Chapter 95. Substance Dependence and Abuse in Anesthesia Care Providers

1. The incidence of opioid abuse among anesthesia providers is 1%.

2. Addiction is a treatable chronic, relapsing disease with a genetic predisposition.

3. Seventy-five percent of addicts declare themselves by age 27 years.

4. Drugs change the brain. The brain changes associated with addiction are independent of drug of abuse.

5. Addiction among anesthesia providers is most commonly detected through observed behavioral changes.

6. Interventions should be based on concern over wellness. Substance abuse is only one of several potential diagnoses that threaten wellness.

7. Intervention at an early stage of addiction is associated with greater treatment success than intervention at later stages.

8. Evidence-based treatment of addiction reduces relapse rates.

9. Most addicts have psychiatric comorbidities that must be treated concurrently to prevent relapse.

10. Reentry requires a contract between the abstinent colleague and his employer.

"[T]here is always soma, delicious soma, half a gramme for a half-holiday, a gramme for a weekend … A gramme is better than a damn … I wish I had my soma."

—Aldous Huxley, Brave New World, 1932

The United States manifests a significant ambivalence toward drugs. While the federal government wages war against the illicit drug industry and organized medicine campaigns against the nonmedical use of prescription drugs, the entertainment and advertising industries, in the spirit of Lieutenant Milo Minderbinder in Joseph Heller's Catch-22, bombard its citizens daily with prodrug references. Lyrics from Toby Keith's popular country music song "Get Drunk and Be Somebody" reflect a common mantra:

All week long, we're real nobodies,

But we just punched out: it's paycheck Friday.

Weekend's here, good God Almighty:

People, let's get drunk an' be somebody

Comedians receive knowing laughs and applause from their audiences when they reference or parody recreational drug use. Television, Internet, and movie advertisements suggest that life is more enjoyable in social settings involving alcohol and more manageable with medications ("Ask your doctor if this medication is right for you."). Helping people to "pass" drug tests is also a big business. For example, if "urine luck" is entered into an Internet search engine, almost 2 million citations are offered. Physicians and major pharmaceutical companies tout the benefit of medications to relieve pain. Anesthesia providers make their living daily demonstrating that drugs of high abuse potential can be safely administered. Even a brochure for the Wood Library Museum of Anesthesiology proclaims, "Thank Goodness They Inhaled." Drug and alcohol misuse is commonly associated with youthful experimentation. However, misuse also is triggered by internal and external pressures to improve self-confidence and performance on the athletic field, in the classroom, on competitive examinations, in the workplace, and in the bedroom to achieve a better body image, to feel more comfortable socially, to cope with the stresses of highly competitive work and study environments, and to escape from dysfunctional social relationships and unsatisfactory living situations. College students call this "pharming."1 The economic costs of drug misuse are staggering. If health care, law enforcement, prosecution, incarceration, and loss of productivity are taken into account, the annual cost associated with drug misuse, excluding nicotine and caffeine, is $465 billion, an amount greater than that linked to heart disease ($290 billion), cancer ($187 billion), or obesity ($123 billion).2 The number of people who misuse drugs is also staggering. According to the 2003 National Survey on Drug Use and Health sponsored by the Substance Abuse and Mental Health Services Administration, 9.1% of the population of the United States older than 12 years, an estimated 21.6 million people, has a substance misuse disorder. Substance misuse disorders run the gamut from smoking marijuana to driving under the influence to chemical dependency. According to the National Institute of Mental Health National Comorbidity Survey Replication, the prevalence of substance misuse disorders in the US English-speaking population is 14.6%.3

Statistics

An argument can be made that the prevalence of substance abuse among physicians is less than for society as a whole. Because 75% of addicts declare themselves by 27 years of age,3 they may be less likely to complete medical training. Supporting this premise is a 1992 study that determined the lifetime prevalence of substance abuse among all physicians, regardless of specialty, to be 7.9% versus the 14.6% previously mentioned.4 There are no published data for the incidence or prevalence of substance misuse or abuse specific to the entire population of practicing anesthesia providers. Surveys of anesthesia residency and student registered nurse anesthetists (SRNAs) training programs consistently report an incidence of 1% to 2%.5 Despite the paucity of data there appears to be a consensus that the incidence (snapshot) and prevalence (professional lifespan) of abuse among anesthesia providers are approximately 1% and 10%, respectively. Thus, if there are 8500 anesthesia residents and SRNAs in training at any given time, then 95 are likely to have a problem severe enough to require suspension from training over any 3-year period or 32 per year. Likewise, if there are 80000 practicing anesthesiologists and CRNAs, then 8000 will have a problem during their professional life.

Vulnerability Hypotheses

Anesthesiologists are consistently overrepresented in analyses of data from physician health programs (PHPs). In Florida, anesthesiologists represent 4.7% of the total physician workforce, but 12% of those enrolled in the PHP.6 It is tempting to accept this data as evidence confirming increased vulnerability of anesthesia providers to substance abuse and propose 5 explanations: (1) Drug abusers seek opportunities. Substance-abusing medical trainees seek to practice anesthesia to gain direct access to opioids and other anesthetic drugs for abuse. (2) Opportunity creates abusers. Unlike most physicians who must write prescriptions or orders for drugs to be administered by persons other than themselves, anesthesiologists personally administer drugs and are daily presented with the opportunity to divert drugs for their own abuse. (3) The stress of administering anesthesia is overwhelming. Susceptible anesthesia providers seek relief from the pressure of administering drugs that can lead to patient death or injury. (4) Personalities of anesthesia providers predispose them to addiction. What makes anesthesia attractive to some are the short-term rewards for effort, the risk of doing something daily with life-or-death consequences, and the adrenaline rush of the occasional need-to-rescue situations. This phenotype is shared by those who view experimenting with illicit drugs attractive. (5) Environment creates abusers. Chronic inhalation of trace amounts of anesthetic drugs in the operating room may increase the susceptibility to abuse. If hypotheses 2 and 5, and possibly 3 and 4, are ever proved, then drug addiction would definitely be considered an occupational hazard. Even though each of these hypotheses has face validity, no studies specifically designed to test any of the first 4 have been performed. Gold et al, exploring the fifth hypothesis, propose that "secondhand" exposure may produce changes in the brain that predispose susceptible anesthesia providers to chemical dependence.6 They have detected propofol and fentanyl in air sampled from cardiac surgery operating rooms. The highest concentrations were in samples from air around the patient's head.7

