Psychotropic medication in chronic spinal disorders

The Spine Journal 4 (2004) 436–450

Contemporary Concepts Review

Psychotropic medication in chronic spinal disorders Peter B. Polatin, MDa,*, Jeffrey Dersh, PhDb a

Department of Anesthesia and Pain Management, University of Texas Southwestern Medical Center at Dallas, Eugene McDermott Center for Pain Management, 6263 Harry Hines Blvd., Dallas, Dallas, TX 75235, USA b Productive Rehabilitation Institute of Dallas for Ergonomics (PRIDE), 5701 Maple Ave., Dallas, TX 75235, USA

Abstract

BACKGROUND CONTEXT: Optimal treatment of nonmalignant chronic spinal disorders (CSDs) may require the use of one or more nonopioid psychotropic medications. Vast research literature has documented high rates of psychiatric disorders in patients with CSDs. Psychotropic medications are one type of effective treatment for these disorders. Many medications of this type are also used as adjuvants to primary analgesic medications. PURPOSE: Physicians treating CSDs may have little training and experience in prescribing psychotropic medications. Further, they may possess limited information about the efficacy of these types of medications in treating psychiatric disorders comorbid with CSDs or as adjunctive analgesics. Because of the wide variety (antidepressants, anxiolytics, sleep-promoting agents, anticonvulsants, neuroleptics, muscle relaxants) and numerous indications for psychiatric medications, a concise review of the use of psychotropic medications with CSD patients is offered for the spine specialist. STUDY DESIGN: A systematic review of the contemporary English literature on psychotropic medications in the CSD population. METHODS: A computerized search of MEDLINE was performed on all English literature published from 1982 to August 2002. RESULTS: Psychotropic medications have been found to be very efficacious in the large subgroup of patients with CSDs with psychiatric comorbidity, particularly patients with the common constellation of depression, anxiety, excessive somatic complaints, insomnia and irritability. Although the type of medication indicated depends on the particular psychiatric syndrome(s), antidepressants have been found to be extremely useful with the constellation described above. There is less evidence to support the use of nonopioid psychotropic medications in the treatment of nociceptive pain, although clinical experience indicates that some individuals demonstrate a marked analgesic response. Research support for the efficacy of particular psychotropic medications in treating neuropathic pain is stronger, although more controlled research is clearly needed. CONCLUSIONS: Psychotropic medications are extremely useful in the treatment of psychiatric disorders comorbid with CSDs and modestly useful as analgesic adjuvants, particularly with pain of neuropathic etiology. Familiarity with these medications will aid the primary treating physician in optimizing outcomes in this difficult group of patients. 쑖 2004 Elsevier Inc. All rights reserved.

Keywords:

Spinal diseases; Spinal disorders; Musculoskeletal diseases; Chronic pain; Analgesics; Mental disorders; Depressive disorders; Anxiety disorders; Psychopharmacology; Psychotropic drugs

FDA device/drug status: not applicable. Nothing of value received from a commercial entity related to this research. This Contemporary Concepts review article has been reviewed by the Board of the North American Spine Society (NASS). As such, it represents the current position of the state of knowledge of the above subject in spine care. This series is edited by Alexander Vaccaro, MD. Before entering the review process for The Spine Journal, the authors were assisted by members of the NASS Committee on Contemporary Concepts, Alexander R. Vaccaro, Chair; Lawrence W. Frank, MD, and James P. Argires, MD, FACS, reviewers. * Corresponding author. 6263 Harry Hines Boulevard, Dallas, TX 75235, USA. Tel.: (214) 648-0660; fax: (214) 648-0693. E-mail address: [email protected] (P.B. Polatin) 1529-9430/04/$ – see front matter doi:10.1016/j.spinee.2004.01.012

쑖 2004 Elsevier Inc. All rights reserved.

Introduction Optimal treatment of nonmalignant chronic spinal disorders (CSDs) often requires the use of one or more psychotropic medications (PMs). Vast research literature has documented high rates of psychiatric disorders in patients with CSDs [1]. PMs are one type of effective treatment for these disorders. Many PMs are also used to manage the pain that is typically associated with CSDs, including pain of both nociceptive and neuropathic origins. Opioid medications, nonsteroidal anti-inflammatory drugs (NSAIDs)

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

and acetominophen are the primary analgesics with onset of action within minutes to hours. PMs have primary Food and Drug Administration (FDA)–approved indications for nonpain diagnoses but have been found to be useful as analgesics in patients with CSDs by direct central effects, by direct peripheral effects or indirectly, by decreasing emotional distress. Onset of action is slower than that of the primary analgesics and generally occurs over weeks. In most cases, PMs are not the only medication(s) used to control pain in CSDs. Instead, they are typically used adjunctively with the primary analgesics for additional pain relief. By increasing the analgesic effect, these agents may allow the patient to reduce the dose of the primary analgesic required for adequate pain control. Psychotropic medications may be divided into specific categories based on primary therapeutic effect. In this review, the following classes of PMs will be discussed: antidepressants, anticonvulsants, muscle relaxants, antianxiety agents, sleep-promoting agents and neuroleptics. Opioid medications, with the exception of tramadol, will not be addressed, because use of this type of analgesic medication has been thoroughly covered in several recent high-quality reviews [2,3]. Information about specific medications should be checked against the Physician’s Desk Reference and package inserts to ensure accuracy before prescribing for specific patients. The clinician should be mindful of the fact that the uses of some of these medications for the treatment of chronic pain are “off-label” indications and therefore may not be found in standard references. However, peer-review journal references do substantiate these applications, many of which are listed in the reference list of this paper.

Historic background Psychotropic medications were first used in patients with chronic pain, including patients with CSDs, to manage coexisting psychiatric disorders. For example, as the moodelevating properties of the monoamine oxidase inhibitors (MAOIs) and imipramine (a tricyclic antidepressant, TCA) became apparent in the 1950s, these medications were increasingly used to treat depression in general psychiatry practice and in patients with pain and coexisting depression [4], although the frequent coexistence of these conditions had not been documented at that time. The initial report of the antinociceptive properties of nonopioid PMs occurred in 1960, when a group of researchers observed pain relief in patients with cancer with administration of TCAs [5]. This discovery, in combination with the increasing recognition of the frequent co-occurrence of chronic pain and depression in the 1970s and 1980s, culminated in widespread use of TCAs in patients with chronic pain. The evolution of the use of PMs in CSDs has largely paralleled that of their use in a wide variety of chronic pain conditions. Advances have largely been incremental, as new PMs have been developed or recognized, as the pathophysiology of pain has been

437

clarified (eg, nociceptive versus neuropathic) and as the co-occurrence of chronic pain and CSDs with a variety of psychiatric disorders has been increasingly documented. Many developments in the field have been serendipitous; others have been the product of rational application of accruing basic knowledge regarding the pathophysiology of pain and the mechanisms of action of PMs.

Anatomy and pathophysiology: chronic spinal disorders and psychiatric disorders Acute spinal disorders often provoke clinically significant generalized anxiety symptoms [4]. Patients describe feeling tense, fearful, edgy or stressed. In addition, patients worry more and have difficulty controlling ruminative mental activity. Most episodes of spinal pain resolve within a few weeks or months, along with these initial reactive emotional symptoms. However, as spinal disorders become chronic, and especially as patients become entrenched in the sick role, many will develop psychiatric symptoms [4,5], most commonly depressive disorders, anxiety disorders, somatoform disorders, substance use disorders and dysfunctional behaviors suggestive of personality disorders. We are aware of only one study that comprehensively examined rates of psychiatric disorders in an exclusive CSD sample, in this case, 200 patients with chronic low back pain [6]. In this study, experienced mental health practitioners using a validated structured interview format found that, excluding somatoform pain disorder (for which 99% of patients met criteria), 59% of patients met criteria for one or more of the following psychiatric diagnoses: major depressive disorder, an anxiety disorder (primarily panic disorder and/or generalized anxiety disorder) and/or a substance use disorder. Major depressive disorder was the most common disorder and the rate of concurrent disorders was high. Personality disorders were identified in 51% of these patients. The core features of each of these disorders are summarized in Table 1. The direction of the causal relationship between CSDs and psychiatric disorders has been debated for many years [5]. However, it appears that the rates of psychiatric disorders are not elevated before the development of CSDs, suggesting that most psychiatric disorders develop as reactions to the stress associated with CSDs [6]. The type of psychiatric disorder(s) that develops in a particular individual is believed to be linked to pre-existing semidormant characteristics of the individual before the onset of the CSD [1,5,7,8]. In other words, some patients are more prone to developing depression under stress, others anxiety disorders and so on. Maladaptive coping styles, commonly associated with personality traits/disorders or with previous experiences of physical and emotional trauma, may also become more clinically evident as a result of chronic pain. Additionally, the worsening of pre-existing biological illnesses, such as schizophrenia and bipolar affective disorder, may require management with psychotropic medication. Therefore, the full range of

438

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

Table 1 Features of common psychiatric disorders in patients with chronic spinal disorders Psychiatric disorder

Features and symptoms

Major depressive disorder

Persistent sad mood and/or decreased interest and pleasure in normal activities for at least 2 weeks, accompanied by at least three of the following symptoms: decreased or increased appetite or weight, insomnia or hypersomnia, psychomotor agitation or retardation, fatigue or loss of energy, feelings of worthlessness or excessive guilt, difficulty thinking or concentrating, recurrent thoughts of death or suicide

