|Home | About | Journals | Submit | Contact Us | Français|
We tested 2,655 Peruvian MSM for retroviral infection: HTLV-1 was detected in 48 (1.8%), HTLV-2 in 28 (1.1%), and HTLV-1 and -2 in 5 (0.2%); HIV infection was detected in 329 (12.4 %); 24 (7.3%) were coinfected with HTLV. Risk factors for HTLV-1 and -2 varied with sexual role.
Worldwide, 10–20 million people are infected with HTLV-1 . In Peru, HTLV-1 seroprevalence rates vary from 1–2% in pregnant women and indigenous groups to 4–25% in female commercial sex workers (CSW) to 6% in men who have sex with men (MSM) [2–6]. In MSM, risk factors for HTLV-1 infection included older age, greater number of lifetime male partners, and receptive anal intercourse. [6–10]. The study examining concomitant STI, did not find an association between HTLV-1 infection and gonorrhea or syphilis .
HTLV-2 infection is also endemic in many parts of the world. In Peru, HTLV-2 seroprevalence rates vary from 2–4% in indigenous groups to 1.3% in non-IDU MSM. Risk factors for HTLV-2 infection in MSM included older age, unprotected anal intercourse, syphilis and herpes simplex type-2 (HSV-2) infection [4, 11].
HIV infection in Peru is largely confined to MSM, with seroprevalence rates between 13.9% and 22.3% [12, 13]. Seroprevalence rates are 0.2%, in the general population, 0.5% in pregnant women and up to 6.9% in indigenous Amazonian groups [14–16]. In MSM, risk factors for HIV infection included syphilis, HSV-2 infection, employment as a sex worker, unprotected sex, oral receptive sex and self-identifying as homosexual .
Transmission of retroviral infections in Peru occurs within a complex mix of sexual roles and sexual networks. MSM typically segregate into insertive, receptive or versatile (both insertive and receptive) roles during anal sex, with a higher percentage of MSM adopting a versatile role in Lima and higher percentage of receptive role in the provinces . Receptive role was strongly correlated with self-identification as homosexual and insertive role with self-identification as heterosexual . In Peru, 11% of sexually experienced men have participated in same-sex contact, with nearly all self-identifying as heterosexual . Versatile MSM may promote greater spread of HIV infection as they can become infected while receptive and transmit when insertive . We examined prevalent infections of HTLV-1, HTLV-2 and HIV in MSM to determine risk factors and sexual roles associated with specific retroviral infections.
Males 18 years of age or older who reported sexual intercourse with one or more men during the past year and living in Arequipa (Highlands), Iquitos or Pucallpa (Amazon Jungle), Lima, Piura or Sullana (Coast), were eligible to participate in an HIV and STI sentinel surveillance conducted 2002–2003 . Subjects were recruited using outreach work and ‘snowball’ techniques by trained recruiters and peer educators, whereby recruited subjects were invited to receive training to recruit additional subjects. After obtaining consent, we collected information regarding demographics, sexual identity and behavior, and sexual partners. A study physician performed medical history and targeted physical examination. Blood and genital samples were obtained for retroviral and STI testing. The study protocol was approved by the Peruvian National AIDS and STD Control Program, and Institutional Review Boards of the INMENSA, the U.S. NMRCD, and University of Washington.
Blood samples were tested for anti-HIV-1 antibodies using ELISA (Vironostika, Organon Tecnica, Durham, North Carolina) with Western Blot confirmation (Biorad Laboratories, Hercules, California). Blood was tested for Treponema pallidum by quantitative rapid plasma reagin (RPR, Organon Tecnica) with confirmation by micro hemagglutination assay (MHA-TP, Organon Tecnica); syphilis was defined as RPR titer ≥ 1:1 with reactive MHA-TP. Herpes simplex virus type 2 (HSV-2) infection was detected using type-specific ELISA (Focus Technology, Cypress, California), with seropositivity defined as ELISA index ratio ≥ 3.5.
