In recent years, 2 large randomized,controlled studies have documentedthe efficacy of the anticonvulsantgabapentin in the management ofpainful diabetic neuropathy (PDN)1and post-herpetic neuralgia (PHN).2Although vastly different in origin,these 2 neuropathies have exhibitedsome similarities in their response totherapeutic agents of various classes.The discovery that yet another typeof pharmaceutical is useful in treatingpain from either PDN or PHN hasraised questions about the similaritiesand differences in the managementof these 2 painful neuropathicsyndromes.
In recent years, 2 large randomized, controlled studies have documented the efficacy of the anticonvulsant gabapentin in the management of painful diabetic neuropathy (PDN)1 and post-herpetic neuralgia (PHN).2 Although vastly different in origin, these 2 neuropathies have exhibited some similarities in their response to therapeutic agents of various classes. The discovery that yet another type of pharmaceutical is useful in treating pain from either PDN or PHN has raised questions about the similarities and differences in the management of these 2 painful neuropathic syndromes. Here, I focus on therapeutic approaches used to manage these common pain disorders. I will highlight pertinent research findings regarding treatment efficacy. I also provide specific treatment algorithms for effective pain management in patients with PDN and PHN.
EPIDEMIOLOGYPainful diabetic neuropathy. Diabetes mellitus affects 3% to 5% of the general population in the United States; approximately 10% to 15% of persons over age 50 are affected. The majority (90%) have type 2 diabetes. At some time, some form of diabetic neuropathy will develop in a large proportion of patients (especially those with inadequate glycemic control). This is manifested either by peripheral neurologic deficits or by autonomic dysfunction involving a variety of organ systems. 3 A distal symmetric neuropathy affects about 50% of persons with either type 1 or type 2 diabetes.4 Pain is the most distressing symptom and is thought to affect at least 10% of diabetic patients with neuropathic damage. The pain is frequently described as burning, accompanied by numbness and tingling. There is frequently loss of proprioception and vibration sense. Although nerves throughout the body are affected, pain is generally concentrated in the hands, legs, and feet.
Post-herpetic neuralgia. Infection with human herpesvirus type 3 (varicella- zoster virus) is also common; estimates suggest that more than 90% of the population is infected at some time.5 The virus causes varicella (chickenpox) and, when reactivated, shingles (herpes zoster or herpetic neuralgia).5 Approximately 1 million cases of acute shingles are reported annually2; PHN develops in about 10% to 15% of those affected. Increasing age and compromised immune response are associated with a greater incidence of shingles; likewise, increasing age and greater severity of acute pain are independent predictors of progression to chronic PHN (ie, pain following resolution of the rash).5 Sharp, shooting pain seems to predominate initially in PHN; later, burning pain is most common. Throbbing pain and tenderness are also frequently reported by patients with PHN. Mechanical allodynia (the sensation of pain from a light touch) is common. Shingles and PHN are almost always unilateral and involve a limited set of dermatomes (55% thoracic, 25% cranial, 14% lumbar, 12% cervical).6 PDN and PHN are rarely confused. PDN most commonly involves the feet and hands, and the patient must have diabetes. PHN is typically unilateral; 89% of cases involve the thorax or face, and the patient has a history of shingles.
Many therapeutic modalities have been used to treat the pain of PDN or PHN (Table 1). The efficacy of these therapies ranges from established to variable to unknown. Lidocaine. Transdermal lidocaine in the form of a patch has been shown to be effective in PHN.7 It is associated with few adverse effects, because the lidocaine impregnated into the patch acts topically and there is minimal systemic absorption. In addition, the patch acts as a physical barrier, decreasing complaints of allodynia in the patch area. The lidocaine patch has not been studied in PDN. A topical cream containing lidocaine and prilocaine has been of limited effectiveness. Some patients find the cream less convenient because it requires a separate occlusive dressing that may or may not stay in place.
