35 Chronic Opioid Therapy, Drug Abuse, and Addiction
The Massachusetts General Hospital Handbook of Pain Management
Chronic Opioid Therapy, Drug Abuse, and Addiction
Barth L. Wilsey and Scott M. Fishman
He jests at scars that never felt a wound.
—Romeo and Juliet, act 2, scene 2, by William Shakespeare (1564–1616)
I. Opioid use in the atmosphere of regulatory oversight
II. Prescription drug abuse
III. Distinguishing between physical dependence, tolerance, and addiction
IV. Differentiating addiction from efficacy in opioid therapy
V. History of substance abuse
VI. Long-acting versus short-acting opioids
VII. Monitoring adherence to an opioid regimen
VIII. Opioid agreements
IX. Opioid responsiveness
Using opioids for chronic nonmalignant pain (CNMP) remains controversial in the midst of growing awareness of the necessity to help patients manage pain. Much of the controversy surrounding the prescribing of these medications is related to addiction. State boards of medical examiners have proposed that the mere exposure of patients to opioids induces psychological dependence. Within the last decade, a national survey of members of these boards revealed a prevalent belief that physicians who prescribe opioids to patients with CNMP for over 6 months should undergo investigation. Probably no factor impedes the appropriate prescribing of controlled substances as much as the periodic disciplining of practitioners mandated by these misconceptions. Attempts by regulatory bodies to prevent addiction produce significant barriers to effective pain management, and clinicians who treat pain are often fearful of their efforts to deliver adequate analgesia.
Beyond regulatory scrutiny, treating pain has its own intrinsic difficulties stemming from pain’s subjective nature and the lack of conclusive objective markers. For instance, a history of substance abuse or the manifestation of prescription drug abuse are two areas that require special attention, for they pose special dilemmas for even the most experienced physician. Working together with patients on these and other issues that involve potential problems can be an arduous task. Questions arise daily and specific answers remain elusive. This chapter will explore the rational use of opioids in relation to addiction and monitoring of adherence. We hope to help the clinician develop an approach to opioid use that allows substantial suspicion of potentially adverse effects while suspending judgment and maintaining compassion in the service of effective analgesic intervention.
I. OPIOID USE IN THE ATMOSPHERE OF REGULATORY OVERSIGHT
States regulate the prescription of controlled substances, and all physicians prescribing these drugs should be familiar with the regulations in their states. The California triplicate prescription law, established in 1939 for controlled substances, is the longest continuously running multiple-copy prescription program in the United States. The rationale for multiple-copy or serialized prescriptions is to provide a means of tracking these medications, thereby reducing their illicit use. Unfortunately, this law has not led to a reduction in illicit drug trafficking. One possible reason for this is that the vast majority of drug abusers [i.e., those meeting the criteria of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III) for the diagnosis of abuse of schedule II drugs] obtain the drugs from sources other than their physician. Regulatory scrutiny seeks to prevent the misuse associated with addiction, but it risks producing suffering by its secondary disincentive for adequate pain management. Probably the most difficult question is what degree and type of regulation actually controls addiction and illicit drug use, and what merely stands in the way of adequate pain treatment.
Concerns over forgery, theft, excessive dosages, regulatory investigation, and addiction are cited as reasons why pharmacists are asked to uphold the “letter of the law” and are often reluctant to fill prescriptions for strong opioids. Prohibiting preferred drug regimens, restricting allotment to a 30-day maximum supply, and allowing only a 3-day emergency supply limits access to medications. It is not surprising that regulations and their ramifications have discouraged the prescribing of schedule II drugs. Recently, there have been proposals that would seek the enactment of electronic data transfer (EDT) of pharmacy information to centralized processing points so that unscrupulous physicians and patients with multiple prescribers can be identified. Individual state governments are also considering modifying their approaches to drug abuse by adopting the revised Uniform Controlled Substances Act and/or establishing state pain initiatives. Taken together, these programs may someday alleviate the need for regulations requiring restriction of pain prescriptions to a specific number of dosage units and/or using multiple-copy prescriptions for controlled substances.
