1. Institut d’Epidémiologie Neurologique et de Neurologie Tropicale, Limoges, France

E-Mail Contact - PREUX Pierre-Marie : preux (at) unilim (dot) fr


Retroviruses are common in the African continent. The roles of HIV-1 & 2 HTLV - I in the pathogenesis of disorders of the nervous system in Africa have not been well elucidated. To evaluate the individual roles and the interplay of these retroviruses in the pathogenesis of the nervous system morbidity, we studied 1872 patients admitted with neurological disorders in the national hospitals in 6 African countries in a period of 4 years. Myelopathies formed the largest group of disorders encountered (n=425), among which 107 had TSP. Peripheral neuropathies formed the second largest group of disorders observed (n=254). Meningoencephalitis was the third most frequent cause of hospitalisation and isolated cranial nerve palsies were less commonly seen(n=74) in this hospital based study. HIV-1 was significantly associated with meningoencephalitis. The relationship between HIV-1 was significantly associated with TSP/HAM (18.7%) but with a lesser preponderance than what has been described in other regions in the world endemic for HTLV-1.

Keywords: HTLV-1, HIV, neurology, Africa


Les rétrovirus sont fréquents sur le continent africain. Les rôles du VIH-I et de HTLV-1 dans la pathogénèse des troubles du système nerveux en Afrique n’ont pas été clairement élucidés. Pour évaluer leur role et les interrelations existantes entre ces retrovirus dans cette pathogénèse, nous avons étudié 1872 patients atteints d’affection neurologique hospitalisés dans les hopitaux nationaux de six pays africains sur une période de quatre ans. Les myélopathies formaient le groupe d’effectif le plus important (n==425), parmi lequel 107 patients avaient une paraparésie spastique tropicale. Les neuropathies périphériques représentaient 254 patients et les méningo-encéphalites 130 patients. Les paralysies isolées des nerfs crâniens étaient moins souvent vues (74 patients) dans cette étude hospitalière. Le VIH-1 était significativement associé à la survenue de méningo-encéphalite. Les relations entre le VIH-2 es les affections neurologiques n’atteignaient pas la significativité. HTLV-1 était significativement associé à la survenue d’une paraparésie spastique tropicale ( 18,7%) mais avec une moins grande fréquence que ce qui est décrit habituellement dans d’autres régions du monde, endémiques pour HTLV-1.

Mots clés : Afrique, HTLV1, HIV


The Human Adult T Cell Leukemia Virus type-1 (HTLV-1 ) has been definitely associated with two diseases: adult T cell leukemia/lymphoma (ATL) and the HTLV - 1 associated myelopathy/Tropical spastic-paraparesis(HAM/TSP), although most infected persons are asymptomatic. The syndrome of HAM/TSP has been reported in both HTLV -1 endemic and non-endemic zones throughout the world with an estimated lifetime risk of 0.25% in persons infected with HTLV - I (Aboulafia 1995). As these viruses have similar routes of transmission (Preux et al. 1995, Suzuki and: Gojobori, 1998), it is that patients infected with HTLV - I will be coinfected with HIV -I, or 2. These viruses can be transmitted in utero and by transfusion of contaminated blood and cellular products, sharing of needles among the intravenous drug users, sexual intercourse and breastfeeding.

In many parts of Africa, the use of unsterile or inadequately sterilized - needles and other surgical equipment in under equipped medical centers, and by traditional medicinemen, are important means of viral transmission that are particularly confined to this part of the world. The role of coinfections is not clear: HTLV- I may accelerate the progression of HIV but whether the emergence and the spread of HIVs has had any effect on HTLV - I is still debated (Preux et al. 1995). Beilke et al. (1998) suggested that the presence of one virus may activate the emergence of clinical disorders in the other probably due to up-regulation of the coinfecting retrovirus.

In HTLV - I infection, histopathological studies of the spinal cord have demonstrated demyelination lesions; a chronic inflammatory reaction with mononuclear cells and perivascular cuffing. This means that HTLV-1 may not only be lymphotrophic but also neurotrophic (Aboulafia 1995). Due to the long latency between the viral infection and the development of disease, the implication of this virus in the pathogenesis of neurological disorders in Africa may be very important considering its relatively elevated endemicity in some parts of this continent (de The et al. 1989 and Verdier et al. 1994). Recently, for instance Bartholomew and collegues (1997) suggested a possible causal association between this virus and lower motor neuron facial nerve palsy, mainly among the people of African descent, living in Tobago and Trinidad. In Africa, the range of HTLV - I associated neurological syndromes remains to be determined.

