Peripheral neuropathy in people with HIV in resource-limited settings

This article originally appeared in HIV & AIDS treatment in practice, an email newsletter for healthcare workers and community-based organisations in resource-limited settings published by NAM between 2003 and 2014.
This article is more than 15 years old. Click here for more recent articles on this topic


This clinical review was kindly supported by the Diana, Princess of Wales Memorial Fund.

Contributors and reviewers:

Dr Gretchen Birbeck, Chikankata Epilepsy Care Team, Mazabuka, Zambia; Dr Sarah Cox, Chelsea & Westminster Hospital, London, United Kingdom; Assoc Prof. Somnuek Sungkanuparph, Mahidol University, Bangkok, Thailand, Dr Ana-Claire Meyer, University of California San Francisco (USA) and Kemri (Kenya); Dr Ned Sacktor, Johns Hopkins University, Baltimore, USA; Dr Jonathan Mermin, CDC Kenya; Dr Annelies van Rie, University of North Carolina, USA; Dr Siobhan Crowley, WHO; Dr Zvi Bentwich, Rosetta Genomics, Israel; Chris Green, Spiritia Foundation, Indonesia; Dr Marco Vitória, WHO; Dr Anthony Harries, International Union Against Tuberculosis and Lung Disease; Dr John T Brooks, US Centers for Disease Control; Dr Liz Corbett, London School of Hygiene and Tropical Medicine & Biomedical Research and Training Institute, Harare, Zimbabwe; Dr Henry Dudungu, African Palliative Care Association; Dr Alexandra Calmy, Unité VIH/SIDA, Geneva, Switzerland; Professor Bruce James Brew, University of New South Wales, Sydney, Australia; Dr David Simpson, Dept. of Neurology, Mount Sinai Medical Center, New York, USA; Prof Gabriel Anabwani, Baylor College of Medicine, Director of the Botswana-Baylor Children's Clinical Center of Excellence; Dr Liz Gwyther, Hospice Palliative Care Association of South Africa; Dr Zipporah Ali, Kenya Hospices and Palliative Care Association.


It may start with a slight tingling or numbness in the toes or ball of each foot— something so minor that you don’t think anything of it.

Whether caused by HIV, the toxic effects of medications (like ddI or d4T), alcoholism, diabetes or other factors, damage to the peripheral nerves — peripheral neuropathy — can be insidious. In fact, people commonly attribute their own early symptoms to poor blood circulation, a poor night’s sleep, being cold, or an injury from work or exercise. 

Different people may experience different symptoms, ranging from a loss of sensitivity, to paraesthesias (pins and needles or burning sensations) and an extremely heightened sensitivity to touch. But as it worsens and moves up from the feet to the ankles and legs, the sensory changes become increasingly persistent. And yet, the most common reaction to the first subtle symptoms is to ignore them or perhaps to self-medicate, taking things like paracetamol, herbal remedies or an alcoholic beverage.

Peripheral neuropathy is the most frequent neurological complication in HIV infection. But all too commonly people don’t seek out medical attention or tell their care provider until the condition starts to become debilitating and interferes with sleep and the ability to work.  And since walking and balance may remain normal until the condition becomes severe, healthcare workers who don’t routinely screen for it may not become aware that their patient has peripheral neuropathy until it become very difficult to manage, if not irreversible.

This can even happen to some of the best-informed people with HIV, possibly because their health is otherwise improving on antiretroviral therapy (ART). For instance, the following case study describes the experience of Zachie Achmat , of South Africa’s Treatment Action Campaign.

‘‘I felt so much better, my headaches and infections had cleared up, my energy was flooding back, my CD4 count was rising and my viral load was coming down. My treatment was working. Going into my fifth month I started to feel a tingling sensation in my feet. At first, I dismissed it, thinking I’d done something at the gym.

When I finally told the doctor about my symptoms my feet were so sensitive I could hardly walk. My friends and my doctor had warned me about taking d4T in my combination as I already had mild peripheral neuropathy before I started ARVs because of HIV…

All that was required was for me to change a drug – from d4T to AZT. I didn’t do that quickly and instead nearly incapacitated myself for two, three months and only now am I starting to come out of it.” 1

But at least Achmat’s doctor warned him. In many clinical settings, it is doesn’t seem to be on the radar screen.

“We did some work on d4T toxicity, and the experience in Kinshasa is that most patients do not complain and that nurses are not well trained in recognising it,” said Dr Annelies van Rie of the University of North Carolina.

“In general I don't think neuropathy is so much on the clinician's agenda - up till recently the focus has been on getting drugs to keep people alive,” Professor Bruce Brew, Head of Neurology at the University of New South Wales in Australia told HATIP. “But now that these drugs are increasingly available there has to be a shift in mind-set. Additionally, neurological complications including neuropathy are paid little attention - it is a problem with any disease that spans several specialities.”



Damage to the nerves.

peripheral neuropathy

Damage to the nerves of the hands and/or feet, causing symptoms ranging from numbness to excruciating pain.


A group of diseases characterized by high levels of blood sugar (glucose). Type 1 diabetes occurs when the body fails to produce insulin, which is a hormone that regulates blood sugar. Type 2 diabetes occurs when the body either does not produce enough insulin or does not use insulin normally (insulin resistance). Common symptoms of diabetes include frequent urination, unusual thirst and extreme hunger. Some antiretroviral drugs may increase the risk of type 2 diabetes.


Having symptoms.


Cytomegalovirus (CMV)

A virus that can cause blindness in people with advanced HIV disease.

But recent data suggest that as people with HIV age and spend a longer time on treatment, more and more people may develop peripheral neuropathies, despite effective ART. Although data are currently limited, the situation may prove worse in many resource-limited settings — where antiretroviral drug options are limited — and where people often have multiple risks factors for neuropathy including poor nutrition, diabetes, and the toxic effects of commonly used medications.

As increased vigilance for peripheral neuropathy seems more than warranted, especially since early detection and intervention could help avoid a lot of unnecessary pain and suffering. And at present, finding ways to avoid peripheral neuropathy may be the best way to manage it, because the available treatment options appear to be woefully inadequate.

Background on the spectrum of neuropathies seen in people with HIV or in resource-limited settings:

The peripheral nervous system (PNS) consists of the nerves connecting the central nervous system to the rest of the body. In a general sense, peripheral neuropathy refers to a dysfunction or damage of one or more peripheral nerves.

