Opioids in Headache
- Conflict of Interest: None.
Opioid analgesics have long been used to treat head pain of various types. This has been increasing to a significant degree over the past 25 years because of a trend for more liberal use of opioids in non-malignant pain. Opioid treatment for acute headache, as well as prophylactically for refractory chronic headache, is controversial. There are a number of adverse effects associated with acute and chronic opioid treatment. Tolerance, dependence, and addiction are prominent issues. This article attempts to analyze the benefits and disadvantages for opioids in the management of migraine and other headache disorders, relying on known properties of this class of medication as well as clinical data. It will mainly focus on 2 topics: the use of opioid medication for the acute treatment of migraine attacks and continuous prophylactic use for refractory chronic migraine.
Introduction and Background
Opioids have been used to treat pain for thousands of years. The first recorded use was by the ancient Greeks who derived them from the opium poppy Papava somniferum. The use of opioids probably predates recorded history, however, and while it is hard to know when they were first used for headache treatment, this likely has a history spanning millennia as well.
The therapeutic use of opioids is as controversial as any topic in medicine. The use of prescription opioids has dramatically risen over the past 20 years, although may be leveling off and probably will decrease over the next several years.[1, 2] The prescribing of opioids for acute and chronic treatment of headache disorders appears to have followed this cyclic trend. Careful analysis of the benefits and disadvantages of opioid therapy in migraine and other disorders seems especially called for in this current time of change. This article will attempt to do that, focusing on the role of opioids in the acute management of headache as well as chronic treatment of refractory migraine and other chronic disabling headache disorders.
Pharmacological Effects of Opioids
Opioid analgesics are often divided into (1) the naturally occurring alkaloids derived from the opium poppy, (2) semisynthetic agents, and (3) synthetic compounds. The term opiate is supposed to be applied only to the first group, while opioids refer to any analgesic resembling or sharing key properties with this group. Opioids interact with opioid receptors of which there are several known types: mu, delta, and kappa. All 3 receptors when activated can promote analgesia. Opioid receptors are present on both sides of the first synapse in the nociceptive pathway (spinal cord dorsal horn for trunk and limb nociception; spinal trigeminal nucleus for facial and anterior head nociception), so both pain signal transmission and release of excitatory neurotransmitters are reduced by medications with opioid agonism.
Another way of classifying opioids, which is more helpful in assessing effects on pain, is to consider 3 groups divided as to their agonistic/antagonistic effects on opioid receptors: full agonists, partial agonists, and antagonists. Morphine and other opioid agonists probably work for head pain by modifying nociceptive input to the spinal trigeminal nucleus (nucleus caudalis), but it is fairly clear that they have no effect on the source of migraine pain – ie, neurovascular. For this reason, and others that will be covered later, their effectiveness in primary headache is limited.
Adverse Effects of Opioids
There are opioid receptors in virtually all parts of the body, with the highest concentration found in 2 places: the nervous system and the gastrointestinal tract. In the nervous system, areas of concentration include the periaqueductal gray matter, rostral ventral medulla (RVM), locus ceruleus, and dorsal horn regions of the spinal cord. Side effects of opioids are numerous and presumably related to systems that contain these receptors (Table 1). Supraspinal effects include euphoria, sedation, sleep disturbance, respiratory depression, cough suppression, pupillary constriction, truncal rigidity, nausea and vomiting, and temperature dysregulation (hyperthermia or hypothermia). They can also lower seizure threshold by some as yet unknown mechanism. Peripheral effects include bradycardia (although meperidine causes tachycardia), hypotension, constipation and gastroparesis, renal function depression, and pruritus. There is also ample evidence that there is an effect by many opioids on the endocrine and immune systems.[4, 5] Interestingly, unlike the analgesic and euphoric properties, some of these effects do not seem to abate with continued use, including gastrointestinal dysfunction, miosis, and, to some extent, respiratory depression.