Anxiety disorders* Generalized anxiety disorder

Panic disorder

Somatoform disorders† Pain disorder‡ Substance use disorders§

Personality disorders||

Excessive anxiety and worry (apprehensive expectation) and difficulty controlling the worry for at least 6 months, accompanied by at least three of the following symptoms: restlessness or feeling keyed up, being easily fatigued, difficulty concentrating, irritability, muscle tension, sleep disturbance Recurrent unexpected panic attacks. Panic attacks are discrete periods of intense fear or discomfort, in which four or more symptoms develop abruptly and reach a peak within 10 minutes. Common symptoms include heart palpitations, sweating, trembling, shortness of breath, feeling of choking, chest pain, nausea, dizziness and fears of losing control, going crazy or dying. Pain is the predominant focus of clinical presentation, causes clinically significant distress or disability, with psychological factors judged to have an important role in the onset, severity, exacerbation or maintenance of the pain. Abuse of or dependence on one or more substances (eg, opioids, alcohol, cannabis). Abuse is a less malignant diagnosis and involves a maladaptive pattern of substance use leading to clinically significant impairment or distress. Dependence or “addiction” includes the symptoms of abuse in addition to behavioral dependence (excessive substance-seeking activities and related evidence of pathological use patterns), usually accompanied by physical dependence (ie, tolerance and withdrawal syndromes). Note that physical dependence (eg, on opioids) is often present in the absence of addiction in patients with CSDs. There are 10 recognized personality disorders, including paranoid, borderline and dependent. Personality disorders are diagnosed when personality traits are judged to be inflexible and maladaptive, causing significant functional impairment or subjective distress.

CSD⫽chronic spinal disorder. * Other anxiety disorders (posttraumatic stress disorder, specific phobia, obsessive compulsive disorder and social anxiety disorder) are relatively uncommon in patients with CSDs. † Other somatoform disorders (somatization disorder, hypochondriasis, conversion disorder and undifferentiated somatoform disorder) are relatively uncommon in patients with CSDs. ‡ It appears that almost all patients with CSDs with pain complaints, by definition, will meet the criteria for pain disorder. § Controversy is ongoing regarding the diagnosis of opioid dependence in patients with chronic pain. However, the preponderance of evidence suggests that only about 10% of opioid-using chronic pain patients become “addicted” to opioids, despite much higher rates of symptoms of tolerance and withdrawal. || Patients with CSDs with certain personality disorders are probably the most difficult and frustrating to treat. Psychiatric/psychological evaluation and treatment is typically necessary for moderate to severe personality disorders.

clinical psychopharmacology is required to treat secondary symptoms of emotional distress in the chronic pain population. Treatment of psychiatric disorders in patients with CSDs is important for several reasons. Most importantly, the effectiveness of treatment, whether surgery or rehabilitation, is compromised when these disorders are not properly treated [5]. An obvious example is the hopeless and anergic depressed patient, who will not engage in rehabilitation in any meaningful way until his or her symptoms are addressed. Depression and anxiety are known to enhance perception of pain [9], thus serving to perpetuate pain-related dysfunction and disability. For example, anxiety has been found to decrease pain threshold and tolerance [10], and emotional distress has been linked to physical symptoms by means of autonomic arousal, vigilance and misinterpretation [11] or somatic amplification [12]. In summary, effective treatment of psychiatric disorders increases patients’ ability to engage in treatment and may also have an indirect effect on pain level by decreasing emotional distress.

Clinical research: methods A computerized search of MEDLINE was performed on all literature published in English in the last 20 years up to August 2002 using the following medical subject headings or keywords: spinal diseases, spinal disorders, musculoskeletal diseases, pain, pain (intractable), back pain, low back pain, analgesics, psychopathology, mental disorders, depressive disorders, anxiety disorders, somatoform disorders, personality disorders, psychopharmacology, psychotropic drugs and antidepressive agents. Standard texts covering spinal disorders, chronic pain, psychiatry and psychopharmacology were also reviewed. In addition, our clinical experience with thousands of patients with CSDs informed the content of this review. In terms of importance, studies and reviews that focused specifically on the use of psychotropic drugs in patients with CSDs were given highest priority. However, these were few in number. Consequently, we also examined the more general chronic pain and chronic musculoskeletal disorders literature. Of course, conclusions based on these studies are more tentative, and this is noted in the text when

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

it occurs. It should also be noted that we placed greater emphasis on recent meta-analytic and review articles. Most of these articles considered the quality of the original research reports, thereby increasing the scientific merit of the findings. The major classes of psychotropic medications Antidepressants Antidepressants have multiple indications in patients with CSDs, including depression, nociceptive pain, neuropathic pain, anxiety, insomnia, irritability, substance use disorders and somatization. Because of their effectiveness for so many syndromes associated with CSDs, antidepressants are often the most important type of medication in this patient population (in addition to the primary analgesics: opioids and nonsteroidal anti-inflammatory drugs). Antidepressants may be divided into several general classes: tricyclic antidepressants (TCAs), selective serotonin reuptake inhibitors (SSRIs), MAOIs and novel antidepressants. All of the antidepressants are believed to exert their therapeutic effect by selectively increasing the activity of biogenic amine neurotransmitters in multiple pathways within the central nervous system. These neurotransmitters include serotonin, norepinephrine and dopamine, although it is believed that the former two are the neurotransmitters most specifically involved in depression. The antidepressants vary in their specificity for serotonin and/or norepinephrine. These drugs have two different dosage ranges, depending on whether pain or depression is the symptom being treated. Most experts believe that pain responds at a lower dose and at a shorter time after initiation of therapy than does depression [13,14], although it is also recognized that some patients will require higher doses for effective analgesia [14]. This is particularly important when treating a chronic pain patient without depression, where the lower dose will be used. It is less relevant when both symptoms are present, because the antidepressant dosage will be used. It should be noted that some experts believe that analgesia is maximized with increased doses as well as with the duration of therapy, with increasing benefits observed up to 4 months [15]. Therefore, as long as side effects are tolerated, dosage maximization is recommended. Table 2 describes the antidepressants most commonly used with CSD patients, including indications for use and proposed dose ranges for pain and depression. This review will not describe the use of MAOIs because of their limited use in the CSD population, which is related to a high rate of adverse interactions with other types of psychotropic medications (including some opioids) and food products, potentially resulting in a fatal hypertensive crisis if dietary restrictions are not followed or drug interactions not prevented. Adverse effects vary by class of medication. TCAs and SSRIs are associated with different side-effect profiles. In general, the side effects of TCAs are more pronounced,

439

often limiting the dose that can be tolerated by the patient. The SSRIs are better tolerated and have therefore surpassed TCAs as first-line treatment for depression. The novel antidepressants have different mechanisms of actions from the TCAs and SSRIs as well as from each other. Consequently, each has a unique side-effect profile. Table 3 summarizes the common initial adverse effects, common persisting adverse effects, uncommon (but dangerous) adverse effects and contraindications of the antidepressants. All of the antidepressants, with the possible exception of Desyrel (which is generally regarded as less effective), are believed to be about equally effective in the treatment of depression, although dual-action (ie, serotonin and norepinephrine) drugs are at least theoretically more effective (see Table 2). Decisions about which agent to use are therefore based on a variety of other factors, including the side-effect profile and efficacy with associated symptoms and syndromes (eg, pain, anxiety, insomnia). Most patients with CSDs present with a constellation of symptoms/syndromes in addition to pain and depression. The most common are anxiety (generalized anxiety, panic attacks), insomnia, irritability and “somatization.” Somatization refers to an unconscious tendency to experience and communicate emotional distress in terms of somatic discomfort [16], although it is now recognized that somatic and emotional distress can cooccur rather than substitute for one another [12]. Almost all patients with CSDs will meet the criteria for the somatoform (pain) disorder, but there appears to be a subset of this population in which somatization is amplified, [17]. These patients present with a variety of somatic complaints, most of which cannot be medically explained. Antidepressants have been found to be effective in reducing the intensity of somatization symptoms [18]. This is not surprising, given the high rates of somatization in depressed and anxious patients without chronic pain conditions [12]. The constellation of symptoms/syndromes associated with depression and pain are responsive to many of the antidepressants. In our practice, we find that one antidepressant medication, usually at a higher dose range (titrated up over a period of time), will result in amelioration or remittance of depression and the associated symptoms/syndromes, with sleep and irritability often improving during the first week. The other symptoms, including depression and anxiety, begin to improve after 2 to 3 weeks. Therapeutic response to antidepressants increases for at least 2 months (and for a similar amount of time with each increase in dose). Response is often dose related, with a better response at higher doses in most cases. Although it is always prudent to use the lowest effective dose, it has been well established that nonpsychiatric physicians typically do not use adequate antidepressant doses and tend to discontinue these agents too quickly. In general, for depressive and anxiety disorders, antidepressants should be continued for at least 3 to 6 months after the symptoms have remitted (ie, are no longer clinically significant). Many patients do not respond to the initial antidepressant at the starting dose. In general, it is

440

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

Table 2 Antidepressants commonly used to treat patients with chronic spinal disorders Class TCA

SSRI

Novel

Medication, proprietary (nonproprietary)

Dose for pain (mg)

Dose for depression

Neurotransmitter specificity

Other uses

Anafranil (clomipramine) Elavil (amitriptyline) Norpramine (desipramine) Pamelor (nortriptyline) Sinequan (doxepin) Tofranil (imipramine) Celexa (citalopram) Paxil (paroxetine)* Prozac (fluoxetine) Zoloft (sertraline) Lexapro (escitalopram) Desyrel (trazadone) Effexor XR (venlafaxine)† Remeron (mirtazapine) Serzone (nefazadone) Wellbutrin SR (buproprion)‡

? 75–100 75 50–100 50–100 50–75 ? 40–60 ? ? ? ? ? ? ? ?