There are several reasons not to accept at this time an increased vulnerability of anesthesiologists to substance abuse based solely on PHP enrollment data. (1) The percentage enrollment in PHPs has not been verified as a valid cross-sectional sample for addicted physicians. The percentage of individuals enrolled in treatment programs is less than those who need such treatment. The cumulative lifetime probability of patients with substance abuse disorders receiving any treatment is 52.7% to 76.9%.8 (2) The effect of the drug abused on admission to a PHP has not been studied. Anesthesia providers could have the same prevalence of substance misuse disorders as other medical practitioners, but the time span over which they are diagnosed and treated may be compressed into a younger age range because they are more likely to abuse drugs with a higher addictive potential (eg, fentanyl vs alcohol). (3) The effect of the preferred drug of abuse (eg, high with opioids, cocaine, or methamphetamine vs low with alcohol) on peer and legal pressure to enroll in PHPs has not been studied. Alcohol is a legal substance, socially acceptable if consumed without intoxication and legally acceptable if blood alcohol levels while driving are below a certain threshold. "An estimated 8 million adults in the United States are alcoholics. Of this number, only a minority ever receive treatment for the disorder, even when treatment is defined broadly to include participation in Alcoholics Anonymous."9 (4) The effect of the diagnosis of abuse versus dependence, made according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), is not taken into account in most of these analyses. In the National Epidemiologic Survey on Alcohol and Related Conditions, which included 42392 respondents, 22% of those who met the DSM-IV definition of drug dependence did not meet the definition of drug abuse.10 An unknown percentage of dependent individuals carefully avoid behaviors at work associated with abuse so as to protect their primary source of drugs and to escape detection and subsequent enrollment in PHPs. (5) The effectiveness of specialty-dependent substance abuse awareness efforts has not been determined. Educational initiatives have not reduced the incidence in anesthesia training programs over the last 4 decades, perhaps because most of the trainees who declare themselves during training were in early stages of addiction before the start of their residency. But awareness programs may have increased the detection and intervention rates for anesthesiologists more than for other physicians.

Tolerance, Dependence, Withdrawal

Terms associated with both appropriate and inappropriate drug use are tolerance, physical dependence, and withdrawal.11 Each represents an aspect of normal physiologic adaptations seen in all patients after repeated administration of certain drugs. The American Academy of Pain Medicine, the American Pain Society, and the American Society of Addiction Medicine issued a consensus document in 2001 defining physical dependence and tolerance.12 "Tolerance is a state of adaptation in which exposure to a drug induces changes that result in a diminution of one or more of the drug's effect over time." Either a given dose of a drug yields a decreased effect after repetitive administration or an increasingly larger dose of the drug is required to produce an effect of intensity comparable to the first dose. "Physical dependence is a state of adaptation that is manifested by a drug class specific withdrawal syndrome that can be produced by abrupt cessation, rapid dose reduction, decreasing blood level of the drug, and/or administration of an antagonist." All chronic pain patients who are treated with opioids develop tolerance and physical dependence. All patients initiate the physiologic processes that lead to tolerance and dependence with their first exposure to the drug.

Abuse, Intoxication, Addiction

Terms specifically applied to inappropriate drug use are abuse, intoxication, and addiction.11 Abuse is the inappropriate use of a drug. Intoxication refers to reversible "clinically significant maladaptive behavioral or psychological changes developing during or shortly after use of" a drug of abuse. Addiction is not used in the first through the fourth editions of the DSM, copyright by the American Psychiatric Association, or the International Classification of Diseases (ICD-9 or ICD-10), copyright by the World Health Organization. These 2 systems classify all substance misuse disorders with the terms abuse, intoxication, withdrawal, and dependence (Table 95-1). In this setting what the layperson considered addiction was what the psychiatrist called dependent use, typically associated with withdrawal after intoxication and recurrent use either to limit withdrawal symptoms or give into cravings. Recognizing that the term dependence is misused for addiction when it is actually a normal physiologic response to both licit and illicit drugs, DSM will replace it with addiction in its fifth edition due to be published in 2013. In DSM-V, "Substance-Related Disorders" will become "Addiction and Related Disorders"; tolerance and withdrawal will not be counted as symptoms in the diagnosis of substance use disorders when they occur within the context of appropriate medical treatment; and drug craving and compulsive drug seeking will be added to the list of symptoms for addiction.13 The National Institute on Drug Abuse and the National Institute on Alcohol Abuse and Alcoholism differentiate dependence and addiction as follows: "Physical dependence refers to adaptations that result in withdrawal symptoms when drugs such as alcohol and heroin are discontinued. Those are distinct from the neuroadaptations that result in addiction, which refers to the loss of control over the intense urges to take the drug even at the expense of adverse consequences."14

Pseudoaddiction

Pseudoaddiction is a diagnosis made in chronic pain patients who develop drug-seeking behaviors because their pain is inadequately treated.15 The diagnosis requires 4 conditions: an appropriately diagnosed and documented pain condition, inadequate pain treatment unrecognized by the medical staff, analgesia-seeking behavior misinterpreted by the medical staff as inappropriate non–pain-related drug seeking, and resolution of drug-seeking behavior when sufficient analgesia is provided. If the drug seeking is not pain related, then the diagnosis becomes either substance (opioid) abuse or dependence.

Opioid versus Opiate

Although the terms opiate and opioid are frequently used interchangeably, opioid actually refers to any agent that binds to opioid receptors. There are 4 classes of opioids: (1) endogenous peptides, such as endorphins, enkephalins, and dynorphins; (2) opiates, such as morphine and codeine, which are the naturally occurring alkaloids derived from the opium poppy (Papaver somniferum); (3) semisynthetic derivates of opiates, such as heroin, hydrocodone, hydromorphone, oxycodone, and oxymorphone; and (4) synthetic drugs, such as meperidine, methadone, fentanyl, sufentanil, alfentanil, remifentanil, butorphanol, nalbuphine, and buprenorphine.

Opioid versus Narcotic

Narcotic is a general term that refers to substances that induce a sleep state or narcosis. Before the twentieth century, opium was used to create a "daydreaming" state of drowsiness rather than the "high" or "rush" now sought from heroin, cocaine, and methamphetamine. Thus this definition of narcotic was more appropriate then than now. Narcotic is also a legal term that applies to drugs identified in the Harrison Narcotics Tax Act of 1914, the Controlled Substances Act of 1970, and the United Nations Single Convention on Narcotic Drugs of 1961. In this context, opioids, the stimulants cocaine and methamphetamine, and cannabinoids are referred to as narcotics (Table 95-2).

