Frequency and characteristics of HTLV in migrants: results from the +Redivi collaborative network in Spain

Abstract

Introduction

The human T-lymphotropic viruses 1 and 2 (HTLV-1 and HTLV-2) are retroviruses with demonstrated tropism for T lymphocytes, which cause chronic infection in humans.¹ HTLV-1 and HTLV-2 can be transmitted among humans (vertically, through breast feeding, sexual contact, contaminated blood transfusions/sharing of contaminated needles by intravenous drug users) and are postulated to have originated in Africa and then spread along migratory routes to Europe, Asia and America.^2,3

Estimates for the number of people living with HTLV-1 infection globally range from 5 to 10 million, and prevalence of infection is higher in specific geographical areas.¹ Known areas of high prevalence include Japan, several countries in Africa, the Caribbean, Europe, Latin America, Oceania and the Middle East and migration has contributed to increased detection of cases in lower prevalence countries. HTLV-2 is endemic in certain African populations and in indigenous groups in the American Continent and there are an estimated 800 000 infected individuals worldwide.^4,5

Although most people infected with HTLV-1 will not develop conditions associated with the virus, some may develop severe disease including adult T-cell leukaemia/lymphoma (ATL) and HTLV-1-associated myelopathy or tropical spastic paraparesis (HAM/TSP).¹ HTLV-2 has not been definitively associated with lymphoproliferative disease although neurological syndromes have been described in association with the infection.^6,7

A nationwide registry for HTLV infections in Spain has contributed to the study of these viruses.⁸ However, several gaps in epidemiological data have been identified both in endemic and non-endemic areas, the public health burden of the infection may be difficult to assess and effective preventive measures may not be adequately implemented.¹ The objective of this study was to describe the main epidemiological and clinical characteristics of immigrants diagnosed with HTLV infection within the +Redivi Spanish network.

Methods

The current descriptive analysis was based on a cohort included in +Redivi (‘Red Cooperativa para el Estudio de las infecciones importadas por viajeros e inmigrantes’). +Redivi includes 22 health centres (1 primary care centre and 21 hospitals, both specialized and non-specialized in travel medicine) in 8 different regions of Spain that share a common online database where new cases of imported infections are registered (the network was initially comprised of 4 centres and new centres have been included progressively over the years). A unique data collection sheet was developed to collect information on demographic data, trip characteristics (destination, return date, duration, type of traveller), pre-travel advice and anti-malarial prophylaxis when applicable, reason for consultation and final diagnosis. Information from the database could not be linked back to the patient as these were deidentified at inclusion. Cases recorded were those where the final diagnosis was considered to be an imported disease associated with travel (both for symptomatic and asymptomatic patients). Returning travellers/immigrants could account for more than one case in the database if they had travelled more than once and had a new diagnosis of an imported disease. Persons attending a first consultation were classified as follows: immigrant (person living in Spain but born in any other country), visiting friends and relatives (VFR)-immigrant (immigrant travelling back from his/her country of birth after visiting friends and relatives), VFR-traveller (Spanish traveller who returns from his/her first-degree relative’s country of birth) and travellers (conventional international tourist returning from travel, expatriates and missionaries).

Cases recorded in +Redivi from October 2009 to December 2020 were included in the present analysis and patients with a diagnosis of HTLV type 1 or 2 infection were identified. Diagnosis was based on positive HTLV serology (enzyme-linked immunosorbent assay (ELISA)/chemiluminescent immunoassay (CLIA)) with line immunoassay (LIA)/Western blot with/without polymerase chain reaction (PCR). Strongyloidiasis was diagnosed based on positive serology (ELISA) and/or stool examination/agar plate cultures for larvae. Characteristics for migrants with HTLV infection were described including gender, age, type of case, country of birth/area of origin, time from arrival to consultation, immunosuppression, associated symptoms, comorbidities and outcome. Regarding possible mode of acquisition, sexual transmission was considered if the partner was positive for HTLV, vertical/breastfeeding associated transmission was considered if mother was positive for HTLV and transmission following transfusion/transplantation was considered if infection was documented in the donor. Qualitative variables were expressed as relative and absolute frequencies, and quantitative data were expressed as median and interquartile range (IQR) (when appropriate).