100–250 150–300 75–200 75–150 150–300 150–300 20–80 20–80 20–80 50–200 10–30 150–400 75–375 15–60 300–600 200–400

Ser, Ser, Nor Nor Ser, Ser, Ser Ser, Ser Ser, Ser Ser Ser, Ser, Ser Ser,

Anx, Ins, Som Anx, Ins, Som Som Som Anx, Ins, Som Anx, Ins, Som Anx, Irr, Som Anx, Ins, Irr, Som Anx, Irr, Som Anx, Irr, Som Anx, Irr, Som Anx, Ins Anx, Irr, Som, ADHD Anx, Ins, Irr, Som Anx, Ins, Irr, Som SUB, obesity, ADHD

Nor Nor

Nor Nor Nor Dop

Nor Nor Dop, Nor

* Paxil is now available in a controlled-release preparation (Paxil CR). This new preparation, with a slightly different dose range, is associated with less gastrointestinal distress. However, it may be a less effective sleep aid. † Effexor was originally released as a twice daily medication. The extended release (XR) is used almost universally at the present time because it requires only once-a-day dosing. ‡ Wellbutrin SR (sustained release) is administered twice daily. The original Wellbutrin required three-times-a-day dosing. Wellbutrin SR is also indicated for smoking cessation under the proprietary name Zyban. Wellbutrin and Zyban should never be administered concurrently. ADHD⫽attention deficit hyperactivity disorder; Anx⫽anxiety; Dop⫽dopamine; Ins⫽insomnia; Irr⫽irritability; Nor⫽norepinephrine; Ser⫽serotonin; Som⫽somatization; SR⫽sustained release; SUB⫽substance use disorders; XR⫽extended release.

recommended that the physician maximize the dose of the initial medication (barring intolerable side effects), before tapering and switching to another antidepressant. It should be noted that the SSRIs, some of the TCAs and venlafaxine (Effexor) are first-line treatments for generalized anxiety disorder, panic disorder and other anxiety disorders, such as posttraumatic stress disorder, although not all are FDA approved for such indications. Mirtazapine (Remeron), nefazodone (Serzone) and trazadone (Desyrel) also appear to be effective for anxiety symptoms, although there is less supportive evidence in the controlled research literature [19]. As noted above, choice of antidepressant is usually based on side-effect profile, efficacy with associated symptoms and syndromes and specificity of biogenic amine activity (serotonin and/or norepinephrine). Patients with irritability, anxiety and insomnia will benefit from a more sedating antidepressant, such as doxepin (Sinequan), amitriptyline (Elavil), trazodone (Desyrel), mirtazapine (Remeron), nefazadone (Serzone) or paroxetine (Paxil). Those with anergia and psychomotor retardation would do better on a drug with a more energizing profile, such as nortriptyline (Pamelor), fluoxetine (Prozac), sertraline (Zoloft) or buproprion (Wellbutrin). In practice, the SSRIs and novel antidepressants are used more often, and the TCAs less often, because of concerns about intolerable side effects and safety of the TCAs, especially at higher doses. The SSRIs are particularly effective with irritability, which is most often an associated symptom of depression and/or anxiety. When seen in association with impulsive

behavior, irritability is often a symptom of an Axis II (personality) disorder [20], such as borderline or antisocial personality disorder. SSRIs have demonstrated efficacy for irritability associated with these personality disorders [21], but if the irritability is a symptom of an underlying bipolar disorder (previously called manic depression), it can exacerbate the manic episode. Therefore, close monitoring of behavioral and affective changes is important after initiating antidepressant medication. The degree of norepinephrine versus serotonin activity of any individual antidepressant will influence its side-effect profile. Antidepressants with a high degree of noradrenergic activity are associated with primary autonomic, cardiac and ocular side effects, which may be particularly problematic for the older patient with heart disease, glaucoma, prostatic hypertrophy or cognitive impairment. Primarily serotonergic agents may be preferable in this age group and in patients who have had a particularly refractory reaction to a previously tried noradrenergic antidepressant, such as increased appetite, sedation or intolerable dry mouth. However, the serotonergic agents are more frequently associated with headaches, gastric symptoms and agitation. Side effects of antidepressants are common and require careful clinical monitoring. Doses, with the exception of Remeron, should be started low and titrated upward, guided by improvement in pain or depression and by emergence of side effects. Remeron is often started at a higher dose (ie, 30 mg at night before sleep), because it has been found to produce less daytime sedation at higher doses. With several of the older TCAs (Elavil, Sinequan, Tofranil), as well as

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

441

Table 3 Adverse effects associated with antidepressants Uncommon dangerous side effects

Common initial side effects

Common persisting side effects

TCAs

Sedation, dry mouth, blurred vision, urinary retention, constipation, postural hypotension

Agranulocytosis, seizures

SSRIs

Headache, insomnia, anxiety, dizziness, tremors, sedation, nausea, diarrhea

Same as initial side effects, although may attenuate weight gain, sexual dysfunction, slowed cardiac conduction Decreased libido, anorgasmia

Trazodone

Sedation, dry mouth, postural hypotension

Same, although may attenuate

Priapism

Effexor XR

Dizziness, hypertension, tachycardia, nausea, dry mouth, sweating

Sexual dysfunction, hypertension

Remeron

Sedation, increased appetite Sedation, dizziness, headache Agitation, insomnia, dry mouth, blurred vision, headache, tremor, dizziness, nauea, vomiting, constipation

Same, weight gain

Class/medication

Serzone Wellbutrin (including SR and XL)

Suicidal ideation

Contraindications

Comments

History of myocardial infarction, seizure disorder, benign prostatic hypertrophy, narrow angle glaucoma, cardiac arrhythmia or cardiac disease Cautious use with renal or liver impairment

Withdrawal syndrome, overdose can be fatal, efficacy can be limited by side effects

Liver failure

Liver disease

Seizures

History of seizures

Sometimes withdrawal syndrome, generally safe in overdosage, caution when prescibing along with drugs that undergo extensive liver metabolism and have narrow therapeutic index Not well tolerated at antidepressant dose, no sexual side effects, good sedative Prolonged withdrawal syndrome, requiring supervised taper; extended release (XR) recommended May need to be discontinued if weight gain No sexual dysfunction Dose should be raised slowly and not exceed 300 mg; QD to reduce seizure risk. Less effective for anxiety and insomnia; SR or XI preferred

TCAs⫽tricyclic antidepressants; SSRIs⫽selective serotonin reuptake inhibitors; XR⫽extended release.

trazadone (Desyrel), a therapeutic level is associated with some dry mouth or morning “hangover.” However, any side effect that is associated with distress or impaired functioning will require dosage modification or a change of medication. The drawing of blood levels to monitor clinical response is useful when 1) an agent with a “therapeutic window” is being prescribed, such as imipramine; 2) the clinical response is ambiguous, raising the questions of absorption and compliance; 3) an older or more medically complicated patient is being treated. The routine use of blood levels is, however, not essential for a safe and effective administration of these antidepressant medications. There is a documented adjuvant role for the antidepressants in the treatment of chronic pain syndromes, particularly headache and neuropathic pain, less so in CSDs associated with nociceptive pain [14,22–24]. Nociceptive pain occurs with ongoing tissue damage. The putative nociceptive analgesic mechanism of action of antidepressants is increased activity of serotonergic and noradrenergic projections descending from the brain stem to the spinal cord, where they

have an inhibitory effect on ascending pain signals. It has been proposed that these agents may have additional central and peripheral analgesic mechanisms of action, including an effect on opioid receptors [25]. The issue of which neurotransmitter is primarily affected by the antidepressant agent has both conceptual and clinical applications. Heterocyclic agents with the best-documented nociceptive effects appear to be those agents with the ability to increase both serotonin and norepinephrine activity (see Table 2) [26]. Despite the evidence for the efficacy of antidepressants in various chronic pain syndromes, relatively few studies specifically examine the analgesic effectiveness of these agents in patients with CSDs. Those that do fail to differentiate nociceptive and neuropathic pain (or the co-occurrence of both types of pain), making it difficult to interpret the findings. Because nociceptive pain is likely to be the primary pain mechanism in most patients with CSDs [27], the results of these few studies likely have more relevance for nociceptive than neuropathic pain. The available evidence suggests a statistically significant, but not clinically significant, analgesic effect for TCAs and some SSRIs [28–30]. Although the

