Current Perspectives on Pain in AIDS, Part 1

June 1, 2002
William Breitbart, MD

Oncology, ONCOLOGY Vol 16 No 6, Volume 16, Issue 6

As active participants in the care of patients with acquired immunodeficiency syndrome (AIDS), oncologists need to be aware of the many facets of pain management in this population. This two-part article, which will conclude in the July 2002 issue, describes the prevalence and types of pain syndromes encountered in patients with AIDS, and reviews the psychological and functional impact of pain as well as the barriers to adequate pain treatment in this group and others with human immunodeficiency virus (HIV)-related disease.

ABSTRACT: As active participants in the care of patients with acquired immunodeficiency syndrome (AIDS), oncologists need to be aware of the many facets of pain management in this population. This two-part article, which will conclude in the July 2002 issue, describes the prevalence and types of pain syndromes encountered in patients with AIDS, and reviews the psychological and functional impact of pain as well as the barriers to adequate pain treatment in this group and others with human immunodeficiency virus (HIV)-related disease. Finally, principles of pain management, with particular emphasis on controlling pain in HIV-infected patients with a history of substance abuse, are outlined. [ONCOLOGY 16:818-835, 2002]

Oncologists are actively involved in several aspects ofthe care of patients with acquired immunodeficiency syndrome (AIDS), and so theyshould be aware of important quality-of-life issues such as pain management inthis population. With the introduction of highly active antiretroviral therapies(ie, combination therapies including protease inhibitors), the face of the AIDSepidemic—particularly for those who can avail themselves of and tolerate thesenew therapies—is indeed changing. Even with advances in AIDS therapies, paincontinues to be an important palliative care issue for patients with humanimmunodeficiency virus (HIV)-associated disease. For example, as theepidemiology of the AIDS epidemic changes in the United States, the challenge ofmanaging pain in AIDS patients with a history of substance abuse is becoming anever-growing challenge.

Studies have documented that pain in individuals with HIV infection or AIDSis highly prevalent, varied in syndromal presentation, associated withsignificant psychological and functional morbidity, and alarmingly undertreated.[1-12]Responses from a self-referred sample of AIDS outpatients indicate that theseindividuals experience many distressing physical and psychological symptomsalong with a high level of distress.[13] Clearly, pain management needs to beintegrated more fully into the total care of patients with HIV-related disease.

This two-part article, which will conclude in the July 2002 issue, addressesvarious aspects of pain management in AIDS patients, from assessment totreatment (both pharmacotherapeutic and nonpharmacologic), with special focus onthe challenges of controlling pain in substance abusers with AIDS.

Prevalence of Pain in AIDS

Estimates of the prevalence of pain in HIV-infected individuals have rangedfrom 30% to more than 90%, with this frequency increasing as diseaseprogresses,[4-8,12,14-16] particularly in the latest stages of illness. Studiessuggest that approximately 30% of ambulatory HIV-infected patients in earlystages of HIV disease (pre-AIDS, or category A/B disease) experience clinicallysignificant pain, and as many as 56% have had episodic painful symptoms of lessclear clinical significance.[5,7,12]

In a prospective cross-sectional survey of 438 ambulatory AIDS patients inNew York City, 63% reported frequent or persistent pain of at least 2 weeks’duration at the time of assessment.[5] The prevalence of pain in this largesample increased significantly as HIV disease progressed, with 45% of AIDSpatients with category A3 disease reporting pain, compared with 55% of thosewith category B3 disease, and 67% of those with category C1-3 disease.Patients in this sample of ambulatory AIDS patients also were more likely toreport pain if they had other concurrent HIV-related symptoms (eg, fatigue,wasting), had received treatment for an AIDS-related opportunistic infection, orhad not been receiving antiretroviral medications (eg, zidovudine [AZT,Retrovir], didanosine [ddI, Videx], zalcitabine [ddC, Hivid], stavudine [d4t,Zerit]).

In a study of pain in hospitalized patients with AIDS in a public hospital inNew York City, over 50% required treatment for pain, which was the presentingcomplaint in 30% and the second most common presenting problem after fever.[8]In a French multicenter study, 62% of hospitalized patients with HIV disease hadclinically significant pain.[7] Schofferman and Brody[16] reported that 53% ofpatients with far-advanced AIDS cared for in a hospice setting had pain, whileKimball and McCormick[15] reported that up to 93% of AIDS patients in theirhospice experienced at least one 48-hour period of pain during the last 2 weeksof life.

Larue and colleagues[17] demonstrated that patients with AIDS being cared forby hospice at home had prevalence rates and intensity ratings for pain that werecomparable to, and even exceeded, those of cancer patients. Breitbart andcolleagues[4] reported that ambulatory AIDS patients in their New York Citysample reported a mean pain intensity of 5.4 (on the 0-10 numerical ratingscale of the Brief Pain Inventory) and a mean "pain at its worst" of7.4. In addition, as with pain prevalence, the intensity of pain experienced bypatients with HIV disease increases significantly as disease progresses. AIDSpatients with pain, like their counterparts with cancer pain, describe anaverage of 2.5 to 3 concurrent pains at any given time.[4,6]

Frich and Borgbjerg[18] concluded that the incidence of disturbing pain inAIDS is high, specifically in the extremities, gastrointestinal (GI) tract, andhead. In a study of 95 AIDS patients, the overall incidence of pain was 88%, and69% of the patients suffered from pain that interfered with daily activity to adegree described as moderate to severe.[18]

Among AIDS patients approaching the end of life, 93% report experiencing painand discomfort at some time during their final 2 weeks.[15] This percentage maybe even higher, given that some pain and discomfort is likely to gounrecognized. Most patients in the study by Kimball and McCormick experienced atleast one 48-hour period of pain and discomfort during the last 2 weeks of life;88% received some sort of opioid analgesia, with the majority experiencingrelief afterward.[15]

Overview of Pain Syndromes in HIV/AIDS

Pain syndromes encountered in AIDS patients are diverse in nature andetiology. As shown in Table 1, pain syndromes in HIV disease can be categorizedinto three types: (1) those directly related to HIV infection orconsequences of immunosuppression, (2) those due to AIDS therapies, and (3)those unrelated to AIDS or AIDS therapies.[2,3,6]

In studies to date, approximately 45% of pain syndromes are directly relatedto HIV infection or consequences of immunosuppression; 15% to 30% are due totherapies for HIV- or AIDS-related conditions, as well as diagnostic procedures;and the remaining 25% to 40% are unrelated to HIV or its therapies.[6] The mostcommon pain syndromes reported in studies of AIDS patients include painfulsensory peripheral neuropathy, pain due to extensive Kaposi’s sarcoma,headache, oral and pharyngeal pain, abdominal pain, chest pain, arthralgias andmyalgias, and painful dermatologic conditions.[5,6,8,10,12,14,16,17,19]

In a sample of 151 ambulatory AIDS patients who underwent a researchassessment including a clinical interview, neurologic examination, and review ofmedical records,[6] the most common pain diagnoses included headaches (46% ofpatients, 17% of all pains), joint pains (arthritis, arthralgias, etc: 31% ofpatients; 12% of pains), painful polyneuropathy (distal symmetricalpolyneuropathy: 28% of patients; 10% of pains), and muscle pains (myalgia,myositis: 27% of patients; 12% of pains). Other common pain diagnoses includedskin pain (Kaposi’s sarcoma, infections: 25% of patients; 30% of homosexualmales had pain from extensive Kaposi’s lesions), bone pain (20% of patients),abdominal pain (17%), chest pain (13%), and painful radiculopathy (12%).