Serum samples were tested for HTLV-1/2 antibody using ELISA (Vironostika, North Carolina); ELISA positive samples underwent confirmatory testing with Western blot assay (HTLV-1/2 Blot 2.4, Genelabs Diagnostics, Singapore). An individual was considered HTLV-1 seropositive if ELISA assay was positive and confirmatory Western blot assay revealed bands representing gag (p24, p19), gp46, and two env proteins (GD21 and rgp 46-I). An individual was considered HTLV-2 seropositive if ELISA assay was positive and confirmatory Western blot assay revealed bands representing p24, GD21 and rgp46-II. An individual was considered indeterminate if gag and env proteins were absent but other HTLV specific bands, such as p53 or p19, were present; these subjects were excluded from the analysis.
Univariate analyses to identify factors independently associated with retroviral infection included chi-square and Fisher’s exact tests for categorical variables and t-test and Mann-Whitney U tests for continuous variables. All predictors significant at an alpha level of 0.20 in the univariate analyses were included in a stepwise backwards selection process using multivariate logistic regression models. Predictors significant at alpha of 0.05 were retained in the multivariate model. These analyses yielded odds ratios (OR) and 95% confidence intervals (CI). All reported p values represent two-sided tests. Analyses were performed using SPSS (v11.5.0 Chicago, Il) and Stata (v10.0, Stata Corporation, College Station, TX).
Of 3,280 participants participating in the sentinel surveillance, 2,703 consented to future testing of samples. Compared to participants consenting to specimen storage, those not consenting were more likely to report high risk sexual behavior, such as sex work, more HIV-infected male sex partners, and have HIV or HSV infection .
Of 2,703 MSM consenting to specimen storage, 2,655 had samples available for HTLV testing. Of these subjects, 48 (1.8%) had HTLV-1 infection confirmed by Western blot assay, 28 (1.1%) had HTLV-2 infection, 5 (0.2%) had both HTLV-1 and -2 infection, and 329 (12.4%) had HIV infection (Table 1). Of the 329 HIV positive samples, 15 (4.6%) were coinfected with HTLV-1, 6 (1.8%) were coinfected with HTLV-2, and 3 (0.9%) were coinfected with both HTLV-1 and -2. Western blot assay was indeterminate for HTLV in 12.
Compared to MSM without retroviral infection, MSM with HTLV-1 or -2 infection, but not HIV, were older, and more likely to have syphilis, HSV-2 infection, or an anorectal abnormality detected on rectal examination (Table 1). MSM with HTLV-2 infection, regardless of HIV status, were more likely to self-identify as homosexual, report participating in only receptive unprotected anal intercourse (UAI) and be from Iquitos or Pucallpa. Restricting the analysis to MSM with HIV infection, subjects with HTLV-1 coinfection were more likely to consider themselves a sex worker and had a greater number of male partners (Table 2). In multivariate analyses, both HTLV-1 and -2 infections were significantly associated with syphilis and HSV-2 infections, older age and retroviral coinfection, but only HTLV-2 was associated with receptive UAI; HIV was associated with HTLV-1 and HSV-2 coinfections (Table 3).
This study is the first to examine risk factors for retroviral infections in MSM who historically have not used injection drugs. Overall, 81 (3.1%) of 2,655 MSM were infected with HTLV-1, HTLV-2, or both; HIV infection was detected in 329 (12.4%). Compared to MSM without retroviral infection, MSM with any retroviral infection were more likely to be older, self identify as homosexual, have fewer female partners, practice receptive UAI, have anorectal abnormalities detected on physical examination and have syphilis or HSV-2 infection.
Subtle differences in risk factors for each retroviral infection suggest different sexual roles carry different risks for acquisition. As HTLV-2 infection was most prevalent in Iquitos and Pucallpa - where HTLV-2 has been detected in indigenous groups - it is plausible that sexual transmission of HTLV-2 occurred from indigenous groups into those MSM whose receptive UAI roles placed them at highest risk for acquiring HTLV-2 infection and that HTLV-2 may be transmitted less efficiently than other retroviruses. As HTLV-1 infection was strongly associated with bisexuality and self-perception as a sex worker (for HIV coinfected subjects), it is plausible these MSM were sexually active in networks that included female sex workers – who historically have higher prevalence rates of HTLV-1 infection. As expected, HIV infection was more common in MSM who reported a variety of high-risk behaviors, as well as physical and serological markers of sexually transmitted infections, suggesting these MSM are at highest risk for acquiring retroviral infection.
