OPIOID-INDUCED TOLERANCE AND PHYSICAL DEPENDENCE
There is lingering misunderstanding among some physicians about the important differences between physical dependence and addiction. The repeated administration of any opioid almost inevitably results in the development of tolerance and physical dependence. These predictable phenomena reflect counter-adaptations in opioid receptors and their intracellular signaling cascades.29 These short-term results of repeated opioid administration resolve rapidly after discontinuation of the opioid (i.e., in a few days to a few weeks, depending on the duration of exposure, type of opioid, and dose). In contrast, addiction will occur in only a small percentage of patients exposed to opioids. Addiction develops slowly, usually only after months of exposure, but once addiction develops, it is a separate, often chronic medical illness that will typically not remit simply with opioid discontinuation and will carry a high risk of relapse for years without proper treatment. The molecular processes responsible for addiction are also distinct from those underlying tolerance and physical dependence, and so are the clinical consequences.
Tolerance leads to a decrease in opioid potency with repeated administration. Thus, prescribing opioids long-term for their analgesic effects will typically require increasingly higher doses in order to maintain the initial level of analgesia — up to 10 times the original dose.30 Similarly, tolerance with respect to the rewarding effects of opioids leads to the characteristic dose escalation seen in opioid addiction, which can result in daily doses of up to 800 morphine milligram equivalents (MME, the conversion factor used to facilitate comparison of potency among opioids).31
Some opioid effects show tolerance after a single dose,32 whereas for others, tolerance occurs more slowly.29 In particular, tolerance to the analgesic and euphoric effects of opioids develops quickly, whereas tolerance to respiratory depression develops more slowly,33,34 which explains why increases in dose by the prescriber or patient to maintain analgesia (or reward) can markedly increase the risk of overdose.
Physical dependence underlies the physiological adaptations that are responsible for the emergence of withdrawal symptoms on the abrupt discontinuation of opioids. Withdrawal symptoms (e.g., piloerection, chills, insomnia, diarrhea, nausea, vomiting, and muscle aches) vary appreciably in severity (from not noticeable to quite uncomfortable) and duration (1 to 14 days) on the basis of the type, dose, and duration of opioid prescribed.35,36
In the context of chronic pain management, the discontinuation of opioids requires dose tapering in order to prevent the emergence of such withdrawal symptoms. In some patients, the repeated use of opioids can also lead to hyperalgesia, which is a state of heightened pain sensitivity.37,38 In the clinical context, hyperalgesia can lead to inappropriate increases in opioid doses, which further exacerbate rather than ameliorate pain.39 In the case of hyperalgesia, dose tapering or tapering to discontinuation is a better pain-relief strategy.40
Unlike tolerance and physical dependence, addiction is not a predictable result of opioid prescribing.Addiction occurs in only a small percentage of persons who are exposed to opioids — even among those with preexisting vulnerabilities (Table 3TABLE 3Factors Associated with the Risk of Opioid Overdose or Addiction.). Older medical texts and several versions of the Diagnostic and Statistical Manual of Mental Disorders(DSM) either overemphasized the role of tolerance and physical dependence in the definition of addiction or equated these processes (DSM-III and DSM-IV). However, more recent studies have shown that the molecular mechanisms underlying addiction are distinct from those responsible for tolerance and physical dependence, in that they evolve much more slowly, last much longer, and disrupt multiple brain processes.57
Cardinal features of addiction include a pronounced craving for the drug, obsessive thinking about the drug, erosion of inhibitory control over efforts to refrain from drug use, and compulsive drug taking (DSM-5). These behavioral changes in turn are associated with structural and functional changes in the reward, inhibitory, and emotional circuits of the brain.58,59 Clinical studies have also shown that the ability of opioids to produce addiction is genetically modulated, with heritability rates similar to those of diabetes, asthma, and hypertension.60,61 For these reasons, we do not know the total dose or the duration of opioid administration that will reliably produce addiction. However, we do know that the risk of opioid addiction varies substantially among persons, that genetic vulnerability accounts for at least 35 to 40% of the risk associated with addiction,62-64 and that adolescents are at increased risk because of the enhanced neuroplasticity of their brains and their underdeveloped frontal cortex, which is necessary for self-control.52,62 Hence, in adolescents, the risks and benefits of prescribing opioids for pain management need to be even more carefully weighed than in adults.