Table 1. Opioid Side Effects
- Respiratory depression
- Sleep disturbance
- Cough suppression
- Pupillary constriction
- Truncal rigidity
- Nausea and vomiting
- Temperature dysregulation
- Renal function depression
- Lowered androgen and estrogen levels
Addiction and Overuse
In the United States, recreational use of opioids was not common until the Civil War years (1861-1865) and became even more widespread when heroin was synthesized in 1874 and marketed as a tonic for many symptoms including pain. Intravenous heroin use blossomed after World War II, which became most problematic in the 1950s, 60s, and 70s. There has been a resurgence in opioid overuse and addiction because in part of the increased use of opioids in the management of non-malignant chronic pain as promoted by Portenoy, Foley, and others since the late 1980s.
Opioids clearly lead to tolerance that then often leads to overuse, further tolerance, and addictive behavior. The mechanism of tolerance was initially thought to involve receptor downregulation and/or receptor population/location changes. The process probably revolves around changes in receptor linkage to second messengers and resulting ion channel effects. In particular, the N-methyl-D-aspartate (NMDA) ion channel complex seems very important because NMDA blockers (eg, ketamine) reduce tolerance (they also seem to reduce central sensitization).
Tolerance to analgesic, euphoric, and relaxing effects seems to be inevitable for most patients taking opioids chronically. Depending on the specific opioid medication, tolerance generally occurs after 2 weeks or so of continued use, and potency reduction can eventually be as great as 35-fold. And it is essential to remember that cross-tolerance is the rule in the opioid family – ie, tolerance to 1 opioid generalized to virtually all others with the possible exception of some effects of mixed agonist-antagonist agents. Tolerance to constipation and slowing of gastrointestinal function generally does not seem to happen. Respiratory depressant effects likewise seem more persistent over time with only moderate tolerance in some patients; hence, the high risk of anoxia and death with inadvertent overdose, even in some chronic users.
When tolerance leads to escalation of use, it almost invariably leads to some degree of dependence, defined as the physiological state of (1) requiring the substance for function and (2) leading to a withdrawal syndrome with abstinence. The withdrawal syndrome occurring with cessation of chronic opioid use consists of rhinorrhea, lacrimation, altered thermoregulation, mydriasis, generalized pain, vomiting, diarrhea, anxiety, and agitation. The withdrawal syndrome usually begins around 6-12 hours after cessation of opioids and is generally over in 2-3 days. This can vary, however – methadone withdrawal can peak after several days and lasts for 2 weeks – and craving for opioids can persist for very long periods of time.
Drug addiction, perhaps best defined as continued use despite negative consequences, occurs with opioid use because of a change in reward system activity and is notoriously difficult to reverse because of the resulting powerful reinforcement of drug use. Tolerance and dependence of course play a significant role as well. Additionally, opioids have strong mood elevating and anxiolytic properties that draw many to overuse.
The recently released Diagnostic and Statistical Manual of Mental Disorders, 5th Edition, avoids the terms addiction and dependence, choosing instead to define the syndrome of 292.9 opioid use disorder, requiring the features of craving, behaviors aimed at obtaining opioids, tolerance, and potential for withdrawal (Table 2). Interestingly, the criteria concerning tolerance and potential for withdrawal are not considered met if the patient is taking opioids under “appropriate medical supervision.” This makes assigning this diagnosis impossible for some patients whom many would consider to have a clear opioid use problem, as long as they are in an opioid maintenance program. Of course, the key phrase “appropriate medical supervision” may be difficult to define. While marijuana is the most prevalent initial drug of abuse in the United States (56%), opioids, including pharmaceutical and non-pharmaceutical forms, are the next most common at around 22%. Easy availability of oral opioids is certainly a factor here, but it may also be related to the relatively rapid development of tolerance in some patients. For example, many cases of opioid addiction began after using several opioid analgesics following third molar extraction or for other short-term uses.
Table 2. Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM V) 292.9 Opioid Use Disorder (abridged)
|A problematic pattern of opioid use leading to clinically significant impairment or distress, as manifested by at least 2 of the following, occurring within a 12-month period:|
- Opioids taken in larger amounts of over a longer period of time than was intended
- Persistent desire or unsuccessful efforts to control or cut down opioid use
- A great deal of time is spent using, obtaining, or recovering from effects
- Failure to fulfill major role obligations at work, school, or home
- Continued use despite problems because of opioids
- Important activities are given up
- Recurrent use in situations that are physically hazardous
- Continued use despite knowing of physical or psychological problem because of opioids
- Tolerance (unless used under “appropriate medical supervision”)
- Withdrawal (unless used under “appropriate medical supervision”)
So, as we consider the actions, advantages and disadvantages of the opioid group, can we draw conclusions about whether or not opioids have a place in the management of headache disorders? We might pose 3 key questions:
- 1.Are opioids useful when taken acutely to abort a migraine headache?