In HIV disease, when people say peripheral neuropathy, they are usually referring either to:

  • distal symmetrical sensory polyneuropathy (DSPN), sensory neuropathies starting in the extremities (spreading up from the toes, and rarely, from the fingers) more or less on both sides of the body equally and/or
  • antiretroviral toxic neuropathy (ATN), which is clinically indistinguishable.

But a host of things can go wrong with the peripheral nervous system ranging from minor transient neuropathies (such as one’s foot falling asleep) to neurological emergencies that can lead to paralysis, loss of essential bodily functions and then death.

The most common conditions are focal or mononeuropathies (disorders of a single nerve or nerve group) and radiculopathies (disorders affecting the roots of the spinal nerve) that may cause sensory disturbances and/or weakness in one part or on one side of the body, including injury-related mononeuropathies such as carpal tunnel syndrome (compression of the median nerve in the wrist). 

In people with HIV, a number of PNS disorders seem to be either caused by the virus, opportunistic infections or neoplasms, by commonly used medications, or by the immune system’s reaction to infection. These can be roughly distinguished by how the condition is localised (whether the disorder is focal or multifocal, and whether or not it is symmetrical), and whether it is primarily sensory (like DSPN or ATN) versus whether it also causes weakness and loss of motor/autonomic functions — though there can be some overlap in some situations. [This review will not address spinal cord problems, myelopathies or myopathies.]

What do all those initials stand for? A neuropathy glossary

AIDP: acute inflammatory demyelinating polyneuropathies

ATN: antiretroviral toxic neuropathy

BMI: body mass index

CIDP: chronic inflammatory demyelinating polyneuropathies

CMV: cytomegalovirus

CROI: Conference on Retroviruses and Opportunistic Infections

ddI: didanosine

d4T: stavudine

DSPN: distal symmetrical sensory polyneuropathy

EMG: electromyography

GBS: Guillain-Barré Syndrome

MM: mononeuritis multiplex

mtDNA: mitochondrial DNA

PN: peripheral neuropathy

PNS: peripheral nervous system

SPN: symptomatic peripheral neuropathy

Focal/asymmetrical deficits

Focal neuropathies are those which focus on one nerve or one place in the body. People with HIV are more prone to develop lymphomas, which can compress nerves or directly involve the nerve roots, and cause different focal disorders depending upon the site of the lymphoma.2, 3  Similarly, tuberculosis appears to cause focal neuropathies.4, 5, 6 

But the pathogen with the greatest predilection for the nervous system is varicella zoster virus (VZV), which can cause postherpetic neuralgia and radiculopathies with lingering asymmetric pain and motor weakness and sometimes more extensive polyradiculopathies too.7, 8  However, VZV related conditions are generally recognisable because they tend to be associated with the characteristic painful shingles rash.

In people with HIV, a number of things appear to cause mononeuritis multiplex (MM), painful asymmetric sensory and motor peripheral neuropathies involving isolated damage to more than one independent nerve area, usually as the result of a lack of oxygen due to decreased blood flow or inflammation of blood vessels. For instance, this has been reported in people with hepatitis B and C in people with or without HIV.9, 10, 11 

HIV itself can also trigger MM; for instance, a number have been reported to develop a brachial neuritis (Parsonage-Turner Syndrome), characterised by the acute onset of shoulder pain followed by weakness of the related muscles, usually during primary infection.12, 13 Other forms of MM can occur later but as long as the CD4 cell count is above 200, the condition tends to be less severe and self-limiting.14, 15 Bell’s palsy and other isolated cranial/facial nerve palsies have also been observed to occur in people with HIV in early and late stage disease.16, 17, 18 

However, in patients with less than 50 CD4 cells, CMV can cause a severe life-threatening form of MM affecting multiple nerves and muscles, often surrounding the pectoral/shoulder girdle.19, 20 

Also in very late stage HIV disease, CMV can infect and destroy nerve roots around the spinal cord causing CMV polyradiculopathy.21, 22, 23, 24, 25  This is a neurological emergency with rapid progression of symptoms ranging from loss of feeling and reflexes in the lower limbs and back pain to paralysis, urine retention, and loss of normal bowel function. CMV polyradiculopathy can be fatal within days or weeks of presentation unless it is promptly treated with empiric ganciclovir, which can also lead to partial recovery. The condition is usually observed in people with some concurrent CMV disease (such as retinitis, encephalopathy or gastroenteritis). However, there have also been reports of neurosyphilis, VZV, and lymphoma causing polyradiculopathies.26, 27, 28 

Symmetrical sensory and motor polyneuropathies

Other symmetrical neuropathies may also begin with tingling and numbness in the hands or feet, including acute inflammatory demyelinating polyneuropathies (AIDP), such as Guillain-Barré Syndrome (GBS), and chronic inflammatory demyelinating polyneuropathies (CIDP) — but these soon progress with marked muscle weakness and then paralysis in the leg, arms, face, and even the respiratory system (this is rare, but represents another neurological emergency).29, 30 

GBS is an autoimmune reaction that typically occurs within a week or two of a gastrointestinal or respiratory infection — and it appears that HIV can trigger a similar reaction in susceptible people. With AIDP, progression occurs within hours, days or weeks but reaches a maximum within four weeks; with CIDP deterioration may continue over months. GBS can be triggered in people by relatively minor infections, but has been observed in people soon after infection with HIV, including one series from Zimbabwe.31

A similar syndrome of sensory neuropathies and muscular weakness, now being called HIV-associated neuromuscular weakness syndrome, has also been associated with mitochondrial toxicity after taking nucleoside analogues, particularly d4T (stavudine) and is commonly a sign of lactic acidosis.32, 33 Of note, the antibiotic dapsone can also produce a sensory/motor neuropathy with marked muscular atrophy but with a delayed onset.34, 35, 36

Distal sensory (axonal) polyneuropathies

It is usually possible to distinguish most of the above neuropathies from the more common sensory bilateral neuropathies such as DSPN or ATN (which both primarily affect the axon or ganglion) on the basis of clinical features.

If a neuropathy is asymmetric, or if there is a pronounced or rapid development of weakness and loss of motor or autonomic functions, then by definition, it isn’t a sensory neuropathy. The stage of HIV disease at which the symptoms occur is also an important consideration.37

HIV-related sensory neuropathies (or HIV DSPN) can start to develop when CD4 cell counts are higher, but they are more commonly seen in people with CD4 cell counts below 200 (see more below).