To avoid misinterpretation by regulatory agencies, physicians contemplating long-term opioid therapy for patients with chronic pain may be well advised to follow clear and consistent procedures to limit diversion of medications and drug abuse. At the very least, it is advisable to perform a thorough initial history and physical examination, to maintain a written treatment plan, and to consult as needed with knowledgeable colleagues. Minimum assessment should include substance abuse, social, and psychiatric histories. This is significant because a history of substance abuse or a lifestyle in which drug use is accepted or pervasive might indicate the need for additional measures before and after initiating opioid therapy.
On the physical exam, note the patient’s affect and mood, evidence of loss of interest in personal grooming, needle marks, and any signs of intoxication or withdrawal. If any of these signs are detected, proceed with a laboratory evaluation. This examination should include a gamma glutamyl transpeptidase level for evidence of hepatocellular damage, a red cell volume for evaluation of megaloblastic anemia associated with alcoholism, and, in the case of positive signs of intravenous drug use, hepatitis B and C antigen titers and a human immunodeficiency virus (HIV-1) ribonucleic acid level. Because opioids never cure the underlying disorder that causes pain, consultation with a specialist in the area of the patient’s pain problem may be necessary when initiating opioid therapy. Although all these steps go a long way toward minimizing the risk of regulatory action for the patient and the clinician, the practitioner must also maintain skills for recognizing and responding to possible prescription drug abuse.
II. PRESCRIPTION DRUG ABUSE
Careful assessment for possible prescription drug abuse is essential to limit a physician’s liability with regard to regulatory scrutiny. Many practitioners rely on their impression of the patient’s “drug-seeking behavior” to provide them with a rationale to refuse prescribing opioids. But there is controversy about the meaning of “drug-seeking behavior,” as the term is often used pejoratively and signs of these behaviors can easily be based on false impressions and lead to false conclusions. Repeated prescription loss, multiple prescribers, and requests for early refills may simply be manifestations of inadequate analgesia by a patient who is attempting self-medication to alleviate pain.
Pseudoaddiction, a related phenomenon first described in patients with cancer, also results in “drug-seeking behavior” when the patient suffers and cannot endure his or her pain any longer. Unlike addiction, this behavior resolves once adequate opioid therapy is prescribed.
The psychiatric and addiction literature has, until recently, been a source of confusion regarding addiction in the patient with chronic pain. To diagnose addictive disease, the DSM-IV Diagnostic Criteria for Substance Dependence required evidence of certain drug-seeking behaviors whereby “important social, occupational, or recreational activities were given up or reduced because of substance use.” But classic evidence of compulsive opioid use may be missing in pain patients because opioid medication is being prescribed and thus readily available. In addition, pain patients usually do not have to compromise their lifestyle or run the risk of endangering their lives to obtain the prescribed opioid. Likewise, an illicit lifestyle (i.e., involvement in criminal activity, drug diversion) is generally not seen in patients with chronic pain. The form of addiction seen in the pain patient is different from the type seen in the street addict. The subtle signs of prescription drug abuse (Table 1) are deciphered from multiple observations and encounters.
Table 1. Signs of prescription drug abuse
If there is evidence of emotional distress accompanying prescription drug abuse, visits to a mental health provider should be encouraged to evaluate psychosocial issues. In cases of comorbid addiction and chronic pain requiring opioid therapy, it may be prudent to coordinate care with both a pain and an addiction specialist.
III. DISTINGUISHING BETWEEN PHYSICAL DEPENDENCE, TOLERANCE, AND ADDICTION
Recently it has become possible to decipher the chemical “trigger zones” in which individual drugs of abuse initiate their habit-forming actions. Addiction and physical dependence are believed to be subserved by distinct anatomic areas within the central nervous system. Dissimilar drugs such as heroin, cocaine, nicotine, alcohol, phencyclidine, and cannabis all activate a common reward circuitry in the brain. The area of the rat brain responsible for opioid reward, the ventral tegmental dopaminergic area (mesolimbic pathway) is anatomically distant from the locus coeruleus, a noradrenergic area in the periventricular gray matter thought to have a major role in maintaining physical dependence. Several lines of evidence support the involvement of noradrenergic neurons in the development of withdrawal phenomena. Norepinephrine levels change in the brain following opioid dependence. Furthermore, administration of an alpha-2-agonist, such as clonidine, or a beta-antagonist, such as propranolol, reduces the severity of opioid withdrawal. The disparate anatomic and biochemical bases of addiction and withdrawal complement their differentiation in the clinical setting.
