Diabetic Neuropathy: An Intensive Review

Jeremiah John Duby; R. Keith Campbell; Stephen M. Setter; John Raymond White; Kristin A. Rasmussen


Am J Health Syst Pharm. 2004;61(2) 

In This Article


The treatment of diabetic neuropathy may be classified as primary prevention, symptom management, and disease modifying. The DCCT and the UKPDS demonstrated that the risk of neuropathy and other complications can be dramatically reduced or delayed by intensified glycemic control in patients with type 1 and 2 diabetes, respectively.[3,6] These results are especially important in light of the fact that treatments directed at symptom management are polypharmacy nightmares fraught with adverse effects and are only moderately effective. Diseasemodifying agents, which target the underlying pathologies, have a disappointing history but remain the critical focus of secondary intervention and are a hope for the near future. Table 1 gives information on dosage adjustments, targeted dosages, drug interactions, and adverse effects.

Tri-cyclic antidepressants (TCAs). TCAs are commonly prescribed and are one of the most investigated classes of medications for the treatment of diabetic sensorimotor neuropathy.[39] The analgesic effect of TCAs appears dependent on inhibition of the re-uptake of norepinephrine and serotonin, which each agent does to various degrees. The sum of research on TCAs in diabetic neuropathy includes less than 200 patients and indicates only a modest improvement of symptoms versus placebo.[40] The combined number needed to treat (NNT) for the six placebo-controlled trials of TCAs was 2.4 (2.0-3.0); however, the study designs were markedly different, especially with respect to dosing, patient populations, and pain evaluation methods.[41] Desipramine, amitriptyline, imipramine, and clomipramine have demonstrated the best efficacy.[40,42] One head-to-head comparison of desipramine with amitriptyline showed no difference in efficacy but suggested that desipramine was better tolerated.[43] A second study comparing desipramine with placebo demonstrated improvement in a majority of patients who had previously failed to receive pain relief from amitriptyline or had discontinued taking it due to bothersome adverse effects.[42] Clomipramine exhibited efficacy comparable to that of desipramine in a small crossover study.[40] Consideration of nortriptyline, which is an active metabolite of amitriptyline, for the treatment of diabetic neuropathy is confounded by its combination with fluphenazine in both existing trials.[44,45] The combination reduced pain and paresthesia by ≥50% in over 80% of the patients in one trial, but -- as might be expected from the potential drug interaction -- greater than 75% of patients (14/18) experienced adverse effects.[44]

The severity of adverse effects associated with TCAs is attributed to their relative affinities for muscarinic, histaminic (H1), and α1-adrenergic receptors. Amitriptyline exhibits the greatest affinity for muscarinic receptors, followed by protriptyline, clomipramine, trimipramine, doxepin, imipramine, nortriptyline, and desipramine.[42] The anticholinergic adverse effects commonly associated with TCAs are dry mouth (xerostomia), constipation, dizziness, blurred vision, and urinary retention. The tertiary amine tricyclics (i.e., amitriptyline, imipramine, clomipramine, and doxepin) also have the greatest affinity for histamine and α1-adrenergic receptors, resulting in extreme sedation and orthostatic hypotension, respectively. Choosing a TCA should be based chiefly on individual patient tolerability and the risk of adverse drug reactions. Desipramine is a reasonable choice for diabetic sensorimotor neuropathy on the basis of efficacy data, relative tolerability compared with other TCAs, and cost. Tertiary TCAs, including amitriptyline, are considered second-or third-line agents in geriatric patients and should be used with extreme caution in this patient population.

