INTRODUCTION — There are a number of distinctive neuropathic syndromes, which can be classified according to the timing of their appearance during HIV infection, their etiology, and whether they are primarily axonal or demyelinating. The most common of these is distal symmetric peripheral neuropathy [1,2].

This topic will cover the pathogenesis, clinical manifestations, diagnosis, and treatment of distal symmetric peripheral neuropathy in patients with HIV.

The approach to peripheral neuropathy in general is discussed in detail elsewhere. (See "Approach to the patient with sensory loss" and "Overview of polyneuropathy".)

EPIDEMIOLOGY

Prevalence — The prevalence of distal symmetrical polyneuropathy (DSPN) in different series has varied from 9 to 63 percent [3-8]. This variability reflects differences in the degree of immunosuppression (higher prevalence with more advanced disease), in the definition of the neuropathy (symptomatic or asymptomatic), and in exposure to neurotoxic antiretrovirals [4,9,10]. Because of known neurotoxicities, didanosine and stavudine are no longer recommended for the treatment of HIV. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach".)

Risk factors — In the era prior to potent antiretroviral therapy (ART), DSPN usually occurred in the setting of advanced immunosuppression [3,9,11-13]. In one report, for example, the mean CD4 count was 113/microL (range 26 to 275 cells/microL) [9].

In addition to immunosuppression, the level of HIV viremia was also correlated with the development of DSPN and the severity of symptoms [5,14,15]. In the Multicenter AIDS Cohort Study, the risk of DSPN was increased 2.3-fold in patients with an HIV RNA level >10,000 copies/mL at baseline [14].

However, in the era of potent ART, immunosuppression or high levels of viremia have not been associated with the development of DSPN in the vast majority of studies [5,6,8,16-19]. There are conflicting data on whether coinfection with hepatitis C is associated with DSPN [6,20,21].

Other factors associated with DSPN include aging, longer duration of HIV infection, host factors such as diabetes, hypertriglyceridemia [22], nutritional deficiencies, mitochondrial polymorphisms, substance use [23], and the use of older nucleoside reverse transcriptase inhibitors such as didanosine and stavudine [5,6,19,22-28]. In a large prospective study, 2141 patients with HIV were followed longitudinally for seven years with annual screening for symptoms and signs of peripheral neuropathy [27]. The risk of peripheral neuropathy was associated with aging and nucleoside analog use while sensory loss was associated with older age, nucleoside analog use, and diabetes. In ART-experienced patients, female sex, taller height, alcohol consumption, higher plasma creatinine, smoking, and use of isoniazid have been identified as additional risk factors for peripheral neuropathy [21,29].

The role of certain antiretroviral medications is discussed below. (See 'Role of drugs' below.)

Effect of antiretroviral therapy on prevalence — In most studies, the incidence of HIV-associated DSPN appears to have decreased in the era of potent ART compared with earlier cohorts, suggesting that effective suppression of HIV itself may have a beneficial effect on peripheral nerve function (figure 1) [5,7,15,16,30]. As an example, in a large cohort of 2515 patients with HIV, certain drugs (didanosine, stavudine, nevirapine, and certain protease inhibitors) were associated with the development of DSPN in the first year of use [5]. However, patients who did not develop DSPN in the first year of ART had a decreased risk of developing this complication with continued drug exposure. In another cohort of 2165 patients followed for more than 3 years, incidence rates of peripheral neuropathy also declined with the initiation of ART [31]. These data suggest that immune restoration, or viral suppression of HIV, lead to a decreased risk of DSPN [5].

PATHOGENESIS — The pathogenesis of distal symmetrical polyneuropathy (DSPN) is incompletely understood and may be multifactorial. DSPN is termed a "dying-back" neuropathy to reflect the pattern of distal fiber loss [32]. It involves myelinated and unmyelinated axons of all sizes; this pattern of axon loss is indistinguishable from that seen with other toxic neuropathies.

There is a paucity of virus and associated inflammation in the peripheral nerves of patients with HIV [32]. Although there are some case reports of HIV being cultured from peripheral nerve, it is widely accepted that almost all recovered virus is from the resident macrophages and monocytes that migrate to areas of injury [32-36]. HIV itself may lead to local axonal injury through two separate mechanisms, both of which appear to be triggered by envelope protein gp120 [37]. One indirect route is via neuronal apoptosis; the other is through direct, local toxicity mediated through activation of mitochondrial caspases. Mitochondrial DNA damage has been shown to accumulate in distal mitochondria of long axons in HIV patients with DSPN, also supporting the possibility that distal mitochondrial dysfunction may play a role [38]. Likely more important, however, is the role that viral antigens play in provoking immune activation and inducing a microenvironment that is toxic to the peripheral nerve, as demonstrated by the following in vitro data:

●Immunohistochemical studies show macrophage and T cell infiltration of peripheral nerves and dorsal root ganglia [36,39].