- 2.Are opioids useful taken regularly for prevention of migraine?
- 3.Are opioids useful for any other kind of head pain?
And, as we ask these questions about opioid use in headache medicine, we have other questions to consider:
- 4.What are the negative consequences to opioids in patients?
- 5.What are the negative consequences to opioids in families and communities?
Are Opioids Useful When Taken Acutely to Abort a Severe Migraine?
Many opioids are available for acute treatment of pain, and some seem to be of use to some patients (Table 3). The most commonly studied opioid is meperidine. A review by Friedman et al that considered several studies with a total of 254 patients noted that meperidine was less effective than dihydroergotamine (DHE) (odds ratio [OR] 0.30), suggested less efficacy than the anti-emetics (OR 0.46), and was most similar to the efficacy of ketorolac (OR 1.75). Further, meperidine tended to cause more sedation and dizziness than DHE, less extrapyramidal side effects than the anti-emetics, and similar sedation and gastrointestinal adverse effects to ketorolac. A recent review by Kelly and Tepper collected and analyzed 75 studies of acute pharmacological treatment of migraine. They found that the opioids meperidine, tramadol, and nalbuphine were superior to placebo in relieving migraine pain, although interestingly, meperidine combined with promethazine was not. Although difficult to assess fully for a number of reasons, they attempted to rate relative effectiveness of a number of acute interventions for migraine and found droperidol, sumatriptan, and prochlorperazine all to be optimally effective in the 77-82% range (ie, % of patients achieving relief). DHE followed in efficacy at 67% and chlorpromazine followed closely at 65%. Ketorolac and meperidine were a bit less effective at 60% and 58%, respectively. It is important to remember, however, that different members of the opioid family have different properties – affinities for opioid receptors (leading to variable analgesic potency), different bioavailabilities, and differences in abilities for crossing the blood-brain barrier – and the comparative studies earlier have not been exhaustive. Additionally, some patients may be opioid non-responders for as yet unclear reasons.[12, 13]
Table 3. Opioid Analgesics Used for the Acute Treatment of Headache
|Oxycodone||Percocet® (with acetaminophen)||NA||15-60||2-4|
|Hydrocodone||Vicodin® (with acetaminophen)||NA||10-30||4|
|Codeine||Many combination preparations||NA||100||3|
Despite the relative effectiveness of opioids for acute migraine, recurrence of pain is felt by some to be a significant problem. This too has been difficult to assess with estimates ranging from 23% to 71%,14-16 so recurrence of headache after opioid treatment may not differ from that after other acute pharmacological treatment. Friedman et al studied 309 patients treated acutely for primary headache and were able to reassess 94% of them at 24 hours. Moderate-to-severe headaches recurred within 24 hours after emergency department (ED) discharge in approximately one third of patients, and contrary to expectations, the recurrence proportion did not vary by analgesic treatment used.
Adverse effects of opioids have also led many authors to suggest that they be granted only a minor role in acute treatment of headache. Significant considerations here when choosing treatment in the ED setting include sedation, respiratory depression, bradycardia, and hypotension, as well as the long-term tendency over time to produce dependency. For home use, similar issues arise but to a lesser extent as administration will generally be oral and thus less likely to produce the more serious systemic effects. However, patients given more than very limited quantities of opioids can certainly be at risk for increasing usage and rapid development of tolerance, dependency, and addiction. Buse et al in an epidemiological analysis of a large telephone survey study of patients with migraine found that nearly 17% of responders were currently using opioids in a pattern highly suggestive of dependence. Not surprisingly, patients in this group were 6 times more likely to meet criteria for depression, had significantly higher levels of disability, and had nearly 9 times the rate of ER visits.