Addiction as a Disease

Addiction has 3 hurdles to overcome before it achieves the ultimate goal of most addictionologists, its recognition as a disease that, once contracted, does not prevent anesthesia providers from returning to work if it is appropriately treated. The first hurdle, disease designation, was cleared in 1956 and 1987, respectively, when the American Medical Association declared that alcoholism and then substance abuse satisfied all 5 criteria for a disease: (1) identified vector: the drug of abuse; (2) an organic lesion: same "core" pathology in the brain regardless of drug of abuse; (3) typical signs and symptoms: acute withdrawal signs may be drug dependent, but abstinence behavior is the same for all drugs of abuse (ie, there is a "core" addiction syndrome); (4) predictable natural course; and (5) evidence-based treatments that alter the natural course. The second hurdle, chronic disease recognition, has almost been cleared. For years addiction was viewed as an acute disease that resolved after the acute withdrawal symptoms subsided rather than a chronic disease with relapses and remissions that must be managed. "Relapse among patients with diabetes, hypertension, and asthma following cessation of treatment has been considered evidence of the effectiveness of those treatments and the need to retain patients in medical monitoring. In contrast, relapse to drug or alcohol use following discharge from addiction treatment has been considered evidence of treatment failure."16 The third hurdle, reliable treatment consistently provided, has not been cleared sufficiently for all anesthesia providers to return to work in the same capacity as they left. But the discouraging data from decades ago regarding relapse need to be replaced with the more optimistic current data when addiction is diagnosed at an early enough stage and appropriate treatment is provided and sustained.9,17

Drugs of Abuse

Although longer-acting opioids are the drugs most likely to be abused by anesthesia providers in surveys that commonly exclude alcohol, caffeine, cannabinoids, and nicotine, case reports document abuse of inhaled agents (eg, sevoflurane18,19), intravenous inducing agents,20,21 and the shorter-acting opioid remifentanil.22 All addictive drugs cause the release of dopamine and glutamate in specific brain regions depending on the stage of addiction. The mechanisms by which concentration of dopamine increases in these regions vary with the drug of abuse in the early stages of addiction, but the mechanism by which most of the dopamine and glutamate increase is independent of the drug of abuse in later stages of addiction. The acute withdrawal symptoms vary with the drug of abuse, but the abstinence behavior of addiction is essentially the same regardless of the drug of abuse.23,24

Despite the considerable diversity in their molecular structures, all drugs of abuse either increase dopaminergic transmission from the ventral tegmental area (VTA) or exaggerate the effects of normal dopamine released into the nucleus accumbens (NAc), prefrontal cortex (PFC), and amygdala (AMG). Dopaminergic neurons in the VTA form a synaptic cleft with neurons in the NAc. Cocaine, methamphetamine, amphetamine, and methylphenidate directly increase the concentration of dopamine in the synaptic cleft by blocking the reuptake of dopamine into the presynaptic VTA neuron. They accomplish this by interfering with the ability of the dopamine transporter (DAT), a carrier protein, to move extracellular dopamine intracellularly. To induce a "high," 50% of DAT must be blocked. Above this threshold the intensity of the "high" is increased with the increasing percentage of DAT blocked.25 Opioids increase dopamine indirectly by acting on γ-aminobutyric (GABA)ergic interneurons in the VTA. By binding to opioid receptors on GABAergic cells, they stop the inhibitory action of GABAergic neurons on dopaminergic cells. The activity of dopaminergic neurons increases, and more dopamine appears in the synaptic cleft in the NAc. Ethanol releases endorphins that block the inhibition of dopaminergic neurons by GABAergic neurons.23,24

Drug Highs versus Natural Rewards

Because of their pharmacokinetic profiles, drugs of abuse produce a longer and stronger "dopamine high" than do natural rewards or threats. Dopamine homeostasis limits the high from natural rewards. The sustained high concentrations of dopamine in the synapses of the NAc, PFC, and AMG exaggerate the value or perceived importance of the drug experience and imprint cues more strongly to repeat the drug experience than to pursue natural rewards. Individuals become chemically dependent by "overlearning of the motivational significance of cues that predict the delivery of drugs."26 Over the same time period, the sustained high concentrations of dopamine with repeated drug intake lead to a downregulation of dopamine receptors, which devalues natural rewards and creates a "reward deficiency syndrome" or "negative emotional state."24,26 Changes in the PFC and AMG lead to an increase in glutamatergic output to the NAc in response to drug cues (eg, passing a street corner where drug was previously purchased or hearing a pop when vial of medicine is opened), drug reinstatement (eg, social pressure to have a drink or requiring postoperative analgesic therapy), or stress that direct the addict to restrict the focus of his or her activity to drug acquisition and administration. An addict will seek drugs despite intellectually knowing and openly acknowledging that doing so will jeopardize marriage, career, health, or life. There is an actual or relative decrease in descending inhibitory influences due to a direct loss of neural traffic saying "do-not-do-this" behavior, an increase in neural traffic insisting on "got-to-have-it" behavior, or a combination of the two.

Dopamine Receptors

Dopamine is no longer viewed solely as the mediator of pleasure but as the primary neurotransmitter for learning from experience what is pleasurable, aversive, novel, and predictable.23,24 Initially, all experience and all drugs of abuse increase firing of dopaminergic neurons in the VTA. A molecule of dopamine released from VTA neurons into the synapses in the NAc, AMG, and PFC binds to 1 of 5 different dopamine receptor subtypes grouped into 2 families: the D1 family, which includes receptors D1 and D5, and the D2 family, which includes receptors D2, D3, and D4. The dopamine D2 receptor density is important in determining an individual's perception of drug experience as pleasurable or aversive and his or her susceptibility to addiction. The brain lesion associated with substance abuse manifests a greater of number D1 receptors and fewer D2 receptors. The density of dopamine D2 receptors decreases approximately 4% to 6% per decade, but advancing age is not associated with an increase in addiction risk. Within a given age range, there is wide individual variability in the density of dopamine D2 receptors. Those with a high density of dopamine D2 receptors are more likely to view a recreational drug experience as unpleasant and thus not develop behaviors that lead to addiction.27 With increasing chemical dependence, the number of D2 receptors and the ratio of the number of D2 receptors to D1 receptors decreases in the NAc and PFC. High levels of dopamine D2 receptors protect against alcohol abuse. When an adenoviral vector is used to deliver the dopamine D2 receptor into the NAc of rats previously trained to self-administer alcohol, the rats reduce their alcohol intake.28 Thus focusing on means to increase the number of dopamine D2 receptors could produce new therapies to treat alcoholism and possibly other addictions.

Glutamate Receptors

Sustained exposure to cocaine or amphetamine causes endocytosis of the AMPA glutamate receptor in the postsynaptic cells of the NAc. A reduced density of postsynaptic AMPA-type subunit GluR2 in the NAc is associated with drug craving in addicts. When a viral protein vector was used to deliver a peptide decoy of the AMPA receptor into the NAc (but not the VTA) of addicted rats, the NAc cells retracted the receptor decoy rather than the receptor itself. Restoring the postsynaptic density of GluR2 in this manner extinguished craving-like behavior in the rats and suggests a new direction for treatment of addiction.29 Another new direction may focus on the cystine-glutamate transporter mechanism.30 Cystine is the oxidized dimeric form of cysteine. Administration of prodrugs for cysteine may increase the extracellular glutamate in the NAc, decrease the amount of glutamate released into the NAc with PFC stimulation, and reduce drug-seeking behavior.