Selective serotonin-reuptake inhibitors (SSRIs). Clinical investigation of SSRIs for the treatment of diabetic sensorimotor neuropathy is limited to studies of paroxetine, fluoxetine, citalopram, and sertraline in a total of 98 patients. Paroxetine was compared with imipramine and placebo in a double-blind crossover study in 26 patients.[46] The antidepressants proved comparable and better than placebo, but 70% of patients receiving placebo experienced a similar improvement on the neuropathic scale used. Perhaps the most notable result was the withdrawal of 7 of the 26 patients from imipramine treatment due to adverse effects, but none of the patients withdrew due to treatment-related reasons during the paroxetine or placebo phases. Citalopram was more effective than placebo in a similar crossover study in 18 patients; however, the overall relief of symptoms was slight, and 2 patients withdrew due to GI adverse effects.[47] Fluoxetine proved to be no better than placebo in a trial of 46 patients with diabetic neuropathy.[43]

In general, SSRIs are considered better tolerated but less effective than TCAs, and they should not be considered for monotherapy of diabetic neuropathy.[39]

Other antidepressants. Extendedrelease bupropion demonstrated greater efficacy than placebo in a double-blind crossover study in 46 patients with neuropathic pain of mixed etiology.[48] Nearly three out of four patients rated their pain as improved or much improved with bupropion, but placebo produced no significant change. Only two patients withdrew from the study; however, over half of patients taking bupropion experienced significant adverse effects, such as dry mouth, insomnia, and headache.

Extended-release venlafaxine 150-225 mg/day (n = 82) was more effective than placebo (n = 81) and extended-release venlafaxine 75 mg/ day (n = 81) in a randomized trial in 244 patients with diabetic neuropathy.[49] Those patients receiving venlafaxine 150-225 mg/day reported significantly lower pain intensity and greater pain relief than recipients of placebo and venlafaxine 75 mg/day. The Patient Global Scale and Physician Clinical Impression Scale also demonstrated parallel improvements. In a smaller study, venlafaxine produced a 75-100% reduction in pain without producing any adverse effects in 10 patients with severe sensorimotor neuropathy.[50] However, the trial was not blinded or placebo controlled and lasted only two weeks.

Gabapentin. Gabapentin is an adjuvant anticonvulsant that is emerging as a first-line agent for the treatment of painful sensory neuropathy.[39] The first double-blind, placebo-controlled study of gabapentin for diabetic neuropathy enrolled 165 patients and was designed to establish high-dose tolerance and efficacy.[51] Patients were initiated on 900 mg/day for the first week, and doses were rapidly increased to 3600 mg/day over the course of four weeks, followed by four weeks of maintenance at the highest dosage tolerated. Gabapentin was well tolerated, and 68% of the patients achieved the maximum dosage; however, roughly one out of four patients experienced dizziness and somnolence, and 8.3% of patients experienced confusion, versus 1.2% of those taking placebo. Gabapentin demonstrated statistically significant efficacy compared with placebo in reducing pain severity that was supported by clinically significant improvements in global impression scales and quality-of-life assessments.

Gabapentin was compared with amitriptyline in a 12-week, double-blind, crossover study in 25 veterans with diabetic neuropathy.[52] The average tolerated dose of gabapentin was slightly less than 1600 mg/day compared with 59 mg/day for amitriptyline. The study demonstrated comparable efficacy and tolerability for both medications. Eleven (52%) of 21 patients taking gabapentin experienced moderate or greater pain relief, compared with 14 (67%) of 21 patients treated with amitriptyline. Sedation and dizziness were more common with gabapentin than amitriptyline, although dizziness rapidly diminished. Dry mouth occurred with twice the frequency during the amitriptyline phase and worsened with time. Two patients withdrew from amitriptyline treatment, versus one patient during the gabapentin phase.

The remaining studies suggest that gabapentin is generally well tolerated, and 1600 mg/day or greater is necessary to achieve clinically significant pain relief.[53,54,55] The obvious disadvantages of gabapentin are the relative cost and the divided (three times-a-day) dosing needed in most patients. For elderly patients, gabapentin has the advantage of very few drug interactions.