442

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

usefulness of SSRIs awaits further documentation, studies involving other chronic pain syndromes in which TCAs were compared with SSRIs showed TCAs to be more effective in every case [14]. Lack of research involving novel antidepressants makes it impossible to draw conclusions about these agents, although Desyrel has proven to be ineffective in a number of studies of various chronic pain syndromes [14]. Several groups of researchers have proposed that antidepressants that block peripheral 5-HT2a receptors (eg, Remeron, Elavil, doxepin) may have an analgesic effect through this mechanism [25,31]. They have also suggested that topical formulations of these types of antidepressants may be a useful alternative drug delivery system for analgesia (doxepin is available as a topical preparation [32]). However, to our knowledge, this hypothesis has not been addressed empirically, either with systemic or oral agents. As noted above, many researchers believe that, at least theoretically, antidepressants that have dual actions on both serotonin and norepinephrine should be more effective analgesics. Venlafaxine (Effexor) is a dual-action antidepressant that blocks the reuptake of serotonin at low doses and additionally blocks the reuptake of norepinephrine at higher doses (note: dopamine transmission may also be affected at very high doses). Effexor is of special interest for several reasons. First, it shares many similarities with the TCAs but lacks the most troublesome side effects. Furthermore, it is similar in structure to tramadol (Ultram), an analgesic with both opioid agonist and monoaminergic activity [14]. Although there are no published controlled trials with this agent, there are case reports, open trials and preclinical work supportive of the efficacy of Effexor for both nociceptive and neurogenic pain [14]. It should be noted that the SSRI Paxil was recently found to inhibit the reuptake of norepinephrine in addition to the reuptake of serotonin at a daily dose of 60 mg [33], suggesting that, at least theoretically, it may be a more effective analgesic than the other SSRIs. However, the relative analgesic efficacy of Paxil (versus the other SSRIs) has not been investigated. Neuropathic pain, or pain that arises from functional abnormalities or structural lesions in the peripheral or central nervous systems, plays a less prominent role in CSDs, in general, than it does in many other chronic pain syndromes (eg, diabetic neuropathy, carpal tunnel syndrome). However, CSD patients with radicular pain, failed back syndrome (associated with root sleeve fibrosis or arachnoiditis) or nerve root injury (lateral recess syndrome, spinal stenosis and arachnoiditis) typically have a significant neuropathic component to their pain [34,35]. Physicians generally rely on the descriptive nature of the pain complaint, neurological abnormalities on physical examination and diagnostic procedures to determine neuropathic involvement, although a validated instrument (Neuropathic Pain Scale) has been developed to aid in this process [36]. TCAs are effective in the treatment of chronic neuropathic pain [34,37,38]. Most patients experience pain relief in the range of 30 to100 mg/ day, although occasionally patients require doses from 150

to 250 mg/day [39]. Evidence for the effectiveness of SSRIs is generally lacking, with SSRIs no more effective than placebo in most studies. Desyrel was found to be ineffective in several studies [14] and Wellbutrin reduced neuropathic pain in a randomized control study of nondepressed patients [40]. Effexor and Paxil are likely to be effective in neuropathic pain because of the dual effect on serotonin and norepinephrine at higher doses [37,38]. Despite the impressive research support for TCAs, none of the reviewed studies involved patients with CSD. Therefore, although it is likely that TCAs are effective for neuropathic pain in patients with CSD, definitive proof awaits further study. Antidepressants have also been used as adjuncts to opioid analgesics in the treatment of chronic pain [9,41,42]. Patients on this combined regimen tend to tolerate a lower opioid maintenance dosage without dependence or abuse. Pain syndromes previously refractory to narcotics alone may be more responsive to combined therapy. Anticonvulsants Anticonvulsants are FDA approved for the use of seizure disorders and bipolar disorder, both of which are uncommon in the CSD population. However, these agents have also been found to be extremely useful for neuropathic pain, which may affect a significant proportion of patients with CSDs. Anticonvulsants may also be useful in ameliorating mood lability (ie, mood swings) and irritability/agitation in personality-disordered patients. Use of these agents to treat psychiatric conditions is best handled by a psychiatrist. The anticonvulsants that have been found to be most helpful in neuropathic pain, and their therapeutic dosages, are recorded in Table 4. Despite the strong research support for these agents in such chronic conditions as trigeminal neuralgia and diabetic neuropathy, no research has directly examined their impact on neuropathic pain associated with CSDs, such as radicular pain or failed back syndrome. Although anticonvulsants are likely to be effective in this population in light of their effectiveness in a broad array of diverse

Table 4 Dosages of anticonvulsants for chronic pain Medication, proprietary name (nonproprietary name)

Dose (mg/day)

Dilantin (diphenyldantoin) Tegretol (carbamazepine) Depakene (valproic acid) Klonopin (clonazepam) Neurontin (gabapentin) Topamax (topiramate) Keppra (levitiracetam) Gabatril (tiagabine) Trileptol (oxcarbazepine) Lamictal (lamotrogine) Felbatol (felbamate) Zonegram (zonisamide)

150–400 (average 300) 100–1,600 (average 400–600) 250–1,000 1.5–6.0 300–3,600 25–400 500–3,000 2–16 600–2,400 50–250 400–3,600 100–600

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

neuropathic conditions, definitive statements regarding their efficacy await further study. Anticonvulsants are a heterogenous group of medications in terms of mechanism of action, but the mechanisms underlying their anticonvulsant effect most likely contribute to their analgesic effect (eg, the pathophysiology of epilepsy and neuropathic pain may be similar) [43,44]. These mechanisms include inhibition of sodium channels (which reduces ectopic discharges involving fast sodium channels) and inhibition of spinal neurotransmission (which is believed to reduce nerve inflammation). The newer antiepileptic agents, including gabapentin (Neurontin), lamotrigine (Lamictal), and topiramate (Topamax) are rapidly becoming the initial agents of choice for neuropathic pain syndromes [45]. Neurontin has the most favorable side-effect profile as well as the most research support. It lacks significant drug–drug interactions and organ toxicity, has a mild side-effect profile and does not require blood levels [38,45]. A systematic review of the controlled literature involving non-CSD neuropathic syndromes demonstrated the analgesic superiority of Neurontin compared with placebo in four studies, superiority to Elavil in one study and equal efficacy compared with Elavil in the final study [46]. The favorable side-effect profile, the lack of significant drug–drug interactions (because it is not appreciably metabolized by the liver), the demonstrated efficacy for neuropathic pain conditions and new evidence supporting its utility for treating anxiety and insomnia [19] suggest that Neurontin be the first anticonvulsant considered. It should be noted that Neurontin can have side effects (eg, sedation, dizziness, ataxia, gastrointestinal distress), especially with rapid escalation of dose, resulting in a 13% dropout rate in the placebo-controlled studies (versus 6% dropout for placebo), suggesting slower titration than is commonly practiced [46]. Lamictal and Topamax have less favorable sideeffect profiles but have also been found to be clinically effective in uncontrolled studies [38,44]. Lamictal can precipitate a rash and must therefore be slowly titrated upward [40]. Newer and promising agents include levetiracetam (Keppra), tiagabine (Gabitril), oxcarbazepine (Trileptal), felbamate (Felbatol) and zonisamide (Zonegran) [47,48]. Carbamazepine (Tegretol) and diphenyldantoin (Dilantin) were until recently the first-line anticonvulsants used with neuropathic pain syndromes, either alone or adjunctively with antidepressants. Depakote has also been found to be useful in some clinical trials [49]. Klonopin, a benzodiazepine with anticonvulsant properties, may also be useful for neuropathic pain [38] and has a lower side-effect profile. With the traditional first-line anticonvulsants (Tegretol, Dilantin, Depakote), patients are started at a low dose and titrated upward at set intervals with monitoring of therapeutic response and/or emergence of side effects. Blood levels are checked periodically. The total daily dose depends on therapeutic response or intolerable side effects. Side effects can be severe and limit the usefulness of these agents. Central nervous system symptoms, such as nystagmus, ataxia,

443

slurred speech, confusion and drowsiness, are common and dose related. Gastrointestinal effects include nausea, vomiting and constipation. There is a risk of hepatitis or liver damage, particularly with Dilantin. The most severe side effects may be on the hematopoietic system and include anemia and bone marrow suppression. Therefore, patients need to be monitored very carefully with full medical evaluation before the initiation of therapy and interval assessment of blood count and liver functions. Klonopin, although more benign in its side-effect profile, does have the potential for dependence and addiction. Muscle relaxants The term “muscle relaxants” (see Table 5) is a misnomer, because these drugs have no peripheral action on “tight muscles,” but rather act on the central nervous system, as do the benzodiazepines, such as diazepam, which are also prescribed for “muscle spasm.” They are sedating, have addictive potential and a withdrawal syndrome, with very little therapeutic benefit for chronic pain [50], with tolerance quickly developing to the therapeutic properties [51]. They may, however, be useful in the early treatment of acute musculoskeletal pain [50,52]. Baclofen, however, has been found to be effective in relieving the pain of trigeminal neuralgia and may have usefulness in controlling other chronic neuropathic pain syndromes [53], primarily through central facilitation of gamma aminobutyric acid transmission. Baclofen is initiated at a low dose and titrated upward to a range of 50 to 60 mg/day, in divided doses. Side effects include drowsiness, dizziness, ataxia, confusion and epigastric distress. After prolonged use, it should be tapered slowly, to avoid hallucinations, anxiety and tachycardia. Carisoprodol (Soma) is metabolized to meprobamate (a barbiturate-like agent), has a significant association with dependence and abuse [54,55] and has been found to be associated with high rates of automobile accidents [56]. Metaxolone (Skelaxine) acts centrally by suppressing polysynaptic spinal cord reflexes and thereby reduces skeletal muscle spasm [52]. Tizanidine (Zanaflex) is a central alpha2-adrenergic agonist and has numerous pharmacologic properties, including reduction in the release of excitatory neurotransmitters and inhibition of spinal reflexes. It has been used to treat headache, musculosketal pain and neuropathic pain [57,58]. Table 5 Muscle relaxants Flexeril (cyclobenzaprine) Soma (carisoprodol) Lioresal (baclofen) Robaxin (methocarbamol) Skelaxine (metaxolone) Parafon forte (chlorzoxazone) Norflex (orphenadrine) Valium (diazepam) Zanaflex (tizanidine)