Brain Lesions

Addiction is the end result of a pathologic hardwiring of brain circuits that control learned behavior. The addicted brain is a brain reorganized in 3 parallel phases as described in Table 95-3 that yield the lesions summarized in Table 95-4. Three key elements in the reorganization are the development of hypofrontality, the conversion of the NAc from a dopaminergic responsive center to a glutamatergic responsive center, and the shift in VTA dopaminergic activity from the NAc to the PFC and AMG. Frontal lobe volume losses occur in cocaine-, alcohol-, and heroin-dependent subjects, with a negative correlation between normalized prefrontal volumes and years of abuse. The pyramidal neurons in the PFC decrease in size and show diminished baseline glutamatergic activity. But presented with drug cues or the drug itself, the amount of glutamate they release in the NAc is exaggerated. Similarly, the size of dopaminergic neurons in the VTA decrease and their baseline activity diminishes. When stimulated by drug administration or drug cues, the amount of dopamine they release in the PFC and AMG is exaggerated. Finally, the number and density of dendritic spines on the medium spiny cells in the NAc and pyramidal neurons in the PFC increase when the drugs of abuse are cocaine or amphetamine but not opioids. These changes in neuron morphology are long lasting. Synapses and postsynaptic dopamine and glutamate receptors are located on these spines. The increase in dendritic arborization appears to be mediated through induction of the transcription factor ΔFosB and its set of target genes and brain-derived neurotrophic factor.23

Three principles appear to govern the reorganization of the brain with addiction.31 Principle 1: An important pathway for drug-seeking behavior involves glutamatergic projections from the PFC to the NAc. Laboratory support for this principle comes from 4 key observations: (1) inactivation of the PFC prevents resumption of drug-seeking behavior in abstinent rats challenged by relapse stimuli; (2) injection of α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (AMPA) glutamate receptor antagonists into the NAc prevents drug- or cue-induced relapse; (3) increased glutamate release occurs in the NAc following drug- or stress-induced relapse; and (4) treatments that prevent the release of glutamate prevent or reduce drug seeking in both animal and human subjects. Principle 2: Relapse may be triggered by cues from past drug abuse activity, stress, or a single dose of the drug of abuse (reinstatement/priming). Each of these relapse stimuli involves a different neural circuit. Principle 3: Craving and relapse require dopaminergic transmission from the VTA, not to the NAc, as happens with initial or acute administration of a drug of abuse, but to the PFC and AMG. In humans, the greater the increase in dopamine concentration in the PFC and AMG, the greater the craving or desire for drugs. The more advanced the addiction, the greater the increase in dopamine concentration in the PFC and AMG with a given cue. In human addicts, there is no increase in dopamine concentration in the NAc with drug cues. Laboratory support for this principle comes from the following 4 observations: (1) inactivation of the VTA inhibits relapse from all stimuli in laboratory animals; (2) dopamine release in the PFC precedes glutamate release in the NAc; (3) if dopamine release into the PFC is prevented, glutamate release in the NAc is prevented, and relapse does not occur; and (4) if dopamine release occurs in the PFC after injection of glutamate antagonists in the NAc, relapse does not occur.

Natural History

The neuroadaptations that occur in the brain as it responds to abuse and transitions to dependence and addiction can be divided into 4 stages chronologically: (1) initial drug experience (hours to days), (2) repetitive drug exposure (weeks to months), (3) early addiction (months to years), and (4) late addiction (years). In the first 2 "recreational drug" stages, the abuser typically is sharing the drug experience with others. In the last 2 stages, the abuser tends to isolate himself or herself. The rate at which a susceptible individual passes through these stages depends in part on the drug of abuse: slower (years) for alcohol and faster (months to years) for opioids, and even faster for cocaine, and methamphetamine (weeks to months). Not only is the pathophysiology different for each stage, but the success rates for intervention, treatment, and reentry into anesthesia practice are different for each stage. The earlier the intervention, the greater the likelihood of successful treatment and reentry (Fig. 95-1).

Evidence-Based Treatment

Over the last decade, 3 major advances have been made in the treatment of addiction caused by any drug.32,33 First is the recognition that addiction is a chronic relapsing disease. This means that successful treatment requires more than detoxification and graduation from a rehabilitation program. It must include rest-of-life management. Second is the development of anti-craving drugs to combat the changes in the neural circuitry in the brain that create the compulsion to repeat the drug experience. Third is the emergence of addictionologists, physicians who specialize in the treatment management of these complex patients (Table 95-5).

Treatment regimens for anesthesia providers seeking reentry are designed to maintain abstinence with opioid antagonists and control cravings with drugs prescribed by addictionologists, active participation in PHP-endorsed behavioral health programs, and active treatment of psychiatric comorbidities, such as depression. If this approach is unsuccessful, as defined by short periods of abstinence, multiple relapses, or failure to remain enrolled in treatment programs, then administration of agonist drugs, such as methadone or buprenorphine, are considered. This substitution may enable a colleague to control his or her addiction but precludes the person from returning to the practice of anesthesia.

The Difficult Ones

Inexperienced opioid misusers and end-stage abusers often present intoxicated, comatose, or dead, but experienced users who do not meet the DSM-IV criteria for abuse are very difficult to identify. They compose approximately 20% of the dependent drug population.14 They remain outwardly competent and professional but inwardly miserable as they contrive to sustain their addiction while hiding it from family, friends, and colleagues and protecting their job and source of drugs. Only late in the addictive process, when the amount of opioid these dependent individuals require becomes too great to acquire at work, when their drug-seeking behavior overcomes their ability to sustain their charade, or when they give in to suicidal ideation do they declare themselves.

Existing Surveillance Systems

Early identification of an errant colleague requires a surveillance system. The Drug-Free Workplace Act (DFA) requires employees to comply with their institution's substance abuse policy and employers to document efforts to maintain a drug-free workplace. However, the DFA does not mandate any specific surveillance system for health care organizations. Some non–health care businesses and institutions have adopted drug testing as their primary surveillance system. In alcohol and drug rehabilitation programs, drug testing alone misses many noncompliant participants. More accurate results in this setting are obtained when behavioral observations are combined with some form of drug testing. How well behavioral assessment and drug testing in a homogeneous population, 100% of whom are known to have a substance abuse problem, translates to the inhomogeneous population in surgical suites, only 1% of whom may have a substance abuse problem, remains to be determined. The primary monitoring activity in health care institutions is for quality of care and compliance and not for drug misuse behavior. Unlike the case for surgeons, few outcome data tend to be collated specific to individual anesthesia providers. Thus surveillance systems that rely on quality assessments are unlikely to detect gradually deteriorating performance or subtle changes in behavior (Table 95-6). There is considerable reluctance on the part of health care workers to report colleagues whom they only suspect of drug abuse. Random drug testing has the potential to reveal a problem earlier than quality of care or behavioral assessments. The complexities of the collection process, the relative ease by which addicts can subvert the system, and the high cost-to-benefit ratio in large systems with low anticipated positive findings have deterred most programs from implementing random urine drug testing.34,35 Most large health care institutions have a drug awareness program and a response system rather than a true surveillance system. They rely on an alert, educated, concerned health care provider reporting suspicious or abnormal behavior to the department head, the professional advisory or wellness committee, or employee assistance program for subsequent investigation. Thus our current systems do not favor early detection of a colleague with a substance misuse problem, especially if he or she meets DSM-IV criteria for dependence but not abuse.