The neuropathies tend to have a gradual onset, usually starting with numbness, a sensation of pins and needles or pain in the most distal part of the longest nerves, in other words, in the toes or anterior plantar surface (balls) of each foot, which then spreads up to the rest of each foot, to the ankles and beyond. 38  Similar sensations may start to develop in the fingertips and hands, at about the same time pain begins to reach the middle of the leg, so sensory neuropathies are said to have a stocking and glove distribution.39

People with similar degrees of (measurable) sensory loss may have very different perceptions of the pain, anything from a prickling or tingling discomfort, to tightness and aching, burning sensations, sporadic shooting, stabbing pain or even electrical shocks.40

The pain can be spontaneous or in response to touch. Some people with sensory neuropathy become so sensitive to touch that they cannot tolerate wearing shoes or socks or lying under bedclothes. Pain is often worse at night and can disrupt sleep. At the same time, ability to accurately sense heat and vibrations may also become impaired — which seems to support the hypothesis that the pain actually originates from uninjured nerve fibres after injury to neighbouring fibres.41

Motor function is preserved for the most part, and walking and balance can appear normal. If the gait is changed, it is usually as the result of an effort to avoid pain. Weakness is uncommon, though there can be somewhat reduced or absent deep tendon reflexes at ankle jerks.

Case study (paraesthesias)

A 42-year-old man has been taking didanosine, stavudine and nevirapine for five months. He also takes cotrimoxazole 960 mg daily. He has gained weight and is pleased with his progress.

For several months, he has experienced tingling, numbness and pain in both his lower extremities, especially his toes. At first he took paracetamol, which offered little relief. At the last visit, a month ago, you added amitriptyline to his regimen. Today, he can hardly walk because of his foot pain. He states that the pain keeps him awake at night.

On examination, his lower extremities appear normal. Dorsalis pedis and posterior tibial pulses are present bilaterally. Ankle jerk reflexes were diminished bilaterally. 42

The man in the case study above has ATN, which is clinically indistinguishable from HIV-related sensory neuropathy 43, 44. However, the onset may be more acute, and the pain somewhat more severe; for instance, there could be more reports of deep aching pain, particularly across the top of the foot.45 The only real way to tell ATN apart from HIV sensory neuropathy is that it appears one week to six months after starting an ART regimen containing d4T or ddI (especially when used in combination) — and may resolve eventually in up to two-thirds of patients upon discontinuing the offending drug.46 Again, the most notable clinical feature is pain — if instead, there is rapidly progressive weakness and high lactate the clinician should be alerted to HIV-associated neuromuscular weakness syndrome and lactic acidosis.

Other toxic sensory neuropathies

Many toxins, solvents, insecticides and medications can also cause peripheral neuropathies. Perhaps most importantly for people with HIV, this includes isoniazid when it is given without supplemental vitamin B6 (pyridoxine), but also ethambutol, ethionamide, vincristine, thalidomide, metronidazole, high-doses of vitamin B6 and other drugs. The Merck Manual Online has a table listing the toxic causes of neuropathy, categorised by the types of neuropathy each drug causes. Most of the medications that may be used by people with HIV cause sensory neuropathies, though, as already mentioned, motor problems are a big part of the neuropathy on dapsone.

“Although [d4T and other d-drugs] are undoubtedly major players in the risk for PN there are other contributors,” Dr John T Brooks, of the US Centers for Disease Control told HATIP. “Alcohol or solvent (e.g. glue sniffing) abuse are well known causes and these behaviours are not rarely encountered among HIV-infected persons. Indeed, persons seeking symptom relief for peripheral neuropathy might engage in these behaviours, which in turn might exacerbate their peripheral neuropathy.”

Indeed, a history of alcohol overuse is associated with gradually progressive sensorimotor neuropathies, probably due to both toxic effects of ethanol and metabolites, and because of nutritional deficiencies that are often associated with alcoholism (in particular thiamine deficiency).47, 48 Malnutrition and nutritional deficiencies — either inadequate intake or an inability to absorb or properly utilise certain nutrients such as B vitamins, especially vitamin B12 — have also been linked with neuropathy.49

In addition, many other things commonly cause, or could aggravate sensory neuropathy in resource-limited settings. For instance, diabetes (type I and II) is associated with many different types of neuropathy (mononeuropathies, diffuse neuropathies, loss of motor or autonomic functions, as well as distal sensory polyneuropathies, etc) that can lead to severe pain, foot injuries/ulcers (so severe as to require amputation), and death in cases of autonomic neuropathy.50, 51 Globally, about 10% of the population with diabetes have developed symptomatic neuropathy. Symptoms are more likely to occur in diabetics with poor glycaemic control, and there is often evidence of concurrent diabetic complications such as nephropathy, retinopathy etc. Although muscle weakness and loss of reflexes tend to be more common in people with diabetic neuropathy, many cases can be difficult to distinguish from HIV-related neuropathies.

In resource-limited settings, leprosy is still one of the most common causes of neuropathy with a range of clinical presentations.52, 53 Since symptoms of neuropathy can occur in the absence of skin lesions, it may take a nerve biopsy to be able make a definitive diagnosis of leprosy.54, 55 

It must be noted that even when these other conditions cause primarily sensory neuropathies that are linked to axonal damage, they don’t all cause the disorder in the same way. In fact, there are far more theories about how HIV and each of these illnesses might cause sensory neuropathies than there are ideas about how to treat them. It is safe to say that the pathogenesis of sensory neuropathies is multifactorial. Nevertheless, having any of these other conditions at the same time as HIV may increase the likelihood of developing HIV-related DSPN or ATN.56

Epidemiology and risk factors for sensory neuropathies (DSPN and ATN)

It is difficult to pin down just how common HIV-related DSPN and ATN really are because of changes in the standard of care, differences in populations involved in the studies, and how neuropathy was measured in a study.

For instance, some studies look at easily measured clinical symptoms, others measure loss of vibration or heat sensation and reflexes the patient may not be aware of, and still others use electromyography (EMG), and nerve conduction studies that can detect subclinical nerve damage. One early review which included some autopsy studies found evidence of peripheral nerve damage in almost everyone with HIV — even though only 40% had symptoms.57, 58

Based mostly on studies from the US, it is generally estimated that at least one-third of people with advanced HIV disease develop symptomatic HIV-related DSPN.59 Another 15 to 30% of those who take ddI, d4T or ddC appear to develop ATN.60, 61 However, with the discontinuation of ddC (by far the most toxic nucleoside analogue) and recommendations against using ddI and d4T together, those figures could be lower.

