby taking the drug? And what might be the costs oflitigation that might arise should medical errors bemade that either a patient’s attorney, or the attor-ney of the clinician, ascribe to the use of the drug?

Light TherapyThe application of higher levels of ambient light

at the right time periods has been shown in labora-tory studies to reset circadian rhythms and could beused to aid shift workers.8 Special facilities might beneeded to allow workers to obtain light therapy atthe right time. The timing of light therapy is verycritical to its success—thus, as in the case of nap-ping, relief personnel might be required to allowothers to obtain light therapy.

In conclusion, one thing is certain. The emergingresearch indicates strongly that, both before andafter the ACGME rule changes for residents, ourhealth care industry and our professions haveignored substantial costs to patients and society dueto fatigue-induced errors, poor performance, andautomobile accidents by sleepy clinicians. Afteryears of clamoring for change in the work and dutyhours for physicians in training, significant but lim-ited changes have been imposed. Further changesseem warranted, for this group as well as for expe-rienced health care personnel. We must acknowl-edge that these reforms carry a cost. For thestrongest interventions, the costs are likely to behigh. Modeling the direct and indirect costs is not asimple matter, and must take into account a largevariety of factors that may differ strongly in differ-ent work settings, and at different institutions. Theymay also differ based on the health care disciplineinvolved and the administrative structures of thedifferent components of the workforce. Still, costand policy modeling is a mature art, if seeminglyarcane to most clinicians. The time is right for someserious research on the organizational, logistical,and cost implications of the various proposed coun-termeasures to best inform policy choices. We callon our colleagues in health services research tobecome interested in the economic and policyimplications of the fatigue issue, and help us findthe best solutions to protect our patients, andimprove our own health in the most expeditiousand cost-effective manner possible.

Dr. Gaba is a Professor in the Department of Anesthe-sia and the Associate Dean for Immersive and Simulation-Based Learning at Stanford University, Stanford, CA.

Solutions to Fatigue Require Resources

See “Costs References,” Page 22


Corporate DonorsFounding Patron ($400,000 and higher)American Society of Anesthesiologists (asahq.org)

Foundation Patron

($50,000 to $99,999)GE Healthcare

(gemedical.com)

Benefactor Patron ($25,000 to $49,999)

Abbott Laboratories (abbott.com)

Aspect Medical Systems (aspectms.com)

AstraZeneca Pharmaceuticals LP(astrazeneca.com)

Philips Medical Systems (med-ical.philips.com)

Tyco Healthcare (tycohealthcare.com)

Grand Patron ($15,000 to $24,999)Baxter Anesthesia and Critical Care (baxter.com)Dräger Medical, Inc. (nad.com)Preferred Physicians Medical Risk Retention Group (ppmrrg.com)

Patrons ($10,000 to $14,999)Endo Pharmaceuticals (endo.com)Spacelabs Medical (spacelabs.com)

Sustaining Members ($5,000 to $9,999)Alaris Medical Systems (alarismed.com)Arrow International (arrowintl.com)First Professionals Insurance Company (fpic.com)B. Braun Medical (bbraun.co.uk)Becton Dickinson (bd.com)Datascope Corporation (datascope.com)Kimberly-Clark (kimberly-clark.com) LMA of North America (lmana.com)Medical Education Technologies, Inc. (meti.com)Organon (organon.com)Roche Laboratories (roche.com)Smiths Medical (smiths-medical.com)The Doctors Company (thedoctors.com)Vance Wall FoundationVital Signs (vital-signs.com)

Sponsoring Members ($1,000 to $4,999)Anesthesia Business Consultants (anesthesiallc.com)Cerner Corporation (cerner.com)Cook Critical Care (cookgroup.com)Docusys, Inc. (docusys.net)King Systems Corporation (kingsystems.com)Masimo Corporation (masimo.com)Oridion (oridion.com)Somnia, Inc. (somniainc.com) Trifid Medical Group, LLC (trifidmedical.com)

Corporate Level Members ($500 to $999)Armstrong Medical, Ltd. Belmont Instrument Corporation (belmontinstrument.com)Boehringer Laboratories (autovac.com)Hoana Medical, Inc. (hoana.com)Lippincott Williams and Wilkins

Participating AssociationsAmerican Association of Nurse Anesthetists (aana.com)

Subscribing SocietiesAmerican Society of Anesthesia Technologists and Technicians

(asatt.org)American Academy of Anesthesiologist Assistants (anesthetist.org)

Anesthesia Patient Safety Foundation

Community Donors (includes Anesthesia Groups, Individuals, Specialty Organizations, and State Societies)

Grand Sponsor ($5,000 and higher)American Association of Nurse AnesthetistsAmerican Medical FoundationFlorida Society of AnesthesiologistsIndiana Society of AnesthesiologistsRobert K. Stoelting, MD

Sustaining Sponsor ($2,000 to $4,999)Alabama State Society of AnesthesiologistsAnesthesia Associates of MassachusettsAnesthesia Consultants Medical GroupAnesthesia Resources ManagementAnesthesia Service Medical GroupArizona Society of AnesthesiologistsNassib G. ChamounGeorgia Society of AnesthesiologistsIndianapolis Society of AnesthesiologistsIowa Society of AnesthesiologistsMassachusetts Society of AnesthesiologistsMichigan Society of AnesthesiologistsMinnesota Society of AnesthesiologistsOhio Society of AnesthesiologistsOld Pueblo Anesthesia GroupPennsylvania Society of AnesthesiologistsSociety of Cardiovascular Anesthesiologists