Of note, the mixed mu agonist-antagonist opioids including nalbuphine and butorphanol tend not to be abused because perhaps in part of their analgesic “ceiling” properties. Their typical opioid side effects seem also to be less pronounced as well, although they do have respiratory depressant properties. Tramadol, because of its relatively weak mu receptor binding properties, tends not to produce respiratory, cardiac, or gastrointestinal effects with typical doses. It can, however, produce tolerance and dependence, and unlike other opioids, it inhibits serotonin and norepinephrine reuptake so must be used cautiously in patients taking similarly acting medications chronically. Meperidine, still one of the most commonly used opioids in emergency rooms, has a unique metabolite, normeperidine, which is notable for a particularly long half-life (up to 24 hours) so with repeated doses, levels can accumulate leading to severe toxicity including respiratory compromise and seizures.
A final cautionary issue regarding the use of opioids for the acute relief of migraine is the propensity of virtually all of them to lead to medication overuse headache (MOH) (Table 4) and/or progression of episodic migraine to chronic migraine (CM) (“chronification”).19-21 Bigal et al documented the association of opioid usage with progression of migraine fairly convincingly, with the critical frequency of use approximately 8 days per month. This may underlie the findings in several studies that prior opioid use leads to headache unresponsiveness to other acute medications,[22, 23] as well as to a higher likelihood of an emergency room visit.
Table 4. 8.2.4 Opioid-Overuse Headache – International Classification of Headache Disorders (ICHD III)
- Headache occurring on ≥15 days per month in a patient with a pre-existing headache disorder
- Regular overuse of 1 or more opioids on ≥10 days per month for >3 months
- Not better accounted for by another ICHD-III diagnosis
Are Opioids Useful When Taken Prophylactically for Intractable Migraine?
Beginning in the 1990s, a dramatic increase in opioid treatment for non-terminal chronic pain conditions has been seen. This turnaround from a previously very hesitant approach to opioid prescribing by the medical community was largely fueled by pharmaceutical companies and a small group of investigators who asserted that fears of tolerance and addiction were exaggerated, and proselytized the daily use of opioid medications for painful illnesses including arthritis, back pain, fibromyalgia, and chronic headache disorders. Despite relatively sparse evidence for efficacy and safety, a nearly religious movement seemed to take hold, leading to the concepts of “Pain as the fifth vital sign” and that undertreatment with opioids was essentially unethical.[25, 26] Assessments of pain management programs nationally and internationally were predominantly assays of access to opioid-based treatment. A 2008 review discovered that of all long-term opioid therapy at that time, more than 90% was being prescribed for chronic non-cancer pain. Between 1997 and 2002, oxycodone prescriptions alone quadrupled and a 2009 study reported that more than 3% of all adults in the US were receiving long-term opioid therapy for chronic non-cancer pain. During the same period, opioid addiction and its consequences, including deaths from unintentional overdose, markedly increased. Between 1985 and 2005, data from the National Vital Statistics System of the Centers for Disease Control and Prevention show that the death rate from unintentional drug overdose increased by nearly 600%, much of this is due to prescription opioid abuse.
During the same roughly 20-year period, trends in treating patients with frequent headaches paralleled the dramatic rise in opioid use for non-malignant pain. Guidelines published by the American Pain Society in 2009 proposed chronic headache disorders as one of the 4 common chronic pain conditions where chronic opioid therapy might be considered. And a number of regimens for continuous opioid therapy have been devised for patients with refractory CM and other intractable chronic headache disorders (Table 5). However, evidence for the effectiveness of chronic opioid treatment of CM patients is lacking. Saper et al have followed a large cohort of refractory headache patients treated with daily opioid therapy, and while initially promising, results have begun to look bleak.[31, 32] While about one quarter of the 160 enrolled patients seemed to attaining a 50% or better improvement (using an index of severe headache activity), other measures were much less encouraging, and there was serious question as to the validity of patients' self-reporting. Disability scores, for example, did not improve even for this group, and behaviors such as drug-seeking and dose violation seemed to persist for many. Other reports suggest better results for opioid therapy in headache,[33, 34] but all are fraught with a number of pitfalls. First, when comparing active and placebo responses, maintaining good blinding is probably impossible because of the euphoric and sedating properties of opioids. Related to this is the presumed tendency for patients to exaggerate improvement with opioids do to the anxiolytic and other beneficial effects on mood, not to mention the potential impact of habituation.