In one early pre-ART report, clinical and EMG evidence of a distal symmetric polyneuropathy was found in 35% (13/37) of hospitalised patients without other risk factors for neuropathy. Most of the symptoms were mild, however.62 Another prospective study from the pre-ART era in people with less than 200 CD4 cells, reported that 36% developed symptomatic sensory neuropathy within the first year of follow-up, 52% within 24 months.63 Soon after the introduction of ART, the yearly incidence of peripheral neuropathy was reported to fall to about 21%. It is not clear that the news remains as good for people on prolonged treatment however

Recent evidence from two major studies from the US, presented this year at the Conference on Retroviruses and Opportunistic Infections (CROI) suggests the degree of immunosuppression at the time ART is commenced remains an important predictor for sensory neuropathies — while the current CD4 cell count and/or viral load is not.

At entry into the first study, called CHARTER, 57% of 1539 participants were reported to have at least one sign of DSPN (and/or ATN).64, 65 Signs of DSPN included diminished ability to recognise vibration on bilateral great toes, reduced ability to reliably discriminate sharp from dull in the lower legs, feet and toes, and absent or weakened bilateral ankle reflexes compared to knees as assessed by a trained clinician using a standardised trial protocol. However, only 61% of these actually reported experiencing symptoms — in other words, a little over one-third of the entire cohort. Only 15% of those with at least one sign of DSPN reported moderate or severe pain, although reduced quality of life was reported in some subjects with only slight and mild pain (overall, 23% of those with signs of DSPN).

It is important to note that CHARTER had sought to enrol typical clinic attendees at six sites in the US. At baseline, the typical study participant was 43 years old. 77% were male; 49% were African American; 63% had already received an AIDS diagnosis, 37% had a history of previous (but not current) opportunistic infections. 26% were positive for hepatitis C. The baseline CD4 cell count was 420 (256-603), but the nadir CD4 (the lowest CD4 cell count before going onto ART) had been around 174 (49-300). The median plasma HIV RNA load was 2.3 log/mL (1.7-4.0), but 59% had detectable viral loads above 50 copies/mL. 71% were currently on ART.

In multivariate analysis, significant risk factors for DSPN (in order of importance) was older age, prior d-drug use (i.e., d4T, ddI or ddC), currently being on ART (generally a sign of having had more advanced disease), a history of opiate abuse or dependence, and having a very low CD4 cell count before starting ART.

Factors not significant in the multivariate regression were gender, CD4 recovery, plasma viral load, and history of alcohol abuse or dependence and, notably, current d-drug use (although it should be noted that the use of all the ‘d’ drugs, including d4T has fallen out of favour in the US and most of the industrialised world).

Nevertheless, it is worrisome that the effects of DSPN and ATN seem to be lingering, particularly in the older subjects who once had more advanced disease. While 20% of 20-29 year olds had at least one symptom of DSPN, this proportion climbed to 60% in 40-49 year olds and 75% in 50-59 year olds.

And what appears clear from an analysis of the AIDS Clinical Trials Group (ACTG) Longitudinal Linked Randomized Trials (ALLRT) cohort is that DSPN and/or ATN is growing more common despite suppressed viral loads, higher CD4 cell counts on ART — and a decline in d-drug usage.66

In one of the ALLRT substudies, trained non-neurologist staff administered the Brief Peripheral Neuropathy Screen (BPNS), which assesses signs (vibration sensation and ankle reflexes) and symptoms (pain, “pins and needles”, and numbness) in the study participants every 48 weeks. PN was defined as at least mild loss of vibration sensation in both great toes bilaterally, or absent or hypoactive ankle reflexes bilaterally relative to knees. Symptomatic PN (SPN) was defined as PN plus bilateral symptoms. 

At present, the study involves 2135 participants who were ART-naïve at study entry: 81% male, 44% white, 32% black, median age of 39 years, median viral load 4.9 log/ml and median CD4 count of 206 at ART initiation.  Of note, the following analysis presented at CROI reported only on those participants with a good response to treatment (CD4 cell counts over 350 and viral loads below 400 copies/ml (undetectable for this study).

For instance, at week 48, 29.5% of those with undetectable viral load (n=1253) had PN, 10.1% had SPN (slightly less for those with CD4 cell counts over 350). But neuropathy was worse after each 48 week period of treatment. At week 384, follow-up data are available for 331 subjects with undetectable viral load: 44.4% have PN, 14.2% SPN.

Of note, d-drug use in the cohort peaked at week 144 (at 25%) dropping to 10.9% at week 384. Even so, PN appears to increase over time after initiation of ART. The study notes that most of this PN is not causing much pain (although around 15% of those with PN reported moderate to severe pain). However, it bears repeating, these are just the people with good virologic responses to treatment.

In multivariate analyses, age, d-drug use and viral load at baseline were significantly associated with both the likelihood of PN and SPN.

Neuropathies in non-white populations

Another finding from both CHARTER and the ALLRT studies, is that race did not seem to affect the likelihood of developing peripheral neuropathy.

This is in contrast to the findings from an earlier North American study, the HIV outpatient study (HOPS) where 490 (22.5%) out of 2178 people with HIV had peripheral neuropathy, and the data suggested that white race increased the risk of peripheral neuropathy (adjusted odds ratio (aOR) 1.26; 95% CI: 1.02-1.56; p=0.033).67

However, other factors were much more strongly associated with peripheral neuropathy in HOPS, including (once again) age over 40 years, nadir CD4 cell count <50 cells/mm, nadir CD4 50-199 cells/mm, viral load >10,000 copies/ml (all p<0.001) and diabetes (p=0.012).