Results

In total, 23 937 records from the +Redivi database were analysed (October 2009–December 2020). These included 13 150 (54.9%) immigrants, 3857 (16.1%) VFR-immigrants, 371 (1.6%) VFR-travellers and 6559 (27.4%) travellers.

A total of 107/17 007 cases (0.6%) had a diagnosis of HTLV infection: 83 (77.7%) HTLV-1 infections, 6 (5.6%) HTLV-2 infections and 18 (16.8%) non-specified HTLV infections. Of these, 67 (62.6%) were diagnosed by serology, 31 (29%) by serology and PCR and for 9 (8.4%) the diagnostic method was not specified. The majority of patients were female (76, 71%), median age was 42 years (IQR 33.4–51.6) and median time from arrival to Spain until first consultation was 10 years (IQR 4.6–13.6). Median duration of follow-up was 5 years (IQR 2–7). The group included 100 (93.5%) immigrants and 7 (6.6%) VFR-immigrants. As for geographical area of origin, 71 (66.4%) patients were from South America, mainly from Peru (25, 23.4%), Ecuador (14, 13.1%), Colombia (13, 12.1%), Bolivia (9, 8.1%), Chile (3, 2.8%) and Venezuela (3, 2.8%). There were 15 (14%) patients from Sub-Saharan Africa, mainly from Equatorial Guinea (5, 4.7%) and Nigeria (3, 2.8%), 13 (12.1%) from the Central America-Caribbean area (mainly from the Dominican Republic 9, 8.4% and Honduras 2, 1.9%), 4 (3.7%) from Europe, 3 (2.8%) from West Asia and 1 (0.9%) from North Africa. Patients with HTLV-1 were also mainly of South American (59, 71.1%), Sub-Saharan African (9, 10.8%) and Central American-Caribbean (8, 9.6%) origin. Immigrants with HTLV-2 were from South America (5, 83.3%) and Europe (1, 16.7%). The main characteristics of migrant patients with HTLV infection are shown in Table 1.

For most patients mode of acquisition of HTLV infection could not be determined (94, 87.9%), for six patients (5.6%) infection was considered to have been acquired sexually, for four patients (3.7%) vertically/through breastfeeding and for three patients (2.8%) following transfusion/transplantation.

Regarding reason for medical consultation, 96 (89.7%) were asymptomatic and were mainly diagnosed as part of screening programs (55, 51.4%, of these following donation at blood banks, and 9, 8.4%, following contact tracing). For 18 patients (16.8%), cause of screening was not specified, 9 (8.4%) were screened following an abnormal/non-specified laboratory test and 5 (4.7%) were tested due to a reported previous history of HTLV infection). Other six (5.6%) patients referred neurological symptoms and three (2.8%) gastrointestinal symptoms at initial consultation. Two patients presented with other non-specific symptoms (1.9%). The majority of patients were not immunosuppressed (99, 92.5%), four patients had HIV infection (3.7%), one patient was on immunosuppressive therapy (0.9%) and three patients (2.9%) were classified as immunosuppressed but cause was not specified.

Ninety patients (84.2%) did not have/develop any HTLV-associated conditions, seven (6.5%) had a final diagnosis of TSP, five (4.7%) had other associated conditions (dermatitis, uveitis, pulmonary disease), three (2.8%) had other neurological symptoms and two (1.9%) had ATL. No patients with HTLV-2 had HTLV-associated conditions.