Despite the substantial differences between addiction and the pharmacological states of physical dependence and tolerance, these concepts and labels are frequently misunderstood and used inappropriately. Physical dependence is characterized as a physiologic state in which abrupt cessation of a drug results in a strong counterreaction called withdrawal. Such reactions are common to many drugs such as alcohol, benzodiazepines, and caffeine. Physical dependence also occurs with drugs that have almost no abuse potential such as clonidine.
Opioid withdrawal can result from abrupt cessation of administration of an opioid antagonist. The withdrawal syndrome for opioids is often characterized as a “flulike” condition, with runny nose, chills, yawning, sweating, aching muscles, abdominal cramps, nausea, and diarrhea. It is self-limited, usually lasting 3 to 7 days. To avoid the syndrome, medications can be tapered by 10% to 15% every 48 to 72 hours. Usually, a 2- to 3-week period is necessary for completion of the taper. Occasionally, it is also necessary to add clonidine 0.2 to 0.4 mg per day to ward off particularly bothersome symptoms of withdrawal in selected patients.
The Committee on Pain of the American Society of Addiction Medicine defined tolerance as a form of neuroadaptation to the effects of chronically administered opioids (or other medications). Tolerance occurs when exposure to the opioid results in a higher dose requirement to sustain the same level of effect. Although this is a common feature of chronic opioid therapy in animal models, in clinical circumstances it is not commonly a barrier to effective opioid analgesia. Dose escalation can alternatively indicate other problems such as disease progression. Fortunately, tolerance to the nonanalgesic effects of opioids (e.g., sedation, cognitive impairment, decreased motor reflexes) appears to occur more reliably.
Addiction in the context of pain treatment with opioids was defined by this same Committee on Pain of the American Society of Addiction Medicine as being characterized by a persistent pattern of dysfunctional opioid use. This could involve adverse consequences, loss of control over the use of opioids, and/or preoccupation with obtaining opioids despite the presence of adequate analgesia. Addiction implies a psychiatric or behavioral state in which the subject pursues a self-indulgent drug effect despite its damaging impact. Although the term addiction may include the signs and symptoms of physical dependence and tolerance, physical dependence and/or tolerance are not synonymous with addiction. In the patient with chronic pain who is chronically taking opioids, physical dependence and tolerance can be anticipated; however, the maladaptive behavior changes associated with addiction do not necessarily follow.
Most authorities seem to agree that the presence of problem behaviors in the patient with chronic pain is tantamount to addiction. The occurrence of dysfunction appears to be a critical component of addiction. It is of major importance to recognize the distinction between the dysfunction that marks addiction and the improved function that marks effective pain management. Thus, addiction and effective pain treatment have diametrically opposite endpoints and are distinguishable.
Addiction to opioid analgesics is estimated to occur in between 3% and 16% of patients with chronic pain. The exact number is difficult to calculate because of unclear terminology and ongoing changes in the nomenclature. Doctor shopping, multiple prescribers, prescription loss, visiting without an appointment, frequent telephone calls to the clinic, self-reported multiple drug intolerances or “allergies,” and frequent dose escalations are the common manifestations of addiction in patients with pain. However, there is rarely a single behavior or event that confirms the diagnosis of addiction. Making this diagnosis requires careful consideration of diverse information and firm conclusions cannot always be supported. The diagnosis of addiction can range from crystal clear to murky and elusive. Often, the decision to alter or discontinue opioid therapy is based on partial suspicion of dysfunction and addiction but may be more securely based on the collateral finding of insufficient gains in function from the therapeutic trial of opioids.