But it would be premature to conclude that the risk of sensory neuropathy will be the same across the world, without significant differences between populations. There are a number of inherited characteristics that affect susceptibility to disease, drug metabolism and toxicity. For instance, several recent studies have been exploring whether inherited differences in the haemochromatosis gene, involved in iron absorption and regulation in the body, and genetic polymorphisms (mutations) affecting the production of inflammatory cytokines such as TNF-alpha or interleukin-12B might be involved in susceptibility to neuropathy.68, 69, 70 

Researchers have also proposed that inherited differences in mitochondrial DNA (mtDNA, the maternally inherited DNA inside of mitochondria, organelles that are involved in cellular energy production (ATP synthesis), managing oxidant stress and apoptosis) may also play a role in susceptibility to d-drug related mitochondrial toxicity and thus, possibly, peripheral neuropathy.71

At CROI this year Dr Todd Hulgan of Vanderbilt University presented the first data from African-Americans with HIV suggesting that people from one subgroup of mtDNA (L1c) could be three times more like than other African-Americans to develop peripheral neuropathy when taking a d-drug containing regimen.72 Dr Hulgan has previously published data suggesting that people with the mitochondrial haplogroup T (one of about a dozen mitochondrial haplogroups found in Europe), were almost five times as likely as other Caucasian participants in ACTG 384 to develop peripheral neuropathy when taking ART regimens containing ddI/d4T.73

Dr Hulgan cautioned that the sample size is rather small, and that no clear mechanism has been identified that might explain increased susceptibility to peripheral neuropathy.

“Studies to replicate this association in other African-American and African populations, are warranted and also in other ART-associated toxicity phenotypes,” he said.

Indeed, there is no replacement for doing studies within Africa. The handful of studies that have been done report a significant burden of DSPN and PN in Africa and most other resource-limited settings — but rates do vary a bit and it is difficult to make comparisons between studies using different methods to assess neuropathy.

According to a pre-ART study from Zimbabwe 60% of people with HIV had peripheral neuropathies, 37% were symptomatic, but these included some neuropathies other than DSPN (such as AIDP).74, 75 In a more recent study in Uganda, 47% of people with HIV were reported to have signs of peripheral neuropathy, though only 9% were symptomatic.76 In a more recent report from the same cohort, symptoms of neuropathy developed in 38% of previously asymptomatic HIV+ patients after initiation of d4T-containing ART.77

Also in Uganda, Forna et al reported that peripheral neuropathy was by far the most commonly reported toxicity among 1029 subjects receiving home-based ART in Tororo district, with 36% developing ATN, 9% severe.78 Severe neuropathy was defined as having pain, moderate weakness in the feet, mild weakness in the hands, or severe sensory loss in the extremities, causing moderate interference with ambulation.  Note, this analysis excluded 135 patients (13.1%) who had DSPN before going onto ART. In a multivariate analysis, age >35 years was associated with increased hazard of any peripheral neuropathy. Age >35 years and tuberculosis treatment at baseline were associated with increased hazard of severe peripheral neuropathy. Note TB treatment was given with pyridoxine.

“We believe that clinicians should pay increased attention to neuropathy, especially in persons older than 35 years and in those receiving treatment for tuberculosis,” Forna et al wrote.

In Zambia, 32.2% of the HIV-positive clinic patients self-reported experiencing peripheral neuropathy symptoms (of tingling, burning, or numbness in their feet or hands) before going onto ART according to a report from Dr Gretchen Birbeck, of the Chikankata Epilepsy Care Team, Mazabuka, Zambia.79  17.3% more reported developing new symptoms after going onto ART (Triomune).

HATIP asked Dr Birbeck why these rates of symptomatic neuropathy seem so much higher than reported elsewhere in Africa.

“Standard Triomune is used here without any [d4T dose] adjustments, though the BMI of our patient population is quite small relative to the US. There are no population-based data at all out there that I know of regarding [the prevalence of] malnutrition here. We published a hospital-based period prevalence study of neurologic conditions here and have a similar one almost ready for submission from the University Teaching Hospital in Lusaka where 24.1% of outpatient visits to neurology were for neuropathies - the most common single reason for neurology consultation.

“The key issue is that neuropathies are common, but folks often have multiple reasons for neuropathies so aetiology is hard to sort out without punch biopsies. For example, people might have poor nutritional status but have also been on isoniazid (INH) without vitamin B 6 (pyridoxine) — which isn't routinely available for the TB cases on long-term INH who are already nutritionally marginalised.  Diabetes isn't that uncommon here. Leprosy is still a problem too.”

She noted that a new EMG/NCV machine is now available in neighbouring Malawi, and that she would like to do “some decent work on this issue in the future.”

In South Africa, there are surprisingly few data. One study reported that 6% of patients put onto a d4T-based regimen switched due to peripheral neuropathy.80 In a recent study at Kalafong Hospital HIV Clinic in Pretoria, a questionnaire was administered to 354 patients, 20.9% of whom reported having neuropathic pain prior to starting ART.81 This pain was significantly more frequent in patients who were male, had lower CD4 counts or higher viral load levels, and those on TB treatment. 80% of these reported significant pain. Note this was assessed by a questionnaire and was not a diagnosis of DSPN, since postherpetic neuralgia and other conditions can also cause neuropathic pain. However, the pain was localised in the lower limbs in 84% and thus more likely to be peripheral neuropathy. 

Patients with lower CD4+ counts (p=0.009) and those with higher viral load values (p=0.006) showed a significantly higher prevalence of neuropathic pain, as did patients with a history of recent or current treatment for tuberculosis (p<0.001). The authors concluded that the prevalence of neuropathic pain, “20.9% is lower than expected in comparison with previous studies in Africa and the USA.” Even so, 20% with pain before going onto ART is nothing to sneeze at. Furthermore, it is hard to compare cohorts directly. For instance, simply having a younger cohort could result in finding less symptomatic pain.

Indeed, Professor Gabriel Anabwani of Baylor College of Medicine and Director of the Botswana-Baylor Children's Clinical Center of Excellence told HATIP: “We have about 600 children who have been treated with a d4T-containing regimen for up to six years in an ongoing RCT. Only a small number have had lipodystrophy and probably none has experienced peripheral neuropathy.”

Furthermore, an analysis of the toxicity of dual nucleoside analogues in over 2233 children with HIV under 13 years of age in the US study, AIDS Clinical Trials Group 219C, reported that less than 1% developed peripheral neuropathy — even on ddI/d4T.82 The question is, whether they will start to develop neuropathies if they remain on the same regimens when they are older.

But there do appear to be some significant differences by population.

Associate Professor Somnuek Sungkanuparph of Mahidol University, in Bangkok told HATIP:

“I always had a belief, but no time to prove, that ethnicity has a big influence on peripheral neuropathy. I have cared for HIV-infected patients for more than eight years and now we have a big HIV cohort in my hospital [>2000 patients]. We find that the rate of peripheral neuropathy is very low (2-3%), compared to that reported in whites and Africans, although we always look for it. Most of the cases with peripheral neuropathy also had diabetes mellitus (fair to poorly controlled) and older age (40+).”