Contributing Sponsor ($750 to $1,999)Academy of AnesthesiologyAffiliated Anesthesiologists, Inc.American Society of Critical Care AnesthesiologistsJ. Jeffrey Andrews, MDAnesthesia Associates of Northwest Dayton, Inc.Anesthesia Services of BirminghamAnesthesia Service of Eugene, PCArkansas Society of AnesthesiologistsAsheville Anesthesia Group

Association of Anesthesia Program DirectorsCapital Anesthesiology AssociationFrederick W. Cheney, MDConnecticut State Society of AnesthesiologistsJeffrey B. Cooper, PhDJohn H. Eichhorn, MDDavid M. Gaba, MDIllinois Society of AnesthesiologistsKentucky Society of AnesthesiologistsEdward R. Molina-Lamas, MDMadison Anesthesiology ConsultantsMaine Society of AnesthesiologistsMaryland Society of AnesthesiologistsMembers of the Academy of AnesthesiologyMichiana Anesthesia CareMissouri Society of AnesthesiologistsRobert C. Morell, MDNebraska Society of AnesthesiologistsJohn B. Neeld, MDNevada State Society of AnesthesiologistsNew Mexico Anesthesia ConsultantsNorthwest Anesthesia PhysiciansNeshan Ohanian, MDOklahoma Society of AnesthesiologistsOregon Anesthesiology GroupOregon Society of AnesthesiologistsPhysician Anesthesia ServicePittsburgh Anesthesia AssociatesSociety of Academic Anesthesia ChairsSociety for Ambulatory Anesthesia Society of Neurosurgical Anesthesia & Critical CareSociety for Pediatric AnesthesiaSouth Carolina Society of AnesthesiologistsSouth Dakota Society of AnesthesiologistsTennessee Society of AnesthesiologistsTexas Society of AnesthesiologistsWashington State Society of AnesthesiologistsWest Jersey Anesthesia Associates, PAWisconsin Society of Anesthesiologists

Sponsor ($100 to $749)Anesthesia Associates of BoiseAnesthesia Associates of Chester CountyAnesthesia Associates of St. CloudSusan Bender, CRNASorin J. Brull, MDCalifornia Society of AnesthesiologistsRobert A. Caplan, MDCardiovascular Anesthesia, LLCCHMC Anesthesia FoundationMelvin A. Cohen, MDStuart M. Cohen, MDColorado Society of AnesthesiologistsCommonwealth Anesthesia AssociatesW. Keith Conerly, CRNAPaula A. Craigo, MDGlenn E. DeBoerWalter C. Dunwiddie, MDNorig Ellison, MDJan Ehrenwerth, MDBarry J. Friedberg, MDGeorgia Anesthesia AssociatesBarry M. Glazer, MDGriffin Anesthesia AssociatesPeter Hendricks, MDJames S. Hicks, MDDr. and Mrs. Glen E. HolleyVictor J. HoughIdaho Society of AnesthesiologistsIndependence Anesthesia, Inc.Sharon Rose JohnsonDaniel J. Klemmedson, DDS, MDRoger W. Litwiller, MDThomas T. McGranahan, MDMedical Anesthesiology Consultants CorporationMississippi Society of AnesthesiologistsRoger A. Moore, MDErvin Moss, MDNew Hampshire Society of AnesthesiologistsNew Jersey State Society of Anesthesiologists

Beverly K. Nichols, CRNAL. Charles Novak, MDDenise O’Brien, RNJill Oftebro, CRNAPhysician’s Accounts ReceivableMukesh K. Patel, MDPhysician Specialists in AnesthesiaRichard C. Prielipp, MDDebra D. Pulley, MDRhode Island Society of AnesthesiologistsSociety for Technology in AnesthesiaSouthern Tier Anesthesiologists, PCSouth County Anesthesia AssociationUniversity of Maryland Anesthesiology

AssociatesUtah Society of AnesthesiologistsVermont Society of AnesthesiologistsVirginia Society of AnesthesiologistsDrs. Mary Ann and Mark A. Warner, MDMatthew B. Weinger, MDWest Virginia Society of AnesthesiologistsDr. and Mrs. WetchlerLawrence Wobbrock, JDWaterville Anesthesia AssociatesWest River Anesthesiology ConsultantsWoodland Anesthesia Associates

In MemoriamIn memory of Gale E. Dryden, MD (Friends of the

Dryden family)In memory of Margie Frola, CRNA (Sharon R.

Johnson, MD)In memory of Charles L. Maimbourg, MD (Texas

Society of Anesthesiologists)In memory of Walter H. Mannheimer (Texas Soci-

ety of Anesthesiologists)In memory of Robert A. Sandoval, MD (Texas

Society of Anesthesiologists)In memory of Morton V. Sinclair (Russell S.

Peterson, MD)

Note: Donations are always welcome. Send to APSF; c/o 520 N. Northwest Highway, Park Ridge, IL 60068-2573 (Donor list current through March 14, 2005)


• Identify the most important research questions,and the key gaps between what is being studiedand what needs to be studied

• Develop an agenda for possible research

• Determine if existing interventions requiregreater attention for evidence-basedguidelines/practice parameters

• Identify new interventions in need of study

• Develop a plan for dissemination and follow-up.

