Article Text


Prevalence of HTLV infection in pregnant women in Spain
  1. Ana Machuca1,
  2. Concepción Tuset2,
  3. Vincent Soriano1,
  4. Estrella Caballero3,
  5. Antonio Aguilera4,
  6. Raul Ortiz de Lejarazu5,
  7. The Htlv Spanish Study Group
  1. 1Service of Infectious Diseases, Hospital Carlos III, Instituto de Salud Carlos III, Madrid, Spain
  2. 2Hospital General Universitario, Valencia
  3. 3Hospital Vall d'Hebron, Barcelona
  4. 4Hospital Xeral, Santiago
  5. 5Hospital Clínico Universitario, Valladolid
  1. Dr Vincent Soriano, C/ Rafael Calvo 7, 2° A 28010-Madrid, Spain vsoriano{at}


Objective: To estimate the prevalence of HTLV infection among pregnant women in Spain.

Methods: A commercial ELISA incorporating HTLV-I and HTLV-II antigens was used for HTLV antibody screening. Repeatedly reactive samples were further examined by western blot. Moreover, confirmation with PCR was performed when cells were available.

Results: 20 366 pregnant women in 12 different Spanish cities were tested in a 3 year period (July 1996 to August 1999). 32 samples were repeatedly reactive by ELISA, and 10 of them were confirmed as positive by western blot (eight for HTLV-II and two for HTLV-I). In addition, three of 13 women who had an indeterminate western blot pattern yielded positive results for HTLV-II by PCR. All 11 HTLV-II infected women had been born in Spain, and all but one were former drug users. Seven of them were coinfected with HIV-1. One HTLV-I infected woman was from Peru, where HTLV is endemic and where she most probably was infected during sexual intercourse.

Conclusion: The overall prevalence of HTLV infection among pregnant women in Spain is 0.064% (13/20 366), and HTLV-II instead of HTLV-I is the most commonly found variant. A strong relation was found among HTLV-II infection and specific epidemiological features, such as Spanish nationality and injecting drug use. Although HTLV-II can be vertically transmitted, mainly through breast feeding, both the low prevalence of infection and its lack of pathogenicity would not support the introduction of HTLV antenatal screening in Spain.

  • HTLV
  • pregnant women
  • adult T cell leukaemia/lymphoma
  • tropical spastic paraparesis
  • epidemiology

Statistics from


Human T lymphotropic virus types I and II (HTLV-I and HTLV-II) are two old human retroviruses with a close genetic sequence. HTLV-I, the first human retrovirus identified, is endemic in Japan, west Africa, the Caribbean, some areas of South America, and Melanesia.1,2 In contrast, HTLV-II infection is common among several Amerindian tribes of North, Central, and South America,3,4 pygmies in Central Africa,5 and among injecting drug users (IDUs) in North America and Europe.6–8 In Spain, HTLV-I/II infection is rare. Up to December 1999, 240 individuals with HTLV-II infection and 40 with HTLV-I had been identified.8–10 Most HTLV-I infected people were immigrants from endemic areas, or natives who had travelled or had had sexual partners from those areas, whereas HTLV-II infection was mainly found in Spanish IDUs, often coinfected with HIV-1.8–10

HTLV-I/II infection is transmitted during sexual intercourse (more efficiently from male to female),11 blood transfusion,12 sharing of contaminated needles,13 and from mother to child, mainly through breast feeding,14 although transplacental transmission can also occur.15 In endemic areas, HTLV is mainly transmitted heterosexually and from mother to child. Up to 25% of babies born to HTLV-I infected mothers become infected if they are breast fed, and this rate declines to 5% if they are bottle fed.16,17

HTLV-I infected individuals are prone to develop lymphoproliferative and neurological disorders as adults, classically adult T cell leukaemia/lymphoma (ATL),1 and tropical spastic paraparesis or HTLV-I associated myelopathy (TSP/HAM),18 although other inflammatory conditions have also been linked to HTLV-I infection. Overall, HTLV-I carriage is associated with a 2–5% lifetime risk of developing disease.19 HTLV-II has not yet been definitively associated with any disease, although several reports of neurological diseases mimicking TSP/HAM have appeared in the literature.20,21

HTLV-I infection in Europe is thought to be unusual except among immigrants from regions where HTLV is endemic. The screening of blood donors, which is currently mandatory in some European countries, has revealed a very low prevalence of infection across Europe.22 However, recent serosurveillance studies have pointed out that the rate of infection in pregnant women can be 50–100-fold higher than in blood donors.23–25 Of note, HTLV-I rather than HTLV-II is the virus implicated in most instances. These data might support the cost effectiveness of the antenatal HTLV screening.