He notes that they had slightly higher rates, 5-8%, when they used ddI/d4T, which they no longer do.

Published studies from Thailand do show somewhat lower rates of peripheral neuropathy.83 For instance, a study from Bangkok reported an overall frequency of probable HIV DSPN of 10% (symptomatic) and that of possible DSPN at 28% among HIV-infected subjects attending an outpatient ID clinic in Bangkok, Thailand.

Are short people better off?

But perhaps the difference isn’t genetic. Perhaps it is something much more basic, like height.

Another study presented at CROI reported that, along with age and increasing treatment exposure, being taller puts one at increased risk of developing ATN — whether you are from Australia, Indonesia or Malaysia (and time will tell whether this observation bears out in other settings such as Africa).84

Dr Catherine Cherry of Monash University in Melbourne, Australia, presented findings from the study, which sought to identify risk factors associated with sensory neuropathy amongst HIV patients in the Asia-Pacific region — and to assess whether these risk factors could be used to predict sensory neuropathy in patients who were asymptomatic before d4T exposure. The idea being that, if there is no choice but to use d4T in most patients, those who are most likely develop sensory neuropathy can be offered an alternative regimen — if they can be identified in advance.

So screening programmes for sensory neuropathy were set up in Melbourne (where d4T is very rarely used now and TB and therefore, isoniazid use is uncommon), Kuala Lumpur (where about half the patients use d4T, TB is a common opportunistic infection and therefore isoniazid is widely used), and Jakarta (routine d4T use and TB the most common opportunistic infection). The ACTG Brief Peripheral Neuropathy Screen (BPNS) was used to look for the presence of symptoms and signs of sensory neuropathy, and the participant’s height, age, and weight were recorded along with demographic, laboratory, and treatment data obtained from the medical file. Close to 100 participants were enrolled at each site.

“What we found is that overall, neuropathy is highly prevalent throughout our region, ranging from a low of 19% in Kuala Lumpur up to 42% in Melbourne and 34% in Jakarta," said Dr Cherry. In addition to treatment exposures, increasing age (p = 0.002) and height (p = 0.001) were independently associated with sensory neuropathy risk.

An analysis of the participant characteristics suggested “cut offs” of ≥170 cm height and age ≥40 years might be useful for predicting patients at risk of neuropathy. These were applied retrospectively to 181 d4T-exposed patients who were asymptomatic before starting d4T. Patients who were younger and shorter had a sensory neuropathy risk of 20%. A third of those who were younger but taller developed neuropathy,  compared to 38% of older but shorter patients.

“But what I find really horrifying is that amongst patients who did not have neuropathy symptoms, but who are older than 40 and taller than 170 cm, fully two-thirds - 66% - developed neuropathy if given stavudine,” said Dr Cherry.

These observations were consistent across sites — neither weight nor ethnicity affected the risk.

“So the take home message that I get from this is that we could have used age and height to predict neuropathy risk, prior to stavudine use, in the countries where we were looking.”

Dr Cherry currently has a project underway in South Africa to investigate whether these observations hold true in African settings as well.

This is one of two articles on peripheral neuropathy published in HIV & AIDS treatment in practice in March 2009. Click here for the second article: Peripheral neuropathy: diagnosis and treatment.

Key points

  • Neuropathy is a common symptom in people with HIV. It may be caused by antiretroviral drugs, by HIV disease, by an opportunistic infection, by isoniazid, or by the effects of concomitant conditions such as diabetes, alcoholism, leprosy.
  • Neuropathy that develops fast, or which is focused in one place, is usually related to diabetes, to an infection such as CMV, to lymphoma, to an autoimmune reaction or an infectious cause.
  • HIV or drug-related neuropathy tends to emerge more slowly, affecting both limbs (although diabetes or leprosy may also have this effect in some).
  • Drug-related neuropathy is caused by d4T (stavudine) or ddI.
  • HIV-related neuropathies occur more frequently in people with severe immune suppression, even after they start treatment and even if they do not take d4T
  • The likelihood of developing peripheral neuropathy may vary by ethnic group, but more evidence is needed.
  • There is evidence that older people are at higher risk of developing peripheral neuropathy, and taller people also have a higher risk. d4T appears to exacerbate this risk further in older, taller people. Age and height may prove to be useful screening tools for determining whether patients should avoid d4T altogether.

[1] TAC. ARV’s in our lives. A handbook for people living with HIV and treatment advocates in support groups, clinics and communities (

[2] Berger JR et al. Cranial neuropathy heralding otherwise occult AIDS-related large cell lymphoma. J Clin Neuroophthalmol 13: 113-18, 1993.

[3] Enting RH, Essenlink RA, Portegies R. Lymphomatous meningitis in AIDS-related systemic non-Hodgkin’s lymphoma: a report of eight cases. J Neurol Neurosurg Psychiatry 57:150-3, 1994.

[4] Andersson MI, Willcox PA. Tuberculous tenosynovitis and carpal tunnel syndrome as a presentation of HIV disease. J Infect.;39(3):240-1, 1999.

[5] Orrell R et al. Peripheral nerve granuloma in a patient with tuberculosis. J Neurol Neurosurg Psychiatry. 73(6): 769–771, 2002.

[6] Hernando I et al. Sensory-motor polyneuropathy associated with lung tuberculosis: presentation of two new cases]. Rev Neurol. 24(131):841-2, 1996.

[7] Nagel MA, Gilden DH. The protean neurologic manifestations of varicella-zoster virus infection. Cleve Clin J Med. 74(7):489-94, 496, 498-9, 2007. (available online)

[8] Mueller NH et al. Varicella zoster virus infection: clinical features, molecular pathogenesis of disease, and latency. Neurol Clin ;26(3):675-97, viii, 2008.

[9] Caniello M et al. Confluent peripheral multiple mononeuropathy associated to acute hepatitis B: a case report. Rev Inst Med Trop Sao Paulo;44(3):171-3, 2002.

[10] Heckmann JG et al. Neurological manifestations of chronic hepatitis C. J Neurol;246(6):486-91, 1999.

[11] Boukhris S et al. Polyneuropathy with demyelinating features in mixed cryoglobulinemia with hepatitis C virus infection. Eur J Neurol. ;13(9):937-41, 2006.

[12] Louis E et al. Bilateral amyotrophic neuralgia (Parsonage Turner syndrome) with HIV seroconversion. Rev Neurol (Paris)159(6-7 Pt 1):685-7, 2003.