Session 1: Epidemiology of andRisk Factors for Long-Term

OutcomesRobert Lagasse, MD, began by reviewing some of

the history of research on short-term, adverse out-comes in anesthesiology, illuminated issues in study-ing long-term outcomes, and suggested that theadvent of Anesthesia Information Management Sys-tems (AIMS) could make it easier to link intraopera-tive events to both short- and long-term outcomes.Next, Shukri Khuri, MD, described the VeteransHealth Administration’s National Surgical QualityImprovement Program (NSQIP). Dr. Khuri is a car-diac surgeon who has led the NSQIP project formany years. NSQIP tracks surgical outcomes (typi-cally out to 30 days post-op) for a number of majorsurgical procedures, in a variety of surgical special-ties, across all VA facilities that perform them. TheNSQIP database depends on data entered by a spe-cially trained nurse at each site, who is able to lever-age the VA’s electronic medical record system. Dataare collected on a large number of preoperative vari-ables, on a few intraoperative variables (e.g., dura-tion of surgery, surgical procedure), and then on asuite of postoperative outcome variables. The NSQIPdatabase now has more than 1.2 million records andfor selected patient groups has assessed mortality out

to 10 years. The NSQIP program has expanded to theprivate sector, first in a set of studies and now in ajoint program with the American College of Sur-geons–ACS-NSQIP. Finally in this session, TerriMonk, MD, discussed the history of studies of long-term mortality following anesthesia and surgery indifferent patient populations. She provided anoverview of intervention studies such as those usingbeta-blockade in the perioperative period. She alsopresented preliminary but provocative data from heroriginal research examining perioperative cognitivedysfunction in patients having general anesthesia.Data on intraoperative vital signs and blinded bis-pectral index (BIS) data were recorded for thesepatients as were various postoperative outcomes,including mortality at 1 year. The data on mortalitywere then subjected to multiple regression modelingto determine which underlying preoperative andintraoperative factors correlated with death at 1 year.Not surprisingly, underlying medical problems werethe major risk factors (odds ratio 16.1), as was timespent with a systolic blood pressure less than 80mmHg (odds ratio 1.04/minute <80). A BIS value<40 was also identified as a significant risk factor. Dr.Monk also described results of a similar study per-formed on a larger number of patients in Sweden byLennmarken et al. that also showed that BIS <45 wasa statistically significant risk factor. Dr. Monkdescribed all of these results as surprising, since therewas no obvious mechanism to account for these find-ings. In addition, approximately one-half of thedeaths in her study were due to cancer, and the cor-relation of time with hypotension or with BIS <40 todeath due to cancer was even harder to explain.

Discussion of Session 1Robert Stoelting, MD, APSF President, chaired

the discussion of Session 1, at which the panelreviewed the epidemiology. There was particularfocus on the data from Weldon and Monk’s studyand from Lennmarken et al. because these resultswere unexpected. It was acknowledged that thesedata are preliminary and are strictly observational.Nonetheless, they are provocative data, as they sug-gest that some factor occurring during the briefperiod of the anesthetic may be linked to mortalityremote from the perioperative time frame. Therewas spirited debate about these data and what theycould mean if they are confirmed with furtherstudies. One suggestion was that the occurrenceof BIS <45 represents merely a marker for patientswho are particularly vulnerable for some reason.Another possibility discussed was that thesepatients may have a higher adrenergic state andare subsequently treated with higher levels of

by David M. Gaba, MD, and Robert C. Morell, MD

Anesthesiologists have long been pioneers inseeking ways to improve the safety of patients. Todate, most of our efforts have been directed towardreducing the likelihood of adverse events in theimmediate perioperative period. Over the last fewyears several threads of information have been coa-lescing that suggest there is another patient safetyissue involving adverse outcomes occurring longafter the traditional perioperative period. Theseoutcomes are not directly tied to a specific compli-cation of the surgery and may be referred to as“long-term outcomes” of anesthesia and surgery.

The APSF has begun to take an active interest inthis topic. In the APSF Newsletter (Fall 2003 andSpring 2004), a number of these issues were intro-duced, along with the hypothesis that the underly-ing biological mechanism to explain the occurrenceof such long-term outcomes might have to do withinflammatory processes triggered in the periopera-tive period. As detailed in those articles there are avariety of reasons to think that mortality, and pre-sumably morbidity, can be affected by periopera-tive events, and that inflammation could be a keyelement in such occurrences.

Because the threads were disparate, from differ-ent medical domains, and still uncertain, the APSFdecided to convene a multidisciplinary experts’workshop. Following is a brief report of some of thehighlights from that workshop. Dr. David Gabawas selected as the principal investigator for theworkshop because of his extensive research experi-ence, expertise in patient safety, role as secretary ofthe APSF, and the fact that he does not conductresearch in this particular area. Thirty expertsattended the conference in Boston, MA, on Septem-ber 21-22, 2004. The names of attendees and theirbiographies are available on the APSF website. Thefull program for the workshop is also available onthe site. The program was divided into 4 sessionsincluding epidemiology of long-term outcomes fol-lowing anesthesia and surgery, 2 sessions oninflammatory processes and other underlying bio-logical mechanisms, and a final session reviewingthe issues discussed and the pitfalls of conductingoutcomes research, and debating the best course ofaction for future research.

The goals for the conference were to:

• Define the problem(s) of long-term outcomesafter anesthesia and surgery

• Estimate the scope and nature of the problem(s)

• Assess the state of the science of the putativeinflammatory mechanisms See “Workshop,” Next Page

Report of APSF Workshop onPostoperative Long-Term Outcome


hypnotics or volatile anesthetics. It was also sug-gested that the patients with the low BIS valuesmight have a greater or more extended inflamma-tory response to anesthesia and surgery, althoughthere are no studies yet that have investigated sucha putative mechanism. It was widely agreed thatthese questions will need to be examined in largerstudies that specifically investigate these issues.

