Co-infection with HTLV-1 is associated with a shorter survival time for HIV-1-infected patients in Bahia, Brazil

This is the authors' abstract. We don't add key points for this paper.

Co-infection with HTLV-1 reaches 20% among patients infected by HIV-1 in Bahia, Brazil. To evaluate its impact on survival, we conducted a retrospective, case–control study involving 198 patients (63 cases). Co-infection was associated with parenteral exposure (P= 0.0001) and female sex (P= 0.02). Co-infected patients had a shorter mean survival (1849 days) than controls (2430 days,P= 0.001), regardless of sex or baseline CD4 cell count. In Bahia, Brazil, co-infection with HIV-1 and HTLV-1 is associated with a shorter survival time. HIV and HTLV-1 share the same routes of transmission (sexual, parenteral and vertical). In consequence, co-infection with these two agents is a common finding in certain areas of the world, where they are prevalent. Although there is still some controversy about the reciprocal effects of co-infection by these agents, it has already been well documented that such interaction can modify the biological properties of both viruses [1,2]. The available in-vitro information suggests that co-infection with HTLV-1 in HIV-infected patients can promote HIV replication, increase CD4 cell counts, and make cells that are usually not susceptible to HIV susceptible to infection by that virus [3–6]. Although these reports support the hypothesis that co-infection with HTLV-1 could accelerate AIDS progression, there are scarce data about its impact on the mortality of AIDS patients. In Salvador, a city with 2.5 million inhabitants in the northeast region of Brazil, the estimated prevalence of such infections among the general population is 2% for HTLV-1 and 0.3% for HIV-1 [7,8]. However, among patients infected by HIV-1, the prevalence of HTLV-1 infection is as high as 20%[8,9]. We evaluated a cohort of HIV patients, in order to assess the impact of HIV–HTLV co-infection on survival. We reviewed the medical charts of patients attending the AIDS clinics of the Federal University of Bahia Hospital, which cares for approximately 65% of all HIV-infected patients in the state. We conducted a retrospective, case–control study for an evaluation of survival in a cohort of HIV-infected patients, co-infected (cases) or not (controls) with HTLV-1. All patients having an available serology result for HTLV-1 and a follow-up period greater than 180 days were eligible for the study. A definition of cases included a Western blot confirmed infection by both agents, whereas controls were those with a confirmed infection by HIV-1, but a negative serology (enzyme immunoassay) for HTLV-1. Cases and controls were matched by the date of diagnosis of HIV infection. The medical charts of all patients diagnosed with HIV infection during a 10 year period (1989–1999) were revised. All patients with at least 180 days of follow-up, and with available data on serology for HTLV were included. We evaluated the mortality rates for cases and controls, according to sex and age. Risk factors for the acquisition of retroviral infection and, when available, CD4/CD8 cell counts and plasma viral loads were also recorded. When more than one value was available, the first result obtained after the diagnosis date was recorded. For analysis purposes, survival time was censored at 3000 days. A total of 198 patients (63 cases) were included for analysis. Two-thirds (75%) of patients were men. Women were more likely to be cases than controls [relative risk 1.93; 95% confidence interval (CI) 1.26–2.95;P = 0.002, Yates corrected]. The mean age was 32.6 years (median 33), and it was similar for cases and controls. Sexual exposure was the likely route of acquisition of retroviral infection for 38% of patients. The use of intravenous drugs (10% of patients), contaminated blood transfusion (4%), mother-to-child transmission (3%), and a history of a previous sexually transmitted disease (3%) were other detected risks for infection. In addition, 43% of patients reported at least two of the risks listed above. Co-infection with HTLV-1 was strongly associated with parenteral exposure [odds ratio (OR) 4.77; 95% CI 2.3–9.9;P = 0.00001, Yates corrected]. Patients using intravenous drugs were also significantly younger (mean age 28.9 ± 9.5 versus 33 ± 8.5 years;P = 0.009, Kruskal–Wallis test), and had a lower probability of being alive (OR 0.42; 95% CI 0.2–0.85;P = 0.02) than those reporting sexual exposure. Men were significantly less exposed to infection through the parenteral route (OR 0.31; 95% CI 0.16–0.6;P = 0.0008) than women. They were also less likely to be co-infected (OR 0.56; 95% CI 0.4–0.79). CD4 cell counts were available for 115 subjects, and it was similar for cases and controls (446 ± 402 versus 443 ± 357 cells/mm3, for cases and controls, respectively;P = 0.7). Co-infected women had a mean CD4 cell count greater than men (274 ± 224 versus 918 ± 411;P = 0.0003), but these values were similar for both sexes among controls (442 ± 368 cells/mm3 for men and 449 ± 316 cells/mm3 for women;P = 0.7). Also, the levels of CD4 cell were comparable for patients with parenteral or sexual exposure (P = 0.2). The overall mortality rate was 44%. The overall mean survival time was 1772 ± 1058 days. Cases had a shorter survival (mean 1849 days; median 2378 days) than controls (mean 2430 days; median 3000 days;P = 0.001, log rank test), as shown in Fig. 1. We found no difference in mortality according to sex within cases or control groups.Fig. 1.: Kaplan–Mayer curve displaying the cumulative survival rates for cases and controls.Patients who died during the follow-up period had an initial CD4 cell count similar to survivors (481 ± 417 versus 371 ± 244cells/mm3 for alive and dead patients, respectively;P = 0.4). In conclusion, in Bahia, Brazil, co-infection with HTLV-1 is associated with a shorter survival time for HIV-infected patients. The major risk for co-infection appears to be the use of intravenous drugs, mostly among women. Although the retrospective design of this study did not permit a definitive answer to the question about the real impact of co-infection on AIDS progression, HTLV-1 co-infection seems to be a major risk factor for mortality among HIV-infected patients. Carlos Britesa,b Roberta Alencara Rafael Gusmãoa Célia Pedrosob Eduardo M. Nettob Diana Pedral-Sampaiob Roberto Badaróa,b