Capsaicin cream. Clinical trials have shown the 0.075% formulation to be useful in PHN8; the product has been marketed for this indication. The degree of pain relief is variable, however, and overall is modest. The cream is usually applied 3 or 4 times per day. Results are less encouraging in PDN; the response to 0.075% capsaicin cream and to placebo creams tends to be equivalent.
NSAIDs. Preparations using topically- applied NSAIDs are available in the United States only as over-thecounter salicylate compounds, but their efficacy has not been formally evaluated. Similar preparations have been used in Europe, however, and have been beneficial in PHN.9 Topical NSAIDs have not been studied for PDN.
Antidepressants. Approximately half of patients with PHN or PDN benefit from therapy with a tricyclic antidepressant, such as amitriptyline.10,11 These agents are relatively inexpensive; unfortunately, however, they have many adverse effects that limit their usefulness-especially in elderly persons. Nortriptyline is the initial drug of choice: it has the same efficacy as amitriptyline, but fewer side effects. The newer selective serotonin reuptake inhibitors have not been shown to be as effective as the tricyclics in treating painful conditions; in particular, fluoxetine was not of benefit in PDN.12 Paroxetine has been shown to be superior to placebo, 8 but it has not been formally compared to a tricyclic.
Anticonvulsants. Gabapentin has been shown to be efficacious in both PHN and PDN.1,2 Two randomized, double-blind, placebo-controlled studies were done that both aimed for a dosage of 3600 mg/d of gabapentin, although the actual average dosage was closer to 1800 mg/d. Primary efficacy measure was a change in the average daily pain score (based on an 11-point scale). In those studies, gabapentin monotherapy was effective in the treatment of pain and sleep interference associated with PHN and diabetic neuropathy. Another study compared the effects of gabapentin with amitriptyline in patients with PDN.13 The drugs had similar efficacy as well as similar frequency of intolerable adverse effects. Carbamazepine has been shown to be effective in treating PDN,13 but controlled trials are lacking for PHN. Phenytoin has been studied in PDN with conflicting results; trials of this drug are lacking for PHN. Other anticonvulsants have not been studied in PHN and PDN. The benzodiazepine lorazepam was found to be less effective than amitriptyline in PHN.14
Antiarrhythmics. Controlled trials show that intravenous lidocaine provides long-term pain relief for PDN. However, its efficacy has not been confirmed by clinical experience.10 One study showed oral mexiletine to be efficacious in this setting, with a response similar to intravenous lidocaine. 15 Data regarding efficacy of IV lidocaine in PHN are minimal. α2-
Agonists. Oral clonidine has been shown to decrease the pain of PHN. There are conflicting reports regarding benefit from clonidine in PDN. Adverse effects such as postural hypotension, dizziness, and sedation are common with this drug.10
NSAIDs. Oral ibuprofen lacked efficacy in alleviating the pain of PHN.9 There are conflicting data regarding the efficacy of NSAIDs in PDN. Studies of the new cyclooxygenase-2 selective agents are lacking. Overall NSAID monotherapy appears to be of minimal benefit in controlling neuropathic pain.
Opioids. Opioids appear to be of some benefit in 30% to 50% of patients with neuropathic pain. Oxycodone has been shown to be efficacious in treating PHN.11Tramadol, a weak opioid with some tricyclic effects, is beneficial in PDN.16 However, the dosage of opioid needed in PHN or PDN has not been established, and there is probably significant interpatient variability. In addition, tolerance frequently develops when opioids are used to treat chronic pain conditions. Tolerance is frequently managed by rotating to another opioid, because cross-tolerance is not complete.
Transcutaneous electrical nerve stimulation. TENS is moderately beneficial in PDN, and may be of some benefit in PHN.17 Because this form of therapy has minimal side effects and can be self-administered by the patient, there are few reasons not to try it.
Invasive procedures. For patients with PHN, intrathecal corticosteroids are potentially promising. Although epidural corticosteroids appear to have little benefit in cases of established PHN, dramatic responses to intrathecal methylprednisolone have recently been reported in Japanese patients with this condition.18,19 However, the exact preparation used in Japan is not available in this country. By contrast, there are no published data supporting either intrathecal or epidural corticosteroids in PDN. As a general rule, corticosteroids should be avoided in persons with diabetes.