To determine the roles ofHTLV-l, HIV - 1 and 2 and their coinfections in the etiology of the nervous system morbidity in Africans, we studied 1872 neurological patients admitted in the neurology and general medical wards in six African countries between 1987 and 1991. The initial phase of this work was published by Ramiandrisoa et al. (1991).


Clinical Analysis

Between March 1987 and July 1991, all patients with neurological disorders admitted to neurology wards or general medical wards in University Hospitals of Brazzaville, Lome, Dakar, Abidjan, Ouagadougou, and Conakry were evaluated. These hospitals are found in the administrative capitals of Congo, Togo, Senegal, Ivory Coast, Burkina Faso and Guinea respectively. The neurological diagnoses were established by the responsible neurologist in each of these centers. In total, 1872 patients were included in this series. The sex ratio was 1.8 with 1214 (64.9%) males and 658 (35.1%) females. The mean age was 43.9 ±17.5 years.

After the clinical and paraclinical data (total blood count with and white cells differential count, urea and electrolytes analysis, cervical-spine and/or thoraco-lumbar X-rays, CSF studies and myelography when indicated) were analysed, patients were grouped in six clinically homogeneous disease groups . Those with spatic paraparesis of unknown aetiology having an insidious onset, slowly and chronically progressive, without remission or acute on chronic episodes, frequently associated with paraesthesiae and hypoesthesiae of the lower limbs, lumbago and sphincter dysfunctions were classified as tropical spastic paraplegia or paraparesis (TSP). Peripheral neuropathies from diverse causes including vitamin deficiencies, chronic alcoholism and dietary toxicities were identified. These mainly presented as tropical ataxias (the principal presentation of tropical myeloneuropathies). Isolated cranial nerve palsies were notably represented by facial palsy of lower motor neuron type and the oculo-motor nerve palsy. Meningo-encephalitis was diagnosed on clinical and CSF study and classified appropriately. All medullary disorders of diverse causes such amyotrophic lateral sclerosis, idiopathic myelopathies other that TSP were classified as myelopathies. Strokes, parkinsonism, Parkinson’s disease, dementias, epilepsies, and other diseases were classified together.

Laboratory analysis

Ten milliliters of venous blood were drawn from each patient and immediately centrifuged. The sera were then deep-frozen to negative 20°C and airlifted to France in dry-ice isothermal containers. The cold chain was maintained as such, and the samples were analysed in the Department of Bacteriology and Virology in the Faculty of Medicine in Limoges University.

Screening for HTLV-l
For screening purposes ELISA Abbott HTLV-l (Abbott Laboratories, Chicago, IL) technique was used. Each positive or borderline sera were subjected to the Western Blot (Western Blot Commercial Diagnostic Biotechnology, Singapore). The WHO positively for HTLV-I criteria were used as the standard, i.e., presence of active antibodies against the products of at least two different genes; gag and env. Indeterminate Western Blots were considered as negative in the statistical analysis.

Screening for HIV
Recombinant HIV-1 & 2 ELISA test (Abbott GmbH, Diagnostika, Wiesbaden, Germany) was used. The positive and borderline samples were confirmed using Western Blot for HIV-1 & 2 (New Lav Blot I and 2, Diagnostic Pasteur, Marnes la Coquette, France). Positivity for HIV-1 was confirmed by presence of antibodies against one or more of the enyelop gene products gp 160, gp 110, or gp 41), and one or more core proteins gag (p55, p24, pi 5). HIV-2 was determined by the presence of antibodies against the products from the two genes, i.e., eny(gp 140, gpl05, gp36) and gag (p56, p26, pi 6). The double positivity for HIV-1 & 2 was considered when a clear reaction occurred simultaneously against the envelop glycoproteins of the two viruses.

Statistical Analysis

The Mantel-Haenzel chi(2) method was used for group comparisons. Fisher’s exact-test was applied when the theoretical size was below 5. Given the high numbers of comparisons to be done, the threshold of significance was maintained at 1%o.