●Activated cytokines are found in the dorsal root ganglia of patients with HIV and distal symmetrical polyneuropathy raising the possibility of a multifocal, immunologically mediated inflammatory disease [36].

●HIV gp120 activates Schwann cells via its chemokine receptor (CXCR4), leading both to neuronal apoptosis [37] and to production and release of tumor necrosis factor and other proinflammatory cytokines, which are directly toxic to neurons [40,41].

●HIV infection is associated with a reduction in mitochondrial DNA content and changes in morphology [42].

Not only does HIV infection result in nerve damage, but recent data suggests that reinnervation is impaired in patients with HIV infection, limiting the ability of the peripheral nervous system to heal itself [43].

Role of drugs — Many cases of distal symmetrical polyneuropathy are iatrogenic, due to intrinsic neuronal toxicities of certain antiretroviral medications [44-49]. The neuropathy is indistinguishable electrophysiologically from HIV-associated DSPN, although the hands may be affected more often in drug-induced cases [10,44-47]. The incidence of neuropathy is dose-dependent and increases with the duration of drug exposure [48]. The onset is typically seven to nine weeks after beginning therapy.

Dideoxynucleosides — The incidence of drug-related neuropathy has been shown to correlate directly with the degree of mitochondrial toxicity of particular nucleoside reverse transcriptase inhibitors, although a direct link between toxicity and oxidant stress has not been demonstrated [30,50-53]. Commonly implicated agents include stavudine (d4T) and to a lesser extent didanosine (ddI) [6,50].

The neurotoxicity of the combination of didanosine and stavudine was illustrated in a multicenter, randomized, partially double-blind trial of 620 antiretroviral-naive patients who were assigned to sequential three-drug regimens with different nucleoside analogues [54]. At a median of 2.3 years, symptomatic peripheral neuropathy was significantly more likely in patients treated with regimens containing didanosine and stavudine compared to those containing zidovudine and lamivudine (27 versus 10 percent) [54].

Genetic determinants may also play a role in risk in drug-induced DSPN:

●In a study of 509 patients with HIV, mitochondrial haplotype T was more common in those patients who developed DSPN [55]. Among 137 White subjects randomized to receive didanosine and stavudine, 21 percent of those who developed peripheral neuropathy belonged to mitochondrial haplogroup T compared to five percent of control subjects (odds ratio, 5.4).

●In a case-control study, hemochromatosis gene mutations were associated with a decreased risk of developing DSPN during dideoxynucleoside therapy [56]. This protective effect may be related to the requirement of iron for mitochondrial function.

The potential neurotoxicity of antiretroviral drugs does not preclude their use, since the beneficial effects on viral load suppression and immune function recovery appears to outweigh their potential neurotoxicity [4,5,16,17].

Protease inhibitors — Some early data had suggested a potential role of protease inhibitors (PIs) in the pathogenesis of DSPN. One small study, which enrolled 101 patients with DSPN from 1998 to 2004, suggested an association with early generation protease inhibitors (ie, indinavir, saquinavir, and ritonavir) compared to later generation PIs [57]. The implications of these findings were unclear since most of these patients were also taking agents with known neurotoxicity, such as stavudine.

These findings prompted a much larger prospective, observational, multicenter study of current and past exposure to PIs as a risk factor for DSPN in 1159 patients with HIV [58]. Although PI use was associated with an increased risk of DSPN in the univariate analysis, this association disappeared after adjusting for previously validated concomitant risk factors, such as dideoxynucleoside use.

Furthermore, the clinical importance of these findings is less relevant now since none of these agents are used as first line agents for the treatment of HIV. (See "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach".)

Other drugs — Other drugs that can lead to the onset of DSPN include:

●Vincristine, which may be used to treat Kaposi's sarcoma

●Dapsone, which may be used to treat or prevent Pneumocystis jirovecii (formerly carinii) pneumonia

●Thalidomide, which may be used to treat aphthous ulcers

●Isoniazid and ethambutol, which are used to treat tuberculosis

●Nevirapine, a non-nucleoside reverse transcriptase inhibitor, used to treat HIV infection.