Spinal cord stimulation. Data from case series indicate moderate efficacy of spinal cord stimulation in PDN. Results for PHN are less encouraging. 20 Spinal cord stimulation is expensive and requires a long-term commitment from both the patient and practitioner implanting it. As such, it is probably an option only for those persons with intolerable pain who have been unable to obtain relief from less drastic means of therapy.
Multimodal analgesia (polypharmacy). There are surprisingly few data regarding the efficacy of combinations of treatments for PDN or PHN. The gabapentin trial for PHN2 allowed patients to continue taking stable doses of tricyclics; however, entry criteria required that study participants have moderate to severe pain. Therefore, those patients who were taking tricyclics could not have had major benefit from them. The gabapentin trial for PDN,1 on the other hand, required that patients be weaned from tricyclics. Lidocaine patch studies have allowed patients to continue other therapies at stable doses.7 Both acetaminophen and NSAIDs have been shown to potentiate the analgesic effect of opioids in many painful conditions, but studies of these common combinations are lacking for PDN and PHN. Single-agent therapy has the advantage of simplicity, but the probability of adverse effects rises as the dosage is increased. Multimodal therapy with 2 or 3 drugs may provide equivalent analgesia while minimizing the adverse effects from any single agent. On the other hand, multimodal therapy increases the probability of adverse drug interactions. Table 2 lists common adverse effects and interactions for medications most commonly used for treating neuropathic pain.
SUGGESTED CARE ALGORITHMS
Figures 1 and 2 depict my approach to patients with PDN and PHN. This approach employs therapies with the fewest adverse effects and the highest probability of benefit first. For patients with moderate pain from either PDN or PHN, TENS can be tried early on. For PHN, try a transdermal lidocaine patch. If these are ineffective or not tolerated by the patient, consider a trial of capsaicin cream. If topical and physical modalities do not adequately control the pain, then a systemic agent should be added. I usually start with nortriptyline because it can be dosed once a day at bedtime and may improve sleep. I increase the dosage over 2 to 4 weeks, depending on the patient’s age and overall condition. Nortriptyline is taken in the evening, usually 1 hour before bedtime to maximize its sedative effect. Begin with a lower dosage (10 mg qhs) in patients who are older than age 65 or frail and increase the dosage weekly. Start younger patients on higher doses (25 mg qhs). The goal is to progressively increase the dose to 75 mg/d or until dose-limiting side effects develop. Check plasma levels; if they are very low, absorption problems or compliance issues may need to be addressed. The target plasma level of nortriptyline is 300 nmol/L.21
If pain is not adequately controlled, add gabapentin. Start with a low dose at bedtime (300 mg) and increase to 2 and then 3 times per day. Increase the number of capsules progressively until the dosage is 600 mg tid (1800 mg/d). This can be accomplished over a 2-week period. Progressively increase the dosage (up to 3600 mg/d) in patients who still have moderate or severe pain and who do not have dose-limiting adverse effects.
Try opioids early on for patients with severe pain. Otherwise, use them only when the approach outlined above fails to provide adequate analgesia. I prefer oxycodone, either with acetaminophen or without, and titrate to the point of efficacy or unacceptable side effects. Oxycodone is readily available and relatively inexpensive; in addition, in my experience, oxycodone appears to have a slightly better sideeffect profile than morphine, and a better analgesic potential than codeine or propoxyphene. Start with 5 to 10 mg of oxycodone every 6 hours as needed. The amount of opioid for patients using acetaminophen-containing compounds is limited by the acetaminophen dose (which should be restricted to less than 4000 mg/d).
Patients who benefit from opioids and who are using short-acting opioids more often than every 6 hours should be switched to a long-acting agent. Direct conversion from regular oxycodone to the controlled-release formula is the most straightforward. Charts are available for conversion to sustained-release morphine, transdermal fentanyl, or methadone. The great majority of patients will obtain adequate pain control from this sequence of therapies. However, an occasional patient either will not respond satisfactorily or will have intolerable adverse reactions to all of the above. These patients may benefit from referral to a pain specialist.