In summary, the current evidence demonstrates that gabapentin is effective for diabetic sensorimotor neuropathy, and its use requires no less consideration and vigilance than more traditional medications, especially in elderly diabetes patients.[39,51,53,54,55]

Carbamazepine and oxcarbazepine. Carbamazepine is chemically related to the TCAs, and its anticonvulsant and analgesic mechanisms of action are thought to depend on neuron stabilization by inhibition of ionic conductance.[54] The NNT for sensorimotor neuropathy was 3.3 in two placebo-controlled crossover studies.[41,54,56,57] The first study, published in 1969, demonstrated relief of symptoms in 28 of 30 patients within two weeks of initiation of treatment.[54,56] The results of the second trial were equivocal, because carbamazepine failed to demonstrate efficacy versus placebo in the first arm of the study but proved superior to placebo in the second.[54,57] A more recent trial compared the efficacy and tolerability of carbamazepine and nortriptyline-fluphenazine in a double-blind crossover study.[39] Patients treated with carbamazepine experienced a clinically significant improvement in pain and paresthesia from baseline. There was no significant difference in symptom improvement between the two medications, but patients reported more adverse effects for nortriptyline- fluphenazine than carbamazepine. The adverse effects of carbamazepine (e.g., somnolence, dizziness, and ataxia) tend to limit its use in clinical practice.

Oxcarbazepine is chemically related and has a mechanism of action similar to that of carbamazepine, and it was recently proposed that it may have a much better adverse-effect and drug interaction profile.[39,41] There are no published studies examining oxcarbazepine for the treatment of diabetic neuropathy, but it has demonstrated efficacy comparable to that of carbamazepine in the treatment of trigeminal neuralgia.[39,58]

Other anticonvulsants: Phenytoin, lamotrigine, and zonisamide. Evidence of the efficacy of phenytoin in the treatment of diabetic sensorimotor neuropathy would have to be considerable to justify the clear risk of adverse effects and drug interactions. The results of two double-blind studies in neuropathy patients with diabetes provide conflicting results that substantiate the hazard but fail to satisfy the consequent burden of efficacy.[39,54,59]

Conversely, lamotrigine was relatively well tolerated in an eight-week, double-blind, placebo-controlled study in 59 patients with diabetic neuropathy.[60] The 27 patients randomized to lamotrigine experienced a lower rate of adverse effects than for placebo at dosages of up to 400 mg/day. Pain severity was moderately reduced from an average of 6.4 on the Numerical Pain Scale to 4.2, but there were no significant changes in any of the global pain assessment scales or questionnaires. Nearly a third of the patients taking lamotrigine considered it highly efficacious, which raises hope for a subset of patients with intractable chronic pain.[58] Patients were immediately withdrawn from the study if they developed a rash of any kind; toxic epidermal necrolysis is a current concern.

In an open-label study of zonisamide, less than half of the patients completed the eight-week trial, 10 patients withdrew due to adverse effects, and there were no significant changes in pain or global assessment scores.[61]

Opioid analgesics, tramadol, and nonsteroidal antiinflammatory drugs (NSAIDs). The nature and chronicity of neuropathic pain complicate the use of opioid analgesics for diabetic sensorimotor neuropathy. There have been very few trials evaluating the long-term safety and efficacy of opioids in neuropathic pain, and it is generally accepted that these drugs provide only marginal relief at the risk of severe adverse effects and physical dependence.[41] In fact, no published studies have formally examined the treatment of diabetic neuropathy with an opioid. One eight-week, double-blind, placebo-controlled, crossover trial examined controlled-release oxycodone in the treatment of 38 patients with post-herpetic neuralgia.[62] Reductions in the intensity of steady pain, brief pain, and allodynia were statistically and clinically significant versus placebo. Despite the predictable occurrence of constipation, sedation, and nausea, the patients preferred oxycodone to placebo 67% to 11%. Of course, the introduction of opioids to a patient's drug regimen should be considered only on an individual basis. Periodic follow-up is necessary to adjust the dosage, ensure tolerance of adverse effects, and taper the dosage at the conclusion of treatment.