Patients in this sample had a total of 405 pains (averaging 3 concurrentpains); 46% were diagnosed with neuropathic pain, 71% with somatic pain, 29%with visceral pain, and 46% with headache (classified separately because ofcontroversy as to pathophysiology). When classified by pain type (as opposed topatients), 25% of episodes were neuropathic, 44% were nociceptive-somatic, 14%were nociceptive-visceral, and 17% were idiopathic pains.

Study patients with lower CD4-positive cell counts were significantly morelikely to be diagnosed with polyneuropathy as well as headache. Hewitt andcolleagues[6] demonstrated that while pains of a neuropathic nature (eg,polyneuropathies, radiculopathies) certainly comprise a large proportion of thepain syndromes encountered in AIDS patients (see Table 2), pains of asomatic and/or visceral nature are also extremely common clinical problems.

Pain in Women With AIDS

Our group at Memorial Sloan-Kettering has reported on the experience of painin women with AIDS.[6,20] While preliminary in nature, our studies suggest thatwomen with HIV disease experience pain more frequently than men with HIV diseaseand report somewhat higher levels of pain intensity. This may be, in part, areflection of the fact that women with AIDS-related pain are twice as likely tobe undertreated for their pain compared to men.[4] Women with HIV disease haveunique pain syndromes of a gynecologic nature specifically related toopportunistic infections and cancers of the pelvis and genitourinary tract.[21]Moreover, women with AIDS were significantly more likely to be diagnosed withradiculopathy and headache in one survey.[6]

Pain in Children With AIDS

Children with HIV infection also experience pain.[22] HIV-related conditionsin children that are observed to cause pain include meningitis and sinusitis(headaches), otitis media, shingles, cellulitis and abscesses, severe Candidadermatitis, dental caries, intestinal infections such as Mycobacteriumavium-intracellulare and Cryptosporidium, hepatosplenomegaly, oral andesophageal candidiasis, and spasticity associated with encephalopathy thatcauses painful muscle spasms.

Impact of Pain on Quality of Life

In patients with HIV disease, pain has a profound negative impact on physicaland psychological functioning, as well as overall quality of life.[7,11] In astudy of the impact of pain on psychological functioning and quality of life inambulatory AIDS patients,[11] depression was significantly correlated with thepresence of pain. In addition to being significantly more distressed, depressed,and hopeless, individuals with pain were twice as likely to have suicidalideation (40%) as those without pain (20%). HIV-infected patients with pain weremore functionally impaired.[11] Such functional interference was highlycorrelated with levels of pain intensity and depression. Patients with pain weremore likely to be unemployed or disabled and reported less social support.

Larue and colleagues[7] reported that HIV-infected patients with painintensities greater than 5 (on a 0-10 numeric rating scale) reportedsignificantly poorer quality of life during the week preceding the survey thanpatients without pain. Pain intensity had an independent negative impact on HIVpatients’ quality of life, even after adjustment for treatment setting, stageof disease, fatigue, sadness, and depression. Singer and colleagues[12] alsoreported an association between the frequency of multiple pains, increaseddisability, and higher levels of depression. Psychological variables, such asthe amount of control people believe they have over pain, emotional associationsand memories of pain, fears of death, depression, anxiety, and hopelessness,contribute to the experience of pain in people with AIDS and can increasesuffering.[23,11]

Our group also reported that negative thoughts related to pain are associatedwith greater pain intensity, psychological distress, and disability inambulatory patients with AIDS.[24] In our study, AIDS patients who felt thatpain represented a progression of their HIV disease reported more intense painthan those who did not see pain as a threat.

Vogl and colleagues[13] assessed 504 ambulatory AIDS patients in order tomeasure symptom distress, physical and psychosocial functioning, and demographicand disease-related factors. As opposed to those who reported sexual contact asa means of HIV transmission, patients who reported transmission via IV drug useindicated higher levels of distress including distress about physicalsymptoms.[13] Furthermore, Vogl and colleagues[13] showed that both the numberof symptoms and symptom distress were highly associated with psychologicaldistress and poorer quality of life.

Management of Pain in AIDS

Assessment Issues

The initial step in pain management is a comprehensive assessment of painsymptoms. The health professional working in the AIDS setting must have aworking knowledge of the etiology and treatment of pain in AIDS. This wouldinclude an understanding of the different types of AIDS pain syndromes discussedabove, as well as a familiarity with the parameters of appropriate pharmacologictreatment. A close collaboration of the entire health-care team is optimal whenattempting to adequately manage pain in the AIDS patient.

A careful history and physical examination may disclose an identifiablesyndrome (eg, herpes zoster, bacterial infection, or neuropathy) that can betreated routinely.[25,26] A standard pain history[27,28] may provide valuableclues to the nature of the underlying process and, indeed, may disclose othertreatable disorders. A description of the qualitative features of the pain, itstime course, and any maneuvers that increase or decrease pain intensity shouldbe obtained.

Pain intensity (current, average, at best, at worst) should be assessed todetermine the need for weak vs potent analgesics and as a means of seriallyevaluating the effectiveness of ongoing treatment. Pain descriptors (eg,burning, shooting, dull, or sharp) will help determine the mechanism of pain(nociceptive-somatic, nociceptive-visceral, or neuropathic) and may suggest thelikelihood of response to various classes of traditional and adjuvant analgesics(eg, nonsteroidal anti-inflammatory drugs [NSAIDs], opioids, antidepressants,anticonvulsants, oral local anesthetics, or corticosteroids).[29-31]

Additionally, detailed medical, neurologic, and psychosocial assessments(including a history of substance use or abuse) must be conducted. Whenpossible, family members or partners should be interviewed and included in thepain management treatment plan. Patient and family education pamphlets on themanagement of HIV/AIDS pain are available.

During the assessment phase, pain should be aggressively treated while paincomplaints and psychosocial issues are subject to an ongoing process ofreevaluation.[27]

Pain Measurement

An important element in the assessment of pain is the concept that assessmentis continuous and needs to be repeated over the course of pain treatment. Theuse of readily available, simple, and clinically validated pain self-reportmeasures or tools, including the Brief Pain Inventory and the Memorial PainAssessment Card, can make this process simpler and more reliable. These toolscan facilitate the assessment of four basic aspects of pain experience in AIDSthat require ongoing evaluation: (1) pain intensity, (2) pain relief,(3) pain-related functional interference (eg, impact on mood state, generaland specific activities), and (4) monitoring of intervention effects.

Three commonly used self-report pain intensity assessment tools include (1) asimple descriptive pain intensity scale; (2) a 0-10 numeric scale; and (3) thevisual analog scale. A pain faces scale is used in children and in non-Englishspeaking or illiterate populations. The Memorial Pain Assessment Card[32] is ahelpful clinical tool that allows patients to report their pain experience. Thisinstrument consists of visual analog scales that measure pain intensity, painrelief, and mood. The Brief Pain Inventory[33] is another pain assessment toolthat has been widely used in cancer and AIDS pain research and clinicalsettings. This test has a useful pain interference subscale that assesses pain’sinterference in seven domains of quality of life and function. There are manyother pain assessment tools available for adults and children.