Myelopathies form the most frequently observed disorder (n=425) in this hospital based study, among which 107 had TSP. Peripheral neuropathies are the second commonest presenting disorder (n=254). These were mainly due to chronic alcoholism and dietary toxicities. Isolated cranial nerve palsies were less frequently encountered (n=74) probably because people will tend not to consult for this seemingly minor health problem infront of other serious medical and economic difficulties that engulf majority of people living in these countries.

Table 1 and Table 2 illustrated the retroviral seroprevalences; 246 (13.1%) wee seropositive for at least one of the viruses. There was a high co-infection rate as 23.6% seropositive patients had multiple infections.

HIV-1 was significantly responsible for the majority of the neuromorbidity. Patients mainly presented with meningoencephalitis with 27.7% being seropositive. 13% of those with cranial nerve palsies were seropositive for HIV-1, while a non significant 1.9% of patients with TSP was HIV - I positive. A marginal, statistically nonsignificant group of patients with myelopathies were HIV-2 positive (1.6%).

Only 0.3% were indeterminate for HTLV-1. These were treated as negative in subsequent analysis. HTLV-1 alone, or in association with HIV-1, or HIV 2, was found in 20 (18.7%) patients with TSP. This was significantly high when compared to the other clinical categories (p < 0.00l).

In 28% of the cases infected by HIV-1, another retrovirus was present. This was the case in 64% of HIV - 2 positive patients and 43% of HTLV-1 positive patients. The co-infection.rates differed significantly (p<0.0001) among the three retroviruses.


Retroviruses so called because of their genome which is composed ofRNA were among the first viruses to infect mammals. Their implication in human disease was not known untill early 80s when HTLV - I was identified by Poiesz and coworkers (1980) from a patient with cutaneous T cell lymphoma. Since then, the development of sero-epidemiological and molecular studies has provided great insight into the biology and transmissibility of these viruses.

Though the present study was not random, it was quite exhaustive as it covered a large zone of.the Sub-Saharan Africa endemic for these retrovirusesand involved a large number of patients. Moreover strict clinical and laboratory criteria applied minimised the possible inclusion biases. WHO criteria (September 1990) for positivity by Western Blot method was used. As from 1991 the new method targeting the recombinant envelop proteins has allowed the possibility to differentiate between the formally cross reacting HTLV -1 & 11.

It is admitted that some samples which were actually HTLV - 11 positive could have been classed as indeterminate using the older method. We excluded all indeterminate samples in the final analysis, and therefore HTLV - Tl effect would not affect the final results. Apart from a few clusters found among the Congolese Bambuti Efe Pygmies (Vandamme, 1998), the occurrence of HTLV - II is very low in Africa (Bonis et al. 1994) and this virus has not been etiologically linked with any disease although an increasing number of case reports suggests its possible associations with several clinical conditions (see Abiad and Hershaw, 1997)

Several workers have in the past demonstrated a significant association between HTLV- I andTSP though the discordance rate has been quite high among authors. In Martinique, Gessain et al (1985) described for the first time the association between HTLV - I and TSP. In their series, 59% of people with TSP were seropositive for HTLV - 1. In high endemicity zones for HTLV -4, the endemicity for TSP was also high. In a separate study, in Martinique, Vernant and his colleagues (1987)found78%positivityforHTLV-l inTSP patients. Roman and co-workers (1987) in Seychelles, found 85% of the patients seropositive for HTLV - I presented with TSP. In Jamaica Rodger-Johnson and co-workers (1988) found 67%.

In Africa the relationship HTLV - I /TSP is much less evident than in other regions with high endemicity for this virus. Only 18.7% ofthe patients with TSP in this study were seropositive for HTLV - 1. Delaporte et al. (1989) had found only between 3 and 7% of TSP patients to be seropositive for HTLV -I in the humid tropical forest in Gabon. None ofthe TSP patients investigated by de The .and co-workers (1989) in Ivory Coast and Zaire were seropositive. Hugon and colleagues (1990) reported only some sporadic cases. The reasons for this discrepancy between Africa and other regions are not clear but it may be that the HTLV - I is not be playing an important role in the causation of TSP in the Africans.