Ideal Surveillance System

The ideal surveillance system should not be created specifically to detect substance misuse. Rather, it should focus on anesthesia provider well-being in general. If behavioral changes are found to be due not to substance abuse but to some other cause, such as sleep deprivation, stress, a medical illness such as diabetes mellitus, or side effects of medications prescribed for a medical illness, they still must be addressed. The ideal system should have observers who can witness the behavior of the same providers frequently enough to recognize a change. The observational standard should be a change in behavior, a more sensitive but less specific finding, rather than a specific behavior that has reached a certain threshold for inappropriateness or diagnosis. The ideal system collects and integrates different kinds of information from different sources: (1) observation of signs and symptoms of acute drug intoxication or withdrawal; (2) behavioral changes related to dress, work habits, professionalism, personal life, competence, and drug acquisition; (3) signs and symptoms of psychiatric comorbidities associated with addiction, especially depression; (4) analyses of information from anesthesia information systems (AIMS) and evaluations; and (5) drug testing. For example, do AIMS queries reveal anesthesiologists whose opioid use is 2 standard deviations greater than the others, especially if they also record high delivered concentrations of potent inhaled agents, or whose patients consistently have higher pain scores in the recovery room than expected? Without an aggressive surveillance system, the barriers to recognition (Table 95-7) are formidable.

Signs and Symptoms

Most people have no difficulty recognizing the intoxicated individual. Alcoholics are more likely to come late to work intoxicated, but "experienced" users chemically dependent on intravenous opioids are more like to come early to work, check out drugs for the first case, and immediately divert opioids for acute self-administration to treat withdrawal symptoms. These same "experienced" users also request breaks during the day to sequester themselves to inject opioids to prevent withdrawal symptoms. Table 95-8 lists the signs and symptoms of withdrawal.42 When experienced addicts run out of viable veins or when "inexperienced" addicts miss their veins, they frequently resort to subcutaneous or intramuscular injections. Because of poor sterile technique, the sites of this "skin popping" become infected. Abscesses, especially in the lower extremities of anesthesia providers, are red flags suggesting possible substance abuse.36 Intravenous drug users who inject primarily in the upper extremity may develop the puffy hand syndrome.37 Other red flags include keeping injection sites away from public eyes by wardrobe changes, such as switching to wearing long pants rather than short pants to and from work, wearing long-sleeved shirts in clinical settings, or wearing operating room scrubs to and from work, or changing clothes in the privacy of their office or in dressing room stalls rather than out in the open.

Chest Pain and Cocaine

The appearance of chest pain, malignant dysrhythmias (ventricular fibrillation, ventricular tachycardia, supraventricular tachycardia), or signs and symptoms of an acute coronary syndrome in a young colleague should include substance abuse in the differential diagnosis. The average age of patients presenting with cocaine-associated chest pain was 37.6 ± 9.3 years, considerably younger than the typical coronary heart disease population.39,40 Both cocaine and crystal methamphetamine reach peak blood concentrations within seconds to minutes after smoking or intravenous injection, minutes after nasal insufflation or "snorting," and too long after oral ingestion for significant abuse. They increase the catecholamine concentration in the synaptic cleft by interfering with the ability of the catecholamine transporter, a carrier protein, to return extracellular catecholamines intracellularly to presynaptic neurons. The increased catecholamine residence time in the synapse leads to α-adrenergic effects, such as hypertension and coronary spasm, and β-adrenergic effects, such as tachycardia and dysrhythmias. The combined increase in oxygen demand and decrease in oxygen supply leads to anginal symptoms, frequently in the absence of occlusive coronary lesions, myocarditis, and dilated cardiomyopathy. Chronic use accelerates atherosclerosis and produces hypertrophic cardiomyopathy. Although the risk of these myocardial infarctions is greatest within 1 hour of drug administration, they may occur up to 1 to 2 days later.

Detecting Diversion of Anesthetic Drugs

Two papers evaluated the use of an AIMS to monitor for diversion of controlled substances. In a retrospective study in 2007, Vigoda et al43 compared the medications secured from the automated medication dispensing system to administration records on an AIMS record. Discrepancies, found in 15% of cases, occurred mostly due to documentation errors in AIMS or amount of drug waste in the dispensing system. The authors recommended a utilization of an electronic interface between the AIMS and dispensing system to alert users of medication entry errors before closing a record. Epstein et al44 used records from the drug-dispensing machine to identify behavioral patterns associated with abuse of controlled substances. Based on 2 known cases of drug diversion and their onset, the authors used database queries to identify atypical drug transactions. Examples include frequent transactions an hour or more after the end of a procedure or checking out drugs from a machine in a location different than the anesthetizing site. Use of a surveillance system could result in recognition of diversion. In the accompanying editorial, Dexter45 suggests collaboration by anesthesiologists and data analysts to improve drug surveillance. A survey conducted by Wischmeyer et al20 reported that 71% of anesthesia training programs had no established system to control or monitor propofol use. Because propofol abuse had been implicated in 7 anesthesia provider deaths (6 of whom were residents), a monitoring system for its use was recommended. As part of their Substance Abuse Prevention Protocol, Tetzlaff et al46 report the use of an interface between the AIMS and the electronic controlled substance dispensing system to detect "excessive use, excessive wastage, inappropriate transaction site identification, frequent transactions after the close of the case and any regular pattern of failure to correctly reconcile use/wastage/returns." Chisholm and Harrison47 run chart methodology, plotting standardized units of opioid usage against anesthetists' operating room time, to demonstrate a means to detect abnormal patterns of use.

Observing that the frequency of substance abuse remains unchanged despite improved education efforts, Fitzsimmons et al48 initiated a program for random urine drug testing of all residents. Brock and Roy35 commented that their use of a positive urine test as surrogate marker for the reduction in incidence of substance abuse is not dependable. They also questioned the integrity of urine samples brought in any time during a 36-hour response window. Depending on the chain of custody, this allows time for the individual to dilute, contaminate the sample, or supply another person's urine. In response, Fitzsimmons responded that the window had subsequently been reduced to 10 hours.49 Acknowledging that drug screening will not detect all incidents of misuse, the authors reiterated that their intent was to reduce the incidence of abuse as well as to provide a means of earlier detection of abuse. Recent ingestion of poppy seeds can lead to false-positive results for opioid abuse, diminishing the usefulness of random urine drug screens. Due to short half-lives and diluted concentrations, many of the drugs abused by anesthesia providers can be missed by urine drug screens (UDSs). Hair analysis provides a reliable, although expensive, means to detect chronic exposure to drugs, including fentanyl.50 Limitations to the use of hair include lack of hair (shaving, trimming scalp hair), in which case alternative areas (pubic area, underarms, chest) can be used; and environmental exposure to particular agents (sample should be taken gloved, after thorough hand cleansing and in a site away from the tested drug's use). According to 2008 information from a New York State laboratory,51 costs for a UDS including a fentanyl assay were $32.50, including propofol assay were $290, and were over $1000 for hair analysis. Oral fluid testing is also undergoing validation.52