444

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

Antianxiety agents A number of pharmacological agents are useful to control core symptoms of anxiety. As discussed above, many of the antidepressants, particularly the SSRIs and Effexor, have become first-line treatments for anxiety disorders, including generalized anxiety disorder, panic disorder, posttraumatic stress disorder, social anxiety disorder and obsessive-compulsive disorder. The anticonvulsant Neurontin is also being increasingly used as an antianxiety agent [19]. These agents are being used more and more often, and the traditional antianxiety agents (anxiolytics), which will be discussed in this section, are playing a less prominent role. The traditional anxiolytics include the benzodiazepines, barbiturates, Inderal (propranolol), Equanil (meprobamate) and antihistamines. Inderal (propranolol), a β blocker, may be useful on an as needed basis for patients with prominent autonomic symptoms of anxiety (eg, tachycardia, sweating, tremor) and specific phobias [19]. In rehabilitation settings, we have found Inderal to be useful for autonomically aroused patients who avoid physical exercise because of fears of pain or reinjury. A newer agent, unrelated to the traditional anxiolytics, is Buspar (buspirone). This section will focus on the benzodiazepines. The other traditional anxiolytics are rarely used because of significant side effects, heightened risk of addiction or limited efficacy. Although Buspar has proven to be effective in treating generalized anxiety disorder (GAD), many experts believe that there are better agents (SSRIs and Effexor) [59,60]. Generalized anxiety disorder is so often comorbid with full-blown major depressive disorder or subsyndromal depression (up to 80% of patients by some estimates [59]) that, barring intolerable side effects or lack of efficacy, SSRIs or Effexor are better choices. The primary indications for the benzodiazepines in patients with CSDs are anxiety syndromes (particularly panic disorder) and insomnia, although they also have musclerelaxant and anticonvulsant properties [60]. The sleeppromoting qualities of these agents will be discussed in the next section on sedatives/hypnotics. With the exception of clonazepam (Klonopin), which is indicated for neuropathic pain, the issue of whether benzodiazepines are useful as adjuvant analgesics in chronic pain syndromes has not been addressed. It has been proposed, however, that these agents may have analgesic properties, either by reducing muscle tension or indirectly, by reducing emotional distress [61]. Benzodiazepines other than Klonopin are also likely to have a beneficial effect on neuropathic pain, but this has not been investigated. The benzodiazepines are used frequently in acute pain to decrease muscle spasm or to reduce anticipatory anxiety before a procedure. They depress central nervous system activity at the levels of the limbic system, brainstem reticular activating formation and cortex by binding to and facilitating the action of the inhibitory neurotransmitter, gamma aminobutyric acid [13,15]. A variety of benzodiazepines (see Table 6) are useful for the short-term treatment of anxiety. They

are best used over a period of a few weeks or months to control initial symptoms, in conjunction with other biobehavioral interventions, and then tapered slowly as the patient stabilizes with other PMs (ie, antidepressant medication) or nonpharmacological cognitive-behavioral techniques. Side effects, such as drowsiness and ataxia, are common. Less frequently seen are psychomotor impairment (including compromised driving skills), short-term memory loss, cognitive impairment and behavioral disinhibition. These agents do have addictive potential and should be used carefully in patients with a history of drug abuse [13,59,60]. However, although physical tolerance and withdrawal symptoms are common with regular use, the abuse of benzodiazepines, with the exception of patients with pre-existing substance abuse, is rare [19,60]. In general, the warning signs of misuse of benzodiazepines are similar to those for opioid medications: self-escalation of dose, repeated prescription loss, multiple prescribers, frequent telephone calls to the office, multiple drug intolerances described as “allergies,” focusing mainly on medication (ie, benzodiazepines, muscle relaxants or opioids) issues during visits and office visits without an appointment [62]. A withdrawal syndrome characterized by rebound anxiety, agitation, irritability, insomnia and muscle tension [63] requires a tapering dosage under medical supervision. Tapering can take an extended period of time, particularly for longer-acting or higher-potency (ie, more milligrams) benzodiazepines. Withdrawal symptoms generally improve significantly within 1 week but may take up to 3 weeks to fully resolve [19]. Xanax has been linked with “delayed withdrawal” in which irritability or increased anxiety may develop 3 or 4 weeks after the medication has been discontinued. Some experts discourage the use of Xanax for these and other reasons (eg, higher potential for abuse, need for frequent dosing) [64]. In practice, we generally reserve benzodiazepines for patients with anxiety that is acute, disabling because of accompanying severe emotional distress or unresponsive to other treatments. Unlike the antidepressants, these agents have their therapeutic effect almost immediately (minutes to hours). Consequently, they are often used on an as needed basis, with some patients using as few as one or two doses per month and others keeping the medication on hand for “emergencies” but rarely if ever taking it. Other patients, particularly those with panic disorder, may require regular use. Although some patients develop tolerance to the anxiolytic effects after weeks or months, other patients do not show a diminished therapeutic response over extended periods of time [19]. Most patients are able to decrease or discontinue (by tapering) use of benzodiazepines after 1 to 2 months of successful treatment with an antidepressant. Several studies have found that, despite early improvement in anxiety symptoms, the effects of benzodiazepines are not significantly different from placebo after 4 to 6 weeks of treatment [60]. Further, in one study directly comparing a benzodiazepine with an SSRI in generalized anxiety disorder, the onset of

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

445

Table 6 Anxiolytic benzodiazepines Medication, proprietary (nonproprietary)

Dose (mg/day)

Rapidity of effect

Duration of action*

Ativan (lorazepam) Serax (oxazepam) Xanax (alprazolam) Librium (chlordiazepoxide) Tranxene (chlorazepate) Valium (diazepam) Klonopin (clonazepam)

1–6 15–90 0.5–6 10–100 7.5–60 5–40 0.5–3

Intermediate Intermediate–slow Intermediate Intermediate Rapid Rapid Intermediate

Intermediate Intermediate Intermediate Long Long Long Intermediate–long

* Benzodiazepines with an intermediate duration of action require more frequent dosing, are associated with a more pronounced withdrawal syndrome and demonstrate greater potential for “breakthrough” anxiety before the next dose. One strategy for tapering patients off a benzodiazepine with an intermediate duration of action is to switch to an equivalent dose of a longer-acting agent before tapering. Note, however, that longer-acting benzodiazepines may cause a delayed withdrawal reaction (symptoms develop days to 1 week after decreasing or discontinuing medication).

action of Paxil was slower, but the reduction in anxiety symptoms was significantly greater after 4 weeks of treatment [65]. In contrast to benzodiazepines, which tend to become less effective over time, the efficacy of anxiolytic antidepressants increases over at least the first 2 months of treatment [59,60,64].

to cause morning hangover and withdrawal symptoms, although Ambien may lose its effectiveness with regular long-term use. Sonata is unique in several respects: it is not associated with the development of physical tolerance, and it may be taken in the middle of the night (desired wakeup time 4 or more hours later) because of its extremely short half-life [66].

Sleep-promoting medications Insomnia is an epidemic problem in patients with CSDs. It has been reported that 50% to 70% of patients with heterogenous chronic pain conditions report impaired sleep [66]. It has also been found that sleep disturbance is associated with higher pain intensity, greater levels of depression and anxiety and reduced activity levels [66]. The sedating antidepressants (amitriptyline, imipramine, doxepin, trimipramine, Paxil, Serzone, trazadone, Remeron) are effective sleep-promoting agents (also called sedatives or hypnotics). Other antidepressants are also likely to benefit sleep as they have a positive impact on depression, anxiety and pain, although with a slower onset of action. Neurontin is also used as a sleep-promoting agent [19], particularly in patients with prominent neuropathic pain complaints. Other PMs are primary sedatives, including several benzodiazepines, and several other classes of medication (see Table 7). Patients with chronic pain and insomnia, with or without depression, should first be treated with a sedating antidepressant, because these agents may mitigate the sleep-disruptive pain, as well as improve sleep patterns even if depression is not present. In chronic pain, primary sedatives should be reserved for those patients whose insomnia has been refractory to a trial of several antidepressants, or who are unable to tolerate any antidepressants secondary to side effects. The benzodiazepine sedatives do have potential for oversedation (“morning hangover”), addiction, dependence and withdrawal, and should be used cautiously, particularly in patients with risk factors for drug abuse. It should be noted that Benadryl and the sedating antidepressants may also cause a “morning hangover.” This phenomenon generally decreases or disappears over time. The newer omega-1 receptor agonists (Ambien and Sonata) are much less likely

Neuroleptics The neuroleptics are also called “major tranquilizers” or “antipsychotic agents.” The traditional neuroleptics are called “typical,” and the newer agents are called “atypical.” The primary indication for these medications is psychosis (delusions, hallucinations, loss of contact with reality) and

Table 7 Sedatives/hypnotics Class Benzodiazepines

Omega-1 receptor agonists Antihistamines Antidepressants†

Medication, proprietary (nonproprietary)

Dose (mg/day)

Dalmane (flurazepam) Restoril (temazepam) Halcion (triazolam)* Ambien (zolpidem) Sonata (zaleplon) Benadryl and others (diphenhydramine) (See section on antidepressants)

15–30 15–30 0.125–0.25 5–10 5–10 25–100

Not recommended Barbiturates Chloral derivatives Doriden (glutethimide) Noludar (methyprylon) Quaalude (methaqualone) Placidyl (ethchlorvyne) * Halcion has an extremely short half-life and therefore is less likely to cause morning “hangover.” However, it should be regarded as a last resort because of multiple adverse effects (severe amnesia, behavioral disinhibition, etc.). † Elavil and trazadone are often used exclusively for sleep. Other antidepressants that help with sleep include the tricyclic antidepressants, Serzone and Remeron.