Host genetics — Human mitochondrial DNA sequences have diverged over time because of natural selection and human migration leading to distinct patterns of single nucleotide polymorphisms, called haplogroups [59]. Two studies in both non-Hispanic White [55] and non-Hispanic Black patients [60] suggest that variations in these haplogroups may explain host susceptibility to mitochondrial drug injury.

CLINICAL MANIFESTATIONS — DSPN usually manifests as bilateral tingling, and numbness in the toes. The neuropathy gradually spreads proximally in the lower extremities, with only rare involvement of the upper extremities. The spread of sensory symptoms usually occurs over weeks to months. Neuropathic pain is common and may be the presenting symptom [61].

Neurologic examination shows sensory loss to all sensory modalities (vibration, pinprick, temperature) in a stocking distribution, while deep tendon reflexes are reduced or absent at the ankles and occasionally at the knees in more severe cases [4]. Distal weakness in the lower extremities can occur, although most patients have only sensory symptoms and signs. Sensory findings in the hands are more commonly associated with drug toxicity. HIV-related DSPN may evolve from painful to painless numbness.

The presence of brisk knee reflexes in patients with sensory loss raises the possibility of coexistent myelopathy, while the presence of proximal weakness or diffuse areflexia should prompt consideration of acquired inflammatory demyelinating polyradiculoneuropathy, such as Guillain-Barre syndrome. (See "Guillain-Barré syndrome in adults: Pathogenesis, clinical features, and diagnosis".)

DIAGNOSIS — The diagnosis of peripheral neuropathy syndromes in patients with HIV is based mainly upon the clinical picture and physical examination.

Features that would prompt further evaluation, such as electromyography (EMG) and nerve conduction studies (NCS), may include significant weakness or asymmetry of signs. These findings may raise the possibility of alternative diagnoses (eg, acquired demyelinating polyradiculoneuropathy or vasculitic neuropathy). (See 'Other modalities' below.)

Laboratory testing — The evaluation of distal symmetrical polyneuropathy should include blood work to screen for other causes of this type of neuropathy. A typical panel would include:

●Hepatitis C antibody

●Vitamin B12 and folate levels

●Thyroid stimulating hormone assay

●Blood glucose

●Blood urea nitrogen and creatinine

●Serum protein electrophoresis and immunoelectrophoresis

●Syphilis testing

Although these laboratory tests are considered routine in the evaluation of DSPN, they are usually unremarkable in HIV-related or drug-induced polyneuropathy.

Other modalities — Other testing modalities include a subjective peripheral neuropathy screening test, electrodiagnostic studies, skin biopsy for epidermal nerve fiber density analysis, and nerve biopsy.

●Electrodiagnostic studies show a sensorimotor polyneuropathy, which is predominantly axonal [1]. Nerve conduction studies usually confirm a length-dependent axonal polyneuropathy, distinguishing DSPN from acquired inflammatory demyelinating neuropathy [1].

●Skin biopsy for epidermal nerve fiber density analysis has been shown to correlate with neuropathy severity, level of neuropathic pain, and sensory amplitudes on electrodiagnostic studies [62]. An abnormal skin biopsy predicts conversion to symptomatic DSPN in patients with no or asymptomatic neuropathy [63]. Skin biopsy can be positive, whereas electrodiagnostic studies may be negative, in patients with predominantly small nerve fiber involvement and thus can be a useful confirmatory test for early or predominantly small fiber DSPN.

●Nerve biopsy is not usually required but occasionally is performed in severe cases to exclude a confluent presentation of mononeuropathy multiplex. Biopsies show axonal loss with frequent foci of inflammation in the endoneurium or around perineurial blood vessels [33,44,46,64]. The severity of cases is judged clinically (eg, significant weakness on examination) and by electrodiagnostic tests (eg, significant axon loss).

TREATMENT — Treatment options for HIV-related and drug-induced distal symmetrical polyneuropathy are limited.

Impact of antiretroviral therapy — Antiretroviral therapy (ART) is recommended for all patients with HIV infection. ART appears to reduce the risk of distal symmetrical polyneuropathy (DSPN). (See 'Effect of antiretroviral therapy on prevalence' above and "When to initiate antiretroviral therapy in persons with HIV".).

For patients with established DSPN, however the effect of ART on its severity is unclear. There is some evidence showing improved quantitative sensory measures in patients responding to ART [64-66].