Errors arising from the sampling and analytical techniques, false negativity of the screening methods (notably in some subjects seropositive by PCR), the long latency going to several years be- tween infection and first clinical symptoms, the short life expectancy among the Africans are some of the factors that could explain these discrepancies. However, other unidentified factors like toxic substances, nutritional disorders, genetic susceptibility, climatic and geographical conditions, infections or co-infections with bacteria or other viruses or even parasitic infestations could be responsible for the pathogenesis of TSP/HAM rather that HTLV-1 (Montgomery, 1993).

Though HTLV-1 has been implicated in other disorders like polyradiculoneuropathies and polymyositis (Vernant et al, 1991) its exact role in the pathogenesis of these illnesses has not been clearly demonstrated. We did not find in this study any demonstrable relationship between HTLV-1 and other neurological disorders.

The other retroviruses on the other hand seem to play a relatively bigger role in pathogenesis of disorders of the nervous system. HIV is responsible for peripheral neuropathies, myelopathies and meningoencephaUtis. We found a significant association between HIV-1 and meningoencephalitis. HIV - 2 did not seem to be of any significant contributory value.

The interplay of HTLV-1 virus in the disease pathogenesis in the HiV epidemic remains unclear. While there are several reports of pathogenic potentialization by this virus in HIV infection, viable evidence is lacking. The role of the interplay of these viruses in the pathogenesis of neurological disorders in the African continent, where there is a major coexistence of these viruses, is a subject of major interest, particularly because of their neurotropism. Superinfection withHIVs may accelerate the occurrence of clinical symptomatology of paraplegia in HTLV - I infected patients probably secondary to.an immunological reaction (Preux et al. 1995).

Our study found a high coinfection rate in a small group of patients, and even where coinfection was found, higher chances of developing a neurological complication in patients was not evident. Other independent cofactors are probably responsible for the clinical evolution in patients infected with these retroviruses and need to be investigated.

We acknowledge the financial support given by the Ministry of Co-operation, France (decisions FAC no 339/CD/88VI/IGE/41 FSP 2410402C of 27th September 1988 and 113/Cd/87 FSP 210901 V of 19th November 1990) and by "Conseil Regional du Limousin", France. We thank all those kind people who made this study possible.

Table 1: Patients distribution against towns in different countries.

Clinical Categories Abidjan Conakry Brazzaville Ouagadougou Dakar Lome TOTAL
Tropical Spastic Paraplegia 47 3 1 32 6 18 107
Peripheral Neuropathies 72 18 10 21 47 86 254
Isolated Cranial nerve palsy 10 8 0 13 3 40 74
Meningo-encephalitis 33 4 23 29 16 25 130
Myelopathies 90 27 14 18 117 52 318
Other 268 160 58 120 180 203 989
TOTAL 520 220 106 233 369 424 1872

Table 2: Seroprevalence by clinical categories against the retroviruses.

Clinical categories N HTLV 1 (%) HTLV1 & HIV1 (%) HTLV1 & HIV2 HTLV1 & HIV1 & HIV2 HIV1 HIV2 HIV1 % HIV2
Tropical Spastic Paraplegia 107 11,20 4,70 2,80 0,00 1,90 0,00 1,90
Peripheral Neuropathies 254 1,60 0,80 0,00 0,00 7,50 0,80 2,80
Isolated cranial nerve palsy 74 1,40 0,00 0,00 0,00 13,50 0,00 0,00
Meningo-encephalitis 130 2,30 3,90 0,00 0,80 27,70 0,80 6,90
Myelopathies 318 0,60 0,90 0,00 0,30 3,10 1,60 1,90
Other 989 0,50 0,10 0,00 0,00 6,20 1,50 1,30
TOTAL 1872 1,40 0,90 0,20 0,10 7,40 1,20 2,00