Multimodal Approach

Federal guidelines developed by the Agency for Health Care Policy andResearch (AHCPR) for the management of cancer pain[29] also address the issue ofmanagement of pain in AIDS, stating, "The principles of pain assessment andtreatment in the patient with HIV positive/AIDS are not fundamentally differentfrom those in the patient with cancer and should be followed for patients withHIV positive/AIDS." In contrast to pain in cancer, pain in HIV disease maymore commonly have an underlying treatable cause.[10]

Optimal management of pain in AIDS is multimodal and requires pharmacologic,psychotherapeutic, cognitive-behavioral, anesthetic, neurosurgical, andrehabilitative approaches. A multidimensional model of AIDS pain, whichrecognizes the interaction of cognitive, emotional, socioenvironmental, andnociceptive aspects of pain, suggests a model for multimodal intervention.

Pharmacotherapies for Pain in AIDS

The World Health Organization (WHO)[30] has devised guidelines for analgesicmanagement of cancer pain that the AHCPR has endorsed for the management of painrelated to cancer or AIDS.[29] These guidelines, also known widely as the WHOanalgesic ladder, have been well validated.[34] This approach advocatesselection of analgesics based on severity of pain, as well as on the type ofpain (ie, neuropathic vs nonneuropathic pain).

For mild to moderate pain, nonopioid analgesics such as NSAIDs andacetaminophen are recommended. For pain that is persistent and moderate tosevere in intensity, opioid analgesics of increasing potency (such as morphine)should be utilized. Adjuvant agents, such as laxatives and psychostimulants, areuseful in preventing and treating opioid side effects such as constipation orsedation, respectively. Adjuvant analgesic drugs, such as the antidepressantanalgesics, are suggested for considered use, along with opioids and NSAIDs, inall stages of the analgesic ladder (mild, moderate, or severe pain), but havetheir most important clinical application in the management of neuropathic pain.

Although not yet validated in AIDS, this WHO approach has been recommended bythe AHCPR and clinical authorities in pain management and AIDS.[8,10,12,16,29,35,36] Clinical reports describing the successful application of theprinciples of the WHO analgesic ladder to the management of pain in AIDS, withparticular emphasis on the use of opioids, have also recently appeared in theliterature.[9,15,16,37-40]

Nonopioid Analgesics

The nonopioid analgesics (Table 3) are prescribed principally for mild tomoderate pain or to augment the analgesic effects of opioid analgesics in thetreatment of severe pain. The use of NSAIDs in patients with AIDS must beaccompanied by heightened awareness of toxicity and adverse effects. NSAIDs arehighly protein-bound, and the free fraction of available drug is increased inAIDS patients who are cachectic, wasted, and hypoalbuminic, often resulting intoxicities and adverse effects. Patients with AIDS are frequently hypovolemic,on concurrent nephrotoxic drugs, and experiencing HIV nephropathy, and so are atincreased risk for renal toxicity related to NSAIDs. The antipyretic effects ofNSAIDs may also interfere with the early detection of infection in patients withAIDS.

Adverse Effects of NSAIDs

The major adverse effects associated withNSAIDs include gastric ulceration, renal failure, hepatic dysfunction, andbleeding. Patients should be informed of these symptoms, issued guaiac cardswith reagent, and taught to check their stool weekly. The nonacetylatedsalicylates, such as salsalate, sodium salicylate, and choline magnesiumsalicylate, theoretically produce fewer GI side effects and might be consideredin cases where GI distress is an issue. Agents for prophylaxis againstNSAID-associated GI symptoms include H2-antagonist drugs (cimetidine, 300 mgthree or four times daily, or ranitidine, 150 mg twice daily); misoprostol(Cytotec), 200 mg four times daily; omeprazole (Prilosec), 20 mg once daily; oran antacid.

NSAIDs affect kidney function and should be used with caution. These drugscan cause a decrease in glomerular filtration, acute and chronic renal failure,interstitial nephritis, papillary necrosis, and hyperkalemia.[41] In patientswith renal impairment, NSAIDs should be used with caution, since many (ie,ketoprofen, fenoprofen, naproxen, and carprofen) are highly dependent on renalfunction for clearance. The risk of renal dysfunction is greatest in patientswith advanced age, preexisting renal impairment, hypovolemia, concomitanttherapy with nephrotoxic drugs, and heart failure.

Prostaglandins modulate vascular tone, and their inhibition by NSAIDs cancause hypertension as well as interference with the pharmacologic control ofhypertension.[42] Caution should be used in patients receiving beta-adrenergicantagonists, diuretics, or angiotensin-converting enzyme inhibitors.

Many NSAIDs, including indomethacin and sulindac, undergo substantial biliaryexcretion. In patients with hepatic dysfunction, these drugs should be used withcaution. With the exception of the nonacetylated salicylates (eg, sodiumsalicylate, choline magnesium trisalicylate), NSAIDs inhibit plateletaggregation (an effect that is usually reversible, but irreversible withaspirin). NSAIDs should be used with extreme caution, or avoided altogether, inpatients who are thrombocytopenic or who have clotting impairment.

Opioid Analgesics

Opioid analgesics are the mainstay of pharmacotherapy of moderate to severepain in the patient with HIV disease (Table 4). Several reports describing thesafe and effective use of opioid drugs in the management of moderate to severepain in populations of patients with HIV disease (including patients with ahistory of injection drug use as their HIV transmission factor) have begun toappear in the literature.[15,37-39,40,43]

Kaplan and colleagues[43] conducted a multicenter study in which 44 patientswith moderate to severe AIDS-related pain were treated with sustained-releaseoral morphine in an open-label prospective study of patients treated for up to18 days. Pain intensity decreased by 65% in the patients who completed thestudy, quality of life was good in 80%, and therapy was acceptable in 96%. Inaddition, 92% of side effects were resolved, and total morphine dose remainedstable through the course of the study.

In a pilot study, Newshan and Lefkowitz[44] reported similar findings on theeffectiveness and safety of the transdermal fentanyl patch (Duragesic) in asmall sample of patient with AIDS-related pain. With transdermal fentanyl, painseverity scores decreased, mean pain relief scores increased, and dailyfunctioning measures improved significantly.[44] Furthermore, Newshan andLefkowitz[44] reported that transdermal fentanyl was effective for chronic painin chemically dependent and non-chemically dependent AIDS patients.

In persons with AIDS near the end of life, the use of opioid analgesiaremains common practice. The medical records of 185 adult AIDS patientsreceiving hospice care were reviewed by Kimball and McCormick.[15] Most patients(93%) experienced at least one 48-hour period of discomfort during the last2 weeks of life; the majority (88%) received some form of opioidanalgesia.[15] Of these patients, 62% experienced some relief of painthereafter.[15]

Key Principles of Opioid Use

Principles that are useful in guiding theappropriate use of opioid analgesics for pain[31,36,45] include the following:(1) choose an appropriate drug, (2) start with the lowest dose possible,(3) titrate the dose, (4) use "as needed" doses selectively, (5)use an appropriate route of administration, (6) be aware of equivalent analgesicdoses, (7) use a combination of opioid, nonopioid, and adjuvant drugs, (8) beaware of tolerance, and (9) understand physical and psychological dependence. Inchoosing the appropriate opioid analgesic for cancer pain, Portenoy[31]highlights the following important considerations: opioid class,"weak" vs "strong" opioids, pharmacokinetic characteristics,duration of analgesic effect, favorable prior response, and opioid side effects.