There was additional discussion of other periop-erative factors and treatments that have an impacton long-term outcomes. In particular, the data onperioperative beta-blockers are complex. Whilethere have been multiple randomized trials, andthere is a consensus for beta-blockade in patientswith known cardiac disease having vascularsurgery, whether this is beneficial for broader use isstill open to considerable debate.

Sessions 2 and 3 – InflammationPotential biological mechanisms for such occur-

rences were the topics of Sessions 2 and 3. At thebeginning of Session 2, Dr. Steffen Meiler proposeda set of hypotheses concerning the perioperativeinflammatory/immune response as a potential bio-logical link to long-term outcomes after anesthesiaand surgery. Dr. Meiler proposed a “two-hit”model, which states that the inflammatory responseto surgery may amplify pro-inflammatory cellmechanisms of certain disease states, such as coro-nary artery disease, hence contributing to diseaseacceleration and adverse postoperative events. Theevidence that inflammatory processes are criticalfor the progression of atherosclerosis is undeniable.Similarly, inflammatory processes and infectionsare known to play a key role in cancer biology (e.g.,hepatitis leading to hepatocellular carcinoma). Therole of inflammation in the degenerative centralnervous system diseases, such as Alzheimers, ismore tentative, but a growing body of evidence def-initely points in this direction.

Furthermore, Dr. Meiler proposed that certainpatients or patient populations may exhibit anexaggerated inflammatory response to surgeryand/or delayed resolution to the preoperativeimmune status. Limited human data are in supportof this notion. If true, these patients may be at evengreater risk to experience postoperative complica-tions. What would cause an abnormal inflamma-tory response to surgery is not known, nor are allthe factors that might be triggers beyond the surgi-cal procedure itself. Whether anesthetic drugs,other aspects of anesthetic technique, or physiologicoccurrences during surgery could be potent triggersfor abnormal inflammation is not well established.There are threads of evidence that anxiety, fear, andpain can trigger inflammation.

This led Dr. Meiler to propose that perioperativecare is a key nexus for affecting both short- andlong-term outcomes. A common goal between anes-

thesiologists, surgeons, internists, and others willbe to identify which patients are at risk, to defineadjuvant treatments and modified patient careprocesses to prevent the negative outcomes, and toapply them throughout the continuum of the peri-operative period. According to these models, takingan inflammation/ immune-based approach to dis-secting the biological interactions between anesthe-sia, surgery, and postoperative complicationstherefore promises to yield important insights.

Last in the session, Rod Eckenhoff, MD, dis-cussed the effects of volatile anesthetics on theoligomerization of brain proteins. This line ofresearch was triggered by the speculation thatmany neurodegenerative diseases may be causedby the aggregation of normal and abnormal pro-teins (similar to what occurs in mad cow disease).Halothane and other volatile anesthetics do causeoligomer formation in amyloid precursor protein atclinically relevant levels, and this process lasts along time. A single exposure of desflurane caused 3days of differences in protein expression. Otherproteins could be affected similarly. One examplecited is ferritin, which binds volatile anesthetics atlow concentrations. New animal models are beingestablished in rats, which are thought to be a goodmodel system. Finally, although these mechanismssuggest that exposure to volatile anesthetics mightbe linked to the occurrence of dementia, this conclu-sion is very speculative at this point.

Discussion of Session 2The discussion was led by Dr. Carl Rosow.

There was extensive discussion about what ismeant by “stress,” and whether “anxiety,” “stress,”or other terms really describe the same state. Therewas further discussion of the diverse responses andtimeframes of inflammatory responses that werebeing discussed. Some inflammatory response iscritical for wound healing and warding off surgicalinfection, yet too much or too prolonged a responsemight be deleterious. The panel tried to determinewhether any existing studies show a definitive linkbetween the kinds of inflammatory mechanismssuggested and postoperative complications. Somethreads were drawn from studies of patients havingcardiac surgery and cardiopulmonary bypass, but itwas acknowledged that no studies to date demon-strate this specific putative linkage. There was dis-cussion of whether one could study the long-termoutcome of animals (rodents in particular) that hadundergone anesthesia and surgery, while studyingtheir inflammatory responses. There was wideagreement that understanding the subtle issues oflong-term outcomes would require both animal andhuman studies.

Session 3: InflammatoryMechanisms Redux

Session 3 continued the theme of looking at thebasic biological aspects of inflammatory mecha-

nisms. First, Charles Serhan, MD, presented a fasci-nating description of the active processes thatresolve inflammation after it has been triggered.This resolution is not just a “burnout” of the pro-inflammatory functions, but rather has a set of reso-lution functions that involve resolving mediators.Resolution is thus different than mere “anti-inflam-mation,” and the resolution processes offer anotherpotential target for therapeutic manipulation. Fur-thermore, anti-inflammatory therapies may some-times also inhibit the natural pro-resolutionpathways, thus delaying or blunting their beneficialeffects. Many of the mediators of the resolutionphase are lipid mediators, among which are lipox-ins, resolvins, and neuroprotectins. Resolvins maybe the active ingredients of the beneficial effects ofomega3 fish oil and other dietary elements. Similarmolecules in the nervous system are called neuro-protectins. Dr. Serhan summarized the potentialpromise of this line of investigation as offeringways to mitigate the negative effects of inflamma-tion by turning on resolution rather than byattempting to inhibit the pro-inflammatory phaseitself. However, turning this basic science into ther-apies ready for clinical trials will take some time.