Tramadol has opioid analgesic and serotonergic properties that theoretically make it an attractive medication for neuropathic pain. There have been two key studies of the safety and efficacy of tramadol in the treatment of neuropathy. A randomized, double-blind study in 131 diabetes patients with neuropathy demonstrated a clinically and statistically significant reduction in pain intensity compared with placebo; the NNT was 3.1.[41,63,64] Nausea and constipation occurred in over 20% of patients, and headache and dyspepsia were also more common than in placebo recipients. The second double-blind study compared tramadol with placebo for the treatment of pain, paresthesia, and touch-evoked pain associated with polyneuropathies. Tramadol 200-400 mg/day proved to be more effective than placebo; however, although the difference was statistically significant, the clinical improvement was fairly modest, and the NNT for >50% pain relief was 4.3.[41,62] The major adverse effects -- tiredness, dizziness, and dry mouth -- were reported by roughly half of the patients during the tramadol phases, and constipation occurred in over 40%, versus 12% for placebo. Adverse effects were generally reported by patients as absent to mild.

In conclusion, tramadol is a safe and effective medication for diabetic sensorimotor neuropathy, and the dosage required for therapeutic effect is relatively high.

The use of NSAIDs in diabetes patients must be prefaced with the caution that they can impair renal function in vulnerable individuals by inhibiting prostaglandin synthesis; also, long-term use, especially in the elderly, can cause GI bleeding. Both ibuprofen (600 mg four times daily) and sulindac (200 mg twice daily) demonstrated statistically significant reductions in paresthesia scores compared with placebo in a 24-week, single-blind study in 18 veterans with diabetic neuropathy.[65] The paresthesia scale is no longer commonly used, and only those patients with "moderate pain" experienced relief. However, there were no major adverse effects or changes in renal variables in the 16 weeks that the patients received the ibuprofen and sulindac. NSAIDs should not be considered for monotherapy of diabetic neuropathy.[64] They may have a limited, adjuvant role for short-term pain relief in patients at low risk of renal dysfunction and GI bleeding who receive adequate education and monitoring.

Mexiletine and lidocaine. Information on the use of mexiletine, a type Ib antiarrhythmic medication, for painful diabetic neuropathy is based on a study demonstrating moderate pain relief with intravenous lidocaine.[66,67] Mexiletine is the oral analogue of lidocaine and is thought to exert its analgesic and antiarrhythmic effects by membrane-stabilizing Na+ -channel antagonism. Four well-designed studies failed to show a statistically significant improvement in pain, paresthesia, or global assessment compared with placebo.[66,68,69,70] One trial suggested that a subset of patients with burning or stabbing pain or those with a high degree of pain benefitted from higher dosages of mexiletine.[68] The studies did show that mexiletine is very well tolerated in dosages up to 675 mg/day; however, patients with heart disease were generally excluded. In the absence of clinical data demonstrating efficacy, mexiletine can be recommended only as an alternative agent for patients with extreme, refractory symptoms and no cardiac risk.

Levodopa. One double-blind, placebo-controlled study in 25 patients with sensorimotor neuropathy demonstrated a statistically significant reduction in pain.[71] The study was too small to make any conclusions about clinical significance, but no adverse effects were observed with the 100-mg dose of levodopa given.

Dextromethorphan. Dextromethorphan is a partial antagonist of the N-methyl-D-aspartate receptor, which has been implicated in the mediation of neuropathic pain in animal models. In a double-blind, crossover study in 13 patients, dextromethorphan led to a 24% reduction in pain compared with placebo (p = 0.014), with an average dosage of 381 mg/day.[72] In fact, 7 of the 13 patients described their pain relief as "a lot" or "moderate" while taking dextromethorphan. Every patient experienced adverse effects while receiving dextromethorphan (e.g., sedation and dizziness). Clinically, the extremely high average dosage required and the adverse effects all but prohibit use.