446

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

symptoms associated with bipolar disorder and severe personality disorders (mania, agitation, paranoid ideation). Use of neuroleptics for treatment of these conditions should be handled by a psychiatrist. Nonpsychiatric physicians are probably most familiar with the typical neuroleptics, such as Haldol. The mechanism of action of these agents is dopaminergic antagonism, especially at postsynaptic D2 receptors. Typical neuroleptics are used less often, mostly because of common and potentially irreversible side effects (movement disorders, sedation, weight gain, cardiac toxicity and bone marrow suppression), as well as decreased efficacy compared with atypical neuroleptics. There is some indication of a modest analgesic effect with typical neuroleptics [15]. Tardive dyskinesia, which is an often-irreversible movement disorder, poses a significant risk with long-term continuous use of the typical neuroleptics. Consequently, use of these agents is not recommended. The atypical neuroleptics have the same indications as the traditional agents, with the benefit of a significantly more benign side-effect profile, which includes minimal potential to cause tardive dyskinesia [67]. Their mechanisms of action include, but are not limited to, antagonism of D2 receptors. Their more modest effect on these receptors accounts for the reduced side effects. The potential analgesic effect of atypical antipsychotics has not been explored empirically, although a recent series of case reports has described the effectiveness of olanzapine (Zyprexa) in three patients with chronic musculoskeletal pain [67]. Psychosis is rare in patients with CSDs but typically responds well to these medications. More often, atypical neuroleptics are used in lower doses to treat mild agitation, paranoid ideation and nonpsychotic disturbances of thought process (eg, tangential thinking). Although not indicated for the following conditions, we also use low doses of these agents for severe anxiety and insomnia that are not responsive to other treatments. Despite a comparably benign side-effect profile, atypical neuroleptics are associated with some side effects, most commonly weight gain and sedation (both less pronounced with Geodon), but also various movement disorders, elevated prolactin levels and QT prolongation (Geodon). The atypical neuroleptics, with the exception of Clozapine, are listed in Table 8. Clozapine is not recommended because it has been found to cause fatal agranulocytosis in rare cases. A new atypical neuroleptic, aripiprazole (Abilify) was released in 2003. This medication has a different mechanism of action (and consequently a different side-effect profile) than the other atypical neuroleptics. The utility of Abilify for CSD patients with CSDs is unknown at this time. Ultram (tramadol) Tramadol is a synthetic, centrally acting analgesic with an unusual mode of action. It has weak opioid activity, particularly at mu receptors, with additional inhibition of serotonin and norepinephrine reuptake, thereby giving it both opioid

Table 8 Atypical neuroleptics Medication, proprietary name (nonproprietary name)

Dose (mg/day)

Risperdal (respiridone) Seroquel (quetiapine) Zyprexa (olanzapine) Geodon (ziprasidone) Abilify (aripiprazole)

2–6 300–800 10–20 40–160 15–30

analgesic and adjuvant analgesic properties [68]. This medication has gained popularity because it is believed to be associated with lower risk for misuse and dependence and also because of the dual analgesic mechanisms of action [69]. No systematic studies have examined the use and effectiveness of Ultram exclusively in patients with CSDs. However, it is generally regarded as a less effective analgesic compared with the regular opioid medications, but more effective than the psychotropic medications with analgesic properties reviewed in this article. Possible side effects include dizziness, vertigo, headache, constipation, nausea, vomiting, pruritis, sedation, physical dependence and misuse. Recently, problems associated with Ultram use have been identified. The most serious is the serotonin syndrome, a serious and potentially life-threatening adverse event most often occurring as a result of concomitant treatment with at least two drugs that increase serotonin activity, including tramadol, MAOI antidepressants and other antidepressants (TCAs such as Elavil, SSRIs such as Paxil and Effexor) [70,71]. The serotonin syndrome generally develops quickly and comprises a triad of mental state changes (eg, disorientation, confusion), autonomic instability (eg, fever, shivering, diaphoresis, nausea, vomiting, diarrhea) and abnormal neurological conditions (hyperreflexia, myoclonus, muscular rigidity, tremor, ataxia). It may also be accompanied by agitation, mania and hallucinations. There are also reported cases of seizures [70] and death [72]. The opioid Demerol (meperidine) may also precipitate the serotonin syndrome under similar circumstances. Consequently, the concomitant use of Ultram (or Demerol) and antidepressants is not recommended without patient advisement and close monitoring. Because antidepressants are so efficacious in the CSD population, we recommend avoiding Ultram and Demerol in general. Adjuvant analgesics Several types of medications have been found to enhance the analgesic affect of the nonnarcotic analgesics and the opioids (see Table 9) in chronic pain states. The antidepressants, neuroleptics and anticonvulsants have primary application in chronic pain as described previously but may also be used as adjuvants. Caffeine has been used to enhance NSAID analgesia and shortens the time to onset of effect [73,74]. The usage of

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450 Table 9 Adjuvant analgesics With nonsteroidal anti-inflammatory drugs Caffeine With opioids Tricyclic antidepressants Neuroleptics Anticonvulsants Antihistamines (hydroxyzine) Amphetamines

antihistamines along with opioids in chronic pain states allows the use of a lower dose of narcotic and provides additional sedation [75]. Both of these categories of drugs are relatively nontoxic, with minimal side-effect profiles. Neuroleptics may have some enhancing effect on opioid analgesia but at the very least reduce the nausea commonly seen with use of the narcotics. Psychostimulants, such as dextroamphetamine and Ritalin, augment narcotic analgesia and also counteract sedation seen with the opioids. However, the risk of central nervous system toxicity and addiction is significant in these agents and must be monitored carefully. Discussion Physicians evaluating patients with CSDs for pharmacotherapeutic management need to take a complete medical history, including an inquiry of previous psychiatric or psychological treatment and prior response to medications. A comprehensive physical examination should be performed, including an assessment of general health, cardiovascular, pulmonary, gastrointestinal, genitourinary and neuromuscular function. Assessment of current psychological functioning, including the presence of psychiatric disorders, is also important, given the high rates of such disorders and their impact on treatment success. Although the nonpsychiatrist physician may not feel confident to perform a full mental health assessment when evaluating a patient with chronic pain, inquiry about current depression, anxiety and sleep disturbance is important. The Beck Depression Inventory [76] is frequently used to screen for depression in medical settings. A more promising screening instrument is the Primary Care Evaluation of Mental Disorders (PRIME-MD), which consists of a questionnaire followed by a brief structured interview by the physician or a trained staff member [77]. The PRIME-MD assesses somatization, depression, panic disorder, generalized anxiety disorder and alcohol abuse, which are the most common psychiatric disorders in patients with CSDs. It takes the patient 5 to 10 minutes to complete and about the same amount of time for the interviewer. The Patient Health Questionnaire is a patient self-report version of the PRIME-MD that screens for the same psychiatric disorders but does not require further interviewing of the patient [78]. It is recommended for routine use by physicians treating patients with CSDs.

447

Identification of a psychiatric disorder is only the first step in addressing it. Most nonpsychiatrist physicians treating patients with CSDs are generally not experts regarding psychotropic medications (PMs), although many are skilled with the use of PMs for adjuvant analgesic purposes. We recommend that such physicians who choose to use PMs limit their use to the treatment of pain, depression, anxiety, somatization and insomnia. When disorders such as bipolar disorder, a psychotic disorder (eg, schizophrenia) or severe personality disorders are suspected or identified, referral to a psychiatrist is advised. Indications for psychiatric consultation include 1) questions of, or management issues involved in, definitive psychiatric diagnosis, 2) management of psychiatric emergencies (suicidal/homicidal ideation or severe agitation), 3) assessment of failed prior medication strategies, 4) questions of compliance or medication “abuse” and 5) stabilization of regimens before surgery or definitive therapies. Routine referral is indicated for patients unresponsive to two medication trials and those with two or more psychiatric conditions [67]. Substance abuse or “addiction” is an additional indication for psychiatric consultation. Excessive use of substances, such as opioids and alcohol, can precipitate depressive and anxiety syndromes, as well as personality disorders. It is important to determine if these syndromes are independent of the substance use or are substance induced. In the first case, treatment of the syndromes should follow routine clinical practice, whereas in the latter case, treatment stability in substance use should be the first therapeutic step [79]. Most patients with CSDs are taking multiple medications, and the physician must be aware of possible drug interactions. In particular, many of the medications most commonly used with this population cause sedation, including opioids, muscle relaxants, benzodiazepines, most mood stabilizers and most antidepressants. Oversedation and excessive central nervous system depression are not uncommon. Psychiatric referral is recommended when three or more PMs (including opioids) are being used. Patient education is essential when prescribing PMs to patients with CSDs. How the information is presented, as well as the content of the information, should not be overlooked. Many patients resist psychiatric labels, both because of continued stigma regarding psychiatric illness and because they misinterpret the recommendation for PMs as the physician telling them that the pain is “in their head.” We are careful to emphasize that the pain is “real” and that the medication addresses “chemical imbalances” caused by the stress of a chronic medical condition. We also highlight the expected beneficial impact on nonpsychiatric symptoms (sleep, pain, irritability). It is also important to explain how often the medication should be taken, the transient nature of most side effects and the delayed therapeutic response (with most PMs). On a daily basis, we see patients who either stopped antidepressants after a few days because of side effects or ineffectiveness or are taking the antidepressants on an as-needed basis (ie, in an ineffective manner).