Systematic surveillance of the AIMS interface with medication dispensing units (electronic or manual) combined with random analysis of returned waste drugs is a useful means of screening for drug misuse. Diversion detection should also include continued education on the risk factors for as well as the typical signs and symptoms of substance abuse. The authors also recommend "for-cause" testing rather than random screening. The American Society of Anesthesiologists' (ASA) Web site includes a model department policy for drug and alcohol testing53 that was developed by the Committee on Occupational Health's Task force on Chemical Dependence. A position statement on security of medications in the operating room was approved by the ASA Executive Committee in October 2003.54

Definition and Goals

Intervention is an "advocacy-oriented" confrontation of a colleague whose behavior has suggested that he or she may have a problem with chemical dependence.51,55 Its goal is to convince a colleague with documented detrimental changes in behavior to enter an evaluation program voluntarily. Treating a colleague for apparent signs of intoxication, side effects of the abused drug or its adulterants, or withdrawal and resuscitating a colleague from an overdose are acute emergency medical treatments. These actions are not included in the scope of the term intervention, although they should trigger a subsequent intervention. More commonly intervention is required in the absence of these medical issues. It is divided into 3 phases: investigation of suspicious behavior, preparation of a plan for confronting the suspect, and the confrontation itself. Each phase of intervention should be conducted with the assistance of specifically trained personnel who are not members of the Anesthesia Department. Intervention does not include diagnosis or treatment. It is important to remember that most interventions are prompted by documented observations of inappropriate behavior, which could have many causes, and not the witness of actual drug administration.

Keys to Success

The 3 keys to a successful intervention are a thorough investigation, an experienced intervention team, and rehearsed intervention plan. These factors are well discussed in the review by Silverstein et al51,55 and summarized as follows. The investigation, which may take weeks, is ideally performed confidentially by an experienced team drawn from the human resources, employee assistance, or risk management departments or physician well-being committee of the medical center, with a representative from the Anesthesia Department. The purpose of the investigation is not to make the diagnosis of substance misuse, substance abuse, or chemical dependency but to gather sufficient evidence of behavioral changes, drug diversion, or drug use to mandate an evaluation. No intervention should be attempted based only on suspicion without evidence.

Intervention Team

The intervention team should consist of 2 or more members with experience in confronting people who deny their problems. If behavioral changes are the only evidence of suspected drug misuse, the team cannot assume that drug misuse is the problem. Rather, the team should accept that the evidence suggests it is reasonable and necessary for the individual to have an evaluation before he or she is allowed to return to work. It is the purpose of the subsequent medical and psychiatric evaluation to determine the cause of the behavioral changes. There should never be a one-on-one intervention. The gender and cultural makeup of the team is important to avoid charges of harassment or assault. Ideally, one team member should be someone who is either a certified addictionologist, a former impaired professional, or an experienced employee assistance professional. A spouse or family member could be motivated to either facilitate or sabotage the process. Advice from an experienced interventionalist is recommended to determine suitability of the spouse as a member of the intervention team. Interventions take time, sometimes hours, if diversion is to be successfully accomplished. It is important for team members to devote their full attention to the intervention by turning off their beepers and cell phones and canceling all other commitments until the intervention is completed.

Intervention Plan

The intervention plan should include preparation for immediate drug testing, inpatient admission to a hospital or treatment center, accompanied transfer to testing site and inpatient facility, and contingency plans if the suspect refuses to accept testing and evaluation. The purpose of the intervention is not to accuse the individual of a crime or to make the diagnosis of drug misuse. It is to convince the health professional colleague to submit to drug testing and an evaluation. Drug testing should be required as a routine part of institutional policy and procedure related to risk management of untoward events or inappropriate or unaccountable behavior. Because the suspect may become physically hostile, security personnel should be alerted to be in the vicinity of the intervention, but they should remain out of sight unless needed to avoid the perception of an impending arrest. Finally, an individual once confronted must be regarded as a suicide risk and not be left alone until he or she has been admitted to an evaluation center or accompanied by a responsible individual away from the medical center if he or she refuses evaluation and leaves against medical advice.

Diversion

Unfortunately, diversion is a term used for 2 opposing activities. Diversion is the term used when drugs designated for analgesia, sedation, or anesthesia are diverted by an anesthesia provider for personal use. Intervention with the goal of detoxification, treatment, and monitored reentry into the workforce is also called diversion. Cooperation between the state medical boards and the PHPs is essential to enable the practitioner with a problem to walk the bureaucratic tightrope between confidentiality and ability to return to work versus public safety issues and loss of ability to practice medicine. Unfortunately, the protection afforded by PHPs tends not to extend to CRNAs.

Techniques

Medically supervised opioid withdrawal, or detoxification, can be divided into 4 general approaches in order of decreasing duration of time required to complete the process: (1) substitution of abused opioid with an approved opioid, methadone, or buprenorphine, and a slow taper (weeks to months) of the approved opioid to minimize withdrawal symptoms; (2) complete discontinuation of any opioids and administration of medications, such as clonidine and benzodiazepines, to attenuate withdrawal symptoms (1-2 wk); (3) precipitation of withdrawal by administration of antagonists, naloxone or naltrexone, and the administration of medications to attenuate the withdrawal symptoms (rapid detoxification in 3-5 d); and (4) precipitation of withdrawal by administration of antagonists during general anesthesia or sedation (ultrarapid opioid detoxification [UROD] in 6-8 h).56 When clonidine-assisted, buprenorphine-assisted, and anesthesia-assisted detoxifications were compared with naltrexone as preparation for antagonist therapy, no differences in the rate of retention in treatment programs or frequency of opioid-positive urine specimens were observed. There was a significant increase in morbidity with UROD blamed, perhaps somewhat questionably by nonanesthesiologists, on the use of general anesthesia.57 The accompanying editorial by another nonanesthesiologist stated "anesthesia-assisted detoxification should have no significant role in the treatment of opioid dependence."58 When UROD was combined with subcutaneous implantation of naltrexone pellets, pulmonary edema, aspiration pneumonia, prolonged debilitating withdrawal, delirium, and death occurred after discharge home.59 However, there are experienced advocates in Europe for UROD in select patients.60-62

UROD Position Statement

Currently in the United States, UROD should be considered only if it is part of a clinical study approved by an institutional review board. The American Society of Addiction Medicine (ASAM) has published a public policy statement regarding opioid antagonist agent detoxification under sedation or anesthesia stating: (1) "ASAM does not support the initiation of acute opioid detoxification interventions unless they are part of an integrated continuum of services that promote ongoing recovery from addiction," and (2) "[UROD] is a procedure with uncertain risks and benefits, and its use in clinical settings is not supportable until a clearly positive risk-benefit relationship can be demonstrated. Further research on UROD should be conducted."63 It is unethical to provide general anesthesia or deep sedation for ultrarapid detoxification to patients who are not enrolled in approved treatment programs under the care of credentialed addictionologists and who just want to "get clean" to go back to work. Doing so facilitates illegal drug use, enables addictive behavior, and puts patients at risk for significant morbidity and mortality if they are discharged directly home from the recovery room.