Topicals: Capsaicin cream and isosorbide dinitrate spray. Capsaicin is extracted from capsicum peppers and produces a dose-dependent desensitization of type C nociceptive fibers by depleting the neurotransmitter substance P.[73,74] The largest published study of an analgesic treatment for diabetic sensorimotor neuropathy was conducted by the Capsaicin Study Group.[74] The multi-center study involved 252 patients with painful symptoms severe enough to interfere with daily activities. Patients were randomized to either 0.075% capsaicin or vehicle cream (placebo) for six weeks. Capsaicin proved significantly more effective than placebo in providing pain relief (58.4% versus 45.3%), decreasing pain intensity (38.1% versus 27.4%), and improving global assessment scores (58.4% versus 45.3%) (p < 0.05), although placebo produced very high response rates as well. Burning was the most commonly reported adverse effect, affecting 87 of the 135 patients in the capsaicin group, compared with 23 of the 49 patients in the placebo group; however, the reports of burning tended to diminish and pain relief increased as therapy progressed. Coughing, irritation, and rash were reported by approximately 10% of patients receiving capsaicin.

There are several practical considerations in the clinical use of capsaicin cream. First, it must be applied three or four times daily to the affected area, which may be quite large. Second, application results in temporary burning, which is intolerable for many patients already in pain. Finally, as demonstrated by this and other trials, several weeks of diligent use may be needed for capsaicin to have an effect. The recommendation of capsaicin to a patient must be accompanied by practical instructions about handling, adverse effects, and expected results.

A pilot study of isosorbide dinitrate spray for diabetic neuropathy demonstrated safety and efficacy in 22 patients with refractory pain and burning.[75] In the 12-week, double-blind, placebo-controlled, crossover study, isosorbide dinitrate spray consistently produced a statistically and clinically significant reduction in pain and burning. The treatment, which consisted of spraying the affected area before bedtime, was extremely well tolerated, with only two patients complaining of transient headache. Relief was provided throughout the night and subsequent day until the next treatment. The study was relatively brief and small, but the results suggest great potential benefit with little risk for severely afflicted patients. Isosorbide dinitrate is not currently available in the United States.

Complementary and alternative therapies. Electrostimulation has been shown to provide temporary relief of pain associated with diabetic sensorimotor neuropathy, but the feasibility and efficacy of maintenance therapy remain controversial. A study of transcutaneous electrotherapy randomized 31 patients with diabetic peripheral neuropathy to take a portable electrotherapy machine home for one week of self-administration.[76] Patients were assigned to therapy with either active or inactive electrodes. The therapy for both groups consisted of placing electrodes as instructed and administering electrical shock for 30 minutes every day. Active therapy improved neuropathic symptoms in 15 (83%) of the 18 patients, compared with 5 (38%) of 13 patients receiving sham therapy. No major adverse effects were observed in either group. The therapy had a very low residual effect, and patients' pain returned within one week of stopping treatment.

In a related study, percutaneous electrical nerve stimulation (PENS) was examined for value in the treatment of painful diabetic neuropathy.[77] Therapy consisted of using 10 32-gauge "acupuncture-like" needles to puncture the soft tissue and muscle of the foot and leg to a depth of 1- 3 cm and applying alternating frequencies of electrical shock. The crossover design randomized 50 patients to either electrical stimulation or acupuncture alone, which was considered sham treatment. Patients on active therapy experienced profound reduction in lower-extremity pain (56% versus 14%), increased physical activity (48% versus 13%), and improved sleep quality (41% versus 13%) compared with the sham treatment. Ninety-two percent of patients preferred the active therapy, and a similar majority reported an improved sense of well-being and willingness to pay "extra" for PENS.