Our study likely underestimated the prevalence of HTLV infection and associated high-risk behaviors for two reasons: 1) the 577 participants from the sentinel surveillance who did not agree to have samples stored had higher risk behaviors and were thus more likely to have retroviral infections; and 2) we did not use PCR assay to examine indeterminate HTLV assays – a technique that previously demonstrated HTLV-2 infection in indeterminate samples . In addition, we did not ask MSM about partners from indigenous communities, limiting our ability to examine this factor for HTLV-2 infection. Another limitation is lack of CD4 cell count for many subjects, preventing us from examining associations between severity of HIV infection and coinfection with HTLV-1 or -2.
In conclusion, retroviral infections were common in Peruvian MSM and associated with high-risk behaviors, syphilis and HSV-2 infections. Our analysis suggests different sexual roles and networks and perhaps different transmission efficiency are associated with varying risks for transmission of specific retroviral infections. Future studies will more closely examine the role sexual networks play in transmission of retroviral infections between MSM, female CSW and indigenous communities and examine effects of retroviral coinfection upon the natural history of HIV and HTLV infections.
The authors gratefully acknowledge the contributions of Drs. Jorge Vergara and Silvia Montano for local study leadership and coordination; Mrs. Gloria Chauca (U.S. Naval Medical Research Center Detachment, Lima, Peru) for assistance with laboratory testing; Dr. Jose Sabastien for coordination with the Peruvian National Strategy for HIV Control and to all recruiters and peer educators; and mainly, to all study participants. Special thanks to U.S. NMCRD support in the implementation of the Second Generation Sentinel Surveillance.
Sponsorship: This study was supported by NIH Fogarty International Center grant R21NS048838 and RO1NS55627 to Joseph R. Zunt, and University of Washington Center for AIDS Research and (CFAR) Grant AI27757
PERUVIAN HIV SENTINEL SURVEILLANCE WORKING GROUP
Members of the Peruvian HIV Sentinel Surveillance Working Group include: Luis Suarez, Monica Pun (General Directorate of Epidemiology, Ministry of Health, Lima, Peru); Cesar Cabezas, Patricia Caballero (National Institute of Health, Ministry of Health, Lima, Peru); Jorge Sanchez, Javier Lama, Juan Guanira, Aldo Lucchetti, Pedro Goicochea, Jorge Vergara (Investigaciones Medicas en Salud, Lima, Peru); Martin Casapia, Juan Carlos Hinojosa (Asociacion Civil Selva Amazonica, Iquitos, Peru); Victoria Zamalloa (Instituto Sur Peruano de Infectologia, Arequipa, Peru); Abner Ortiz (Centro Medico Cayetano Heredia, Pucallpa, Peru); Nora Ojeda (Asociacion de Servicios Generales de Salud y Educacion, Sullana, Peru); Anabeli Tataje (Policlinico Daniel Alcides Carrion, Ica, Peru); Pablo Campos, Patricia Garcia, Cesar Carcamo (Universidad Peruana Cayetano Heredia, Lima, Peru); Connie L. Celum, King K. Holmes, Joseph Zunt, William L. H. Whittington, James P. Hughes (University of Washington, Seattle, WA); Jose L. Sanchez (past member), Silvia Montano, Victor A. Laguna-Torres and Tadeusz Kochel (US Naval Medical Research Center Detachment, Lima, Peru).
This work was presented in part at the 53rd annual meeting of the American Society of Tropical Medicine and Hygiene, Miami Beach, Florida, 2004; abstract 2004-L-2044.
The opinions and assertions made by the authors do not reflect the official position or opinion of the government of the Republic of Peru or the Ministry of Health of Peru.
All authors certify that they have no potential conflicts of interest to disclose.