Table 5. Opioid Analgesics That Have Been Used for Prophylactic Treatment of Chronic Refractory Headaches
|Morphine (po, sustained release)||30||Q12 hours||10-12|
|Oxycodone (po, sustained release)||20||Q12 hours||4-5|
|Fentanyl (transdermal)||12.5-25 µg/hr||continuous||–|
Adverse effects to opioids may be amplified when use is daily. Significant gastrointestinal dysfunction in particular has been seen in many patients on continuous opioid therapy. The “opioid bowel syndrome” can include intractable severe abdominal pain, which in some cases leads to inappropriate escalation of opioid medication. The most worrisome potential adverse effects from regular opioid use are respiratory depression and sudden cardiac death presumably because of arrhythmia. These potentially life-threatening phenomena are difficult to quantify but clearly exist. Methadone is particularly worrisome because of its long half-life (as long as 50 hours in some patients) with unnoticed rising blood levels. In addition, methadone is known to prolong the QT interval so can lead to fatal arrhythmia (torsade de pointes). Significant sleep disturbance and sexual dysfunction can also emerge in patients taking daily opioids. Webster et al studied 147 patients receiving daily opioids for various pain conditions and found sleep apnea in 75% (either obstructive or central).
Cognitive and behavioral function must be closely monitored in any patient on daily or even frequent opioid medication. Mood alteration, mental fogginess, and motivational issues are universally known side effects to opioids in humans. Some of these do lessen with tolerance, but symptoms of anxiety and mood change can lead to increased use. Sjogren et al in 2000 studied cognitive function in 40 patients receiving long-term oral opioid therapy for non-malignant pain, primarily sustained-release morphine or methadone, and compared psychometric performance with 40 age-matched healthy volunteers. Clear relative deficiencies in memory, attention, and psychomotor speed were found in patients on opioid therapy. However, other studies have not replicated these results. The obvious confounding issue in attempting to assess cognition and behavior is that any abnormal function in these areas might be the result of pain, anxiety, or depression because of the medical condition rather than the opioids themselves.
Along with the nearly inevitable tolerance to analgesic benefits that seems to accompany frequent opioid use, a seemingly paradoxical phenomenon – opioid induced hyperalgesia (OIH) – has been observed.
This has clearly been shown to occur in many patients taking daily opioids and is diagnosed clinically by noting increased pain despite an increase in dosage (which generally happens if the prescriber incorrectly assumes tolerance has developed). OIH, while not completely understood, probably results from a number of mechanisms including activation of excitatory anti-analgesic pathways, pain facilitation via dynorphin and CCK activation, increased activity of nociceptive pathway excitatory neuropeptides (calcitonin gene-related peptide and substance P), descending facilitation of pain involving the RVM and probably glial activation with release of cytokines that augment nociception. OIH can easily be misdiagnosed as tolerance or disease-worsening, but a clue can be local allodynia, in addition heightened pain with dosage increase.
Finally, as described earlier, MOH can compromise the potential benefits of opioids for relief of chronic headache disorders. With daily opioid use, presumably the risk of MOH rises significantly. While this is difficult to assess, several authors have demonstrated improvement in chronic primary headache following discontinuation of opioid medications.39-42
Despite the many negatives, there will be selected cases for whom one could consider continuous opioid therapy. These might include patients who are truly refractory to a number of properly executed pharmacological and non-pharmacological approaches, or who truly cannot tolerate any of the alternatives. Saper and his team have devised a set of guidelines for choosing patients who might be appropriate for daily opioid therapy (Table 6). These guidelines are based on data from longitudinal studies as well as years of accumulated experience in working with intractable patients and opioid programs. They stipulate that patients be over 30, have very frequent and disabling pain, and have a history of good compliance. They also require that the pain has been refractory or that typical measures are contraindicated, and that the patient is well known to the skilled prescriber. Past addictive disease, serious mental illness, inappropriate drug-seeking behavior, and a home environment that includes drug abuse are all considered contraindications to chronic opioid treatment. Formal monitoring including a thorough written contract, urine drug screening, and regular office visits including psychological counseling are all required.