In theory, electrostimulation could produce analgesia by inducing the release of endogenous opioid-like chemicals, and these trials seem to support a role for electrotherapy. However, there are obvious weaknesses in the study designs and practical obstacles to clinical use. It is almost impossible to compare electrostimulation with a true sham or placebo therapy. In addition, PENS, for all its absence of reported adverse effects, remains relatively invasive. Further, both studies had very strict and extensive exclusion criteria that would prohibit a large fraction of the diabetic population from participating. The results are generally considered preliminary and difficult to imitate and maintain in a clinical setting.[39]

Thioctic acid, or α-lipoic acid, is an alternative therapy that has recently received attention for use in the treatment of symptomatic sensorimotor neuropathy. In a multi-center, double-blind study in 328 patients with diabetic neuropathy, intravenous infusions of α-lipoic acid proved safe and effective for the short-term relief of pain, paresthesia, and burning.[78] Overall, the 600-mg dose produced the optimal statistically and clinically significant pain relief compared with other doses and placebo, and 93% of patients receiving the 600-mg infusions rated their tolerance as either good or very good. The obvious limitation to duplicating this success in clinical practice is the requirement for daily 30-minute i.v. infusions, which also presents certain risks. Many herbal manufacturers are now promoting α-lipoic acid for diabetes patients without studies proving effectiveness. The half-life of α-lipoic acid is two to four minutes, making oral products suspect for treatment of neuropathy.

Prevention of CAN is essential, because there are very few effective treatments. There is persuasive evidence that the risk of diabetic CAN may be reduced by 68% through the management of hyperglycemia, hyperlipidemia, and hypertension and the use of ACE inhibitors and antioxidants.[1,15] Further, effective glycemic control may reverse CAN in the early stages of development. The symptoms of postural hypotension may be reduced by nonpharmacologic measures, such as increasing water consumption and using body stockings.[1] Fludrocortisone effectively increases blood pressure, but consideration must be given to the risk of triggering or exacerbating heart failure, edema, and hypertension. Antihypertensives may produce a paradoxical increase in blood pressure by activating or antagonizing α1-, α2-, and β-adrenergic receptors that are inappropriately expressed due to autonomic denervation or dysfunction.[1,5,14,15,16] Clonidine should be initiated with extreme care in this patient population. Midodrine, dihydroergotamine or caffeine, and octreotide have been suggested for the treatment of severe, refractory CAN.[1,16] Finally, imbalances in sympathetic and parasympathetic activities may be stabilized or improved with ACE inhibitors and β-blockers such as atenolol, metoprolol, and propranolol.[1,15,16]

The primary treatment for diabetic GI autonomic neuropathy is glycemic control. Hyperglycemia has been shown to directly reduce and retard gastric contractions and the GI emptying rate, and the sensation of premature satiety and fullness is also related to elevated blood glucose levels.[20,79,80] Studies have shown that short-term reduction of preprandial hyperglycemia does not lead to a resolution of symptoms, but GI symptoms are much more prevalent in patients with poor glycemic control.[19,20,80] Nonpharmacologic treatments are directed at dietary changes, and it is recommended that symptomatic patients reduce the fat content and increase the fiber content of their meals.[1,20,80] Perhaps the most frustrating factor is that GI dys-function itself makes glycemic control extremely challenging, because it becomes virtually impossible to time nutrient absorption with antidiabetic therapy.[19]

Despite the high prevalence of GERD, constipation, and diarrhea in patients with diabetes, there are conflicting reports about whether the incidence is greater than that in the general population.[79,80] Further, there are no unique treatment algorithms for the symptoms of diabetic GERD, constipation, or diarrhea, except that fectclonidine is suggested for severe cases of diabetic diarrhea to reduce stool volume.[1]