448

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

Many patients fear addiction, and it is important to address this issue by informing patients that these medications have no potential for addiction or (in the case of medications with the potential for addiction) that the risk of addiction is low if they do not have a history of substance abuse and take the medication as prescribed. Pharmacotherapy is an important treatment approach for patients with CSDs, but it should be recognized that behavioral and psychotherapeutic interventions are also effective interventions for patients with comorbid psychiatric disorders. Referral to a psychologist with training and experience in treating medical patients, especially chronic pain patients, should be used more often. Finally, the physician working with patients with CSDs should be aware that treatment is most effective in an interdisciplinary context, along with other biobehavioral interventions aimed at decreasing suffering and increasing functional and personal autonomy. This team effort requires interdisciplinary planning, communication, monitoring and feedback. To assume that medication alone will resolve the complex psychosocioeconomic stressors with which the CSD patient is attempting to cope is both naive and simplistic. Therefore, because the treatment of chronic pain is a collaborative process between health-care disciplines, it is important for not only physicians, but also allied health professionals to be knowledgeable about choices of medication to control pain, efficacy of specific drugs in various chronic pain syndromes, dosage ranges and side effects.

Future topics Existing research does not adequately address the effectiveness of PMs as either primary or adjuvant analgesics in patients with CSDs. Clearly, randomized controlled studies are needed. Potential subjects for such studies should be described in great detail and may have to broken up into subgroups (eg, nociceptive, neuropathic or mixed pain conditions). Ideally, there would be a washout period before initiating treatment. Subjects would be assigned to receive either opioids, NSAIDs or various PMs. After determining the maximal effect of opioids and NSAIDS, PMs would be added to determine if additional pain relief occurs and/or whether opioid doses could be reduced without loss of analgesic effect. Such studies may be difficult to conduct. Potential subjects might be unwilling to undergo a washout period or to give up a medication with percieved efficacy. A multicenter study would increase the subject pool and increase the ability of the researchers to generalize from the findings. Future developments in the pharmacologic treatment of patients with CSDs will be influenced by other processes. Medications initially developed for nonanalgesic purposes may be found to have an analgesic effect. Better understanding of the actual mechanisms involved in the pathogenesis of chronic pain may enable medical researchers to match a

particular pharmacologic agent with a known mechanism of action to a particular pain syndrome with a newly discovered pathogenic mechanism.

Conclusion with key points Pharmacologic treatment of chronic pain has as its goals analgesia and the relief of emotional distress, and uses medications at doses and for applications that may not always be recognized by the medical community. This is particularly true for the antidepressants, the anticonvulsants, anxiolytics and baclofen. Additionally, the use of opioids for the longterm treatment of chronic nonmalignant pain is still considered controversial. It is, therefore, important that clinicians treating patients with chronic pain be familiar with these pharmacological approaches and be prepared to educate colleagues who may challenge some of these prescriptions. Because of the clinical complexities of these syndromes, it is not uncommon to see patients on a number of different pharmacologic agents. Progressing from the most directly analgesic agent to psychotropics for control of emotional distress, the use of an NSAID, an opioid, an antidepressant and an anticonvulsant concomitantly is not uncommon. Optimal communication between treatment disciplines is essential and may be achieved only by frequent interdisciplinary staff meetings in which clinical and behavioral observations, medical assessment and treatment planning are shared responsibilities. Whenever possible, one physician should be in charge of all analgesic and psychotropic medication for a patient with chronic pain. If two or more physicians are collaborating in such a function, communication between them is essential to avoid redundancy or possible toxicity in the medication regimen.

References [1] Dersh J, Gatchel RJ, Polatin P. Chronic spinal disorders and psychopathology: research findings and theoretical considerations. Spine J 2001; 1(2):88–94. [2] Schofferman J. Long term opioid analgesic therapy for severe refractory lumbar spine pain. Clin J Pain 1999;15:136–40. [3] Portnoy RK. Current pharmacotherapy of chronic pain. J Pain Symptom Manage 2000;19(1):S16–20. [4] Gatchel RJ. Early development of physical and mental deconditioning in painful spinal disorders. In: Mayer TG, Mooney V, Gatchel RJ, editors. Contemporary conservative care for painful spinal disorders. Philadelphia, PA: Lea and Feibiger, 1991:278–89. [5] Gatchel RJ. Psychological disorders and chronic pain: cause and effect relationships. In: Gatchel RJ, Turk DC, editors. Psychological approaches to pain management: a practitioner’s handbook. New York: Guilford Publications, 1996:33–52. [6] Polatin PB, Kinney RK, Gatchel RJ, Lillo E, Mayer TG. Psychiatric illness chronic low back pain. Spine 1993;18:66–71. [7] Dersh J, Gatchel RJ, Polatin P, Mayer T. Prevalence of psychiatric disorders in patients with chronic work-related musculoskeletal pain disability. J Occup Environ Med 2002;44(5):459–68. [8] Banks SM, Kerns RD. Explaining high rates of depression in chronic pain: a diathesis-stress framework. Psychol Bull 1996;119:95–110.

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450 [9] Rockers DM, Fishman SM. Pain and affective disorders. In: Ballantyne J, Fishman SM, Abdi S, editors. The Massachussetts General Hospital handbook of pain management. 2nd ed. Philadelphia: Lippincott Williams and Wilkins, 2002:506–12. [10] Sullivan M, Katon W. Somatization: the path between distress and somatic symptoms. Am Pain Soc J 1993;2:141–9. [11] Barsky AJ, Klerman GL. Overview: hypochondriasis, bodily complaints, and somatic styles. Am J Psychiatry 1983;140:273–83. [12] Simon GE. Somatization and psychiatric disorders. In: Kirmayer LJ, Robbins JM, editors. Current concepts of somatization: research and clinical applications. Washington, DC: American Psychiatric Press, 1991:37–62. [13] Polatin PB. Integration of pharmacotherapy with psychological treatment. In: Gatchel RJ, Turk DC, editors. Psychological approaches to pain management: a practitioner’s handbook. New York: Guilford Publications, 1996:305–28. [14] Lynch ME. Antidepressants as analgesics: a review of randomized controlled studies. J Psychiatr Neurosci 2001;26(1):30–6. [15] Rockers DM, Fishman SM. Psychopharmacology for the pain specialist. In: Ballantyne J, Fishman SM, Abdi S, editors. The Massachussetts General Hospital handbook of pain management. 2nd ed. Philadelphia, PA: Lippincott Williams and Wilkins, 2002:138–52. [16] Piccinelli M, Simon G. Gender and cross-cultural differences in somatic symptoms associated with emotional distress: an international study in primary care. Psychol Med 1997;27:433–44. [17] Fishbain DA. Approaches to treatment decisions for psychiatric comorbidity in the management of the chronic pain patient. Med Clin North Am 1999;83:737–63. [18] O’Malley PG, Jackson JL, Santoro J, Tomkins G, Balden E, Kroenke K. Antidepressant therapy for unexplained symptoms and symptom syndromes. J Fam Pract 1999;48(12):980–90. [19] Schatzberg AF, Cole JO, DeBattista C. Manual of clinical psychopharmacology. 4th ed. Washington, DC: American Psychiatric Publishing, 2003. [20] American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association, 1994. [21] Slaby AE, Tancredi LR. Micropharmacology: treating disturbances of mood, thought, and behavior as specific neurotransmitter dysregulations rather than as clinical syndromes. Primary Psychiatry 2001; 8(4):28–33. [22] Atkinson J. Psychopharmacological agents in the treatment of pain syndromes. In: Tollison D, editor. Handbook of chronic pain management. Baltimore, MD: Williams and Wilkins, 1989:69–102. [23] Monks R. Psychotropic drugs. In: Wall P, Melzack R, editors. Textbook of pain. 3rd ed. New York: Churchill Livingston, 1994:963–89. [24] Onghena P, Van Houdenhove B. Antidepressant-induced analgesia in chronic non-malignant pain: a meta-analysis of 39 placebo-controlled studies. Pain 1992;40:205–19. [25] Sawynok J, Esser MJ, Reid AR. Antidepressants as analgesics: an overview of central and peripheral mechanisms of action. J Psychiatr Neurosci 2001;26(1):21–9. [26] Ansari A. The efficacy of newer antidepressants in the treatment of chronic pain: a review of current literature. Harvard Rev Psychiatry 2000;7(5):257–77. [27] Seaman DR, Cleveland C. Spinal pain syndromes: nociceptive, neuropathic, and psychological mechanisms. J Manipulative Physiol Ther 1999;22(7):458–72. [28] Moulin DE. Systemic drug treatment for chronic musculoskeletal pain. Clin J Pain 2001;17:S86–93. [29] Salerno SM, Browning R, Jackson JL. The effect of antidepressant treatment on chronic back pain: a meta-analysis. Arch Intern Med 2002;162:19–24. [30] Dickens C, Jayson M, Sutton C, Creed F. The relationship between pain and depression in a trial using paroxetine in sufferers of chronic low back pain. Psychosomatics 2000;41(6):490–9.