The aim of our study was to estimate the prevalence of HTLV-I/II infection among pregnant women in Spain in the past 3 years, and therefore to provide information for considering whether HTLV antenatal screening should be recommended.

Material and methods


The prevalence of antibodies to HTLV in pregnant women was examined using sera collected in 12 different hospitals, distributed throughout Spain. Women had been referred for routine rubella and toxoplasma antibody testing. HTLV screening was performed by anonymous testing. In some institutions, the technicians were able to analyse those sera which were reactive to HTLV antibodies. In these cases, and after patient's signed consent, new blood was drawn, and peripheral blood mononuclear cells (PBMCs) were available for further polymerase chain reaction (PCR) analyses.

Between July 1996 and August 1999, a total of 20 366 samples were analysed. Their origin was Valencia (n=7523), Barcelona (n=3725), Santiago (n=3238), Valladolid (n=2686), Madrid (n=1786), Cadiz (n=754), Seville (n=479), and Orense (n=128).


Antibodies to HTLV-I/II were screened in pools of five sera, using a commercial enzyme linked immunosorbent assay (ELISA), which incorporates both HTLV-I and HTLV-II antigens (Murex HTLV-I+II; Dartford, Kent). Each reactive pool was further analysed by testing the five corresponding sera separately.

Repeatedly reactive ELISA samples were confirmed by western blot (WB HTLV plus; Genelabs, Singapore), which incorporates on a single nitrocellulose strip HTLV-I and HTLV-II antigens. The HTLV European Research Network (HERN) criteria were used for interpreting western blot patterns.22 Figure 1 summarises the diagnostic algorithm proposed by the HERN for HTLV and other related infections.

Figure 1

Algorithm for the diagnosis of HTLV infection.

The presence of HIV antibodies was examined in specimens showing reactivity against HTLV. A commercial ELISA (Axsym HIV 1+2, Abbott, Chicago, IL, USA), and western blot (HIV blot, Sanofi-Pasteur, Paris, France) were used for screening and confirmatory purposes, respectively. All assays were performed according to the manufacturer's instructions.


Frozen PBMCs were stored from a few HTLV seroreactive subjects, from whom new blood could be drawn, in order to perform genetic analysis by PCR (Amplicor HTLV-I/II PCR kit, Roche, Switzerland).26


Thirty two (0.16%) of the 20 366 tested sera were repeatedly reactive using the screening ELISA HTLV-I+II. However, only 10 samples were confirmed as positive by western blot (eight HTLV-II and two HTLV-I), the remaining samples were either indeterminate (13 sera) or negative (nine sera).

HTLV-I/II PCR analysis could be performed in specimens belonging to 11 women, six of whom had indeterminate western blot patterns, four a positive western blot for HTLV-II, and one for HTLV-I. Three women with HTLV indeterminate western blot were confirmed as HTLV-II positive by PCR. All of them showed a characteristic pattern on western blot, with reactivity against the HTLV-II recombinant protein rgp46-II along with any other isolated band (see Fig 2, cases 1, 4, and 8). Of note, all three were coinfected with HIV-1.

Figure 2

Western blot pattern in 11 infected pregnant Spanish women.

PCR results obtained in the four women with positive HTLV-II western blot and another positive for HTLV-I western blot yielded concordant results, respectively.

In summary, of 20 366 pregnant women tested in Spain since 1997, 11 (0.054%) were found to be infected with HTLV-II and two (0.01%) with HTLV-I. All 11 HTLV-II carriers were Spanish, and all but one had been IDUs. Another was the sex partner of a male IDU. Seven of the 11 were coinfected with HIV-1.

One woman who was positive for HTLV-I was an immigrant from Peru, and was HIV-1 negative. She denied drug addiction practices or blood transfusions, but admitted multiple heterosexual contacts in her country of origin. The other HTLV-I infected woman was HIV-1 negative, and no epidemiological data were available for her. Table 1 summarises the main features of all HTLV infected women identified in this study.