  1. ABIAD H & HERSHOW R Current understanding of HTLV - II and its role in disease Infectious Medicine 1997;14: 815-820.
  2. ABOULAFIA DM Clinical implications of the human T cell leukemia virus type 1/11 associated diseases. The AIDS Reader 199;118-125.
  3. BARTHOLOMEW C, CLEGHORN F, JACK N & EDWARDS J Human T Cell Associated Lymphotrophic Virus type I associated facial nerve palsy in Trinidad and Tobago. Annals of Neurology 1997;1: 806-809.
  4. BEILKEMA,JAPAS&VINSONGDHTLV - I and HTLV - II Virus expression increase with HIV - I coinfection. Journal of acquired immonodeficiency syndromes and human retrovirology l998;17,391-397.
  5. DELAPORTE E, PEETERS M, DURAND PJ, DUPONTAETAL. Seroepidemiological surveys among randomised populations of Western Central African countries. Journal of Acquired Immuno-Deficiency Syndrome 1989;2:410-413
  6. DE THE G, GIORDANO C, GESSAIN A ET AL. Human retroviruses in HTLV I,HIV I, and HIV II and neurological disorders in some equatorial areas of Africa. J Acquired Immune-Deficiency Syndrome 1989;2:550-556.
  7. GESSAIN A, BARIN F, VERNANT JC ET AL. Antibodies to Human T Lymphotrophic viruses type I in patients with tropical spastic paraparesis: Lancet 1985;ii: 407-409.
  8. GIORDANO C, N’DIAYE IP, MANSOUR M, ET AL. HTLV-I and Neurological diseases in Sub-Saharan Africa. In:HTLV I and the nervous system. Roman CG, Vernant JC, Osame M (Eds) Alan R. Liss, (1989) pp 195-202
  9. HUGON J, VAIIAT JM, DUMAS M ET AL. Low prevalence of HTLV I antibodies in the serum of patients with chronic spastic paraplegia from Ivory Coast. Journal of Neurology Neurosurgery and Psychiatry 1990;53:269.
  10. MONTGOMERY RD The epidemiology of myelopathy associated with human T-Lymphotrophic virus type-1. Transactions of the Royal Society of Tropical Medicine and Hygiene 1993; 87:154-159.
  11. POIESZ BJ, RUSCETTI FW, GAZDAR AF, BUNN PA, MINNA JD & GALLO RC Detection and isolation of type C retrovirus particles from fresh and cultured lymphocytes of a patient with cutaneous T-Lymphoma. Proceedings of the National Academy of Sciences of USA 1980;77:7415-7419.
  12. PREUX PM, CRUZ M, DRUET-CABANAC M, HOUINATOD& DUMAS M Coinfections HTLV-HIV. In: Recent Advances in Tropical Neurology , F. Clifford Rose, (Ed),. Elsevier Science BV: (1995) pp 237-243.
  13. RAMIANDRISOA H, DUMAS M, GIORDANO C ET AL. Human Retroviruses HTLV-I, HIV-I, HIV-2 and neurological disorders in West Africa Journal of Tropical and Geographical Neurology. 1991;1:39 - 44.
  14. RODGER-JOHNSON P, MORGAN OSL, MORA C ET AL. (The role of HTLV-I in tropical spastic paraparesis in Jamaica. Annuals of Neurology 1988;23 (supply): S121 - S 126.
  15. ROMAN GC, SPENCER PS, SCHOENBERG BS ET AL. Tropical spastic paralysis : HTLV - I antibodies in patients from Seychelles Island New England Journal of Medicine 1987;316:51.
  16. SUZUKI Y & GOJOBORI T. The origin and evolution of human T-cell lymphotrophic virus type I and II. Virus Genes 1998;16:1,69-84.
  17. VANDAMME AM, SALEMI M, BRUSSEL MV ET AL. African Origin of humanT- lymphotrophic type 2 (HTLV-II) supported by a protential new HTLV-lid subtype in Congolese Bambuti Efe Pygmies. Journal of Virology 1998;72,4327-4340.
  18. VERDIER M, BONIS J & DENIS F (1994) The prevalence and incidence of HTLV in Africa. In: Aids in Africa, Essex M, MboupS,KankiPJ,KalengayiMR, (Eds). Raven Press Ltd, New York, pp 173-193.
  19. VERNANT JC, MAURS L, GESSAIN A ET AL. Endemic tropical spastic paraparesis associated with human T-Lymphotrophic virus type I: A clinical and seroepidemiological study of 25 cases. Annals of Neurology 1987;21: 123-130.
  20. VERNANT JC, BELLANCE R & BUCHER B Manifestations neurologiques associees au virus HTLV-I. In: Manifestations neurologiques et infections retrovirales. Said G, Saimot AG, (Eds). Pradel Editions: Paris, (1991) pp 193-207.

© 2002-2012 African Journal of Neurological Sciences.
All rights reserved. Terms of use.
Tous droits réservés. Termes d'Utilisation.
ISSN: 1992-2647