Discussion of Session 3Don Stanski, MD, chaired the discussion of Ses-

sion 3. There was extensive discussion of how tofind the middle ground between the elegant basicscience work and the interface to the clinical issuesof long-term, post-surgical outcomes. In this regard,there is some work underway to develop more eas-ily run assays for some of the resolution molecules.Another thread was trying to better link the seem-ingly beneficial effects of drugs like beta-blockers,clonidine, statins, and the underlying bioactivemediators.

An interesting question was raised as to theeffects of discontinuing patients’ aspirin or non-steroidal anti-inflammatory therapies prior tosurgery so as to minimize their effects on plateletaggregation, and thus on perioperative blood loss.A side effect of stopping these drugs could be topromote or unmask a more extreme inflammatoryresponse.

In addition, there was considerable discussionabout whether it is beginning to be possible to teaseout which patients are most susceptible to short andlong-term negative outcomes on the basis of theirgenotype or biochemical markers, in addition to thetraditional risk factor analysis.

Session 4: Wrap UpDr. Jeff Cooper chaired this session and reiter-

ated the interest of the APSF in the concept of long-term, postoperative outcome. Active participants inthis session included Drs. Dan Sessler and LeeFleisher who shared their experience, observations,

Inflammatory Response May Be Deleterious“Workshop,” From Preceding Page

See “Workshop,” Next Page


and recommendations concerning future investiga-tions. Dr. Karl Hammermeister proposed a “strawman” sequence of investigations to be consideredby the panel. This sequence would be to:

• Confirm excess late adverse outcomes (e.g.,mortality)

• Identify predictors of late adverse outcomes

• Evaluate mechanisms of excess late adverseoutcomes

• Conduct small-scale trials of interventions forexcess late adverse outcomes

• Conduct large-scale comparative randomizedclinical trials of interventions.

Following this discussion Dr. Cooper called for3 mini-votes of the participants:

• Do you believe that there IS some relationshipbetween inflammatory processes in theperioperative period and long-term survival?

A majority of participants voted yes.

• Should an appropriate study be done to measure“excess” mortality resulting from identifiablefactors of anesthesia and surgery?

Again, a majority of participants voted yes.

• Should such a study demonstrating “excessmortality” be completed before any otherstudies–such as those suggested in the “strawman”–are begun?

Only a few participants voted yes.

Marcel Durieux, MD, PhD, advocated conduct-ing campaigns of basic research and clinicalresearch in parallel, since useful clinical investiga-tions can be done even before the underlying mech-anisms are fully defined in the laboratory. Amongthe clinical questions that currently seem ready forinvestigation, he listed:

• Relationship of intraoperative EEG measures tolong-term mortality. The panel largely agreedthat the data discussed so far are intriguing butvery preliminary. He emphasized that the issueis not necessarily “deep anesthesia,” but that thismight be a marker for patients who havedifferent underlying physiology. We need to findout what causes this relationship, increasedrequirement, or exaggerated response.

• The development of chronic pain after surgery,and use of “preventive analgesia” to preventpostoperative chronic pain

• Perioperative transfusion and its effect on long-term outcome

• The development of a perioperative hyper-coagulable state, which may result in thrombo-embolism and/or myocardial ischemia.

Some comments were made about possible fol-low-up activities stemming from this conference.Dr. Khuri indicated that NSQIP is very interestedabout incorporating AIMS data in NSQIP. Dr. Huntfrom CMS indicated that CMS might well be inter-ested in adding additional variables to the databaseanalysis projects that are currently in design.

Dr. Thomas Russell, the Executive Director ofthe American College of Surgeons, said that it wasvery clear that the entire team caring for a surgicalpatient—surgeon, anesthesiologist, primary carephysician, cardiologists, oncologists, and many oth-ers—needed to engage in serious new efforts toshare and coordinate their knowledge, perspectives,and clinical efforts in order to optimize outcomesfor the patients. Never again should they satisfythemselves to work solely within their own silos, nomatter how expertly.

In summary, the group arrived at a number ofthreads of agreement and observation, which willnow serve as the basis for future analysis andaction. These include

1. Historically, surgeons and anesthesiologists havelargely felt their actions only have immediate ornear-term consequences.Things not directlyrelated to the surgical procedure that occurred"way down the line" (the "long-term outcomes")had to do with the patient's underlying medicalconditions and were just bad luck. But the groupthought it was distinctly possible that there arethings that happen during surgery that havelasting effects and may have a long-term impacton morbidity and mortality.

2. The group acknowledged that there may in factbe excess mortality over the long-term linked tothe process of anesthesia and surgery. But thedata are extremely sparse, complicated, and havemany limitations and pitfalls. The questionshould be pursued further to find moredefinitive answers.

3. There should be more studies of large numbersof patients to better identify risk factors for theoccurrence of adverse long-term outcomes aswell as for short-term complications.

4. Inflammation has been implicated in manydisease processes and it is definitely possible thatthere exists a relationship between inflammationand the long-term outcomes associated withsurgery and anesthesia. Much remains to bedetermined to see if this linkage is present, and if

so, its strength and what can be done about it.Studies are needed both on the basic biology ofinflammation, and on the specifics of this biologyin the setting of anesthesia and surgery.

5. As we collect better data about the nature ofpostoperative outcome, studies are needed toevaluate the mechanisms, and define possibleinterventions. This may happen first in small-scale trials, but ultimately large-scale studies,with thousands of patients, will be needed. Thebenefits of extrapolating existing therapies, suchas perioperative beta-blockade, to broadergroups of patients requires further study as well.