Opioid analgesics are divided into two classes, the agonists and theagonist-antagonists, based on their affinity to opioid receptors. Pentazocine(Talwin), butorphanol (Stadol), and nalbuphine are examples of opioid analgesicswith mixed agonist-antagonist properties. These drugs can reverse opioid effectsand precipitate an opioid withdrawal syndrome in patients who areopioid-tolerant or dependent. They are of limited use in the management ofchronic pain in AIDS.

Oxycodone (in combination with either aspirin or acetaminophen), hydrocodone,and codeine—the so-called "weaker" opioid analgesics—are indicatedfor use in step 2 of the WHO ladder for mild to moderate pain. More severe painis best managed with morphine or another of the stronger opioid analgesics, suchas hydromorphone, methadone, levorphanol, or fentanyl. As a single agent withoutaspirin or acetaminophen, oxycodone is considered a "stronger" opioidand is available in immediate- and sustained-release forms.

Routes of Administration

The oral route has often been described as thepreferred route of administration for opioid analgesics, from the perspectivesof both convenience and cost. However, the transdermal route of administrationhas gained rapid acceptance among clinicians and patients.

Patients with HIV infection are burdened with the task of taking 20 to 40tablets of medication per day and often need to follow complicated regimens, forexample, in which medication has to be taken on an empty stomach. In a study ofpatient-related barriers to pain management in AIDS patients,[4] the vastmajority prefered using a pain intervention that required a minimal number ofadditional pills (eg, sustained-release preparations of oral opioids) orinterventions that did not require taking pills at all (eg, a transdermal opioidsystem).

Immediate-release oral morphine or hydromorphone preparations require thatthe drug be taken every 3 to 4 hours. Longer-acting, sustained-release oralmorphine and oxycodone preparations provide 8 to 12 hours or more of analgesia,minimizing the number of daily doses required for the control of persistentpain. Rescue doses of an immediate-release, short-acting opioid are oftennecessary to supplement the use of sustained-release morphine or oxycodone,particularly during periods of titration or pain escalation.

The transdermal fentanyl patch system also has applications in the managementof severe pain in AIDS.[39,40] Each transdermal fentanyl patch contains a 48- to72-hour supply of fentanyl, which is absorbed from a depot in the skin. Levelsin the plasma rise slowly over 12 to 18 hours after patch placement. As withsustained-release morphine preparations, all patients on the fentanyl patchshould be provided with oral or parenteral rapidly acting short-duration opioidsto manage breakthrough pain. The transdermal system is convenient and canminimize the reminders of pain associated with repeated oral dosing ofanalgesics. In AIDS patients, the absorption of transdermal fentanyl could beincreased with fever, resulting in increased plasma levels and a shorterduration of analgesia from the patch.

It is important to note that opioids can be administered through a variety ofroutes—oral, rectal, transdermal, intravenous, subcutaneous, intraspinal, andeven intraventricular.[39] There are advantages and disadvantages, as well asindications for use of these various routes. Further discussion of alternativedelivery routes such as the intraspinal route are beyond the scope of thisreview. However, interested readers are directed to the AHCPR’s clinicalpractice guideline on the management of cancer pain, which is available free ofcharge through 1-800-4-CANCER.[29]

Opioid Side Effects

Although the opioids are extremely effectiveanalgesics, they commonly produce side effects. Nevertheless, these reactionscan be minimized if anticipated in advance (see Table 5).

Sedation is a common central nervous system side effect, especially duringthe initiation of treatment; it usually resolves after the patient has beenmaintained on a steady dosage. Persistent sedation can be alleviated with apsychostimulant, such as dextroamphetamine, pemoline, or methylphenidate. Allare prescribed in divided doses in early morning and at noon. Additionally,psychostimulants can improve depressed mood and enhance analgesia.[46-48]

Delirium, of an either agitated or somnolent variety, can also develop in apatient on opioid analgesics and is usually accompanied by attentional deficits,disorientation, and perceptual disturbances (visual hallucinations or, morecommonly, illusions). Myoclonus and asterixis are often early signs ofneurotoxicity that accompany the course of opioid-induced delirium. Meperidine,when administered chronically in patients with renal impairment, can lead todelirium due to accumulation of the neuroexcitatory metabolitenormeperidine.[49]

Opioid-induced delirium can be alleviated through any of three possiblestrategies: (1) lowering the dose of the opioid, (2) changing to a differentopioid, or (3) treating the delirium with low doses of neuroleptics, such ashaloperidol or olanzapine (Zyprexa). The third strategy is especially useful foragitation and clears the sensorium.[50] For agitated states, intravenoushaloperidol in doses starting at 1 to 2 mg is useful, with rapid escalation ofdose if no effect is noted.

Gastrointestinal side effects of opioid analgesics are also common. The mostprevalent are nausea, vomiting, and constipation.[31] Concomitant therapy withprochlorperazine for nausea is sometimes effective. Since all opioid analgesicsare not tolerated in the same manner, switching to another narcotic can behelpful if an antiemetic regimen fails to control nausea.

Constipation caused by narcotic effects on gut receptors is a problemfrequently encountered, and it tends to be responsive to the regular use ofsenna derivatives. A careful review of medications is imperative, sinceanticholinergic drugs such as the tricyclic antidepressants can worsenopioid-induced constipation and cause bowel obstruction.

Respiratory depression is a worrisome but rare side effect of the opioidanalgesics. Respiratory difficulties can almost always be avoided if two generalprinciples are adhered to: (1) start opioid analgesics in low doses inopioid-naive patients, and (2) be cognizant of relative potencies whenswitching opioid analgesics, routes of administration, or both.

Adjuvant Analgesics

Adjuvant analgesics are the third class of medications frequently prescribedfor the treatment of chronic pain; they also have important applications in themanagement of pain in AIDS (Table 6). Adjuvant analgesic drugs are used toenhance the analgesic efficacy of opioids, treat concurrent symptoms thatexacerbate pain, and provide independent analgesia. They may be used in allstages of the analgesic ladder. Commonly used adjuvant drugs includeantidepressants, neuroleptics, psychostimulants, anticonvulsants,corticosteroids, and oral anesthetics.[29,47,48,51]


The current literature supports the use ofantidepressants as adjuvant analgesic agents in the management of a wide varietyof chronic pain syndromes, including cancer pain, postherpetic neuralgia,diabetic neuropathy, fibromyalgia, headache, and low back pain.[52-57]

The analgesic mechanisms of these drugs include antidepressant activity,[53]potentiation or enhancement of opioid analgesia,[58-60] and direct analgesiceffects.[61] The leading hypothesis suggests that both serotonergic andnoradrenergic properties of the antidepressants are probably important and thatvariations among individuals in pain may reflect the status of theirneurotransmitter systems.[25] Other possible mechanisms of antidepressantanalgesic activity that have been proposed include adrenergic and serotoninreceptor effects,[62] adenosinergic effects, antihistaminic effects,[62] anddirect neuronal effects, such as inhibition of paroxysmal neuronal discharge anddecreasing sensitivity of adrenergic receptors on injured nerve sprouts.[63]