Table 6. A Screening Checklist for Chronic Opioid Therapy in Refractory Migraine (After Saper et al)
- ○ Patient over 30
- ○ Frequent and disabling pain
- ○ History of good compliance
- ○ Typical treatments are ineffective or contraindicated
- ○ Well known to prescriber
- ○ No history of addictive disease
- ○ No history of serious mental illness
- ○ No inappropriate drug-seeking behavior
- ○ No drug abuse in environment
- ○ Monitoring includes: written contract, urine drug screening, regular office visits, and psychological counseling are all required
Are Opioids Useful for Other Head Pain?
Up to this point, mainly refractory migraine has been considered. Would other primary headache disorders be targets for chronic opioid therapy? The refractory cluster headache patient, for example, might be considered a prime candidate particularly if there is frequent, extremely severe, disabling pain, leading to sleep deprivation and potentially suicidal ideation. However, it is in this type of patient that one can see the inherent dangers of beginning a program of regular opioid treatment. The frequency of headaches might very well lead to rather rapid escalation of dosage, particularly if there has been any history of opioid use and/or tolerance. Prophylactic medications like calcium channel blockers and lithium will have to be carefully prescribed to avoid drug-drug and additive interactions. Similar considerations are probably apt for most patients with other refractory primary headache forms.
Might opioids be an option for acute or chronic pain from secondary headaches? Conceivably yes, particularly if the cause of pain is expected to be self-limited – for example, acute head trauma, post-surgical head pain, otitis media, and cellulitis. However, it has become clear that physical and psychological dependence can occur very quickly and that even OIH can occur with even brief courses of opioids, and there are often reasonably good alternatives. Additionally, acute injuries or infections carry other imperatives. In the case of acute traumatic brain injury, for example, it will be crucial to remember that opioids increase intracranial pressure and may impair the ability to perform accurate mental status exams.
Opioids will continue to be used for acute pain of all types including migraine and other headaches. There are generally better alternatives, such as DHE, non-steroidal anti-inflammatory drugs, and neuroleptic anti-emetics. However, data about relative effectiveness of opioids and other analgesic/abortive medications are by no means complete as different members of the opioid class have different properties and potencies, and exhaustive comparative trials have not been done. Opioids have numerous adverse effects, some life-threatening. There is high risk for tolerance, dependency, and addiction with significant effects on patients, families, and communities. And opioid use seems to make migraines subsequently more frequent and more difficult to treat. However, there are patients for whom opioids on occasion are optimal acute treatments, for example, patients who have contraindications to ergot-type medications because of cardiovascular or cerebrovascular conditions, patients who should not receive neuroleptic medications because of QT prolongation or other contraindications, pregnant women, and patients who cannot tolerate or fail to respond to all other categories of acute medications. In these situations, both clinician and patient should thoroughly understand the pitfalls in using opioids and make a conscious decision to use them sparingly. This will also apply to cases of secondary head pain expected to be of a time-limited course.
As for prophylaxis in patients with refractory CM, regular preventive dosing can seem better than the nightmare of daily or near daily severe pain. And certainly, there are a handful of patients who seem to flourish with a carefully controlled regimen of methadone or long-acting morphine preparation. But in the experience of most headache medicine specialists, these patients are very few in number, and this observation is borne out by the longitudinal studies. Moreover, the recent trend in overprescribing opioids for chronic non-terminal pain has led to serious individual and societal consequences that must now be addressed conscientiously. On the other hand, this approach, if done in a careful way by skilled practitioners, can be viable for selected cases.
Statement of Authorship
- (a) Conception and DesignMorris Levin
- (b) Acquisition of DataMorris Levin
- (c) Analysis and Interpretation of DataMorris Levin
- (a) Drafting the ManuscriptMorris Levin
- (b) Revising It for Intellectual ContentMorris Levin
- (a) Final Approval of the Completed ManuscriptMorris Levin