Concomitant diagnoses/comorbidities were found in 41 (38.3%) patients and 66 (61.7%) patients had no registered associated diagnoses. The main recorded associated conditions included: 15 (14%) strongyloidiasis, 4 (3.7%) HIV co-infection, 3 (2.8%) hypothyroidism, 3 (2.8%) asthma, 3 (2.8%) dyslipidaemia, 3 (2.8%) Chagas disease, 3 (2.8%) diabetes mellitus, 3 (2.8%) latent tuberculosis, 2 (1.9%) treated tuberculosis and 2 (1.9%) anaemia.

Considering patients whose outcome was recorded, 89 (83.2%) remained asymptomatic and 10 (9.3%) were lost to follow-up or no outcome was recorded. Regarding patients with neurological symptoms, three patients had progressive symptoms, with worsening lower limb weakness (one with sphincter involvement) and another had chronic debilitating paraparesis with complications (pressure ulcers). Four patients (4.1% if only patients with a recorded outcome were considered) died, two from complications associated with ATL diagnosed during follow-up of HTLV infection (1 received treatment with zidovudine, lamivudine and chemotherapy and the other patient received interferon-alpha, zidovudine, which was changed to raltegravir and chemotherapy). Another patient died due to a Strongyloides hyperinfection syndrome following corticosteroid use, which led to the diagnosis of HTLV infection and one from oesophageal cancer with no HTLV-associated conditions (screened due to a positive family history of HTLV).

Regarding HTLV screening of potential contacts, this was not performed or not recorded in the majority of cases (75, 70.1%). For 17 (53.1%) patients, screening in contacts was recorded as negative, for 8 (25%) patients a partner was recorded as positive, in 3 cases (9.4%) a child was recorded as positive, in 3 cases (9.4%) the patient’s mother was positive and in 1 case (3.1%) a sibling was recorded as positive.

Discussion

The HTLV viruses affect up to 10 million people worldwide, with the majority of infections being due to HTLV-1.^1,3 Transmission, incidence and morbidity of HTLV-1 may be reduced by incorporating control measures into existing public health strategies; however, even in endemic areas, awareness of this infection and modes of transmission may not be widespread.¹ In this descriptive series from a national network in a European country recording imported infections in travellers and immigrants, HTLV infection was an infrequent diagnosis, although when only immigrants were considered, this accounted for 0.6% of all diagnoses, with over 75% of these being due to HTLV-1. A recent systematic review and meta-analysis found patients with HTLV-1 infection have an increased risk of death compared with those without HTLV-1, although the causes have not been clearly recognized and the burden of the disease may be greater than previously established.⁹

Migrants with a diagnosis of HTLV had a median age of 42 years, < 10% were found to have immunosuppression and the median time from arrival to first consultation was 10 years, probably reflecting the proportion of long-settled asymptomatic immigrants diagnosed through screening, for example at blood banks (nearly 90% of HTLV positive immigrants were asymptomatic at initial consultation). The HTLV positive migrant group also had a high representation of patients from South America and the Central America-Caribbean area. Although for several countries in America there are no reliable HTLV-1 prevalence data, this observation could be in accordance with previous studies describing high burdens of infection in specific population groups in South America (up to 8%).^1,10,11 A recent study in Europe focusing on recently arrived immigrants from sub-Saharan Africa or Asia found a low prevalence of HTLV-1/2 infection (<0.1%).¹² Of note, there were a large proportion of female patients in the HTLV positive group (70%), with the consequent implications regarding recommendations to prevent transmission through breastfeeding and the screening of children.

As expected, the majority of positive patients (over 80%) had no HTLV-associated diagnoses, although neurological conditions were found in 9% of patients (6.5% had TSP) and two patients (1.9%) had ATL. These figures were lower (even when only HTLV-1 infections were considered) than those reported from the Spanish HTLV-1 registry, where TSP was found in 10% of patients and ATL was diagnosed in 8% of cases.¹³ Comparisons with studies from other areas are limited as data on ATL incidence in HTLV-1 infected patients are scarce. A study of 90 HTLV-1 carriers identified following screening at blood bank services in Israel, and followed for up to 15 years, identified 4 cases of ATL.¹⁴ In another study of 153 patients with HTLV-1 infection who were followed up for a median of 4.5 years, 4 developed ATL.¹⁵ In people with HTLV-1 infection, the lifetime risk of TSP has been estimated to be around 2%, emphasizing the need for chronic follow-up of these patients.¹ Median time of follow-up in the current series was 5 years, and future efforts should therefore focus on the retention in care of these patients.