Trials of antiarrhythmics, other anticonvulsants, other antidepressants, or α2-agonists can be undertaken. Patients with PHN may benefit from a series of intrathecal corticosteroid injections, if the efficacy of this treatment is confirmed by studies in the United States. A trial of spinal cord stimulation may be warranted for either PDN or PHN. Neuroablative procedures are a last resort for either condition because they are associated with significant morbidity. There are no controlled studies of neuroablative procedures, and in my experience, pain recurs frequently- even if there is an initial benefit. With long-term pain there are changes in the spinal cord and brain stem that permit the continued perception of pain even after a dorsal root entry zone lesion is induced.
Backonja M, Beydoun A, Edwards KR, et al.Gabapentin for the symptomatic treatment of painfulneuropathy in patients with diabetes mellitus:a randomized controlled trial.
Rowbotham M, Harden N, Stacey B, et al. Gabapentinfor the treatment of postherpetic neuralgia:a randomized controlled trial.
Skyler JA. Diabetes mellitus, types I and II. In:Kelley WN, ed.
Textbook of Medicine
. 3rd ed. Philadelphia:Lippincott-Raven; 1997:2249.
Bennett JC, Plum F.
Cecil Textbook of Medicine
.Philadelphia: WB Saunders Company; 1996:1274-1275,2093-2095.
Yen-Moore A, Severson JL. Viral diseases of theskin. In: Rakel RE, ed.
Conn’s Current Therapy
.Philadelphia: WB Saunders Company; 1999:840-841.
Loeser JD. Herpes zoster and postherpetic neuralgia.
Argoff CE. New analgesics for neuropathic pain:the lidocaine patch.
Clin J Pain
Watson CPN. Topical capsaicin as an adjuvantanalgesic.
J Pain Symptom Manage
Moore RA, Tramer MR, Carroll D, et al. Quantitativesystematic review of topically applied nonsteroidalanti-inflammatory drugs.
Kingery WS. A critical review of controlled clinicaltrials for peripheral neuropathic pain and complexregional pain syndrome.
Watson CPN. The treatment of neuropathic pain:antidepressants and opioids.
Clin J Pain
Max MB, Lynch SA, Muir J, et al. Effects of desipramine,amitriptyline, and fluoxetine on the painin diabetic neuropathy.
N Engl J Med
Morello CM, Leckband SG, Stoner CP, et al.Randomized double-blind study comparing the efficacyof gabapentin with amitriptyline on diabetic peripheralneuropathy.
Arch Intern Med
Backonja M. Anticonvulsants (antineuropathics)for neuropathic pain.
Clin J Pain
Galer BS, Harle J, Rowbotham MC. Responseto intravenous lidocaine infusion predicts subsequentresponse to oral mexiletine: a prospectivestudy.
J Pain Symptom Manage
Harati Y, Gooch C, Swenson M, et al. Doubleblindrandomized trial of tramadol for the treatment of the pain of diabetic neuropathy.
Kumar D, Marshall HJ. Diabetic peripheral neuropathy:amelioration of pain with transcutaneouselectrostimulation.
Kotani N, Kushikata T, Hashimoto H, et al. Intrathecalmethylprednisolone for intractable postherpeticneuralgia.
N Engl J Med
Kikuchi A, Kotani N, Sato T, et al. Comparativetherapeutic evaluation of intrathecal versus epiduralmethylprednisolone for long-term analgesia in patientswith intractable postherpetic neuralgia.
RegAnesth Pain Med
Meyerson BA, Linderoth B. Spinal cord stimulation.In: Loeser JD, ed.
Bonica’s Management ofPain
. Philadelphia: Lippincott Williams & Wilkins;2001:1857-1876.
Leijon G, Boivie J. Central post-stroke pain-a controlled trial of amitriptyline and carbamazepine.