Table 1

Main features of HTLV-I/II infected pregnant women in Spain


HTLV-I infection has been associated with neurological disease or lymphoproliferative disorders in adults. The virus can be transmitted by sexual contact,11 parenteral exposure,12,13 or from mother to child.14,15 Recent studies have suggested that breast feeding accounts for most cases of vertical transmission of HTLV-I infection,16,17 thus providing an opportunity to avoid most infections in children born of seropositive mothers. However, data from blood donors underline that HTLV infections are rare in Spain, like in other European countries, in the range of 0.001%.10,22 Antenatal screening is therefore not offered to pregnant women. However, recent reports have pointed out that HTLV prevalence in blood donors might be misleading, since rates of HTLV infection 50–100-fold higher have been noticed in studies directly targeting pregnant women in Europe.23–25 In our study the prevalence of HTLV infection in pregnant women from several Spanish cities was 0.064%, more than 50 times higher than that found in Spanish blood donors.10,22 In some studies carried out in England and France most HTLV positive pregnant women were infected with HTLV-I.23–25 This could be explained mainly by the fact that these studies have focused on metropolitan areas (London, Paris, etc), which have large African and/or Afro-Caribbean immigrant communities. In contrast, in Spain HTLV-II instead of HTLV-I was the most frequent variant found in pregnant women. Moreover, only one of the two women found to be HTLV-I infected was an immigrant from an endemic region (Peru).

Most HTLV-II infections described so far in Spain have been reported in IDUs, 90% of them coinfected with HIV-1.8 In our study, a strong relation was also found between HTLV-II infection and previous drug addiction practices. All the pregnant women infected with HTLV-II were Spanish, and seven of the 11 were coinfected with HIV-1. Avoiding breast feeding would reduce the risk of both HIV-1 and/or HTLV-II infection in newborns.

Several reports have shown that reactivity to HTLV-II can be diminished in HIV-1 coinfected people suffering a profound immune impairment, leading to a characteristic indeterminate pattern on western blot strips (reactivity to rgp46-I or rgp46-II plus any other isolated band).27,28 We found three HTLV-II infected women who had an indeterminate pattern on western blot. All three were Spanish natives, IDUs, and coinfected with HIV-1.

In conclusion, our results show that the prevalence of HTLV infection among pregnant women in Spain is low, HTLV-II being (with no proved clear pathogenicity) more common than HTLV-I. Moreover, most HTLV-II positive pregnant women were coinfected with HIV-1. Since stopping breast feeding is already advised in all HIV positive women, additional measures to protect their newborns from HTLV-II does not seem to be required. The low prevalence of HTLV-I infection and the lack of pathogenicity of the predominant HTLV-II would not support the introduction of HTLV antenatal screening in Spain. Selective antenatal screening should to be considered in former IDUs or immigrants from endemic regions. Further cost-benefit studies are needed to determine whether HTLV antenatal screening should be recommended in Spain.


Members of The HTLV Spanish Study Group. C Tuset, G Marcaida, and R Gisbert, Hospital General Universitario, Valencia; E Caballero, Hospital Vall d'Hebrón, Barcelona; JM Eirós and R Ortiz de Lejarazu, Hospital Clínico Universitario, Valladolid; J García, Hospital Cristal-Piñor, Orense; A Aguilera, A Cid, A Jato, and B Regueiro, Hospital Xeral, Santiago; G Cilla and E Pérez-Trallero, Hospital Nuestra Señora de Aránzazu, San Sebastián; F Merino and B Esparza, Universidad del País Vasco, Bilbao; R Martínez-Zapico, Hospital La Paz, Madrid; E Calderón, FJ Medrano and M Leal, Hospital Virgen del Rocío, Sevilla; J Capote, Hospital Puerta del Mar, Cádiz; F Dronda, Hospital Ramón y Cajal, Madrid; R Benito, Hospital Clínico, Zaragoza; MA Rodríguez-Iglesias, Hospital Universitario de Puerto Real, Cádiz; J del Romero and C Rodríguez, Centro Sanitario Sandoval, Madrid; AM Martín and O Evora, Hospital Insular, Las Palmas de Gran Canaria; A Machuca, V Soriano, M Gutiérrez, and J González-Lahoz, Hospital Carlos III, Instituto de Salud Carlos III, Madrid.


This work was partially funded by an EC Concerted Action HTLV European Research Network (HERN) Antenatal Seroprevalence and Perinatal Transmission Study; CT97–2710), and grants from Instituto de Salud Carlos III, and Asociación Investigación y Educación en SIDA (AIES).

Contributors: AM, VS, CT, EC, AA, and ROL recruited women, took samples, and performed testing. Other members of the HTLV Spanish study group provided specimens, and participated actively in the final discussion of results, and the conclusions of the manuscript.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.