Despite the large proportion of asymptomatic patients at initial consultation, 38% had concomitant diagnoses and/or comorbidities. Around 14% of patients were diagnosed with strongyloidiasis, representing a higher prevalence than that found in other studies. A recent study of HTLV-1 positive blood donors found a prevalence of strongyloidiasis (based on positive serology) of around 3%.¹⁶ HTLV-1 is a known risk factor for the development of hyperinfection or disseminated strongyloidiasis, as registered in one of the cases in the series, and S. stercoralis may have an impact on the course of HTLV-1 infection, having been considered a co-factor in the development of certain associated diseases.^17–19 Therefore, screening for Strongyloides in patients with HTLV infection should be mandatory and patients with strongyloidiasis should be screened for risk factors. Diagnosis of strongyloidiasis may be difficult due to the low sensitivity of stool microscopy, the low specificity of serology and the limited availability of tests for specific DNA detection in stool. The patient with the Strongyloides hyperinfection syndrome did not receive ivermectin before corticosteroid use and the development of this complication then led to the diagnosis of HTLV infection. During the COVID-19 pandemic, the World Health Organization issued an alert regarding this complication of strongyloidiasis and recommended presumptive treatment with ivermectin (with/without laboratory screening) for those at moderate-high risk of hyperinfection.^20,21 Therefore, ivermectin for strongyloidiasis should be considered in patients from endemic areas with associated risk factors. Four patients (3.7%) were diagnosed with HTLV/HIV co-infection, not an unusual finding as both viruses share transmission routes, and prevalence was similar to that found in other studies.¹³

As the network database serves mainly as an initial registry of imported diagnoses, details on treatment and follow-up of patients were not comprehensively available for all. Patients were not routinely screened for HTLV infection and further studies and specific protocols to decrease the under-diagnosis of this infection would therefore aid in the management of these patients. A high proviral load of HTLV-1 in peripheral blood mononuclear cells has been associated with disease onset and progression for ATL and TSP, and this could be used as a biological marker of disease activity and progression for these patients.^22,23 A more detailed analysis of associated comorbidities and concomitant diagnoses may also aid in improving the knowledge of other health effects of HTLV infection.^9,24 The current analysis of data on imported infections in over 23 500 patients, including 17 000 immigrants and the description of characteristics in migrants with HTLV infection highlights certain important issues. In this series in a non-endemic area, HTLV infections (the majority due to HTLV-1) and associated conditions were mainly diagnosed in asymptomatic patients from Latin America or the Central America-Caribbean area, following screening usually at blood banks. Of note, these were generally long-settled immigrants, and the majority were female with the consequent implications for screening of contacts and prevention. These patients especially should probably undergo a complete work-up including plasma viral load, which has been associated with an increased risk of mother to child transmission, and be included in specific health promotion and prevention strategies.²⁵ Intrafamilial ‘silent’ transmission via vertical and sexual routes within family clusters may occur, and a prevalence of up to 20–30% in relatives of index cases infected with HTLV-1 has been found in studies in certain South American countries.^26,27 In the majority of cases in the current series (>70%), screening of potential contacts was not performed or not registered, representing a missed opportunity for decreasing the under diagnosis of this infection. Specific preventive measures would include cessation of breastfeeding or breast milk freeze–thaw methods as well as considering antibody screening among all blood donors and leucoreduction of blood products.¹

Increased awareness of HTLV infections, mainly due to HTLV-1, and associated conditions are necessary both in endemic and non-endemic areas, where prevalence may be higher in specific groups such as immigrants, to build robust protocolized public health strategies and close the current ‘gaps’ in epidemiological knowledge.