Dr. Gaba is a Professor in the Department of Anes-thesia and the Associate Dean for Immersive and Simu-lation Based Learning at Stanford University, Stanford,CA. He is also Secretary of the APSF.


Hypoxemia in theHospital Frequentlyan Early Sign ofImpendingCatastrophe

To the Editor:

Dr. Stemp makes an interesting point, albeitwith great drama: detection of transient hypoxemiais not nearly as vital to patient outcome as determi-nation of cause. Summarizing the logic of his letter,one would conclude that treating hypoxemia withoxygen but neglecting further investigation enablesthe undetected respiratory pathology to continue.

His points are legitimate. Most humans outsidethe hospital tolerate transient hypoxemia withoutsequelae as it is usually produced by a benign orself-limited process. However, hypoxemia in thehospital frequently is an early sign of impendingcatastrophe.

Perhaps Dr. Stemp would agree with this con-clusion: oxygen should be the first step in address-ing hypoxemia, but never the last.

Samuel Metz, MDOregon Anesthesiology GroupPortland, OR

Working Group Develops Basis for Action“Workshop,” From Preceding Page


data is confidential. In the semi annual risk-adjusted reports the hospitals can only identifythemselves. Confidentiality is maintained for sur-geons, hospitals and patients.

The ACS and the APSF recently collaborated tosubmit a proposal to the National Library of Medi-cine requesting funds to investigate the feasibilityof transferring intraoperative data to the ACSNSQIP database from the Anesthesia InformationManagement System operating in some hospitals.The availability of intraoperative hemodynamic,metabolic, and temperature data could substan-tially improve the capacity of the ACS NSQIP topredict and evaluate surgical outcomes. The ACSNSQIP provides an excellent opportunity for sur-geons, anesthesiologists, and nurse anesthetists towork together effectively to evaluate and therebyimprove surgical outcomes.

The reduction of morbidity and mortality aresufficient justification to employ this program. Ithas the added advantage of allowing surgeons tofulfill their professional responsibility of self-evalu-ation. It is also an excellent opportunity for sur-geons to work in collaboration with the leaders ofhospital administration, anesthesiology, and nurs-ing services to benefit patients.

As representatives of the American Collegeof Surgeons we sincerely hope the administra-tors, surgeons, anesthesiologists, nurse anes-thetists, and nurses of all hospitals choose tointroduce, maintain, and use this unique surgi-cal quality improvement tool. Further informa-tion about ACS NSQIP can be obtained bycontacting [email protected], the ACS website atwww.FACS.org, or www.nsqip.org.

Dr. Jones is Director, Division of Research and OptimalPatient Care, American College of Surgeons, Chicago,IL, and Professor of Surgery University of Virginia,Charlottesville, VA.

community hospitals. With 2 years of complete datathe NSQIP functions very well in these 18 privatesector hospitals.

The private sector hospitals could not use VAresources, facilities, or information systems. Forthat reason, a private company QCMetrix, devel-oped a web-based data collection system andtrained the private sector nurses. The ColoradoHealth Outcomes Program (COHO) affiliated withthe University of Colorado provides biostatisticalservices, data management, and report preparationfor the private sector initiative.

In addition to proven performance in the VA, 4year’s private sector experience has demonstratedthe effectiveness of the NSQIP as a quality improve-ment tool and a source of new clinical knowledge.So the ACS developed a business plan to offer thisprogram, beginning with General and VascularSurgery, to all interested hospitals. The VA pro-gram will continue as the VA NSQIP and the pri-vate sector initiative will become the ACS NSQIP.

The ACS NSQIP will contribute to the reductionof surgical mortality and morbidity. The success ofACS NSQIP depends upon several factors. First, thetrained, dedicated nurses provide reliable data. Theprogram includes measures of reliability of the datacollection so there is minimal variability among thehospitals in the program. Second, each hospitalmust have a surgeon responsible for the conduct ofthe program. The surgeon works closely with thenurse and monitors the data. Third, an ExecutiveCommittee reviews the reports provided to the hos-pitals on a regular basis. Fourth, the ACS NSQIPgenerates Best Practices from the database andmakes them available on the website. Fifth, the sur-geon and the nurse can review their hospital’s data(unadjusted) on the website and compare it withthe means of the other hospitals in the programcontemporaneously. The website provides a veryversatile instrument for drilling down into the data-base to examine the variables in detail. This data-base can form the cornerstone of the morbidity andmortality conference and other quality improve-ment programs in the hospital. All ACS NSQIP datais encrypted when it leaves the participating hospi-tal for management by QCMetrix and COHO. Also,all hospitals are de-identified in the database. All

By R. Scott Jones, MD, FACS

Surgical operations, in contrast to medical treat-ments of chronic diseases, lend themselves to obser-vation by outcome studies. Whether patients live ordie, have complications, are cured, have theirsymptoms relieved, return to work or play, and aresatisfied with their care are very important issuesvital to the assessment of the quality of surgicalcare. The Veterans Administration Health System(VA) addressed surgical quality improvement bydeveloping the National Surgical Quality Improve-ment Program (NSQIP) to employ a prospective,peer-controlled, validated database to quantify 30-day risk-adjusted surgical outcomes.