Relapse

Addiction now is recognized as a chronic relapsing disease. Unfortunately, most of the data regarding relapse come from abstinence programs that either are too short term or too limited with patients who are far less motivated than most health care providers or have been identified and treated too late in the addiction process. The reported relapse rates are so high that antagonist therapy with its goal of no relapse typically is rejected in favor of agonist therapy with methadone or buprenorphine with a goal of fewer relapses. However, there is a group of patients who do get close to optimum therapy, physicians in PHPs.9 In this group, the results for abstinence programs are far more optimistic. In their study of relapse in the Washington Physician Health Program, Domino et al17 observed an overall relapse rate of 25% over 10 years. To express the results more positively, 75% of physicians did not relapse. Three factors predicted relapse: a family history of substance abuse, a coexisting psychiatric illness, and major but not minor opioid use. Major opioids included fentanyl, sufentanil, morphine, meperidine, methadone, heroin, and controlled-release oxycodone. Minor opioids included butorphanol, codeine, hydrocodone, pentazocine, propoxyphene, and tramadol. Relapsed rates decreased with increasing time in treatment. All who avoided relapse in the first 5 years successfully returned to the practice of medicine. With specific regard to anesthesiologists, the study had insufficient power to separate out a contribution of anesthesiology as a profession independent of the abuse of major opioids. However, the authors did refer to data from the New Jersey Physician Health Program and California Physicians Diversion programs that did not reveal a risk for anesthesiologists higher than for other specialists.

Reentry

Both the Joint Commission on Accreditation of Health Care Organizations (JCAHO) and the American Board of Anesthesiology (ABA) support the intent of the Americans with Disabilities Act: protection of individuals with a history of alcohol or substance abuse who are not currently abusing alcohol or using drugs illegally. The JCAHO now mandates that hospital medical staffs have "a process to identify and manage matters of individual health for licensed independent practitioners. The purpose of this process is to help with rehabilitation, rather than discipline, to aid a practitioner in retaining and regaining optimal professional functioning that is consistent with protection of patients."64 The ABA's Booklet of Information65 contains a specific policy admitting qualified applicants with a history of alcohol or illicit drug use into the examination system, if the ABA receives "acceptable documentation" that "they do not pose a direct threat to the health and safety of others" and are "not currently engaged in the illegal use of drugs." After a candidate satisfies the examination requirements, "the ABA will determine whether it should defer awarding its certification to the candidate for a period of time" usually several years after the candidate enters a rehabilitation program, to assure a reasonable period of abstinence, monitored compliance with reentry contracts, and safe care of patients. The ABA receives notifications of actions by state medical boards from The Federation of State Medical Boards and "will initiate proceedings to revoke certification(s) of diplomats with a medical license that is revoked, suspended or surrendered in lieu of revocation or suspension upon notice of such action."

The return of successfully treated anesthesia care providers to the clinical practice of anesthesia (reentry) remains controversial. Currently, the decision to return the practitioner to the operating room is made on an individual basis. Many studies have suggested that upon completion of treatment, anesthesiology trainees should be redirected into another specialty. In 1990 Menk et al published results from a survey of 159 anesthesiology training programs; 34% of the parenteral opioid abusing group successfully reentered training versus 70% of the nonopioid using group.66 Unfortunately death was the presenting relapse symptom in 16% of the reentering parenteral opioid using group. From a survey of academic chairs and program directors, Booth et al67 reported occurrence of addictive disorders in 1% of faculty members and 1.6% of residents. Of note, 18% of these cases were detected by death or near death. In the most recent study of treatment outcomes among 199 residents in anesthesiology involved in drug misuse, 167 returned to medicine, 100 continued as anesthesia residents, 9 died, and 91 completed their anesthesia training, for a 79% overall reentry rate in medicine and a 46% reentry rate into anesthesia.5 Fry provided dismal results from a small 10-year survey of anesthesiologists and anesthesia trainees in Australia and New Zealand.68 Only 19% of abusing trainees were successful in a return to any specialty of medicine, 15% of those reentering anesthesia were successful in training completion, and there was 31% mortality among all the trainees (5 of 16). Collins and McAllister reanalyzed Fry's data to demonstrate 14% mortality for those residents returning to anesthesia.69 Wischmeyer et al conducted a survey of academic chairs that included abuse of propofol; 18% of departments reported abuse of propofol, of which 28% presented by death.20 In 2009, Bryson published another survey of US residency program directors' experience with trainee substance abuse over a 10-year period. He reported 29% relapse rate and 10% incidence of death as the presenting symptom of relapse in residents returning to anesthesia training.70 All of these studies were retrospective studies that relied heavily on the opinion of the program director of departmental chair, who may not have been in position during the survey time period.

Controversy over the subject of routine reentry after successful treatment has recently become a topic of considerable discussion. A review of indexed literature from 1993 to 2008 by Bryson and Silverstein51 concluded that there was no guarantee against relapse after successful completion of treatment. Stating that despite educational and program efforts "deaths from opioid abuse continue" and based on the subjective report of "nearly 100% relapse" of 12 nurse anesthetists over the course of 20 years, the ensuing editorial by Berge et al recommended "one strike, you're out" rather than return to the operating room.71 Six letters to the editor responded to both documents. Cohen described the Physician Health Committee of the Medical Society of the District of Columbia's monitoring program for reentering anesthesiologists. This program includes individual consideration, long-term close surveillance, and aftercare by a specialist in addiction medicine.72 Skipper and Dupont73 included reports by Pelton and Ikeda,74 Paris and Canavan,75 and Domino et al17 in support of individual consideration for return to the operating room. Paris and Canavan used data from the New Jersey PHP to demonstrate no difference in relapse rate between 32 anesthesiologists and 36 physician controls, as well as no significant difference between practicing physicians and residents. In addition, Skipper and Dupont cited the outcomes report of McClellan et al76 after a 5 or more year follow-up of 904 physicians from 16 PHPs, which supported individualized treatment. Earley and Berry77 cited Angres et al78 in support of using specific factors in making the decision to return immediately after treatment. Noting that none of the published studies contained specifics of treatment, follow-up care, or previously mentioned factors, they recommended research evaluating the various assessment and management protocols. Katz79 criticized Berge for not making a distinction between residents and attending physicians, the effect on relapse and death upon redirection to another specialty or not practicing medicine, as well as abuse of different drugs. Citing the "one strike, you're out" default position, Specht was concerned that individuals would be less likely to seek help due to fear of career loss.80 In the only letter of support of "one strike, you're out," Torri mentioned concerns for patient safety under the care of the relapsing anesthesia care provider before detection.81 Berge et al responded that reentry criteria that predict success upon return to anesthesia practice, particularly within the framework of a well-functioning PHP, as well as life-long monitoring may provide a platform for reentry. They also supported research with valid study design, outcome metrics, and appropriate data analyses.82