In a person with an opioid addiction, discontinuation of the opioid will rapidly reverse the tolerance and physical dependence within days or a couple of weeks. In contrast, the underlying changes that are associated with addiction will persist for months and even years after the discontinuation of opioids.65 This finding is clinically relevant, because after abstinence from opioids, addicted patients are particularly vulnerable to overdosing: their intense drive to take the drug persists, but the tolerance that previously protected them from overdosing is no longer present. These effects explain the high risk of overdosing among persons with an opioid addiction after they have been released from prison or from a detoxification program.66,67
The rewarding effects of opioids play a major role in the risks of opioid diversion, overdose, and addiction. However, the likelihood and severity of these risks are largely independent and governed by different factors. All these risks are present to some degree with all opioids and with all pain diagnoses. This means that no single or simple change in prescribing behavior can be expected to alleviate all risks while properly managing pain. For example, these risks cannot be mitigated simply by restricting prescribing to a particular type of opioid or by avoiding the prescription of opioids to a particular type of patient. However, there are common strategies that can help mitigate all risks, including limiting the prescribed opioid to the lowest effective dose for the shortest effective duration (for both acute and chronic pain) without compromising effective analgesia. Regular monitoring and reassessment provide opportunities to minimize the risks associated with long-term opioid use by allowing for the tapering and discontinuing of opioids among patients who are not receiving a clear benefit or among those who are engaging in practices that increase the risk of overdose (e.g., consumption of high doses of alcohol, concurrent use of benzodiazepines, and poor adherence to opiate medications).68
Preventing Drug Diversion
The most common form of diversion is the transfer of opioid analgesics by patients who have received legitimately prescribed opioids to family members or friends who are usually trying to self-medicate a generic pain.69 This type of diversion applies to prescriptions given for the management of either chronic or acute pain and would be best managed by educating patients on the dangers of sharing their medications and on the importance of safe storage and disposal.70
Approximately 7 to 10% of diversion occurs among patients who feign pain to acquire prescribed opioids,71 usually with the goal of maintaining their addiction, and who will often attempt to acquire opioids from multiple physicians (doctor shopping).71-73 Physicians have attempted to identify dissembling or addicted patients through screening instruments or through detection of so-called aberrant behaviors that are thought to be indicative of addiction (Table 4TABLE 4Mitigation Strategies against Opioid Diversion and Misuse.).77However, the most recent review of patient screening efforts showed no evidence that any scale or procedure was effective.8 Risks of diversion through doctor shopping are best mitigated by the full participation of all prescribers in Prescription Drug Monitoring Programs (PDMPs). PDMPs are statewide electronic databases that collect information on prescription and dispensing of controlled prescription drugs (including opioid drugs) and were designed to monitor information pertaining to suspected abuse or diversion.78 Although these data have been shown to help health care professionals reduce doctor shopping and overdoses,79-81 their use by health care providers is inconsistent.82-84 This in part reflects the fact that PDMPs are voluntary programs in many states. Although 25 states and the District of Columbia update their databases daily, as of this writing, only Oklahoma provides real-time reporting.85 In addition, only 22 of 49 PDMPs share information across states.86 Another obstacle is that access to PDMP data requires a computer that is separate from that used to access electronic health records. However, implementation and consistent use will be facilitated by rapid changes in laws to require mandatory consultation of a PDMP before prescribing, advances in electronic technologies to deliver PDMP information in real time, better integration of PDMPs with electronic health records, and access of PDMP data across state lines.87
Reducing Risk of Overdose
The rate of death from opioid overdose has quadrupled during the past 15 years in the United States.88 Researchers at the Centers for Disease Control and Prevention have estimated that 28,647 drug overdose deaths (61%) in 2014 in the United States involved some type of opioid, including heroin.89 Even more prevalent are nonfatal opioid overdoses that require medical care in a hospital or emergency department. Such events have increased by a factor of six in the past 15 years.90
The contributing factors associated with overdose can be divided into those associated with the opioid itself (type, dose, potency, and duration of action) and those associated with critical features of the patient (Table 3). Although the use of any opioid can lead to overdose, research suggests that exposure to higher doses of all opioids increases the risk of overdose. Opioid doses of more than 100 MME91,92 are disproportionately associated with overdose-related hospital admissions and deaths45 (Table S1 in the Supplementary Appendix, available with the full text of this article at NEJM.org). The use of long-acting opioids, such as methadone and oxycodone, has also been associated with an increased risk of overdose.45
Several identifiable characteristics among patients have been reliably associated with an elevated risk of opioid overdose (Table 3). Included among these factors are a history of overdose,51,93 a history of addiction to any substance (but particularly alcohol, benzodiazepines, or opioids),93 and health problems associated with respiratory depression or concurrent prescription of any medication that has a depressive effect on the respiratory system, such as benzodiazepines and sedative hypnotics.88 The presence of renal or hepatic dysfunction also increases the risk of overdose, since in patients with either of these conditions, the clearance of many opioid drugs is impaired, which leads to higher and longer-lasting drug levels in blood.54,55 Finally, because some cases of overdose may be purposeful suicide attempts,94,95 a history of suicidal thoughts or attempts and a diagnosis of major depression are also markers for an elevated risk of overdose.