449

[31] Blier P, Abbot FV. Putative mechanisms of action of antidepressant drugs in affective and anxiety disorders and pain. J Psychiatr Neurosci 2001;26(1):37–43. [32] Sabroe RA, Kennedy CT, Archer CB. The effects of topical doxepin on responses to histamine, substance P, and prostaglandin E2 in human skin. Br J Dermatol 1997;137(3):386–90. [33] Gilmor ML, Owens MJ, Nemeroff CB. Inhibition of norepinephrine uptake in patients with major depression treated with paroxetine. Am J Psychiatry 2002;159(10):1702–10. [34] Watson CPN. The treatment of neuropathic pain: antidepressants and opioids. Clin J Pain 2000;16:S49–55. [35] Fine PG, Vallerand AH. Optimizing treatment of chronic pain with opioid therapy [monograph 1]. In: Appropriate opioid pharmacotherapy for chronic pain management [from the National Pain Education Council]. Deerfield, IL: Discovery International, 2002:11–43. [36] Galer BS, Jensen BP. Development and preliminary validation of a pain measure specific to neuropathic pain: the neuropathic pain scale. Neurology 1997;48(2):332–8. [37] Jenson MG. The role of selective serotonin reuptake inhibitors in the management of neuropathic pain. Pain Clin 2002;3:18–21. [38] Carter GT, Galer BS. Advances in the management of neuropathic pain. Phys Med Rehab Clin North Am 2001;12(2):447–59. [39] Markman JD, Oaklander AL. Neuropathic pain syndromes. In: Ballantyne J, Fishman SM, Abdi S, editors. The Massachussetts General Hospital handbook of pain management. 2nd ed. Philadelphia: Lippincott Williams and Wilkins, 2002:339–59. [40] Semenchuk MR, Sherman S, Davis B. Double-blind, randomized trial of buproprion SR for the treatment of neuropathic pain. Neurology 2001;57(9):1583–8. [41] France R, Houpt J, Ellinwood E. Therapeutic effects of antidepressants in chronic pain. Gen Hosp Psychiatry 1984;6:55–63. [42] Urban B, France R, Sternberger E. Long-term use of narcotic/antidepressant medication in the management of phantom limb pain. Pain 1986;24:191–6. [43] Cluff RS. Adjuvant treatments. In: Ballantyne J, Fishman SM, Abdi S, editors. The Massachussetts General Hospital handbook of pain management. 2nd ed. Philadelphia: Lippincott Williams and Wilkins, 2002:125–37. [44] Attal N. Chronic neuropathic pain: mechanisms and treatment. Clin J Pain 2000;16:S118–30. [45] Mao J, Chen L. Gabapentin in pain management. Anesth Analg 2000;91:680–7. [46] Mellegers MA, Furlan AD, Mailis A. Gabapentin for neuropathic pain; systemic review of controlled and uncontrolled literature. Clin J Pain 2001;17:284–95. [47] Tremont-Lukats I, Megeff C, Backonja MM. Anticonvulsants for neuropathic pain syndromes: mechanism of action and place in therapy. Drugs 2000;60(5):1029–52. [48] Ipponi A, Lamberti C, Medica A, Bartolini A, Malmberg-Aiello P. Tiagabine antinociception in rodents depends on GABAB receptor activation: parallel antinociception testing and medial thalamus GABA microdialysis. Euro J Pharm 1999;368:205–11. [49] Hering R, Kuritzky A. Sodium valproate in the prophylactic treatment of migraine: a double blind study versus placebo. Cephalalgia 1992; 12:81–4. [50] Schofferman J. The use of medications for low back pain. In: Cole AJ, Herring SA, editors. The low back pain handbook. 2nd ed. Philadelphia, PA: Hanley and Belfus, 2003:133–49. [51] Browning R, Jackson JL, O’Malley PG. Cyclobenzaprine and back pain: a meta-analysis. Arch Intern Med 2001;161(13):1613–20. [52] Gross L. Metaxalone: a review of clinical experience. J Neurol Orthop Med Surg 1998;18:76–9. [53] Steardo L, Leo A. Efficacy of baclofen in trigeminal neuralgia and some other painful conditions: a clinical trial. Eur Neurol 1984;23: 51–55. [54] Littrell R, Hayes L, Stillner V. Carisoprodol (Soma): a new cautious perspective on an old agent. South Med J 1987;86(7):753–6.

450

P.B. Polatin and J. Dersh / The Spine Journal 4 (2004) 436–450

[55] Bailey DN, Briggs JR. Carisoprodol: an unrecognized drug of abuse. Am J Clin Pathol 2002;117(3):396–400. [56] Logan BK, Case GA, Gordon AM. Carisoprodol, meprobomate, and driving impairment. J Forensic Sci 2000;45(3):619–23. [57] Berry H, Hutchinson D. Tizanidine and ibuprofen in acute low back pain: results of a double-blind multicentre study in general practice. J Int Med Res 1988;16:83–91. [58] Semenchuk M, Sherman S. Effectiveness of tizanidine in neuropathic pain: an open-label study. J Pain 2000;4:285–92. [59] Ballenger JC, Davidson JRT, Lecrubier Y, et al. Consensus statement on generalized anxiety disorder from the international consensus group on depression and anxiety. J Clin Psychiatry 2001;62(suppl 11):53–62. [60] Gorman JM. Treatment of generalized anxiety disorder. J Clin Psychiatry 2002;63(suppl 8):17–23. [61] France R, Krishnan K. Psychotropic drugs in chronic pain. In: France R, Krishnan K, editors. Chronic pain. Washington, DC: American Psychiatric Press, 1988:323–74. [62] Wilsey BL, Fishman SM. Chronic opioid therapy, drug abuse, and addiction. In: Ballantyne J, Fishman SM, Abdi S, editors. The Massachussetts General Hospital handbook of pain management. 2nd ed. Philadelphia, PA: Lippincott Williams and Wilkins, 2002:495–505. [63] Sussman N. Benzodiazepines: do they still have a clinical role in the treatment of panic disorder. Prim Psychiatry 2001;8:35–8. [64] Hilty DM, Hales RE, Servis ME, Nuovo J. Algorithms for the diagnosis and treatment of anxiety disorders in the primary care setting. Prim Psychiatry 2001;8:58–67. [65] Rocca P, Fonzo V, Scotta M. Paroxetine efficacy in the treatment of generalized anxiety disorder. Acta Psychiatr Scand 1997;95:444–50. [66] Wilson KG, Eriksson MY, D’Eon JL, Mikail SF, Emery PC. Major depression and insomnia in chronic pain. Clin J Pain 2002;18:77–83. [67] Lim RF, Hilty DM, Jerant AF. An algorithm for treating psychotic disorders in primary care. Prim Psychiatry 2001;8(8):68–72.

[68] Uppington J. Opioids. In: Ballantyne J, Fishman SM, Abdi S, editors. The Massachussetts General Hospital handbook of pain management. 2nd ed. Philadelphia, PA: Lippincott Williams and Wilkins, 2002:103–24. [69] Barkin RL, Barkin D. Pharmacological management of acute and chronic pain: focus on drug interactions and patient-specific pharmacotherapeutic selection. South Med J 2001;94(8):756–70. [70] Lange-Asschenfeldt C, Weighmann H, Hiemke C, Mann K. Serotonin syndrome as a result of fluoxetine in a patient with tramadol abuse. J Clin Psychopharmacol 2002;22(4):440–1. [71] Egberts AC, ter Borgh J, Brodie-Meijer CC. Serotonin syndrome attributed to tramadol addition to paroxetine therapy. Int Clin Psychopharmacol 1997;12:181–2. [72] Ripple MG, Pestaner JP, Levine BS, Smialek JE. Lethal combination of tramadol and multiple drugs affecting serotonin. Am J Forens Med Pathol 2000;21(4):370–4. [73] Laska E, Sunshine A. Caffeine as an analgesic adjuvant. JAMA 1984;251:1711–8. [74] Schactel B, Fillingim J. Caffeine as an analgesic adjuvant. A doubleblind study comparing aspirin with caffeine to aspirin and placebo in patients with sore throat. Arch Intern Med 1991;151:733–7. [75] Sunshine A, Olson N. Non-narcotic analgesics. In: Wall P, Melzack R, editors. Textbook of pain. New York: Churchill Livingston, 1994: 923–942. [76] Beck AT. Beck Depression Inventory. Philadelphia, PA: Center for Cognitive Therapy, 1978. [77] Spitzer RL, Williams JBW, Kroenke K, et al. Utility of a new procedure for diagnosing mental disorders in primary care: the PRIME-MD 1000 study. JAMA 1994;272:1749–56. [78] Spitzer RL, Kroenke K, Williams JBW. Validation and utility of a self-report version of PRIME-MD. JAMA 1999;282(18):1737–44. [79] Strain EC. Assessment and treatment of comorbid psychiatric disorders in opioid-dependent patients. Clin J Pain 2002;18:S14–27.