[13] Calabrese LH et al. Acute infection with the human immunodeficiency virus (HIV) associated with acute brachial neuritis and exanthematous rash. Ann Intern Med 107:849-51, 1987.

[14] Said G, Lacroix C. Primary and secondary vasculitic neuropathy. J Neurol 252:633-41, 2005.

[15] Lipkin WI et al. Inflammatory neuropathy in homosexual men with lymphadenopathy. Neurology 35:1479-83, 1985.

[16] Serrano P et al. Bilateral Bell palsy and acute HIV type 1 infection: report of 2 cases and review. Clin Infect Dis.;44(6):e57-61, 2007.

[17] Bélec L et al. Peripheral facial paralysis and HIV infection: report of four African cases and review of the literature. J Neurol. 236(7):411-4, 1989.

[18] Murr AH, Benecke JE Jr. Association of facial paralysis with HIV positivity. Am J Otol. 12(6):450-1, 1991.

[19] Roullet E et al. Cytomegalovirus multifocal neuropathy in AIDS: analysis of 15 consecutive cases. Neurology 44:2174-82, 1994.

[20] McCutchan JA. Clinical impact of cytomegalovirus infections of the nervous system in patients with AIDS. Clin Infect Dis 21 (Suppl 2):S196-201, 1995.

[21] Fuller GN. Cytomegalovirus and the peripheral nervous system in AIDS. J Acquir Immune Defic Syndr 5(Suppl 1):S33-6, 1992.

[22] Kolson DL, Gonzalez-Scarano F. HIV-associated neuropathies: role of HIV-1, CMV, and other viruses. J Peripher Nerv Syst 6:2-7, 2001.

[23] Said G et al. Cytomegalovirus neuropathy in acquired immunodeficiency syndrome: a clinical and pathological study. Ann Neurol 29:139-46, 1991.

[24] Kim YS et al. Polyradiculopathy due to cytomegalovirus: Report of two cases in which improvement occurred after prolonged therapy and review of the literature. Clin Infect Dis 17:32-37, 1993.

[25] Miller RF et al. Acute lumbosacral polyradiculopathy due to cytomegalovirus in advanced HIV disease: CSF findings in 17 patients. J Neurol Neurosurg Psychiatry 61(5):456-60, 1996. (free online download).

[26] Winston A, Marriott D, Brew B. Early syphilis presenting as a painful polyradiculopathy in an HIV positive individual. Sex Transm Infect 81(2):133-4 2005.

[27] Tajima Y, Sudo K Matumoto A, Malignant lymphoma originating in the cauda equina mimicking the Inflammatory polyradiculoneuropathy . Intern Med 46: 1029-1032, 2007. Free online download

[28] Léger JM et al. Lymphoma-induced polyradiculopathy in AIDS: two cases. J Neurol 239(3):132-4, 1992.

[29] Cornblath DR, McArthur JC. Predominant sensory neuropathy in patients with AIDS and AIDS related complex. Neurology 38:794-6, 1988.

[30] Cornblath DR, Griffin GW, Tennekoon, FI. Immunoreactive myelin basic protein in cerebrospinal fluid of patients with peripheral neuropathies. Ann Neuol 20:370-372, 1986.

[31] Thornton CA, Latif AS, Emmanuel JC. Guillain-Barré syndrome associated with human immunodeficiency virus infection in Zimbabwe. Neurology. 41(6):812-5, 1991.

[32] Simpson, D et al. HIV-associated neuromuscular weakness syndrome. AIDS. 18(10):1403-12, 2004.

[33] Vidal JE et al. HIV-associated neuromuscular weakness syndrome in Brazil: report of the two first cases. Arq Neuropsiquiatr;65(3B):848-51, 2007.

[34]Fredericks, EJ; Kugelman, TP; Kirsch, N. Dapsone-induced motor polyneuropathy. A complication of prolonged treatment of subcorneal pustular dermatosis. Arch Dermatol 112(8):1158–1160, 1976.

[35] Gutmann, L; Martin, JD; Welton, W. Dapsone motor neuropathy--an axonal disease. Neurology. Jun;26(6 Pt 1):514–516, 1976.

[36] Ahrens EM, Meckler RJ, Callen JP.Dapsone-induced peripheral neuropathy. Int J Dermatol 25(5):314-6, 1986.

[37] Keswani SC et al. The spectrum of peripheral neuropathies in AIDS. in The Neurology of AIDS, Editors, Gendelman HE et al. Edition: 2, illustrated, Oxford University Press, 2005.

[38] Verma S, Simpson DM. Peripheral neuropathy in HIV infection. Handb Clin Neurol ;85:129-37, 2007.

[39] Keswani, op cit.

[40] Spudich SS, Price RW. Neurological disease, in AIDS Therapy, 3rd Edition, Editors Dolan R, Masur H, Saag M. Churchill Livingstone, Elsevier Inc, 2008.

[41] Wu G et al. Early onset of spontaneous activity in uninjured C-fiber nociceptors after Injury to neighboring nerve fibers. The Journal of Neuroscience, 21:RC140:1-5, 2001.

[42] From, Brown RC et al.  HIV/AIDS. Clinical Case Studies. 101 Cases for Training Clinical Staff. Nanasi, 2006.

[43] Dalakas MC. Peripheral neuropathy and antiretroviral drugs. J Periph. Nerv. Syst. 6: 14-20, 2001.

[44] Simpson DM, Tagliati M. Nucleoside analogue-associated peripheral neuropathy in human immunodeficiency virus infection. J Acquir Immune Defic Syndr Hum Retrovirol 9(2):153-61, 1995.

[45] Spudich, Price. Op cit.

[46] Keswani, op cit.

[47] Koike H, Sobue G. Alcoholic neuropathy. Curr Opin Neurol;19(5):481-6, 2006.

[48] Koike H et al. Alcoholic neuropathy is clinicopathologically distinct from thiamine-deficiency neuropathy. Annals of Neurology, 54:1, 19 – 29, 2003.

[49] Williams FH. Neuromuscular complications of nutritional deficiencies. Phys Med Rehabil Clin N Am. 19(1):125-48, vii, 2008.

[50] Vinik AI, Mehrabyan A. Diabetic neuropathies.Med Clin North Am. 88(4):947-99, xi, 2004.