There is substantial evidence that the tricyclic antidepressants inparticular are analgesic and useful in the management of chronic neuropathic andnonneuropathic pain syndromes. Amitriptyline is the most extensively studiedtricyclic antidepressant. It has been proved effective as an analgesic in alarge number of clinical trials addressing a wide variety of chronic painsyndromes, including neuropathy, cancer pain, fibromyalgia, andothers.[25,53,54,64-66] Other tricyclics that have been shown to have efficacyas analgesics include imipramine,[67,68] desipramine,[26,69] nortriptyline,[70]clomipramine,[71,72] and doxepin.[73]

The heterocyclic and noncyclic antidepressant drugs, such as trazodone,mianserin, maprotiline, and the newer selective serotonin-reuptake inhibitors (SSRIs),fluoxetine (Prozac, Sarafem) and paroxetine (Paxil), may also be useful asadjuvant analgesics for chronic pain syndromes.[25,47,48,54,61,68,69,74-78]Fluoxetine, a potent antidepressant with specific serotonin-reuptake inhibitionactivity,[77] has been shown to have analgesic properties in experimental animalpain models[78] but failed to show analgesic effects in a clinical trial forneuropathy.[69] Several case reports suggest fluoxetine may be a useful adjuvantanalgesic in the management of headache and fibrositis.[79] A newer SSRI,paroxetine became the first antidepressant of this class shown to be a highlyeffective analgesic in a controlled trial for the treatment of diabeticneuropathy.[68]

Newer antidepressants such as sertraline (Zoloft), venlafaxine (Effexor), andnefazodone (Serzone) may also eventually prove to be clinically useful asadjuvant analgesics. Nefazodone, for instance, has been demonstrated topotentiate opioid analgesics in an animal model.[80]

Given the diversity of clinical syndromes in which the antidepressants havebeen demonstrated to be analgesic, trials of these drugs can be justified in thetreatment of virtually every type of chronic pain.[51] The established benefitof several of the antidepressants in patients with neuropathic pains,[65,68]however, suggests these drugs may be particularly useful in populations such ascancer and AIDS patients, who frequently have an underlying neuropathiccomponent to the pain(s).[51] Although studies of the analgesic efficacy ofthese drugs in HIV-related painful neuropathies have not yet been conducted,they are widely applied clinically using the model of diabetic and postherpeticneuropathies.

While antidepressant drugs are analgesic in both neuropathic andnonneuropathic pain models, they are most commonly used (as analgesics) incombination with opioid drugs, particularly for moderate-to-severe pain.Antidepressant adjuvant analgesics have their broadest application ascoanalgesics, potentiating the analgesic effects of opioid drugs.[29] Theopioid-sparing effects of antidepressant analgesics have been demonstrated innumerous trials, especially in cancer populations with neuropathic as well asnonneuropathic pain syndromes.[54,57]

The dose and time course of onset of analgesia for antidepressants when usedas analgesics appears to be similar to their use as antidepressants. There iscompelling evidence that the therapeutic analgesic effects of amitriptyline arecorrelated with serum levels, as are the drug’s antidepressant effects, andthat analgesic treatment failure is due to low serum levels.[65] Therefore, ahigh-dose amitriptyline regimen of 150 mg or higher is suggested.[65] The properanalgesic dose for paroxetine is likely in the 40- to 60-mg range, with themajor analgesic trial utilizing a fixed dose of 40 mg.[68] Anecdotal evidencesuggests that the debilitated medically ill (eg, cancer and AIDS patients) oftenrespond (in terms of depression or pain) to lower doses of an antidepressantthan are usually required in the physically healthy, probably because ofimpaired metabolism of these drugs.[47,48]

As to the time course of onset of analgesia, a biphasic process appears tooccur. Immediate or early analgesic effects occur within hours or days, andthese are probably mediated through inhibition of synaptic reuptake ofcatecholamines. In addition, there are later, longer-lasting analgesic effectsthat peak over a 2- to 4-week period; these are probably due to effects of theantidepressants on receptor function.[64,65]

Neuroleptics and Benzodiazepines

Neuroleptic drugs such asmethotrimeprazine, fluphenazine, haloperidol, and pimozide (Orap) may play arole as adjuvant analgesics in AIDS patients with pain.[70,81-83] However, theiruse must be weighed against what appears to be an increased sensitivity to theextrapyramidal side effects of these drugs in AIDS patients with neurologiccomplications.[84] Anxiolytics such as alprazolam and clonazepam may also beuseful as adjuvant analgesics, particularly in the management of neuropathicpains.[85-87]


Psychostimulants such as dextroamphetamine,methylphenidate, pemoline, and modafinil (Provigil) may be usefulantidepressants in patients with HIV infection or AIDS who are cognitivelyimpaired.[46,85] Psychostimulants also enhance the analgesic effects of theopioid drugs,[88] are useful in diminishing sedation secondary to narcoticanalgesics, and are potent adjuvant analgesics. In relatively low doses,psychostimulants stimulate appetite, promote a sense of well-being, and improvefeelings of weakness and fatigue in cancer patients.

Bruera et al[46] demonstrated that a regimen of methylphenidate, 10 mg withbreakfast and 5 mg with lunch, significantly decreased sedation and potentiatedthe effect of narcotics in patients with cancer pain. Methylphenidate has alsobeen demonstrated to improve functioning on a number neuropsychological tests,including tests of memory, speed, and concentration, in patients receivingcontinuous infusions of opioids for cancer pain.[46] Dextroamphetamine has alsobeen reported to have additive analgesic effects when used with morphine inpostoperative pain.[89]

Pemoline is a unique alternative psychostimulant that is chemically unrelatedto amphetamine but may have similar usefulness as an antidepressant and adjuvantanalgesic in AIDS patients.[48] Advantages of pemoline as a psychostimulant inAIDS pain patients include the lack of abuse potential, the lack of federalregulation through special triplicate prescriptions, the mild sympathomimeticeffects, and the fact that it comes in a chewable tablet form that can beabsorbed through the buccal mucosa and thus can be used by AIDS patients whohave difficulty swallowing or who have intestinal obstruction.

Clinically, pemoline is as effective as methylphenidate or dextroamphetaminein the treatment of depressive symptoms and in countering the sedating effectsof opioid analgesics, but there are no studies of the drug’s capacity topotentiate the analgesic properties of opioids. Pemoline should be used withcaution in patients with liver impairment, and liver function tests should bemonitored periodically with longer-term treatment. The Food and DrugAdministration suggests that patients sign an informed consent document, whichoutlines the potential liver toxicities of pemoline the drug is prescribed.

Modafinil, a novel psychostimulant that has shown efficacy in treatingexcessive daytime sleepiness associated with narcolepsy, has recentlydemonstrated potential for the treatment of depression and fatigue.[90] Thisagent needs further study; however, it appears to be a promising alternative toother psychostimulants in patients who cannot tolerate or have contraindicationsto the use of other stimulants. Modafinil has minimal cardiovascular effects,does not cause tolerance or dependence, has a low abuse potential, and does notrequire a special triplicate prescription.

Anticonvulsant Drugs

Selected anticonvulsant drugs appear to beanalgesic for the lancinating dysesthesias that characterize diverse types ofneuropathic pain.[51] Clinical experience also supports the use of these agentsin patients with paroxysmal neuropathic pains that may not be lancinating and,to a far lesser extent, in those with neuropathic pains characterized solely bycontinuous dysesthesias.