IV. DIFFERENTIATING ADDICTION FROM EFFICACY IN OPIOID THERAPY
Restoration of function should be one of the primary treatment goals for the patient with chronic pain (Table 2). Unlike the patient whose level of function is impaired by substance use, the chronic pain patient’s level of function may improve with adequate, judicious use of medications, including opioids. Analgesic trials for chronic pain should use function as an objective outcome, and lack of functional gain or malfunction should indicate treatment failure. As in the case of treatment failure in any therapeutic trial, the possibility of toxicity being responsible for the failure must be considered. An increase in function means that the patient’s activities of daily living increase or improve in quality as a result of therapy. Specific improvements (e.g., increasing participation in recreation, time spent shopping, socializing with friends and relatives, performing yard work, or doing household chores) should be sought. Return to work is another outcome that might be sought in specific cases.
Table 2. Desirable functional outcomes
Improvements in functionality should be made part of the patient’s treatment plan and reviewed on each visit after initiating therapy with opioids. Often, it is necessary to gather collateral information from family members or others. Using improvement as the main outcome goal of prescribing opioids allows pain to be de-emphasized and provides a behavior-oriented program to induce desirable outcomes. On the other hand, acknowledging that a patient is not becoming more functional may forgo sanctioned dose escalations and eventually lead to tapering the patient from these medications.
V. HISTORY OF SUBSTANCE ABUSE
Until relatively recently, a history of substance abuse has been considered a virtual contraindication to opioid therapy for CNMP. A recent retrospective review shed light on this issue by reviewing the records of 20 patients with a history of substance abuse treated with chronic opioid therapy for CNMP. Almost half the group developed prescription drug abuse. Individuals who developed abusive behavior (see Table 1) were more likely to have a prior history of opioid abuse or a recent history of polysubstance abuse, whereas patients with either isolated alcohol abuse or a remote history of polysubstance abuse were more likely to manage their medications appropriately. Thus, isolated alcohol abuse and past polysubstance abuse do not appear to be absolute contraindications to opioid therapy for CNMP.
Prescription drug abuse has also been evaluated prospectively by comparing patients with a history of substance abuse to those without. Of all the patients studied, 34% met one criterion, and 27% met three or more of the abuse criteria listed in Table 1. On psychological testing, no differences were identified between those who developed drug abuse and those who did not. Most interestingly, the incidence of previous drug or alcohol abuse did not differ between the abusers and the nonabusers.
Together, these studies point to the feasibility of using chronic opioid therapy in certain patients with a remote history of drug abuse. This seems especially true in the case of patients with a remote history of alcohol abuse. Concurrent monitoring by an addiction specialist or psychiatrist is advisable.
VI. LONG-ACTING VERSUS SHORT-ACTING OPIOIDS
Whether long-acting opioids offer less risk of inducing addiction has not been well studied. However, a preponderance of diverted street opioids are found to be short-acting, and it is logical that with their fast onset and high peak serum levels they are better suited than long-acting opioids for inducing psychoactive nonanalgesic effects. Theoretically, at least, this euphoric effect of short-acting opioid preparations (e.g., oxycodone, hydrocodone, codeine) might foster addiction. The use of long-acting opioids (e.g., sustained release preparations of morphine and oxycodone, transdermal fentanyl, and soon-to-be-available sustained-release preparations of hydromorphone) have been championed because of their gradual onset and the reduced chance that a euphoric effect may be produced. To reduce the incidence of prescription opioid abuse, it can be argued that all patients who are receiving opioids for CNMP should be utilizing long-acting opioids. Whether or not the use of long-acting opioids actually reduces the incidence of prescription drug abuse is not definitely known and awaits validation with clinical trials.
For ongoing intractable pain, short-acting opioids require frequent administration; this undermines attempts to improve functionality by minimizing attention to analgesic dosing. Long-acting opioids maintain steady opioid serum levels (and a steady level of analgesia) with fewer doses, and the need for fewer analgesic interventions helps to remove the focus on pain. Although long-acting opioids may be preferable for CNMP in general, there are a few exceptions. For opioid-naive patients or those with significant pulmonary disease or sleep apnea, short-acting opioids may be the most appropriate agents, at least initially. In addition, since opioid-induced disorientation and confusion is common in patients with underlying cognitive deficits, persons suspected of having these conditions should be started, and possibly maintained, on a short-acting preparation.
Patients who reject long-acting opioids and find acceptable analgesia only with frequent daily dosing of short-acting opioids are not necessarily difficult or addicted. However, this behavior suggests abuse and should prompt close observation and the use of methods to monitor adherence discussed next.