Correcting reversible risk factors — If a potentially neurotoxic drug is being used, such as stavudine (d4T), didanosine (ddI), or thalidomide, it should be discontinued. There may be a "coasting phenomenon," in which the neuropathy worsens for one to six weeks following reduction in the dose of a nucleoside analog [44,46,47,67]. Gradual improvement then occurs; the time to recovery depends upon the dose and varies from 3 to 19 weeks [44,67].

It is also important to address any nutritional or metabolic issues that may be contributing to DSPN. (See "Overview of polyneuropathy", section on 'Etiology and pathogenesis'.)

Symptomatic approach — Management of polyneuropathy is largely symptomatic and usually aimed at ameliorating the painful dysesthesias. Since abnormal processing of impulses from neuronal receptors contributes to neuropathic pain, the pharmacologic approach includes multiple possible targets with drugs of different classes. These include anticonvulsants, antidepressants, topical analgesics, anti-inflammatories, and opioids for recalcitrant symptoms, as outlined below.

Our approach is consistent with recommendations from the HIV Medicine Association guidelines for the management of chronic pain in patients with HIV infection [68]. Data informing the use of these agents in populations with HIV are limited [69]. Support from the management approach comes mainly from studies of peripheral neuropathy in other disease states, such as diabetes. (See "Overview of polyneuropathy".)

Gabapentin as initial therapy — We suggest gabapentin as initial therapy (titrated up to 1200 or 2400 per day, divided into two or three daily doses). Gabapentin is widely used in the treatment of neuropathic pain associated with diabetes mellitus and other diseases. (See "Management of diabetic neuropathy", section on 'Gabapentin' and "Pharmacologic management of chronic non-cancer pain in adults", section on 'Gabapentin and pregabalin'.)

Data informing the benefit of gabapentin in patients with HIV are limited. A placebo-controlled trial in 26 patients with HIV and DSPN found that gabapentin, initiated at 400 mg daily and titrated up to 1200 or 2400 mg daily, reduced painful symptoms and sleep disturbance at four weeks of follow-up compared with baseline; however, pain and sleep disturbance scores were not statistically different between the two groups at the end of the study [70]. Two larger and longer multicenter trials did not demonstrate a benefit with pregabalin, a drug with a similar mechanism of action, for painful HIV-associated neuropathy [71].

Second-line therapy — If initial therapy with gabapentin is ineffective, we choose a second-line therapy based on a patient's comorbidities. If there is no improvement, we typically switch gabapentin to a second-line agent; if there was partial improvement, we add a second-line agent. For those with concomitant depression, antidepressants, such as nortriptyline or duloxetine, are attractive options. We use topical capsaicin for those who have contraindications to, cannot tolerate, or prefer to avoid systemic medications.

●Antidepressants – Antidepressant options for symptomatic treatment of polyneuropathy include tricyclic agents (amitriptyline or nortriptyline, starting at 10 mg at bedtime and increased as tolerated to 75 to 100 mg until there is symptomatic improvement), duloxetine (60 mg daily), and venlafaxine (37.5 to 225 mg daily). The choice among these generally depends on comorbidities and cost (eg, tricyclic agents are generally the least expensive but should be avoided in those for whom anticholinergic side effects, such as orthostasis or constipation, would be problematic).

Studies evaluating the effect of antidepressants for HIV-associated DSPN are limited, and two double-blinded trials did not show benefit of amitriptyline over placebo [72,73]. Support for the use of antidepressants in HIV-associated DSPN comes mainly from data showing efficacy for neuropathic pain in other populations. (See "Management of diabetic neuropathy", section on 'Administration and efficacy'.)

●Topical capsaicin – Capsaicin is the active ingredient in hot chili peppers and may desensitize peripheral endings of cutaneous afferent neurons. In a placebo-controlled trial of 307 patients with HIV and painful DSPN, an 8% capsaicin patch, applied directly over the painful site for 90 minutes, reduced pain to a greater degree and in a greater proportion of patients than a low-concentration capsaicin patch [74]. In an integrated analysis of this and a second similar trial, a 30-minute patch application also reduced pain compared with the low-concentration control patch. The optimal application time remains to be determined. Local burning at the site of the patch can be ameliorated by pretreatment with topical 4% lidocaine, which is wiped off prior to capsaicin application. Capsaicin cream (0.075%) has been shown to be no better than placebo [68,75].

Other options for select patients — Other options, such as cannabis, alpha lipoic acid (ALA), and opioids, may have a role in the management of select patients with HIV-associated DSPN.