Although most practitioners prefer to begin with carbamazepine because of theextraordinarily good response rate observed in trigeminal neuralgia, this drugmust be used cautiously in AIDS patients with thrombocytopenia, those at riskfor marrow failure, and those whose blood counts must be monitored to determinedisease status. If carbamazepine is used, a complete blood count should beobtained prior to the start of therapy, after 2 and 4 weeks, and every 3 to 4months thereafter. A leukocyte count below 4,000/µL is usually considered to bea contraindication to treatment, and a decline to less than 3,000/µL or anabsolute neutrophil count of less than 1,500/µL during therapy should promptdiscontinuation of the drug.

Other anticonvulsant drugs, including phenytoin, clonazepam, and valproate,may be useful for managing neuropathic pain in AIDS patients.[51] In addition,several newer anticonvulsants have been used to treat neuropathic pain,particularly in patients with reflex sympathetic dystrophy. These drugs includegabapentin (Neurontin), lamotrigine (Lamictal), and felbamate (Felbatol). Ofthese newer anticonvulsants, anecdotal experience has been most favorable withgabapentin, which is now being widely used by pain specialists to treat varioustypes of neuropathic pain. Gabapentin has a relatively high degree of safety,with no known drug-drug interactions and a lack of hepatic metabolism.[51]Treatment with gabapentin is usually initiated at a dose of 300 mg/d and thengradually increased to a range of 900 to 3,200 mg/d in three divided doses.


Corticosteroid drugs have analgesic potential in avariety of chronic pain syndromes, including neuropathic pains and painsyndromes resulting from inflammatory processes.[51] Like other adjuvantanalgesics, corticosteroids are usually added to an opioid regimen. In patientswith advanced disease, these drugs may also improve appetite, nausea, malaise,and overall quality of life. Adverse effects include neuropsychiatric syndromes,gastrointestinal disturbances, and immunosuppression.


Baclofen is a gamma-aminobutyric acid (GABA) agonist withproven efficacy in the treatment of trigeminal neuralgia.[91] On this basis, atrial of the drug is commonly used in the management of paroxysmal neuropathicpains of any type. Dosing is generally undertaken in a manner similar to how thedrug is used for its primary indication, spasticity. A starting dose of 5 mg twoto three times per day is gradually escalated to 30 to 90 mg/d, andsometimes higher if side effects do not occur. The most common adverse effectsare sedation and confusion.

Oral Local Anesthetics

Local anesthetic drugs may be useful in themanagement of neuropathic pains characterized by either continuous orlancinating dysesthesias. Controlled trials have demonstrated the efficacy oftocainide (Tonocard)[92] and mexiletine (Mexitil), and there is clinicalevidence that suggests similar effects from flecainide (Tambocor)[93] andsubcutaneous lidocaine.[94] It is reasonable to undertake a trial with orallocal anesthetic in patients with continuous dysesthesias who fail to respondadequately to, or who cannot tolerate the tricyclic antidepressants, and inpatients with lancinating pains refractory to trials of anticonvulsant drugs andbaclofen. Mexiletine is preferred in the United States.[51]

Paice and colleagues[95] studied 26 subjects in order to test the efficacy oftopical capsaicin in the management of HIV-associated pain. Results suggest thatcapsaicin is ineffective in relieving pain with HIV-associated distalsymmetrical peripheral neuropathy (DSPN). However, capsaicin has been shown tobe effective in relieving pain associated with other neuropathic painsyndromes.[95]

Analgesic/Anti-HIV Drug Interactions

Many available anti-HIV drugs may interact with other medications prescribedfor pain, depression, anxiety, or other medical conditions. These druginteractions can be dangerous, resulting in drug toxicities due to elevatedlevels of medication, or drug ineffectiveness caused by the lowering of druglevels in the serum.

Opioid analgesics can interact with certain anti-HIV drug therapies and theseinteractions should be kept in mind when prescribing opioids. For example, theprotease inhibitor ritonavir (Norvir) can increase the levels of several opioiddrugs, including codeine, hydrocodone, oxycodone, methadone, and fentanyl.Patients on ritonavir should not be prescribed meperidine or propoxyphenebecause of increased risk of serious toxicity.

Antidepressant and anticonvulsant analgesics may also interact withritonavir, which can increase serum levels of bupropion (Wellbutrin, Zyban),fluoxetine, trazodone, and desipramine, resulting in increased drug toxicities(eg, seizures with bupropion). Both ritonavir and saquinavir (Fortovase,Invirase) may increase levels of anticonvulsants such as phenobarbital,phenytoin, carbamazepine, and clonazepam.


1. Breitbart W: Pharmacotherapy of pain in AIDS, in Wormser GP (ed): AClinical Guide to AIDS and HIV, pp 359-378. Philadelphia, Lippincott-Raven,1996.

2. Breitbart W, Rosenfeld B, Passik S, et al: A comparison of pain report andadequacy of analgesic therapy in ambulatory AIDS patients with and without ahistory of substance abuse. Pain 72:235-243, 1997.

3. Breitbart W: Pain in AIDS, in Jensen J, Turner J, Wiesenfeld-Hallin Z (eds):Proceedings of the 8th World Congress on Pain, Progress in Pain Research andManagement, vol 8, pp 63-100. Seattle, IASP Press, 1997.

4. Breitbart W, Rosenfeld B, Passik S, et al: The undertreatment of pain inambulatory AIDS patients. Pain 65:239-245, 1996.

5. Breitbart W, McDonald MV, Rosenfeld B, et al: Pain in ambulatory AIDSpatients. I: Pain characteristics and medical correlates. Pain 68:315-321, 1996.

6. Hewitt D, McDonald M, Portenoy R, et al: Pain syndromes and etiologies inambulatory AIDS patients. Pain 70:117-123, 1997.

7. Larue F, Fontaine A, Colleau S: Underestimation and undertreatment of painin HIV disease: Multicentre study. Br Med J 314:23-28, 1997.

8. Lebovits AK, Lefkowitz M, McCarthy D, et al: The prevalence and managementof pain in patients with AIDS. A review of 134 cases. Clin J Pain 5:245-248,1989.

9. McCormack JP, Li R, Zarowny D, et al: Inadequate treatment of pain inambulatory HIV patients. Clin J Pain 9:247-283, 1993.

10. O’Neill WM, Sherrard JS: Pain in human immunodeficiency virus disease:A review. Pain 54:3-14, 1993.

11. Rosenfeld B, Breitbart W, McDonald MV, et al: Pain in ambulatory AIDSpatients. II: Impact of pain on psychological functioning and quality of life.Pain 68:323-328, 1996.

12. Singer EJ, Zorilla C, Fahy-Chandon B, et al: Painful symptoms reportedfor ambulatory HIV-infected men in a longitudinal study. Pain 54:15-19, 1993.

13. Vogl D, Rosenfeld B, Breitbart W, et al: Symptom prevalence,characteristics, and distress in AIDS outpatients. J Pain Symptom Manage18:253-262, 1999.

14. Breitbart W, Passik S, Bronaugh T, et al: Pain in the ambulatory AIDSpatient: Prevalence and psychosocial correlates (abstract). 38th Annual Meetingof the Academy of Psychosomatic Medicine, Atlanta, October 17-20, 1991.

15. Kimball LR, McCormick WC: The pharmacologic management of pain anddiscomfort in persons with AIDS near the end of life: Use of opioid analgesia inthe hospice setting. J Pain Symptom Manage 11:88-94, 1996.