Conflict of interest

There are no relevant conflicts of interest to declare.

Funding

None.

Author contributions

F.F.N. designed the study, analysed the data and wrote the paper. B.G. assisted with data analysis. All authors contributed to patient management and critically reviewed the final manuscript.

+Redivi Study Group: Magdalena García Rodríguez, Jose Ignacio Mateo González (Consorcio Hospital General Universitario de Valencia, Comunidad Valenciana); Marta Díaz-Menéndez, Natalia Romero (Hospital Carlos III-La Paz, Comunidad de Madrid); Yolanda Meije, María Alejandra Duarte Borges, Xavier Sanz (Hospital de Barcelona, Cataluña); Ana Lucas Dato, Inmaculada González Cuello, Belén Martínez López, Jara Llenas-García (Hospital de la Vega Baja de Orihuela, Comunidad Valenciana); Mónica Romero, José Lorca Barchin (Hospital General de Elda-Virgen de la Salud, Comunidad Valenciana); Diego Torrús, José Manuel Ramos Rincón (Hospital General Universitario de Alicante, Comunidad Valenciana); Inés Suárez-García (Hospital Infanta Sofía, Comunidad de Madrid); Ana Pérez de Ayala, María Asunción Pérez-Jocoiste, Manuel Lisozoain, Cristina Epalza, Carlos Zarco, Irene Losada (Hospital Universitario 12 de Octubre, Madrid Comunidad de Madrid); Azucena Rodríguez-Guardado, Jonathan Fernández Suárez, Jose Antonio Boga Ribeiro (Hospital Universitario Central de Asturias, Oviedo Principado de Asturias); Josune Goikoetxea Aguirre (Hospital Universitario Cruces de Barakaldo, Bilbao, País Vasco); Miren Zuriñe Zubero Sulibarría, Itxaso Lombide Aguirre (Hospital Universitario de Basurto de Bilbao, País Vasco); José Manuel Ruiz Giardín, Juan Víctor Sanmartín López (Hospital Universitario de Fuenlabrada, Comunidad de Madrid); María Velasco Arribas, Mª Teresa Filigheddu, Leonor Moreno Núñez, Oriol Martín Segarra (Hospital Universitario Fundación de Alcorcón, Comunidad de Madrid); Eva Calabuig, María Tasias Pitarch, Marta Montero Alonso, Sandra Cuéllar Tovar (Hospital Universitario y Politécnico La Fe, Universidad de Valencia); Sara Nístal Juncos, César Henríquez (Hospital Universitario Rey Juan Carlos, Comunidad de Madrid); Ana Mena Ribas, María Peñaranda Vera (Hospital Universitario Son Espases, Islas Baleares); Pau Bosch-Nicolau, Israel Molina, Adrián Sánchez Montalvá, Fernando Salvador (Hospital Universitario Vall d’Hebron, Cataluña); Ángel Domínguez (Hospital Universitario Virgen de la Macarena de Sevilla, Andalucía); Begoña Monge-Maillo, Francesca Norman, Jose Antonio Pérez-Molina, Rogelio López-Vélez, Sandra Chamorro Tojeiro (Servicio de Enfermedades Infecciosas, Hospital Ramón y Cajal, Comunidad de Madrid); Begoña Treviño-Maruri, Nuria Serre Delcor, Antonio Soriano-Arandes, Diana Pou Ciruelo, María Espiau Guarner, Inés Oliveira Souto, María Luisa Aznar Ruíz de la Alegría (Unitat de Salut Internacional Drassanes, PROSICS, Cataluña); Moncef Belhassen García, Amparo López Bernus (Complejo Asistencial Universitario de Salamanca, Comunidad de Castilla y León); Francisco Jover Díaz, Philip Erick Wikman-Jorgensen, Elisabet Delgado (Hospital Universitario San Juan de Alicante, Comunidad Valenciana).

References