Recommended mitigation strategies include an overdose risk assessment (Table 3) and urine drug screening before prescription or represcription of opioids (to verify absence of drugs of abuse). The identification of these risks does not automatically rule out opioids as part of effective pain management. However, these risks do indicate the needs for much greater education of the patient (and the patient’s family) about overdose risks, the use of an opioid treatment agreement,96increased caution in prescribing high opioid doses or long-acting opioids, more frequent clinical follow-up, and, potentially, a prescription for and instruction in the use of naloxone, an opioid antagonist that can reverse an opioid-induced overdose. Indeed, expanding access to naloxone has been shown to significantly reduce the rate of death from opioid overdoses.97
Minimizing the Risk of Addiction
For many years, it was believed that pain protected against the development of addiction to opioid medications. However, epidemiologic studies of opioid addiction among patients in pain, as well as preclinical studies of addiction in animal models of chronic pain,24,98,99 have disproved this belief. Although published estimates of iatrogenic addiction vary substantially from less than 1% to more than 26% of cases,100 part of this variability is due to confusion in definition. Rates of carefully diagnosed addiction have averaged less than 8% in published studies, whereas rates of misuse, abuse, and addiction-related aberrant behaviors have ranged from 15 to 26%.101-103 A small (estimated at 4%) but growing percentage of persons who are addicted to prescription opioids transition to heroin,1 mainly because heroin is typically cheaper and in some instances easier to obtain than opioids.
Clinical efforts to prevent the emergence of addiction can be initiated in primary care settings. Assessment of addiction risks before opiates are prescribed is recommended as a mitigation strategy (Table 3). Emerging signs of addiction can be identified and managed through regular monitoring, including urine drug testing before every prescription is written, to assess for the presence of other opioids or drugs of abuse. Responsible physicians should be prepared to make a referral for specialty addiction treatment when indicated. Although addiction is a serious chronic condition, recovery is a predictable result of comprehensive, continuing care and monitoring.104 In particular, the use of medication-assisted therapy in managing opioid addiction among patients with co-occurring pain significantly improves outcomes.105
On the basis of research and clinical evidence, the Department of Health and Human Services recently launched an initiative to reduce opioid overdoses and addiction that focuses on improving opioid prescribing practices to reduce opioid-use disorders and overdoses, expanding the use of naloxone to prevent overdoses, and extending the use of medication-assisted treatment to reduce opioid-use disorders and overdoses.106
It is no longer possible to simply continue previous practices with respect to the management of chronic pain. The associated risks of opioid diversion, overdose, and addiction demand change. Although there are no simple solutions, we recommend three practice and policy changes that can reduce abuse-related risks and improve the treatment of chronic pain.
Increased Use of Science-Supported Prescribing and Management Practices
The extended prescription of opioids (>8 weeks) for the treatment of chronic pain has questionable benefits for individual patients and presents substantial public health risks.8 The risks of overdose and addiction from this prescribing practice — both among patients with chronic pain and the public at large — increase with higher doses (>100 MME), longer duration of prescribing, and perhaps the use of long-acting opioids. Despite these facts, a Medicaid study showed that more than 50% of opioid prescriptions were for doses higher than 90 MME and for periods of more than 6 months.107Better results can be obtained by using the most contemporary guidelines for pain management.14
Increased Medical School Training on Pain and Addiction
Very few medical schools offer adequate training in pain management, and still fewer offer even one course in addiction. The result is that even experienced clinicians are unsure about how to deal with fundamental and omnipresent clinical issues in their practices. Many motivated, well-intentioned physicians do not know whether to prescribe opioids for pain management and, if so, which ones and for how long. Still fewer understand the pharmacologic or clinical relationships among tolerance, physical dependence, and addiction.108 This education is particularly critical for primary care practitioners, who prescribe more than 70% of opioid analgesics.
Increased Research on Pain
At a recent workshop at the National Institutes of Health on the role of opioids in the treatment of chronic pain, attendants recommended several areas of research that are needed for improved clinical practice guidelines. These areas included how to differentiate the unique properties of acute and chronic pain and how to describe the process by which acute pain transitions into chronic pain.8Discovery-oriented research was also recommended to identify new, potent nonopioid analgesics and other pain-treatment strategies (Table 5TABLE 5Alternative Treatments for Chronic Pain.). Access to biomarkers of pain and analgesia that take advantage of neuroimaging technologies or genetic analyses would accelerate the development of new medications and allow for more personalized clinical interventions for pain management.