16. Schofferman J, Brody R: Pain in far advanced AIDS, in Foley KM, BonicaJJ, Ventrafridda V (eds): Advances in Pain Research and Therapy, vol 16, pp379-386. New York, Raven Press, 1990.

17. Larue F, Brasseur L, Musseault P, et al: Pain and HIV infection: A Frenchnational survey (abstract). J Palliat Care 10:95, 1994.

18. Frich LM, Borgbjerg FM: Pain and pain treatment in AIDS patients: Alongitudinal study. J Pain Symptom Manage 19:339-347, 2000.

19. Penfold R, Clark AJM: Pain syndromes in HIV infection. Can J Anaesth39:724-730, 1992.

20. Breitbart W, McDonald M, Rosenfeld B, et al: Pain in ambulatory AIDSpatients. I: Pain characteristics and medical correlates. Pain 68:315-321, 1996.

21. Marte C, Allen M: HIV-related gynecologic conditions: Overlookedcomplications. Focus: A Guide to AIDS Research and Counseling 7:1-3, 1991.

22. Strafford M, Cahill C, Schwartz T, et al: Recognition and treatment ofpain in pediatric patients with AIDS (abstract). J Pain Symptom Manage 6:146,1991.

23. Breitbart W: Suicide risk and pain in cancer and AIDS patients, inChapman R, Foley KM (eds): Current Emerging Issues in Cancer Pain: Research andPractice, pp 49-65. New York, Raven Press, 1993.

24. Payne D, Jacobsen P, Breitbart W, et al: Negative thoughts related topain are associated with greater pain, distress and disability in AIDS pain(abstract). 13th Annual Meeting of the American Pain Society, Miami, November10-13, 1994.

25. Watson CP, Chipman M, Reed K, et al: Amitriptyline vs maprotiline in postherpetic neuralgia: A randomized double-blind, cross-over trial. Pain 48:29-36,1992.

26. Kishore-Kumar R, Max MB, Scafer SC, et al: Desipramine relievespost-herpetic neuralgia. Clin Pharmacol Ther 47:305-312, 1990.

27. Portenoy R, Foley KM: Management of cancer pain, in Holland JC, RowlandJH (eds): Handbook of Psychooncology, pp 369-382. New York, Oxford UniversityPress, 1989.

28. Foley KM: The treatment of cancer pain. N Engl J Med 313:84-95, 1985.

29. Jacox A, Carr D, Payne R, et al: Clinical Practice Guideline Number 9:Management of Cancer Pain. US Department of Health and Human Services, PublicHealth Service, Agency for Health Care Policy and Research, AHCPR publication94-0592, pp 139-141, 1994.

30. World Health Organization: Cancer Pain Relief. Geneva, World HealthOrganization, 1986.

31. Portenoy RK: Pharmacologic approaches to the control of cancer pain. JPsychosocial Oncol 8:75-107, 1990.

32. Fishman B, Pasternak S, Wallenstein SL, et al: The Memorial PainAssessment Card: A valid instrument for the evaluation of cancer pain. Cancer60:1151-1158, 1987.

33. Daut RL, Cleeland CS, Flanery RC: Development of the Wisconsin Brief PainQuestionnaire to assess pain in cancer and other diseases. Pain 117:197-210,1983.

34. Ventrafridda V, Caraceni A, Gamba A: Field testing of the WHO Guidelinesfor Cancer Pain Relief: Summary report of demonstration projects, in Foley KM,Bonica JJ, Ventrafridda V (eds): Advances in Pain Research and Therapy, vol 16,pp 155-165. New York, Raven Press, 1990.

35. Lefkowitz M, Breitbart W: Chronic pain and AIDS, in Wiener RH (eds):Innovations in Pain Medicine 36:2-3,18. 1992.

36. American Pain Society: Principles of Analgesic Use in the Treatment ofAcute Pain and Cancer Pain, 3rd ed. Skokie, Ill, American Pain Society, 1992.

37. Newshan G, Wainapel S: Pain characteristics and their management inpersons with AIDS. J Assoc Nurses AIDS Care 4(2):53-59, 1993.

38. Anand A, Carmosino L, Glatt AE: Evaluation of recalcitrant pain inHIV-infected hospitalized patients. J Acquir Immune Defic Syndr 7:52-56, 1994.

39. Patt RB, Reddy SR: Pain and the opioid analgesics: Alternate routes ofadministration. PAACNOTES 5:453-458, 1993.

40. Lefkowitz M, Newshan G: An evaluation of the use of Duragesic for chronicpain in patients with AIDS (abstract). 16th Annual Meeting of the American PainSociety, New Orleans, November 5-8, 1997.

41. Murray MD, Brater DC: Adverse effects of nonsteroidal anti-inflammatorydrugs on renal function. Ann Intern Med 112:559-560, 1990.

42. Radack K, Deck C: Do nonsteroidal anti-inflammatory drugs interfere withblood pressure control in hypertensive patients? J Gen Intern Med 2:108-112,1987.

43. Kaplan R, Conant M, Cundiff D, et al: Sustained-release morphine sulfatein the management of pain associated with acquired immune deficiency syndrome. JPain Symptom Manage 12:150-160, 1996.

44. Newshan G, Lefkowitz M: Transdermal fentanyl for chronic pain in AIDS: Apilot study. J Pain Symptom Manage 21: 69-77, 2001.

45. Foley KM, Inturrisi CE: Analgesic drug therapy in cancer pain: Principlesand practice, in Payne R, Foley KM (eds): Cancer Pain Medical Clinics of NorthAmerica, pp 207-232. Philadelphia, WB Saunders, 1987.

46. Bruera E, Chadwick S, Brennels C, et al: Methylphenidate associated withnarcotics for the treatment of cancer pain. Cancer Treat Rep 71:67-70, 1987.

47. Breitbart W: Psychotropic adjuvant analgesics for cancer pain.Psychooncology 7:133-145, 1992.

48. Breitbart W, Mermelstein H: Pemoline: An alternative psychostimulant inthe management of depressive disorder in cancer patients. Psychosomatics 33:352-356, 1992.

49. Kaiko R, Foley K, Grabinski P, et al: Central nervous system excitationeffects of meperidine in cancer patients. Ann Neurol 13:180-183, 1983.

50. Breitbart W: Psychiatric management of cancer pain. Cancer 63:2336-2342,1989.

51. Portenoy RK: Adjuvant analgesics in pain management, in Doyle D, HanksGWC, MacDonald N (eds): Oxford Textbook of Palliative Medicine, 2nd ed, pp361-390. New York, Oxford University Press, 1998.

52. Butler S: Present status of tricyclic antidepressants in chronic paintherapy, in Benedetti C, Chapman CR, Moricca G (eds): Advances in Pain Researchand Therapy, vol 7, pp 173-196. New York, Raven Press, 1986.

53. France RD: The future for antidepressants: Treatment of pain.Psychopathology 20:99-113, 1987.

54. Ventafridda V, Bonezzi C, Caraceni A, et al: Antidepressants for cancerpain and other painful syndromes with deafferentation component: Comparison ofamitriptyline and trazodone. Ital J Neurol Sci 8:579-587, 1987.

55. Getto CJ, Sorkness CA, Howell T: Issues in drug management. Part I.Antidepressants and chronic nonmalignant pain: A review. J Pain Symptom Manage2(1):9-18, 1987.