VII. MONITORING ADHERENCE TO AN OPIOID REGIMEN
A wide variety of prescription opioids are available on the street. The markup from pharmacy cost can be considerable (e.g., from $0.25 to $75 per pill). Given the high value placed on diverted drugs, a degree of suspicion for diversion should always be maintained, particularly if there is evidence of abuse. If suspicions appear to be correct, closer control of opioid prescribing is indicated and clearly defined parameters must be maintained. This may include obtaining random urine samples to determine if the patient is taking the prescribed drug or, alternatively, to establish the taking of illicit substances. Recurrent excuses about lost or stolen prescriptions should increase the index of suspicion; this may suggest drug diversion if random urine testing reveals a lack of evidence of medication intake. Requests for specific drugs with a high street value or requests for “name brands” only (e.g., Dilaudid, Percocet) may also indicate a problem, although occasionally the preference my be a conditioned response to a particular drug and due to a placebo effect (see Chapter 3). A high level of suspicion should be balanced by efforts to avoid erroneous assumptions. Valuable information may be gathered from family members, friends, or pharmacists, and if appropriate, prescription requests may be validated by documentation such as police reports of theft or airline tickets for unexpected travel.
Conventional methods of measuring compliance such as tablet counts, diaries, and patient interviews usually overestimate adherence to prescription regimens. Because of the possibility of deception, laboratory testing plays a larger role in the assessment of compliance in the patient suspected of prescription drug abuse. Although opioid levels can be detected in multiple body compartments (serum, urine, hair, and saliva), urine screening is the most commonly used method for routine drug surveillance. The advantages of testing urine include the relative ease of sampling, simpler testing method, lower cost, and longer duration of a positive result compared to that in serum. Unfortunately, routine urine assays provide only qualitative results (i.e., a representative from a specific drug class—such as opioid, benzodiazepine—was or was not present). The routine urine assays is simply a screening method, which needs to be followed by a second confirmatory test. The preliminary test result must be validated when the consequences of a false-positive result are critical, such as in the case of ongoing litigation. The second confirmatory testing is aimed at providing drug identification (e.g., morphine, Dilaudid, codeine) rather than simply identifying that a class of substances is present.
Screening methods for opioids include urine immunoassays and thin layer chromatography. Technically more complex confirmatory tests include high-pressure liquid chromatography and gas chromatography–mass spectrometry.
Results from urine analysis of opioids must be interpreted with knowledge of each laboratory’s specific procedures because toxicology laboratories have differing handling practices for screening and confirmatory tests. It is imperative to become knowledgeable about local laboratory policies (e.g., whether they automatically proceed with the confirmatory test or if this must be determined by the physician ordering the test). In addition, a negative screen can rule out only those opioids that are detectable by that particular assay. For example, some assays detect oxycodone and oxymorphone only at very high concentrations. Consequently, patients taking normal dosages of oxycodone may test negative by urine opioid screen and might therefore be suspected of diversion of medication. Other opioids (including buprenorphine, butorphanol, and pentazocine) are not detected by common opioid assays. Urine screening may also produce false-positive results; for example, recent ingestion of poppy seeds can return a positive screen for morphine.
Following ingestion, opioids may be detected in urine for approximately 10 days. However, this is only an estimate that depends on many factors. For example, dehydration and impaired renal function may slow drug clearance, thereby prolonging the duration of a positive urine result. Consumption of large amounts of fluid can accelerate clearance, a tactic sometimes employed by individuals wishing to foil drug screen detection.
VIII. OPIOID AGREEMENTS
Agreements or contracts are often employed in the chronic administration of opioids and are intended to improve adherence to a treatment regimen. The Massachusetts General Hospital Pain Center opioid agreement is presented in Figure 1. In addition to enhancement of adherence or compliance, contracts provide education and informed consent. The “opioid contract” often includes clear descriptions of what constitutes medication use and abuse, terms for random drug screening, consequences of contract violations, and measures for opioid discontinuation should this be required.
Figure 1. The Massachusetts General Hospital Pain Center opioid agreement.