Oreskovich and Caldeiro reviewed in 2009 the indexed literature on the subject of reentry of anesthesiologists after successful treatment for chemical dependency and found that most PHPs "did not define the role of monitoring programs, type or duration of treatment, role of hair and nail testing to confirm recent opioid use, or the effect of the use of depot naltrexone."83 Review of "the subset of PHPs that incorporated trimodal monitoring (chemical, behavioral, and workplace), aggressively tested hair and fingernails for high-potency opioids, required administration of depot naltrexone, and followed up anesthesiologists for five years after residential treatment that averaged three months" supported the individualized return to clinical practice.83 They note the need to address the inconsistency among the various states' PHP programs and the absence of PHP in several states. When available, they recommend a requirement for participation in a PHP-mandated monitoring and aftercare program. In the previously cited study from the Washington Physician Health Program,17 only 5 of 22 anesthesiologists who misused fentanyl were able to return to practice (23%), but no information about the 11 anesthesiologists whose drug of choice was not fentanyl was presented. Presumably, they were less likely to relapse and more likely to have successfully reentered practice. The authors concluded that risk of relapse was increased in those who used a major opioid, had a coexisting psychiatric illness, or family history of a substance use disorder.

DuPont et al published results from McClellan's 5-year longitudinal cohort study involving 904 physicians enrolled into 1 of 16 responding PHPs. Over the 5-year period, 78% of the participants had negative drug testing, and posttreatment 72% of all physicians continued to practice medicine.84 Skipper et al conducted an analysis of a subset of 102 anesthesiologists from this study.85 Anesthesiologists had a higher rate of opioid and intravenous drug use in comparison with other physicians. There was no statistical difference in completion of contracts (71% anesthesiologists vs 64% nonanesthesiologists) or continuation in other medical practice specialties (76% anesthesiologists vs 73% nonanesthesiologists). There was no report of physician death or significant patient harm due to relapse. Fitzsimmons and Baker's editorial comments that Skipper et al demonstrate that recovery and successful return to practice are very likely for the trained anesthesiologist who receives the support of a PHP.86 They suggest that residents who are referred to a PHP and successfully complete a course of treatment should be considered for reentry into anesthesiology training. In response to a letter to the editor by Silverstein and Bryson,87 Skipper noted that 12% of the returning anesthesiologists changed specialties.88 Citing DuPont's study, he also refuted their suggestion that many addicted anesthesiologists, including the more severe cases, are not referred to PHPs. Thus successful reentry can occur. The question is when and how.

Based on their experience with 5 addicted anesthesia residents, Bryson and Levine recommend a gradual return to the operating room.89 The accompanying editorial by Tetzlaff and Collins commented that relapse risk stratification could lead to development of case-by-case system for reentry.90 Criteria used by PHPs for consideration for reentry has been published. Talbot's original return to work classification has not undergone research evaluation but remains a useful template.78 The literature supports 3 criteria for a negative outcome: positive family history, comorbid psychiatric disorder, and history of relapse. The Earley Consultancy91 is in the process of developing and implementing a reentry evaluation tool: the Medical Personnel Addiction Recovery Inventory (MPARI). This was initially created for anesthesiologists (APARI). The tool is based on expert consensus only and has not been analyzed for validity or interrater reliability. The ASA Web site contains a model curriculum on drug abuse and addiction for residents in anesthesiology.92 It was developed by the Committee on Occupational Health and includes a useful section on reentry with a sample reentry agreement. Talbot's criteria for reentry are listed on the site. Successful reentry requires completion of an effective treatment program, motivation, supportive environment, involvement of family/significant others, as well as a reentry agreement, including monitoring. Although not all states have a well-functioning PHP, the Federation of State Physician Health Programs contains an active list of state programs.93 The North Carolina Physician's Health Program94 contains a list of available services: training in identification of impairment and intervention procedures, assessment of referrals and treatment recommendations, advocacy for successful treatment participants, monitoring and support, and financial assistant for indigent participants.

Reentry Contract

Reentry should involve a work reentry contract stipulating what is expected of the reentering individual by the employer and what the reentering individual should expect from the employer if he or she meets the contractual obligations. Becoming increasingly common are last chance agreements, which clearly describe how substance abuse has affected the employee's performance in the past, how the employee's performance has put the employer at risk, what is expected of the employee if employment is to continue, and what the employee can expect if his or her performance does not meet expectations.

Substance abuse has gotten the attention of professional organizations in anesthesia. The Committee on Occupational Health for the ASA has prepared 2 useful documents: "Model Curriculum for Substance Abuse" and "Model Department Policy for Drug and Alcohol Testing as Part of a Comprehensive Intervention for Suspected Substance Abuse in Anesthesia Professionals." The Society of Academic Anesthesia Chairs and Academic Anesthesia Program Directors (SAAC/AAPD) has sponsored 2 instructional videos: the original Wearing Masks and Collateral Damage: Drug Abuse and Anesthesiology. The American Association of Nurse Anesthetist (AANA) has sponsored the making of several instructional videos that are sequels to Wearing Masks. Both the ASA and AANA have established wellness committees. Anesthesiologists who have an interest in addressing the problems of substance abuse should become members of the Association of Medical Education and Research in Substance Abuse (AMERSA).95 As C. Everett Koop, MD, former US surgeon general remarked in the McGovern Lecture, Hanover, New Hampshire, May 1, 2003,

I try to remember not to call people with addictions—drug addicts. I propose that you do the same. The word addict carries a very negative connotation and suggests that the person is the problem rather than the disease.

• Bryson EO, Silverstein JH. Addiction and substance abuse in anesthesiology. Anesthesiology. 2008;109:905-917.
• Collins GB, McAllister MS, Jensen M, et al. Chemical dependency treatment outcomes of residents in anesthesiology: results of a survey. Anesth Analg. 2005;101:1457-1462.
• Kalivas PW, O'Brien C. Drug addiction as a pathology of staged neuroplasticity. Neuropsychopharmacy. 2008;33:166-180.
• Koob GF, Volkow ND. Neurocircuitry of addiction. Neuropsychopharmacy. 2010;35:217-238.
• Levine MR, Rennie WP. Pre-employment urine drug testing of hospital employees: future questions and review of current literature. Occup Environ Med. 2004;61:318-324.
• McLellan AT, Lewis DC, O'Brien CP, et al. Drug dependence, a chronic medical illness: implications for treatment, insurance, and outcomes evaluation. JAMA. 2000;284:1689-1695.
• O'Brien CP. Evidence-based treatments of addiction. Phil Trans R Soc B. 2008;363:3277-3286.
• O'Connor PG. Methods of detoxification and their role in treating patients with opioid dependence. JAMA. 2005;294:961-963.
• Samet JH, Galanter M, Bridden C, et al. Association for Medical Education and Research in Substance Abuse (AMERSA). Addiction. 2006;101:10-15.
• Skipper GE, Campbell MD, DuPont, RL. Anesthesiologists with substance use disorders: a 5-year outcome study from 16 state physician health programs. Anesth Analg. 2009;109:891-896.