56. Magni G, Arsie D, Deleo D: Antidepressants in the treatment of cancerpain: A survey in Italy. Pain 29:347-353, 1987.

57. Walsh TD: Controlled study of imipramine and morphine in chronic pain dueto advanced cancer. In Foley KM, Bonica JJ, Ventafridda V (eds): Advances inPain Research and Therapy, vol 16, pp 155-165. New York, Raven Press, 1986.

58. Botney M, Fields HC: Amitriptyline potentiates morphine analgesia bydirect action on the central nervous system. Ann Neurol 13:160-164, 1983.

59. Malseed RT, Goldstein FJ: Enhancement of morphine analgesics by tricyclicantidepressants. Neuropharmacology 18:827-829, 1979.

60. Ventafridda V, Branchi M, Ripamonti C, et al: Studies on the effects ofantidepressant drugs on the antinociceptive action of morphine and on plasmamorphine in rat and man. Pain 43:155-162, 1990.

61. Spiegel K, Kalb R, Pasternak GW: Analgesic activity of tricyclicantidepressants. Ann Neurol 13:462-465, 1983.

62. Gram LF: Receptors, pharmacokinetics and clinical effects, in Burrows GD,Normal T, Davis B (eds): Antidepressants, pp 81-95. Amsterdam, Elsevier, 1983.

63. Devor M. Nerve pathophysiology and mechanisms of pain in causalgia. JAuton Nerv Syst 7:371-384, 1983.

64. Pilowsky I, Hallet EC, Bassett EL, et al: A controlled study ofamitriptyline in the treatment of chronic pain. Pain 14:169-179, 1982.

65. Max MB, Culnane M, Schafer SC, et al: Amitriptyline relieves diabeticneuropathy pain in patients with normal and depressed mood. Neurology37:589-596, 1987.

66. Sharav Y, Singer E, Dione RA, et al: The analgesic effect ofamitriptyline on chronic facial pain. Pain 31:199-209, 1987.

67. Young RJ, Clarke BF: Pain relief in diabetic neuropathy: Theeffectiveness of imipramine and related drugs. Diabetic Med 2:363-366, 1985.

68. Sindrup SH, Gram LF, Brosen K, et al: The selective serotonin reuptakeinhibitor paroxetine is effective in the treatment of diabetic neuropathysymptoms. Pain 42:135-144, 1990.

69. Max MB. Effects of desipramine, amitriptyline, and fluoxetine on pain anddiabetic neuropathy. N Engl J Med 326:1250-1256, 1992.

70. Gomez-Perez FJ, Rull JA, Dies H, et al: Nortriptyline and fluphenazine inthe symptomatic treatment of diabetic neuropathy: A double-blind cross-overstudy. Pain 23:395-400, 1985.

71. Langohr HD, Stohr M, Petruch F. An open and double-blind crossover studyon the efficacy of clomipramine (Anafranil) in patients with painful mono- andpolyneuropathies. Eur Neurol 21:309-317, 1982.

72. Tiegno M, Pagnoni B, Calmi A, et al: Chlorimipramine compared topentazocine as a unique treatment in post-operative pain. Int J Clin PharmacolRes 7:141-143, 1987.

73. Hammeroff SR, Cork RC, Scherer K, et al: Doxepin effects on chronic pain,depression and plasma opioids. J Clin Psychiatry 2:22-26, 1982.

74. Davidoff G, Guarracini M, Roth E, et al: Trazodone hydrochloride in thetreatment of dysesthetic pain in traumatic myelopathy: A randomized,double-blind. placebo-controlled study. Pain 29:151-161, 1987.

75. Costa D, Mogos I, Toma T: Efficacy and safety of mianserin in thetreatment of depression of woman with cancer. Acta Psychiatr Scand72:85-92,1985.

76. Eberhard G, von Khorring L, Nilsson HL, et al: A double-blind randomizedstudy of clomipramine vs maprotiline in patients with idiopathic pain syndromes.Neuropsychobiology, 19:25-32, 1988.

77. Feighner JP: A comparative trial of fluoxetine and amitriptyline inpatients with major depressive disorder. J Clin Psychiatry 46:369-372, 1985.

78. Hynes MD, Lochner MA, Bemis K., et al: Fluoxetine, a selective inhibitorof serotonin uptake, potentiates morphine analgesia without altering itsdiscriminative stimulus properties or affinity for opioid receptors. Life Sci36:2317-2323, 1985.

79. Geller SA: Treatment of fibrositis with fluoxetine hydrochloride(Prozac). Am J Med 87:594-595, 1989.

80. Pick CG, Paul D, Eison MS, et al: Potentiation of opioid analgesia by theantidepressant nefazodone. Eur J Pharmacol 2:375-381, 1992.

81. Beaver WT, Wallerstein SL, Houde RW, et al: A comparison of the analgesiceffect methotrimeprazine and morphine in patients with cancer. Clin PharmacolTher 7:436-466, 1966.

82. Maltbie AA, Cavenar SO, Sullivan JL, et al: Analgesia and haloperidol: Ahypothesis. J Clin Psychiatry 40:323-326, 1979.

83. Lechin F, Vander Dijs B, Lechin ME, et al: Pimozide therapy fortrigeminal neuralgia. Arch Neurol 9:960-964, 1989.

84. Breitbart W, Marotta RF, Call P: AIDS and neuroleptic malignant syndrome.Lancet 2:1488, 1988.

85. Fernandez F, Levy JK: Psychiatric diagnosis and pharmacotherapy ofpatients with HIV infection, in Tasman A, Goldfinger SM, Kaufman (eds): Reviewof Psychiatry, vol 9, p 614. Washington, DC, American Psychiatric Press, 1990.

86. Swerdlow M, Cundhill JG: Anticonvulsant drugs used in the treatment oflacerating pains: A comparison. Anesthesia 36:1129-1134, 1981.

87. Caccia MR: Clonazepam in facial neuralgia and cluster headache: Clinicaland electrophysiological study. Eur Neurol 13:560-563, 1975.

88. Bruera E, Breuneis C, Patterson AH, et al: Use of methyphenidate as anadjuvant to narcotic analgesics in patients with advanced cancer. J Pain SymptomManage 4:3-6, 1989.

89. Forrest H: Dextroamphetamine with morphine for the treatment ofpost-operative pain. N Engl J Med 296:712-715, 1977.

90. Menza MA, Kaufman KR, Castellanos AM: Modafinil augmentation ofantidepressant treatment in depression. J Clin Psychiatry 61:378-381, 2000.

91. Fromm GH: Trigeminal neuralgia and related disorders, in Portenoy RH (ed):Pain: Mechanisms and Syndromes. Neurologic Clinics, vol 7, pp 305-319.Philadelphia, WB Saunders, 1989.

92. Lyndstrom P, Lindbloom T: The analgesic tocainide for trigeminalneuralgia. Pain 28:45-50, 1987.

93. Dunlop R, Davies RJ, Hockley J, et al: Letter to the editor. Lancet1:420-421, 1989.

94. Brose WG, Cousins MJ: Subcutaneous lidocaine for treatment of neuropathiccancer pain. Pain 45(2):145-148, 1991.

95. Paice J, Ferrans CE, Lahley FR, et al: Topical capsaicin in themanagement of HIV-associated peripheral neuropathy. J Pain Symptom Manage19:45-52, 2000.