The efficacy of opioid contracting is not known. Studies reviewing use of contracts for patients in methadone programs suggest a positive benefit. Mandatory structured contingency contracting systems involving weekly urine toxicology screens have been scrutinized in this population. When the participants continued to use illicit drugs, the initial contingency was to lower their methadone dose. Tapering (detoxification) and discharge followed subsequent violations. Illicit opioid use decreased significantly for subjects utilizing such stringent contracts.
Anecdotal reports have described the implementation of similar formal treatment agreements for patients with pain. Key features included acknowledgement that previous treatment strategies had failed, listing of side effects and the risks of opioid therapy (including the potential for addiction), and the contingencies of treatment including the importance of pain relief coupled to enhanced function via the active participation in other therapies. A survey comparing opioid contracts from major academic centers disclosed substantial consistency among many contracts across the country. The major impetus of contracting was to improve care through distribution of information, facilitating a mutually agreed-upon course, and enhancing compliance with medications. Thus, although there is limited scientific evidence to support successful contracting in the pain population, the practice seems to be widespread.
A model for tracking prescription drug abuse with a log for monitoring contract violations has been championed. In these models, known as “three strikes and you’re out,” patients are given a maximum of three minor events (e.g., early refills, multiple prescribers, self-escalation of dosage) before they are tapered and discharged from care. Other prescribers are more stringent, tapering and discharging after one or two instances of prescription drug abuse. Certainly, unlawful activities such as forging prescriptions, selling drugs, or resumption of alcohol or illicit drug intake are considered grounds for such a lower threshold for tapering and discharge.
IX. OPIOID RESPONSIVENESS
There is disagreement about the overall beneficial effect of opioids on the treatment of patients with chronic pain. Many open-label trials have been performed in which opioids have been shown to be effective. Several clinical trials have demonstrated the efficacy of opioids in chronic pain.
However, there are also reports of improvement in pain level when patients are detoxified from opioids. In the latter studies, neither psychological profiles nor a history of substance abuse differentiated the groups that improved after opioid withdrawal. Most patients were said to have experienced an improved sense of well-being with abstinence.
These conflicting outcomes are explicable by assuming that there is a spectrum of patients with chronic pain. Patients who are successfully treated with opioids often experience analgesia without noticeable side effects or functional deterioration. The studies in which patients improved after withdrawal of opioids may have had a selected population referred to them for that purpose—patients who abused their medications, prompting their physicians to refer them for detoxification. It is possible that these patients became dysfunctional while receiving opioids, and only improved after withdrawing from these medications. This supports the practice of using function as a primary determinant of long-term treatment of CNMP with opioids in order to simultaneously gauge analgesic efficacy and addictive side effects.
Regulations and social stigma that seek to prevent addiction have had a “chilling effect” on opioid prescribing. There are many tools and strategies that can make chronic opioid therapy less risky for clinicians and more efficacious for patients. At the heart of rational chronic opioid therapy is the recognition that function is an important outcome measure, and lack of functional improvement (or dysfunction) is a sign of treatment failure and possible addiction.
Chabal C, Erjavec M, Jacobson L, et al. Prescription opiate abuse in patients with chronic pain: Clinical criteria, incidence, and predictors. Clin J Pain 1997;13:150–155.
Dunbar S, Katz N. Chronic opioid therapy for nonmalignant pain in patients with a history of substance abuse: Report of 20 cases. J Pain Symptom Manage 1996;11:163–171.
Fishman SM, Wilsey B, Yang J, et al. Adherence monitoring and drug surveillance in chronic opioid therapy. J Pain Symptom Manage 2000;20:293–307.
Fishman SM, Bandman TB, Edwards A, Borsook D. The opioid contract in the management of chronic pain. J Pain Symptom Manage 1999;18:27–37.
Joranson DE, Ryan KM, Gilson AM, Dahl JL. Trends in medical use and abuse of opioid analgesics. JAMA 2000;283:1710–1714.
Portenoy R. Opioid therapy for chronic nonmalignant pain: A review of the critical issues. J Pain Symptom Manage 1996;11:203–217.
Wilsey B, Fishman SM. Issues of addiction in opioid therapy. Prog Anesthesiol 2000;14:71–83.