[51] Said G. Focal and multifocal diabetic neuropathies. Arq Neuropsiquiatr;65(4B):1272-8, 2007 (available online)

[52] Said G. Leprous neuropathy. In: Mendell JR, Kissel JT, Cornblath DR, eds.  Diagnosis and management of peripheral nerve disorders.  Oxford University Press: 551–64, 2001.

[53] Agrawal A et al. Neurological manifestations of Hansen's disease and their management. Clin Neurol Neurosurg. 107(6):445-54, 2005.

[54] Ooi WW, Srinivasan J. Leprosy and the peripheral nervous system: basic and clinical aspects. Muscle Nerve. 30(4):393-409, 2004.

[55] Grimaud J, Vallat JM. [Neurological manifestations of leprosy [in French] Rev Neurol (Paris).159(11):979-95, 2003.

[56] Verma, Simspon, Op cit.

[57] Griffin JW et al. Predominantly sensory neuropathy in AIDS: distal axonal degeneration and unmyelinated fiber loss [abstract]. Neurology. 1991;41(suppl 1):374. Abstract 900S, 1991.

[58] Wiley CA. Neuromuscular diseases of AIDS. The FASEB Journal 3: 2503-2511, 1989.

[59] Verma, Simspon, op cit.

[60] Moyle GJ, Sadler M. Peripheral neuropathy with nucleoside antiretrovirals: risk factors, incidence and management. Drug Saf 19:481-494, 1998.

[61] Fichtenbaum CJ, Clifford DB, Powderly WG. Risk factors for dideoxynucleoside-induced toxic neuropathy in patients with the human immunodeficiency virus infection. J Acquir Immune Defic Syndr Hum Retrovirol 10:169-174, 1995.

[62] So YT et al. Peripheral neuropathy associated with acquired immunodeficiency syndrome. Prevalence and clinical features from a population-based survey. Arch Neurol 45(9):945-8, 1988.

[63] Schifitto G et al; Dana Consortium on the Therapy of HIV Dementia and Related Cognitive Disorders. Incidence of and risk factors for HIV-associated distal sensory polyneuropathy. Neurology. 58(12):1764-8, 2002.

[64] Heaton R et al. HIV-associated neurocognitive impairment remains prevalent in the era of combination ART: the CHARTER study. 16th Conference on Retroviruses and Opportunistic Infections, Montréal, abstract 154, 2009.

[65] Ellis RJ et al. Persisting high prevalence of HIV distal sensory peripheral neuropathy in the era of combination ART: correlates in the CHARTER study. 16th Conference on Retroviruses and Opportunistic Infections, Montréal, abstract 461, 2009.

[66] Evans SR et al. HIV-Associated peripheral neuropathy in the HAART era: results from AIDS Clinical Trials Group (ACTG) Longitudinal Linked Randomized Trials (ALLRT) Protocol A5001. 16th Conference on Retroviruses and Opportunistic Infections, Montréal, abstract 462, 2009.

[67] Lichtenstein KA, Armon C, Baron A, et al, for the HIV Outpatient Study Investigators. Modification of the incidence of drug-associated symmetrical peripheral neuropathy by host and disease factors in the HIV Outpatient Study cohort. Clin Infect Dis. 40:148-157, 2005.

[68] Kallianpur AR et al. Hemochromatosis (HFE) gene mutations and peripheral neuropathy during antiretroviral therapy. AIDS 20(11)13 p 1503-1513, 2006.

[69] Costarelli S et al. No evidence of relation between peripheral neuropathy and presence of hemochromatosis gene mutations in HIV-1-positive patients. JAIDS 46(2)1 255-256, 2007.

[70] Cherry C et al. Cytokine genotype suggests a role for inflammation in nucleoside analog-associated sensory neuropathy (NRTI-SN) and predicts an individual's NRTI-SN risk. AIDS Research and Human Retroviruses. 24(2): 117-123, 2008.

[71] Hulgan T et al. Hemochromatosis gene polymorphisms, mitochondrial haplogroups and peripheral lipoatrophy during antiretroviral therapy. J Inf Dis197:858–866, 2008.

[72] Canter J et al. Mitochondrial sub-haplogroups and peripheral neuropathy during ART among non-Hispanic Black participants in AIDS Clinical Trials Group Study 384. 16th Conference on Retroviruses and Opportunistic Infections, Montréal, abstract 160, 2009.

[73] Hulgan T et al. Mitochondrial haplogroups and peripheral neuropathy during antiretroviral therapy: an adult AIDS clinical trials group study. AIDS 19: 1341-1349, 2005.

[74] Mielke J. Peripheral nerve disease in HIV infected subjects, 2006. An African perspective. The Second assessment of NeuroAIDS in Africa. Arusha, Tanzania, 2006.

[75] Parry, O et al. Peripheral neuropathy in individuals with HIV infection in Zimbabwe. Acta Neurologica Scandanavica 96(4) 218-222, 1997.

[76] Nakasujja N et al. Human immunodeficiency virus neurological complications: an overview of the Ugandan experience. J Neurovirol;11 Suppl 3:26-9, 2005.

[77] Sacktor N et al. Benefits and risks of stavudine therapy for HIV-associated neurologic complications in Uganda. Neurology;72(2):165-70, 2009.

[78] Forna F et al. Clinical toxicity of highly active antiretroviral therapy in a home-based AIDS care program in rural Uganda J Acquir Immune Defic Syndr 2007;44:456–462, 2007.

[79] Birbeck GL, Am. J. Trop. Med. Hyg., in press.

[80] Boulle A et al. Substitutions due to antiretroviral toxicity or contraindication in the first 3 years of antiretroviral therapy in a large South African cohort. Antivir Ther. 12(5): 753-60, 2007.

[81] Hitchcock SA, Meyer HP, Gwyther E. Neuropathic pain in AIDS patients prior to antiretroviral therapy. SAMJ 98 (11), 2008.

[82] Van Dyke RB et al. Toxicities associated with dual nucleoside reverse-transcriptase inhibitor regimens in HIV-infected children. J Infect Dis 198:1599–608, 2008.

[83] Sithinamsuwan P et al. Frequency and characteristics of HIV-associated sensory neuropathy among HIV patients in Bangkok, Thailand. J Acquir Immune Defic Syndr 49(4), 2008.

[84] Cherry C et al. Predicting neuropathy risk before stavudine prescription: an algorithm for minimizing neurotoxicity in resource-limited settings. 16th Conference on Retroviruses and Opportunistic Infections, Montréal, abstract 161, 2009.