For patients who are interested in using smoked cannabis and have access to medical cannabis, small trials have demonstrated a benefit [76,77]. In one trial of 50 patients with HIV-associated DSPN, smoked cannabis (three times daily, in addition to baseline symptomatic therapy) reduced daily pain from DSPN by 34 percent compared with 17 percent with placebo cigarettes [76]. No serious adverse events were reported, although all patients had prior cannabis exposure. Other considerations for use include the potential of cannabis use disorder and the potential for negative neuropsychiatric effects, particularly in those who are naïve to cannabis. (See "Cannabis use: Epidemiology, pharmacology, comorbidities, and adverse effects".)

For patients with pain refractory to other therapies, ALA (600 mg once daily) is a potent antioxidant that has demonstrated to reduce painful symptoms associated with diabetic neuropathy. Although not evaluated for HIV-associated DSPN, it is a reasonable option as an opioid-alternative given its safety and efficacy in other populations [68].

Otherwise, opioids, such as tramadol (50 to 100 mg three times daily), oxycodone, or fentanyl patch, can be added to the above mentioned therapies for breakthrough pain or to control recalcitrant symptoms, but these should only be used with caution given the uncertain benefit for neuropathic pain [78] and the potential for adverse effects, misuse, and addiction. If used, opiates should be administered as a time-limited course at a low dose that is gradually titrated to alleviate symptoms. Written patient-doctor contracts regarding opioid use are often helpful; referral to a pain specialist for evaluation is strongly recommended [79]. (See "Use of opioids in the management of chronic non-cancer pain".)

Interventions with limited benefit — Non-steroidal anti-inflammatory agents are not very effective for neuropathic pain control and their chronic administration can be associated with adverse events, such as exacerbation of underlying renal disease.

Other options that have been shown to have no or limited efficacy for HIV-associated DSPN include:

●Lamotrigine [80,81]

●Lidocaine gel [82]

●Mexiletine [72,83]

●Memantine [84]

●Prosaptide, a polypeptide [85]

●Acupuncture [73]

SUMMARY AND RECOMMENDATIONS

●Distal symmetric polyneuropathy (DSPN) is the most common neurologic manifestation in patients with HIV. Risk factors for DSPN include advanced immunosuppression, level of HIV viremia, aging, diabetes, and nutritional deficiencies. (See 'Epidemiology' above.)

●Certain antiretroviral agents are associated with DSPN, including stavudine, didanosine, and nevirapine. (See 'Pathogenesis' above.)

●DSPN usually manifests as bilateral tingling, and numbness in the lower extremities. Neuropathic pain is common and may be the presenting symptom. Neurologic examination demonstrates sensory loss to all sensory modalities in a stocking distribution with reduced or absent deep tendon reflexes at the ankles. (See 'Clinical manifestations' above.)

●The diagnosis of peripheral neuropathy syndromes in patients with HIV is based upon the clinical picture and physical examination. (See 'Clinical manifestations' above.)

●Patients with features that suggest an alternative diagnosis, such as significant weakness or asymmetry of signs, should undergo electromyography (EMG) and nerve conduction studies (NCS). (See 'Diagnosis' above.)

●All patients who are being evaluated for DSPN should have laboratory testing for B12/folate, thyroid stimulating hormone assay, random glucose, creatinine, serum protein electrophoresis and immunoelectrophoresis, hepatitis C antibody, and syphilis testing. (See 'Diagnosis' above.)

●Treatment options for HIV-related and drug-induced distal symmetrical polyneuropathy are limited. Antiretroviral agents that are associated with DSPN (eg, stavudine, didanosine, and nevirapine) should be discontinued. (See 'Treatment' above.)

●Management of DSPN is mainly aimed at ameliorating the painful dysesthesias. We suggest gabapentin for initial treatment (Grade 2C). If ineffective, we choose a second-line therapy based on a patient's comorbidities. Options include antidepressants, such as nortriptyline or duloxetine, for those with concomitant depression and topical capsaicin for those intolerant of systemic medications. Opioids are generally reserved for short-term treatment of symptoms refractory to other therapies. (See 'Symptomatic approach' above.)

ACKNOWLEDGMENT — We are saddened by the death of John G Bartlett, MD, who passed away in January 2021. UpToDate gratefully acknowledges Dr. Bartlett's role as section editor on this topic, his tenure as the founding Editor-in-Chief for UpToDate in Infectious Diseases, and his dedicated and longstanding involvement with the UpToDate program.