Dedicated, trained, NSQIP nurses collectspecific preoperative, intraoperative, and post-operative data. Statistical analysis of the preop-erative data identifies risk factors predictive ofoutcome. Further analysis calculates the expectedoutcome of a patient population. The observed out-come/expected outcome (O/E ratio) denotes therisk-adjusted outcome. A low O/E ratio indicatesbetter than expected outcome and a high O/E ratioindicates poorer than expected outcome (mortality,morbidity, or other). The NSQIP O/E ratio, basedon valid, reliable data measures the quality of surgi-cal care. This risk-adjustment allows valid compar-isons of outcomes among all hospitals in theprogram. Also, the NSQIP data identifies sources ofmorbidity and mortality to enable managementpractices and improvement of the processes of careto reduce morbidity and mortality. After implemen-tation of the NSQIP, between 1992 and 2002, the VAsurgical mortality decreased 27% and the morbiditydecreased 45%.

After recognizing the effectiveness of the NSQIPthe leaders in the VA arranged for its preliminaryevaluation in 3 university hospitals, Emory Univer-sity, University of Kentucky, and the University ofMichigan. The program worked very well in thatsetting. Then the VA and the American College ofSurgeons (ACS) began collaboration for furtherapplication of the NSQIP in the private sector. Withsupport from a grant from the Agency for Health-care Research and Quality (AHRQ) to the ACS theNSQIP was introduced into 11 additional universityhospitals. Later data was included from 4 affiliated

American College of Surgeons CollaboratesWith APSF to Develop Quality ProgramThe American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP)

APSF Grant ApplicationInformation Available on

Our Website:www.apsf.org


References1. Landrigan CP, et al. Effect of reducing interns' work

hours on serious medical errors in intensive care units.N Engl J Med 2004;351:1838-48.

2. Barger LK, et al. Extended work shifts and the risk ofmotor vehicle crashes among interns. N Engl J Med2005;352:125-34.

3. Veasey S, et al. Sleep loss and fatigue in residency train-ing: a reappraisal. JAMA 2002;288:1116-24.

4. Gaba DM. Structural and organizational issues inpatient safety: a comparison of health care to other high-

Letters to the Editor:

Reader SoftensView of OR ReadingTo the Editor:

I am replying to your article in the APSFNewsletter and offering you my perspective on theissue. I am in total agreement that the image weproject is important, and the image of Dr. Gieseckewith his feet propped up on the anesthesia machinepresents a poor image.

On the other hand, the level of sensory depriva-tion can be extremely variable in cases, and the abil-ity to multitask varies among anesthesiologists. Inthe past, manpower was sufficient, and case vol-ume was large enough that those in need of con-stant stimulation could just stick with open hearts,AAAs, and so forth. Today, we might do complexcases one day and the next be stuck in a dark roomwith an ASA 1 or 2 patient for 2-4 hours in a casewith almost no potential for blood loss with only aminimal break from a sympathetic colleague.

Reading material downloaded onto a PDA froma site such as Avantgo® is much more discreet andcan be done such that monitors can be still in thefield of view, keeping someone used to multitask-ing in complex cases from being lulled into bore-dom. Useful PDA software such as ePOCRATeS®

and other clinical applications make it a commondevice for clinicians to keep with them.

When I started practice, I only supervised resi-dents and CRNAs in an academic setting and was astaunch opponent of any reading in the OR. In mycurrent practice, I supervise some, but often docases of varying complexity alone. This experiencehas caused me to soften my position on this issue.

Mont Stern, MDEast Amhert, NY

PACU Pulse OxTone ToutedTo the Editor:

I'm in total agreement regarding the specialneed to hear the pulse beep tone for pulse oximetryduring anesthesia. Anesthesiologists can testify tothe importance of the pitch/saturation ratio (and tothe fact that the "dive-bombing falling pitch" willnever go unnoticed).

However, I'm always amazed that when I takea patient to the PACU, the nurses have turned offall "noise-makers" (is it to have a quiet place for therecovering patients?).

I have often asked them to turn the pulseoximetry beep tone on with my patients. I explainto them they can turn their backs on the patient andstill hear the most important beep in the world.

I would like to suggest that the APSF institute apolicy for the PACU which is identical to the OR.

Sincerely,

Heddy-Dale Matthias, MDOne Roses Bluff ParkwayMadison, MS [email protected]

Do Not BreakContact Withthe PatientTo the Editor:

I read with interest, and sadness of course, ofthe 3 examples of anesthetic accidents in the article"Turn Your Alarms On" in the winter issue of theAPSF Newsletter.

I have been retired for almost 15 years, but wasforcibly reminded by the article of my constantadmonition to residents and students: “Do notbreak contact with the patient.”

A precordial stethoscope for all consciouspatients and an esophageal one for all under gen-eral anesthesia was a constant routine for all mypatients. My spiel was,

“You are either hearing the heart beat oryou are feeling it beat at all times. This way,you can turn your back to the patient to drawup meds or write on the chart. You can carry ona conversation with the surgeon, but you neverlose input from the heart. The beeping EKGhelps, but is not a substitute. The oximeter andcapnograph are great, too, but nothing takes theplace of listening to or touching the patient.”

There is no excuse, of which I am aware, for theanesthesiologist to leave the head of the patient; he orshe has no obligation to help anybody do anything;his only obligation is to the patient's safety. Thissounds mundane, but I pounded it into my residents.

Monitors are fine, but these 3 accidents would nothave happened if my rules had been followed.

Terring W. Herionimus III, MD, FACA, FCCPProfessor of Anesthesiology, RetiredUniversity of Virginia Medical CenterWest Virginia University Health Sciences Center

“Costs References,” From Page 16

References Support Fatigue Concernshazard industries. California Management Review2000;43:83-102.