The current focus of research is the use of prokinetic agents in the treatment of gastropathy and gastroparesis, which have been shown to be much more prevalent in diabetes patients than the general population.[1,20,79,80] The most recent development is the failure of the motilin agonist ABT-229 to show greater efficacy than placebo in Phase II FDA trials. Cisapride, metoclopramide, and domperidone have been shown to be effective for the treatment of diabetic gastropathy. The market availability of cisapride is restricted because of the risk of QT-interval prolongation and arrhythmia, to which diabetics with autonomic neuropathy are susceptible.[80] Furthermore, the known risk of using cisapride in patients with diabetic gastropathy cannot be justified by the relatively insignificant improvement in symptoms demonstrated by clinical trials.[79,80] Metoclopramide has been shown to be significantly more effective than cisapride; however, its long-term use carries a risk of CNS adverse effects.[19,79] Domperidone has been found to be as effective as metoclopramide for the relief of symptoms of diabetic gastropathy.[18] Domperidone and metoclopramide are both dopamine antagonists, but domperidone does not penetrate the blood- brain barrier to the extent that metoclopramide does, which may account for its more favorable adverse-effect profile.[18,19,20,79] (Domperidone is not currently available in the United States.) Erythromycin is a macrolide antibiotic that directly interacts with motilin receptors; however, the stimulation of GI motility rapidly diminishes with oral administration and duration of use.[15,19] Erythromycan be recommended only for short-term therapy for gastroparesis and should not be administered concurrently with cisapride.[19,80] The efof erythromycin is reduced by hypoglycemia, a critical observation to the study and use of other prokinetic agents.[80]

In conclusion, the GI transit rate does not correlate with symptoms, and accelerating gastric emptying does not necessarily reduce or resolve symptoms.[20] The existing pro-kinetic agents are only modestly efficacious in symptom relief, and their use is greatly limited by associated adverse effects and rapidly developing tachyphylaxis.[20,80]

Neurogenic bladder does not generally respond well to pharmacologic treatments. Bethanechol and doxazosin may reduce urinary retention but may not be sufficient to induce complete bladder voiding, which represents a high risk factor for recurrent urinary-tract infection.[1,16] Patients may be able to use abdominal massage techniques, such as the Credé method, to initiate micturition, but many eventually require catheterization.[1,16]

The treatment of diabetic sexual dysfunction must begin with glycemic control. Erectile dysfunction has been shown to directly correlate with hemoglobin A1c levels.[81] This correlation of poor glycemic control with severity of sexual dysfunction was not supported in two separate studies of diabetic women; however, both trials demonstrated that the incidence of dysfunction was higher in diabetic women than in the general population.[82,83,84] The next step is to evaluate the patient's current medications for agents that are known to cause sexual dysfunction (e.g., β-blockers, antidepressants, spironolactone). Female diabetic sexual dysfunction has not been adequately studied, and treatments are virtually unknown. Vaginal dryness and dyspareunia may respond well to the use of water-based lubricants.[85] Studies evaluating the effect of phosphodiesterase inhibitors for female sexual dysfunction have not yet been published; however, these agents are generally well tolerated and may provide some benefit in the absence of contraindications.[1,16,86] Diabetic erectile dysfunction can be caused by a complex interplay of vascular and neural organic factors and psychological factors; however, it is generally recommended that patients be given a trial of oral medication (i.e., sildenafil, tadalafil, or vardenafil) to improve penile blood supply as much as possible, barring contraindications.[24,87,88,89] Patients may also try intraurethral suppositories or intracavernosal injections of prostaglandin E1 (alprostadil); however, the use of intracavernosal injections is poorly tolerated, with penile pain and fibrotic changes occurring in roughly 10% of patients.[24] Men with severe dysfunction may also try vacuum-constriction devices, which are better tolerated than intracavernosal injections.

Ruboxistaurin mesylate (LY333531) is a protein kinase Cβ inhibitor in Phase III FDA trials for the treatment of diabetic neuropathy. A year-long, double-blind, placebo-controlled, Phase II study demonstrated clinically and statistically significant improvement of objective and subjective neurologic endpoints, including overall neurologic examination and patient global assessment.[90] Perhaps more important, ruboxistaurin was extremely well tolerated.

Three aldose reductase inhibitors are currently in Phase III trials (zenarestat, minalrestat, and zopolrestat). Numerous clinical trials have failed to demonstrate clinically significant changes in symptoms, and their future use may be as preventive treatment.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: