16 Practical Neurology

REVIEW

Pract Neurol 2009; 9: 1626

The neurology of HTLV-1 infection

Sarah A Cooper, Maarten Schim van der Loeff, Graham P Taylor
HTLV-1 is a significant global health problem but remains largely confined to endemic areas and risk groups. However, increasing migration may mean that the virus will be encountered more frequently in areas traditionally thought of as virtually free of HTLV-1. In this review we discuss the epidemiology, transmission, clinical features, diagnosis and treatment of HTLV-1, focussing specifically on the neurological manifestations. We highlight the circumstances in which HTLV-1 should be suspected and outline the current understanding of HTLV-1-associated myelopathy and other neurological presentations.
n 1977, a 28-year-old black American man developed skin nodules which were diagnosed soon afterwards as a cutaneous T-cell lymphoma. After his death in 1979, he became, in 1980, the first person in whom retrovirus infection in humans was formally proven.1 This momentous discovery of a type C retrovirus, subsequently termed the human T-cell lymphotropic virus type 1 (HTLV-1) predated the isolation of HIV-1 in 1983. Epidemiological studies and phylogenetic analysis of HTLV-1, and related simian strains, have suggested that HTLV-1 has infected humans for thousands of years although it is endemic only in certain areas of the world. Monitoring is needed to map the S A Cooper
Neurology Specialist Registrar, Southern General Hospital, Glasgow, UK

M Schim van der Loeff

Senior Epidemiologist, GGD, Amsterdam, The Netherlands

G P Taylor
Reader in Communicable Diseases, Department of Genitourinary Medicine & Communicable Diseases, Imperial College, London, UK Correspondence to: Dr S Cooper Institute of Neurological Sciences, Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF, UK; sarah.cooper@doctors.org.uk 10.1136/jnnp.2008.167155

spread of the virus as, with changing patterns of migration, it could become more common in areas hitherto nearly free of HTLV-1. The isolation of HTLV-1 was considerably overshadowed by the discovery of HIV-1, largely because of the profound clinical consequences of HIV in all of those infected. Like HIV, HTLV-1 causes a lifelong infection but, in contrast to HIV-1, most of those infected remain asymptomatic throughout life.

THE EPIDEMIOLOGY OF HTLV-1 INFECTION

Estimating the global prevalence of HTLV-1 is difficult because there are so few populationbased studies. However, approximately 1520

Cooper, Schim van der Loeff, Taylor 17

million people worldwide are believed to be infected.2 The infection is endemic in regions of Japan, sub-Saharan Africa, the Americas, Melanesia, the Seychelles and the Middle East (fig 1).3 HTLV-1 has been classified into several viral subtypes based on differences in proviral DNA, and these HTLV-1 sequences have been compared to simian isolates (STLV IIII). For most HTLV-1 subtypes, STLV isolates from the same geographical region cluster closely with HTLV-1 suggesting multiple episodes of simian-to-human transmission, and a probable African origin.4 Africa is the only continent where all HTLVs and STLVs have been found. Phylogenetic studies also support an African origin of HTLV-1 and estimate that the spread of primate and human T-cell lymphotropic virus occurred in Africa some 27 000 years ago.5 Although the trade in African slaves has been implicated in the spread of HTLV-1 to a number of endemic zones, the spread of HTLV-1 to Native Americans (North and South), the Japanese, Melanesians and Australian aborigines occurred much earlier. Most prevalence studies have been carried out on low-risk blood donors, or selected groups such as pregnant women, neurological or haematological patients, relatives of those infected, injecting drug users or sex workers (like many of the early studies in HIV). Studies of pregnant women are considered to be the most representative of the background population, although reports from endemic areas

suggest that HTLV-1 is more common in women and the prevalence increases with age.6 Relatively high prevalence figures in either the general population or pregnant women and/or blood donors have been described in Southern Japan (more than 10% of the general population are reported as seropositive in some areas7) with Caribbean countries such as Jamaica and Trinidad reporting up to 6% prevalence.4 In several sub-Saharan African countries (for example, Guinea Bissau, Cameroon and Benin (fig 2)) prevalence of up to 5% has been described.4 Following increased migration in the last five centuries, HTLV-1 has reached Europe but remains uncommon in the general population Nonetheless, HTLV-1 in Europe is considerably more common in certain groups (immigrants from endemic areas and their sexual partners and offspring, sex workers and injecting drug users).3 Epidemiological studies in indigenous Europeans usually implicate transmission through sexual intercourse with partners from endemic regions,8 or blood transfusion. The seroprevalence among low-risk populations in Europe is less than 0.1%.9 In a 1997 1998 UK population-based study 126 010 neonatal blood samples were tested for HTLV specific antibodies (a reliable marker of maternal infection); there were only 67 with confirmed antibodies. The highest seroprevalence was 17.0 per 1000 in infants whose mothers had been born in the Caribbean. It was concluded that an estimated 223 of the 750 000 pregnant women in the UK are

Figure 1 Endemic areas vary considerably demographically, from developed nontropical areas to less developed tropical zones. (A) Southwest Japan; (B) rural Guinea Bissau.

www.practical-neurology.com

18 Practical Neurology

Figure 2 A map showing the prevalence of HTLV1 (Romania is thought to be the European country with the highest HTLV-1 prevalence).

infected with HTLV-1 (a figure at the time that was comparable with maternal HIV-1 infection).10 It was calculated that in England and Wales in 19931997 more than 22 000 people were infected with HTLV-1.11 HTLV-1 is prevalent among intravenous drug users in New York and Brazil4, 12 but it appears that the serologically distinct HTLV-II may be more prevalent in injecting drug users in Europe8 It is unclear why, after spreading throughout the world, HTLV-1 remains highly prevalent in some populations but not in others. In countries such as Korea, Eastern Russia and China, for example, the retrovirus is found at low levels whereas neighbouring Southwestern Japan has a high prevalence (fig 2).

through exposure to infected lymphoid cells, eg by blood transfusion,13 needle stick injuries or needle sharing by intravenous drug users.12

TRANSMISSION OF HTLV-1
The three major modes of transmission are:

N N
10.1136/jnnp.2008.167155

perinatal (especially via prolonged breastfeeding) sexual (predominantly from male to female)

The risk of vertical transmission ranges from 1525% (with less than 5% overall occurring in utero or at birth).9 Twenty two per cent of children aged between 3 and 10 who had been breastfed by infected mothers were seropositive in one study.14 Transfusion of infected blood leads to seroconversion in more than 40% of recipients;15 pre-existing immunosuppression at the time of transfusion is an important risk factor.15 In recent years, a number of countries, including Japan, USA, Canada, France and the UK, have introduced screening of blood donors to reduce the risk of HTLV-1 transmission. The latent period between infection and manifestation of disease can be as long as 2030 years but shorter incubation periods have been described after infection acquired via blood products, possibly indicating exposure to a higher viral load.16 Kidney

Cooper, Schim van der Loeff, Taylor 19

and liver transplantation has also resulted in transmission.17 Transmission dynamics are not fully understood and additional unknown factors are likely to be influential. As yet unexplained are the observations that most HTLV-1 endemic areas are in the tropics, the virus clusters among neighbours, and the prevalence declines in subsequent generations migrating from endemic to non-endemic areas (although cultural changes particularly relating to breastfeeding might be important). In addition, although both HIV-1 and HTLV-1 have similar transmission routes, in some areas their prevalence trends are divergent HTLV-1 declining and HIV-1 increasing.18

VIROLOGY
HTLV-1 is an enveloped, double-stranded RNA type C virus belonging to the family of Retroviridae, subfamily orthoretrovirinae, genus deltaretrovirus . During the life cycle of the virus the RNA is converted to doublestranded DNA and integrated into the DNA of the human host cell. This integrated viral DNA is referred to as the provirus. HTLV-1 preferentially infects CD4+ T cells but CD8+ T cells are also an important reservoir.9, 19 HTLV-1 exists largely as a cell-associated provirus and is transmitted from cell to cell via a viral synapse whereby virus particles bud from the infected cell into a cleft in the synapse.9 (fig 3) This phenomenon may account for the lack of infection associated with plasma. Multiplication of the provirus occurs mainly via clonal expansion of the infected lymphocytes rather than the production of new virions. It is believed that an equilibrium is reached between cytotoxic Tcell control (lysing infected T cells that express viral peptides) and proviral replication which results in a steady provirus load that varies little over time in any given individual.9 Provirus load can vary between HTLV-1 infected people by more than 10 000-fold.20

Paraparesis (HAM/TSP)) and uveitis have all been linked to HTLV-1 infection by casecontrol studies, cohort studies, animal transfer experiments and the isolation of the virus in lesions.9 The lifetime risk of an HTLV-1 carrier developing myelopathy ranges from only 0.25% in Japan to approximately 3% elsewhere. The risk of ATLL is 26% according to the Japanese Cancer Registry. It is unusual, but not unheard of, for a patient with HAM to develop ATLL as well. Some non-neurological conditions reported (with varying degrees of evidence) in association with HTLV-1 include arthropathy, Sjogrens syndrome, thyroiditis, pneumono pathy, crusted scabies, infective dermatitis and leprosy.9, 19 In addition, Strongyloides stercoralis hyperinfection has been reported in HTLV-1 infected patients at higher levels than in HIV-1,21 and a case-control study from Brazil found an association between tuberculosis and HTLV-1 infection.22

Figure 3 An HTLV-1 virion within a pocket in the virological synapse between an infected and uninfected lymphocyte. This illustrates the cell-associated nature of HTLV-1 transmission. Anti-Gag p19, monoclonal antibody against the nuclear protein p19. (Courtesy of Endre Majorivits and Mohamed Nejmeddine.)

HTLV-1-ASSOCIATED DISEASE
Several clinical conditions have been associated with HTLV-1 infection, some supported by more robust evidence than others. Adult T cell leukaemia/lymphoma (ATLL), a spastic paraparesis (commonly referred to as HTLV1 Associated Myelopathy/Tropical Spastic

HTLV-1-ASSOCIATED MYELOPATHY/TROPICAL SPASTIC PARAPARESIS

Many years before HTLV-1 was isolated, physicians in the tropics were reporting cases of idiopathic spastic paraparesis. In 1969 the
www.practical-neurology.com

20 Practical Neurology
agent cross reacts with a host antigen resulting in disease). HAM/TSP may be mediated by molecular mimicry between domains of some viral proteins and normal cellular targets of the spinal cord involved in the neurodegeneration. HAM/TSP patients make antibodies to heterogeneous nuclear ribonuclear protein A1, a neuron-specific autoantigen. These antibodies have been detected in serum, CSF and the central nervous system. In addition, autoantibodies against other nuclear and perinuclear brain proteins have been found in the serum of HAM/TSP patients.29 Antibodies can cross the blood-brain barrier at sites of inflammation, and perivascular inflammation is one of the features of HAM/TSP. However, the most widely accepted hypothesis is that HAM/TSP is the result of bystander damage. HTLV-1 infected CD4+ and to a lesser extend CD8+ T-lymphocytes migrate through the tissues, including the spinal cord. When virus specific CD4+ and CD8+ cells encounter infected lymphocytes they secrete a multitude of cytotoxins including perforin, granzymes and cytokines including the neurotoxic interferon (IFN)-c and tumour necrosis factor (TNF)a19. It is not clear why these encounters do not cause disease all the time in all infected subjects although clearly the frequency of encounter is likely to be higher in those with a high provirus load. One possibility is that more efficient cytotoxic T-lymphocyte responses will result in lysis of the infected cell, whereas a less efficient response will result in pro-inflammatory chemotaxic and neurotoxic cytokine expression.30 Recruitment of lymphocytes to the scene may increase the likelihood of further infected cell/virus specific cell contact essentially fanning the inflammatory process. In addition to the high proviral load, high TNF-a and IFN-c levels in plasma and CSF characterise patients with HAM/TSP9 and increased HTLV-1 Tax expression is independently associated with HAM/TSP.31

Figure 4 Perivascular lymphocytic infiltrate in the central nervous system of a patient with HTLV-1 infection. (Courtesy of Dr Margaret Esiri.)

term tropical spastic paraparesis was first used to describe these cases because clinical and pathological similarities appeared to link them together.23 The association with HTLV-1 was recognised independently in Japan and the Caribbean in the 1980s. In 1985, Gessain et al found that 60% of patients in Martinique with spastic paraparesis had antibodies to HTLV-124 and shortly afterwards Osame et al reported a group of six patients in Kagoshima with myelopathy and HTLV-1 antibodies in serum and CSF.25 Descriptions of patients from the Seychelles, Jamaica and Columbia soon followed.26 A WHO working group proposed the term HTLV-1 associated myelopathy (HAM)/ tropical spastic paraparesis (TSP) which remains in widespread use today.

The pathogenesis of HAM/TSP

HAM/TSP is characterised by a chronic meningomyelitis of the white and grey matter followed by axonal degeneration that preferentially affects the lower thoracic cord. Histopathological studies have shown that the lesions are associated with perivascular and parenchymal infiltration of T cells and that later in the disease the process becomes less cellular and more atrophic27 (fig 4). It is not clear just how HTLV-1 causes neurological damage but a high viral burden and an exaggerated proinflammatory response are involved. There is no direct evidence that HTLV-1 infects neurons, astrocytes or microglia but infected T cells have been observed within spinal cord lesions and CD8+ T cells directed against HTLV-1 antigens accumulate in the cerebrospinal fluid (CSF) of those with HAM/TSP.28 Autoimmunity may play a significant role and molecular mimicry offers an explanation for the association of infection with autoimmune disease (that is, an immune response to an environmental
10.1136/jnnp.2008.167155

Clinical features
HAM/TSP usually presents as a slowly progressive spastic paraparesis with neurogenic bladder disturbance; in about 60% the first symptom is weakness of the lower limbs.32 Although initially this can be asymmetrical, the weakness progresses without remission,

Cooper, Schim van der Loeff, Taylor 21

or any plateau, to a symmetrical spastic gait. The symptoms usually begin during adulthood, most frequently after the age of 30; the median time from symptom onset to unilateral walking aid is 6 years, bilateral walking 13 years, and wheelchair dependence 21 years.33 HAM/TSP is uncommon in children but cases in adolescence especially in association with infective dermatitis have been reported.34 Bladder symptoms are very common; indeed, urinary frequency, urgency, incontinence and/or retention may be seen very early in the disease and even precede the paraparesis by many years. Back pain, constipation, sexual dysfunction and sensory symptoms (usually of the lower extremities) are also very common.9, 19 Pain is present in about 50%,35 is usually lumbar and often radiates to the legs; this in our (unpublished) experience often seems to have a radicular pattern, presumably because spinal nerve roots can be involved in the inflammatory process. In many cases pain is severe and more distressing than the gait disturbance. Circumstances in which HTLV-1 should be suspected in a non-endemic country are outlined in the box. On examination the patients have a spastic gait with weakness of the lower limbs most evident proximally. Diffuse hyper-reflexia of all four limbs is commonly seen along with extensor plantar responses. Having said this, the ankle jerks may be depressed or absent which may reflect involvement of the peripheral nervous system. Upper limb power is usually preserved throughout the illness. Sensory disturbance is generally mild and consists mainly of paraesthesias of the feet and occasionally of the hands. A clearcut sensory level is unusual, although not incompatible with the diagnosis. The striking discrepancy between the obvious motor signs and the mild sensory components is characteristic. Loss of pain and light touch sensation was reported in between 2753% in three clinical series with impairment of vibration sense recorded in between 348%.3638 Abnormal joint position sense is not considered to be a feature. An exaggerated jaw jerk may be observed and there have been reports of lower motor neuron seventh nerve palsies39 although it is not clear if the association with facial palsy is more than a chance finding.

TABLE 1 Diagnostic criteria for tropical spastic paraparesis/HTLV-1associated myelopathy (TSP/HAM)31

Definite: l A non-remitting progressing spastic paraparesis with sufficiently impaired gait to be perceived by the patient, with or without sensory symptoms or signs, which, when present, remain subtle and without a clearcut sensory level. Sphincter signs or symptoms may or may not be present l Presence of HTLV-1 antibodies in serum and/or CSF confirmed by Western blot and/or HTLV-1 PCR positive in the blood l Exclusion of other disorders that can mimic TSP/HAM* Probable: l Monosymptomatic presentation: spasticity or hyperreflexia in the lower limbs or isolated Babinski sign, or spasticity in the lower limbs with subtle sensory signs, or pure neurogenic bladder confirmed by urodynamic tests l Presence of HTLV-1 antibodies in serum and/or CSF confirmed by Western blot and/or HTLV-1 PCR positive in the blood l Exclusion of other disorders that can mimic TSP/HAM* Possible: l Complete or incomplete clinical presentation l Presence of HTLV-1 antibodies in serum and/or CSF confirmed by Western blot and/or HTLV-1 positive in the blood l Disorders that can mimic TSP/HAM have not been excluded to a satisfactory degree* *In order to minimise a misdiagnosis, the following conditions should be excluded by appropriate laboratory and clinical evaluation: multiple sclerosis; familial spastic paraparesis; primary lateral sclerosis; syringomyelia; B12 and folate deficiency; neurosyphilis; sarcoidosis; collagen vascular diseases; Sjogrens syndrome; amyotrophic lateral sclerosis; spinal cord compression (spinal tumour, cervical spondylosis, etc); endemic regional myelopathies with similar clinical manifestations (including schistosomiasis and neurocysticercosis); carcinomatous meningitis; transverse myelitis; paraneoplastic syndromes; Lyme disease; Behcets disease; neurotuberculosis; HIV vacuolar myelopathy; autoimmune myelopathies; toxic myelopathies; fungal myelopathy.

When to suspect HTLV-1 myelopathy in a non-endemic country

In the presence of: Clinical features compatible with HTLV-1 myelopathy, in essence: l a slowly progressive, symmetrical spastic paraparesis l neurogenic bladder symptoms l sensory symptoms +/2 low back pain and Risk factors for the acquisition of HTLV-1 l born in an endemic country l having a sexual partner from an endemic country l being the offspring of someone from an endemic country A sexual and transfusion history is important when investigating an unexplained spastic paraparesis.
www.practical-neurology.com

22 Practical Neurology A few patients progress more rapidly, particularly those older than 50 years at symptom onset, with a higher provirus load (.105 DNA copies/106 peripheral blood mononuclear cells) and higher antibody titres.33 A subacute progression, leading to severe paraplegia in less than two years has been reported in post-transfusional and posttransplant cases.40 Clinical diagnostic guidelines were first proposed by the World Health Organization in 198941 but have been subsequently updated by an international team with criteria for definite, probable and possible cases (table 1).42 Some authors consider that certain HTLV-1 complications appear to cluster in the same patients, in particular HAM/TSP alongside uveitis, arthritis, Sjogrens syndrome and alveolitis.9 development of disease remains uncertain. Serological confirmatory tests include indirect immunofluoresence assays and western blots (both commercially available, the western blot being preferred44) These may provide some indeterminate results and then PCR on a blood sample is required. PCR also allows proviral load to be quantified, this being the most frequently used marker for prognosis. Provirus load is expressed as the number of HTLV-1 DNA copies per fixed number of peripheral blood mononuclear cells (PBMC). There is significant variability between patientsfor example, in a study from Japan the mean proviral copy number per 10 000 PBMC was 798 in 202 HAM/TSP patients compared with 120 in 200 neurologically asymptomatic HTLV-1 carriers.45 PCR is also used to discriminate between HTLV-1 and HTLV-II. Atypical lymphocytes in peripheral blood films (flower cells) may be observed along with hypergammaglobulinaemia and a false positive VDRL.19 The most common CSF abnormalities are a mild lymphocyte pleocytosis (generally no more than 50 cells/mm3) and a mildmoderately increased protein content (in about one half of patients with levels up to 210 mg/dl reported).36 These abnormalities are generally found in the first few years of disease, tending to lessen gradually although they can persist for as long as 10 years after symptom onset.46 Oligoclonal bands are found in the CSF (and also sometimes the serum) in most patients47 but are of little use clinically. Antibodies against HTLV-1 are found in the CSF (and at higher levels in HAM/TSP than in asymptomatic carriers without neurological symptoms) and the HTLV-1 provirus can be demonstrated in the CSF of HAM/TSP patients by PCR. Measuring the provirus load in blood and CSF simultaneously helps to confirm the diagnosis as the ratio of proviral load in CSF to that in peripheral blood is above 1.0 in those with HAM/TSP.48

Risk factors for developing HAM/TSP in those infected with HTLV-1

A high HTLV-1 provirus load (see below) appears to be associated with the development of HAM/TSP along with a faster progression of the spastic paraparesis.9 And an early age of sexual intercourse and more than five sexual partners was associated with the risk of developing HAM/TSP in one study.43 Although transfusion of infected blood is associated with the highest viral innoculum, the most common transmission risk factor for the development of HAM/TSP is sexually acquired HTLV-1 infection,3 for reasons not fully understood. More women than men develop HAM/TSP (a difference not entirely attributable to the higher prevalence of HTLV1 among women in endemic areas) and the disease progresses faster in women.4

Laboratory diagnosis
Serological screening for HTLV-1 antibodies can be by a commercially available enzyme immunoassay (EIA), or particle agglutination test. Confirmatory testing of EIA positive results is recommended to eliminate false positives and to discriminate between HTLV-1 and HTLV-II. This is important as HTLV-II is common in injecting drug users in North America and Europe but its role in the
10.1136/jnnp.2008.167155

Neuroimaging
Non-specific periventricular and subcortical white matter lesions on brain magnetic resonance imaging (MRI) have been observed in 5080% of patients with HAM/TSP.49, 50

Cooper, Schim van der Loeff, Taylor 23

These lesions do not distinguish HAM/TSP patients from neurologically-asymptomatic HTLV-1 infected individuals,50 and there have not been adequate comparisons with HTLV-1 seronegative controls. Spinal cord MR imaging was abnormal in 3/21 (14%) of those with HAM/TSP in a small series49 (spinal cord atrophy has been reported mainly in the thoracic region (fig 5) in selected populations).40 High-signal intensity and contrast enhancement with or without associated spinal cord swelling is located mainly in the posterior columns, posterior horns, or lateral columns at cervical or thoracic levels51 It has been suggested that patients with a more rapidly progressive disease who are scanned earlier in the disease course are more likely to have spinal cord swelling on MR imaging, possibly because this reflects active cord inflammation.51

A novel approach has been sodium valproate as a histone deacetylase inhibitor; in an open-study in Martinique this was associated with a transient rise followed by a 24-fold fall in HTLV-1 proviral load57 (the theory being that it stimulates transient activation of the latent viral reservoir causing its collapse). Unfortunately, thus far, this impressive result has not been associated with any clinical benefit.

Treatment
The only treatments to have been evaluated in randomised, controlled trials (RCTs) are interferon-alpha and the combination of the two nucleoside analogues zidovudine and lamivudine:

Clinical trials for patients with HAM/TSP are conducted at the National Centre for Human Retrovirology in London (http://www.htlv1.eu) and should be considered for all patients in the UK, especially those with early disease. Symptomatic treatment with antispasmodics, laxatives, phosphodiesterase inhibitors for erectile dysfunction, and chemical and physical management of the urinary dysfunction are all important in improving quality of life. Physiotherapy is also important both to maximise function through training and aids, and to restore power following intercurrent illness. Urinary tract infections (usually with coliforms) are common and often associated with transient neurological deterioration, therefore requiring prompt treatment.

The striking discrepancy between the obvious motor signs and the mild sensory components is characteristic

Interferon-a was of benefit to some patients in the short term (4 weeks) in patients with HAM/TSP.52 In addition, there was some improvement in disability in an uncontrolled observational study of 152 patients where 66% reported varying degrees of benefit.53 In an RCT of 16 patients, zidovudine plus lamivudine showed no significant clinical benefit after 612 months of treatment.54 A recent open cohort study of 39 patients with HAM/TSP with a mean follow-up of 2.2 years showed an improvement in overall disability following pulsed intravenous methylpredenisolone.55 In practice, steroids remain the most commonly prescribed drug despite the poor quality of evidence, perhaps because of the premise that there is a significant inflammatory phase relatively early on in the illness. Plasmapharesis has shown some improvements in gait and sensory disturbance in small, uncontrolled studies only.56

Figure 5 MR imaging showing thoracic atrophy in a patient with HAM/TSP.

www.practical-neurology.com

24 Practical Neurology

OTHER NEUROLOGICAL COMPLICATIONS REPORTED IN ASSOCIATION WITH HTLV-1 Myositis

The quality of epidemiological evidence linking HTLV-1 with polymyositis is very limited because there are no case-control or cohort studies. However, there are a significant number of reports from endemic areas of inflammatory muscle disorders in patients who are HTLV-1-infected. Patients with polymyositis in endemic areas have been examined for the presence of HTLV-1 antibodies (11/13 in one Jamaican study),58 and in nearly 200 HTLV-1-infected Brazilian patients there was myopathy in 4 (2%).59 Mostly the patients with myopathy have had coexisting HAM/TSP and it has been proposed that an inflammatory myopathy should be considered if patients with HAM/TSP develop a new pattern of muscle weakness, myalgia and raised plasma creatine kinase concentrations.19 HTLV-1 provirus has been identified in mononuclear inflammatory cells in skeletal muscle of infected individuals with polymyositis.

study was limited by the fact that the clinical assessment was not blind to the serological status of the patients, and the controls were not matched for age or sex. Polyneuropathy has been reported in HTLV-1 infected patients with or without HAM/TSP. Mononeuritis multiplex and mononeuropathy have also been reported in HTLV-1 infected individuals, but in small numbers and as case reports.

Cognitive impairment
Psychomotor slowing and deficits in attention and visuospatial abilities have been reported in HTLV-1-infected individuals more often than in non-infected controls. Patients with HAM/TSP and asymptomatic carriers do not appear to differ in the degree of impairment observed.19 It has been suggested that the white matter changes on brain MRI in HTLV1-infected individuals may represent chronic perivascular inflammation causing a subcortical cognitive deficit, but there is a need for good clinical and pathological studies.

Autonomic dysfunction
There are reports in HTLV-1-infected individuals of various symptoms and signs that could represent dysautonomia (postural hypotension, hypertension, changes in heart rate variability, impotence and urinary retention). Again, more high-quality studies are required.

Polyneuropathy
Polyneuropathies, usually of the small-fibre sensory variety, have been described in HTLV1 infected patients, mainly in case series. In addition, however, a case-control study found polyneuropathy in 8.6% of 196 HTLV-1infected persons and in 2.6% of 196 noninfected controls (p = 0.015).59 However, this

Motor neuron disease/ amyotrophic lateral sclerosis

There have also been several reports of the features of amyotrophic lateral sclerosis (ALS) in patients with HTLV-1. However, in one case-control study only 1/196 (0.5%) HTLV-1 -infected individuals met the diagnostic criteria for ALS.59 Five of 606 (1.9%) neurologically symptomatic HTLV-1 carriers from Brazil presented with ALS and these patients showed a slower progression of disease.60 More studies are required to clarify the validity of any association.

PRACTICE POINTS
l

HTLV-1-associated neurological disease should be considered in individuals from endemic areas, their offspring or those with sexual partners from endemic zones. Presentation with a slowly progressive spastic paraparesis, often with urinary dysfunction, should prompt testing for HTLV-1 antibodies. Further investigations should include HTLV-1 proviral load in peripheral blood mononuclear cells with PCR and anti-HTLV-antibodies (and/or HTLV1 proviral load) in the CSF. Although only a minority of those infected with HTLV-1 develop any neurological problems, the complete spectrum of any morbidity has not been elucidated. Treatment is largely limited to symptomatic management although corticosteroids are often used (with little evidence) and there is limited evidence for the role of interferon-alpha. Monitoring is needed to map the possible spread of HTLV-1 to previously largely unaffected populations.

ACKNOWLEDGEMENTS
This article was reviewed by Jeremy Farrar, Ho Chi Min City, Vietnam.

REFERENCES
1. Poiesz BJ, Ruscetti FW, Gazdar AF, et al. Detection and isolation of type C retrovirus particles from

10.1136/jnnp.2008.167155

Cooper, Schim van der Loeff, Taylor 25

fresh and cultured lymphocytes of a patient with cutaneous T-cell lymphoma. Proc Natl Acad Sci USA 1980;77:741519. de The G, Bomford R. An HTLV-1 vaccine: why, how and for whom? AIDS Res Hum Retroviruses 1993;9:3816. Taylor GP, Bodeus M, Courtois F, et al. The seroepidemiology of Human T-lymphotropic viruses types I and II in Europe: A prospective study of pregnant women. J Acquir Immune Defic Syndr 2005;38:1049. Proietti FA, Carneiro-Proietti AB, Catalan-Soares BC, et al. Global epidemiology of HTLV-1 infection and associated diseases. Oncogene 2005;24:605868. Van Dooren S, Salemi M, Vandamme AM. Dating the origin of the African human T-cell lymphotropic virus type-1 (HTLV-1) subtypes. Mol Biol Evol 2001;18:66171. Larsen O, Andersson S, da Silva Z, et al. Prevalences of HTLV-1 infection and associated risk determinants in a urban population in GuineaBissau, West Africa. J Acquir Immune Defic Syndr 2000;25:15763. Morofuji-Hiratu M, Kaliyama W, Nakashuma K, et al. Prevalence of antibody to HTLV-1 in Okinawa, Japan after an interval of 9 years. Am J Epidemiol 1993;137:438. The HTLV-1 European Research Network. Seroepidemiology of the human T-cell leukaemia/ lymphoma viruses in Europe. J Acquir Immune Defic Syndr Hum Retrovirol 1996;14:6877. Verdonck K, Gondzalez E, Van Dooren S, et al. Human T-lymphotropic virus 1: recent knowledge about an ancient infection. Lancet Inf Dis 2007;7:26681. Ades AE, Parker S, Walker J, et al. Human T-cell leukaemia/lymphoma virus infection in pregnant women in the United Kingdom: population study. BMJ 2000;320:1497501. Tosswill JHC, Taylor GP, Tedder RS, et al. HTLV-1associated disease in England and Wales 1993 1997: a retrospective study of serology requests. BMJ 1999;320:61112. Lee HH, Weiss SH, Brown LS, et al. Patterns of HIV1 and HTLV-1/II in intravenous drug users from the middle atlantic and central regions of the USA. J Infect Dis 1990;162:34752. Manns A, Hisada M, La Grenade L. Human Tlymphotropic virus type 1 infection. Lancet 1999;353:19518. Hino S, Doi H. Mechanisms of HTLV-1 transmission. In: HTLV-1 and the nervous system. In: Roman G, Vernant J-C, Osame M, eds. New York: Alan R Liss, 495501. Manns A, Wilks RJ, Murphy EL, et al. A prospective study of transmission by transfusion of HTLV-1 and risk factors associated with seroconversion. Int J Cancer 1992;51:88691. Montgomery RD. The epidemiology of myelopathy associated with human T-lymphotropic virus 1. Trans R Soc Trop Med Hyg 1993;87:1549. Gonzalez-Perez MP, Munoz-Juarez L, Cardenas FC, et al. HTLV-1 infection in various recipients from the same donor. Transplantation 2003;75:100611. Tortevoye P, Tuppin P, Carles G, et al. Comparative trends of seroprevalence and seroincidence rates of human T cell lymphotropic virus type 1 and human immunodeficiency virus 1 in pregnant women of various ethnic groups sharing the same environment in French Guiana. Am J Trop Med Hyg 2005;73:5605. Araujo AQC, Silva MTT. The HTLV-1 neurological complex. Lancet Neurol 2006;5:106876. Bangham CR. The immune control and cell to cell spread of human T-cell lymphotropic virus type 1. J Gen Virol 2003;84:317789. Marcos LA, Terashima A, Dupont HL, et al. Strongyloides hyperinfection syndrome: an emerging global infectious disease. Trans R Soc Trop Med Hyg 2008;102:31418. Marinho J, Galvao-Castro B, Rodrigues LC, et al. Increased risk of tuberculosis with human T-cell lymphotropic virus type 1 infection: a case-control study. J Acquir Immune Defic Syndr 2005;40:6258. Mani KS, Mani AJ, Montgomery RD. A spastic paraparesis syndrome in South India. J Neurol Sci 1969;9:17999. Gessain A, Barin F, Vernant J-C, et al. Antibodies to human T-lymphotropic virus type-1 in patients with tropical spastic paraparesis. Lancet 1985;2:40710. Osame M, Usuku K, Izumo S, et al. HTLV-1 associated myelopathy, a new clinical entity. Lancet 1986;1:10312. Roman GC, Navarro-Roman LI. The discovery of HTLV-1 myelitis: 21 years later. Lancet Neurol 2007;6:1045. Umehara F, Osame M. Histological analysis of HAM/ TSP pathogenesis. In: Sugamira K, Uchiyame T, Masao M, Kannagi M, eds. Two decades of adult Tcell leukaemia and HTLV-1 research. Tokyo, Japan: Scientific Societies Press, 2003:1418. Kubota R, Soldan SS, Martin R, et al. Selected cytotoxic T lymphocytes with high specification for HTLV-1 in cerebrospinal fluid from a HAM/TSP patient. J Neurovirol 2002;8:537. Garcia-Vallejo F, Dominguez MC, Tamayo O. Autoimmunity and molecular mimickry in tropical spastic paraparesis/HTLV-1-associated myelopathy. Braz J Med Bio Res 2005;38:24150. Bangham CR. The immune response to HTLV-1. Curr Opin Immunol 2000;12:397402. Ramirez E, Cartier L, Torres M, et al. Temporal dynamics of human T-cell lymphotropic virus type 1 tax mRNA and proviral DNA load in peripheral blood mononuclear cells of htlv-1-associated myelopathy patients. J Med Virol 2007;79:78290. Araujo AQ, Andrade-Filho AS, Castro-Costa CM, et al. HTLV-1 associated myelopathy/tropical spastic paraparesis in Brazil: a nationwide survey. J Acquir Immune Defic Syndr Hum Retrovirol 1998;19:53641. Olindo S, Cabre P, Lezin A, et al. Natural history of human T-cell lymphotropic virus type I-associated myelopathy: a 14 year follow-up study. Arch Neurol 2006;63:15606. Primo JR, Briles CR, Oliveira Mde F, et al. Infectious dermatitis and HTLV-1-associated myelopathy/ tropical spastic paraparesis in childhood and adolescence. Clin Inf Dis 2005;41:53541. Gotuzzo E, Cabrera J, Deza L, et al. Clinical characteristics of patients in Peru with Human T-cell Lymphotropic virus Type I-associated tropical spastic paraparesis. Clin Inf Dis 2004;31:93944. Vernant JC, Maurs L, Gessain A, et al. Endemic tropical spastic paraparesis associated with HTLV-1: a clinical and epidemiological study of 25 cases. Ann Neurol 1987;21:123. Araujo A, Alfonso CR, Schor D, et al. Clinical and demographic features of HTLV-1-associated

19. 20.

2.

3.

21.

4.

22.

5.

23.

6.

24.

25.

7.

26.

8.

27.

9.

28.

10.

29.

11.

30. 31.

12.

13.

32.

14.

33.

15.

34.

16.

35.

17.

36.

18.

37.

www.practical-neurology.com

26 Practical Neurology
myelopathy/tropical spastic paraparesis (HAM/TSP) in Rio de Janeiro, Brazil. Acta Neurol Scand 1993;88:5962. Bhigjee AI, Kelbe C, Haribhai MC, et al. Myelopathy associated with HTLV-1 in Natal, South Africa. A clinical and investigative study in 24 patients. Brain 1990;113:130720. Bartholomew C, Cleghorn F, Jack N, et al. Human Tcell lymphotropic virus type-1-associated facial nerve palsy in Trinidad and Tobago. Ann Neurol 1997;41:8069. Cabre P, Smadja D, Cabie A, et al. HTLV-1 and HIV infections of the central nervous system in tropical areas. J Neurol Neurosurg Psych 2000;68:5507. World Health Organization. Report of the scientific group convened by the WHO regional office for the western Pacific, Kagoshima, Japan, 1015 December 1988 Human T-cell lymphotropic virus type 1, HTLV-1. Wkly Epidemiol Rec 1989;49:3823. De Castro-Costa CM, Araujo AQ, Barreto MM, et al. Proposal for diagnostic criteria of tropical spastic paraparesis/HTLV-1-associated myelopathy. AIDS Res Hum Retroviruses 2006;22:9315. Kramer A, Maloney EM, Morgan O, et al. Risk factors and cofactors for human T-cell lymphotropic virus type 1 (HTLV-1) associated myelopathy/tropical spastic paraparesis (HAM/TSP) in Jamaica. Am J Epidemiol 1995;142:121219. Berini CA, Susana Pascuccio M, Bautista CT, et al. Comparison of four commercial screening assays for the diagnosis of HTLV types 1 and 2. J Virol Methods 2008;147:3227. Nagai M, Usuku K, Matsumoto W, et al. Analysis of HTLV-1 provirus load in 202 HAM/TSP patients and 243 asymptomatic HTLV-1 carriers: high provirus load strongly predisposes to HAM/TSP. J Neurovirol 1998;4:58693. Moreno-Carvalho OA, Nascimento-Carvalho CM, Galvao-Castro B. HTlV-1 associated tropical spastic paraparesis. Cerebrospinal fluid evolutive aspects in 128 cases. Arq Neuropsiquiatr 1995;53:6047. Gessain A. Virological aspects of tropical spastic paraparesis/HTLV-1-associated myelopathy and HTLV-1 infection. J Neurovirol 1996;2:299306. Lezin A, Olindo S, Oliere S, et al. HTLV-1 proviral load in cerebrospinal fluid: a new criterion for the diagnosis of HTLV-1-associated myelopathy/ tropical spastic paraparesis? J Infect Dis 2005;191:18304. 49. Bagnato F, Butman JA, Mora CA, et al. Conventional magnetic resonance imaging features in patients with tropical spastic paraparesis. J Neurovirol 2005;11:52534. Morgan DJ, Caskey MF, Abbehusen C, et al. Brain magnetic resonance imaging white matter lesions are frequent in HTLV-1 carriers and do not discriminate from HAM/TSP. AIDS Res Hum Retroviruses 2007;23:1499504. Umehara F, Nose H, Saito M, et al. Abnormalities of spinal magnetic resonance images implicate clinical variability in human T-cell lymphotropic virus type 1-associated myelopathy. J Neurovirol 2007;13:2607. Izumo S, Goto I, Itoyama Y, et al. Interferon-alpha is effective in HTLV-1 associated myelopathy: a multicenter, randomized, double-blind, controlled trial. Neurology 1996;46:101621. Arimura K, Nakagawa M, Izumo S, et al. Safety and efficacy of interferon-alpha in 167 patients with human T-cell lymphotropic virus type-1 associated myelopathy. J Neurovirol 2007;13:36470. Taylor GP. Goon P, Furukawa Y, et al. Zidovudine plus lamivudine in Human T-cell Lymphotropic Virus type-1 associated myelopathy: a randomized trial. Retrovirology 2006;3:63. Croda MG, de Oliveira AC, Vergara MP, et al. Corticosteroid therapy in TSP/HAM patients: the results from a 10 year open cohort. J Neurol Sci 2008;269:1337. Matsuo H, Nakamura T, Tsujihata M, et al. Plasmapheresis in the treatment of human T-cell lymphotropic virus type-I associated myelopathy. Lancet 1988;2:1109. Lezin A, Gillet N, Olindo S, et al. Histone deacetylase medicated transcriptional activation reduces proviral loads in HTLV-1-associated myelopathy/ tropical spastic paraparesis patients. Blood 2007;110:37228. Morgan OS, Rodgers-Johson P, Mora C, et al. HTLV1 and polymyositis in Jamaica. Lancet 1989;2:11847. Leite AC, Mendonca GA, Serpa MJ, et al. Neurological manifestations in HTLV-1-infected blood donors. J Neurol Sci 2003;214:4956. Silva MTT, Leite ACC, Alamy AH, et al. ALS syndrome in HTLV-1 infection. Neurology 2005;65:13323.

38.

50.

39.

51.

40.

41.

52.

42.

53.

43.

54.

55.

44.

56.

45.

57.

46.

58.

47.

59.

48.

60.

10.1136/jnnp.2008.167155