5. Reason J. Human Error. Cambridge: Cambridge Univer-sity Press. 1990:1-302.

6. Dinges DF, Broughton RJ. Sleep and Alertness: Chrono-biological, Behavioral, and Medical Aspects of Napping.New York: Raven Press, 1989.

7. Rosekind MR, et al. Crew Factors in Flight OperationsIX: Effects of Planned Cockpit Rest on Crew Perfor-mance and Alertness in Long-Haul Operations. NASATechnical Memorandum #108839 Moffett Field, CA:NASA Ames Research Center, 1994.

8. Czeisler CA, Dijk DJ. Use of bright light to treat mal-adaptation to night shift work and circadian rhythmsleep disorders. J Sleep Res 1995;4:70-3.

Improve Safety by Supporting Your APSF

u Individualsu State Societiesu Group Practicesu Safety Advocates

Send Your Contribution To:APSF

520 N. Northwest HighwayPark Ridge, IL 60068-2573

Check Out the

New APSF Website:

www.apsf.org


To the Editor:

In the Fall 2004 issue of the APSF Newsletter, Dr.Nichols et al. stressed the importance of machinecheck, especially of the reusable pieces of equip-ment to avoid a foreign body in the circuit causinghigh airway pressures (HAP) after intubation. Thisproblem is not uncommon and continues to causeanxiety when encountered.

I am sure many of us follow a pattern of checkswhen confronting the problem of HAP after intuba-tion. The order of these checks is decided by thepreexisting problems of the patient, the machine,and also by previous encounters of a similar nature.

The checks have to be quick (time and life-sav-ing), systematic, and cost-effective. The followingsequence, although not exhaustive, may suit mostof the circumstances (Figure 1) and is for the benefitof anesthesiologists who have not encounteredHAP after intubation, and, it is hoped, will neverhave to use this algorithm.

On initial notice, it may be a good practice toput one’s hand under the drapes and feel for theETT outside the patient, which may be obstructedby the surgeon’s hand or kinked due to the plastic’sinherent properties. Following that, perform auscul-tation of the chest to detect breath sounds. Precor-dial stethoscopes placed prior to draping come inhandy. Rhonchi, or bilateral absent breath sounds,are treated by a dose of albuterol. If the HAP con-tinues, start Ambu bag ventilation. If ventilation iseasy, the obstruction is proximal to the ETT. This isapproached systematically from either the anglepiece and upstream or the ventilator and down-stream, not forgetting that some of the causesmissed are a stuck O2 flush valve, wrong vacuumsettings in the scavenging system, and leak fromventilator bellows.

The causes of HAP distal to the angle piece areeither from the patient or the ETT. With continuinganesthetic, by the intravenous route if necessary,pass a suction catheter down the ETT (Figure 2).This will give some information, depending onhow far one is able to pass the catheter. Once theETT is eliminated as a cause for HAP, the patientcauses are approached while having in mind thepreexisting problems of the patient and thesequence of events that led to the HAP. This sys-tematic approach avoids changing the ETT unnec-essarily with its attendant problems of laryngealtrauma, saves the time of reintubation, and is alsocost-effective.

One can suitably modify the scheme shown inthe tables for the individual circumstancesdepending on the degree of suspicion about a par-ticular cause.

Hariharan Shankar, MDMilwaukee, WI


High Airway Pressure Mandates Diagnosis and Remedy

Mucus plug/FBsuctioned

Paralyze/deepen anesthetic Pull ETT out 1 cm.

HAPpresent

HAPpresent

Poss.ETT tip

on carina

Passed easilyCannot passbeyond teeth

Cannot passbeyond

oropharynx

Cannot passbeyond ETT tip

Yes

Suctioncatheterthru ETT

Lightanesthesia

Change ETT

Intra-oralkink

Check ETT kinkoutside patient

Auscultate

Treat if>10% or CV

instability

Thoracostomyneedle/tube

Pull ETT untilbilateral breath

sounds

Systematic checkfrom angle

connector upstream

Suction catheterthru ETT

Unilateral absentbreath sounds

Ventilate with Ambu

. Rhonchi/Absent Breath Sounds

Albuterol

No relief

No relief Relief

HAP notrelieved

Relieve Kink

No Yes

No Yes

No Yes

HAP after intubation

Signs of pneumothorax

Signs of tension

Tubetoo deep

Obstructiondistal to circuit

Obstructionupstream to

ETT

Figure 1: Algorithm for High Airway Pressure After Intubation

Figure 2: Suction Catheter Placement Leads to Diagnostic Scheme & Management Recommendations

Anesthesia Patient Safety FoundationBuilding One, Suite Two8007 South Meridian StreetIndianapolis, IN 46217-2922

NONPROFIT ORG.U.S. POSTAGE

PAIDWILMINGTON, DEPERMIT NO. 1387


In this issue: Fatigue & the Practice

of Anesthesiology

■ PerformanceEnhancing Drugs

■ Costs of Fatigue

■ Management ofFatigue

■ Great Britain andIreland’s Approach to Fatigue

Guest Editor: Steve Howard, MD

Dr. Howard is Director of the Patient Safety Center of Inquiry at VA Palo Altoand Associate Professor of Anesthesia atStanford University School of Medicine,Stanford, CA. We have all had to work “dog-tired”—

is it safe?

This Edition Of The APSF Newsletter - Anesthesia Patient Safety

Documents

Transcript of This Edition Of The APSF Newsletter - Anesthesia Patient Safety