Natural History of Leishmania (Leishmania) chagasi infection in Northeastern Brazil: Long-Term Follow-Up

Leishmania (Leishmania) chagasi produces a spectrum of clinical findings ranging from asymptomatic, self-resolving infection to progressive visceral leishmaniasis characterized by fever, hepatosplenomegaly, hypergammaglobulinemia, wasting, and death, if treatment is not instituted [1, 2]. Latent infection is well documented. Subclinical infections with isolated splenomegaly, weight loss, intermittent fever, or diarrhea have been reported [3]. On some occasions, visceral leishmaniasis has first become manifest many years after a person has moved away from an area of endemicity and then become immunocompromised [4, 5]. This current report is the first to describe the natural history of L. chagasi infection in a cohort of persons followed prospectively over a decade. L. chagasi is endemic in rural areas of the state of Ceará in northeastern Brazil, where it produces substantial morbidity and mortality. In 1987, a prospective study was initiated near the town of Brotas to characterize the epidemiology and natural history of infection [1]. On the basis of a review of previously reported cases of visceral leishmaniasis in the state, an area of ∼100 km2 was identified for study. Preliminary data indicated that 24 of ∼800 children living there had been treated for visceral leishmaniasis during the previous 4 years. Chagas' disease and cutaneous leishmaniasis were not endemic in the region. Children aged <11 years were enrolled in the study. Histories were obtained and physical examinations were performed, and blood was drawn 6 times from June 1987 through June 1989 [1]. A sensitive and specific ELISA using promastigotes from a local L. chagasi isolate was used to detect antileishmanial antibodies [6]. Each month, mothers or guardians were questioned about their children's health and symptoms suggestive of visceral leishmaniasis. When indicators of disease were present [1], children were referred for medical evaluation and diagnostic studies for L. chagasi. Intensive surveillance continued until June 1990. The cohort was re-evaluated in September 1994 [7]. A total of 920 children <11 years of age were enrolled in the study. Seroconversion was documented in 108 children; the cumulative annual incidence of infection was 4.6%. Of the sero-convertors, 12 (11.1%) developed visceral leishmaniasis during the first 3 years of observation. Children living in a household with a prior case of visceral leishmaniasis had a 3-fold increased risk of infection. Of the remaining 96 seroconvertors, 1 had developed visceral leishmaniasis and none had died at the time of evaluation in 1994 [7]. That 1 case developed visceral leishmaniasis 14 months after becoming seropositive. A 10-year follow-up was conducted in November 1997 to further characterize the long-term natural history of infection due to L. chagasi in the remaining seroconvertors. Individuals were located by their previous family unit and by their father's name. Follow-up information was available from family members or reliable sources on all 95 seroconvertors who had been asymptomatic; a complete history was obtained, and physical examination was performed on the 72 who were present at the time of study. None of the remaining seroconvertors had developed visceral leishmaniasis or died. Of those examined, none had fever, hepatosplenomegaly, or other findings suggestive of visceral leishmaniasis. In addition, public health records were reviewed to identify cases of visceral leishmaniasis that had occurred in the area. Only 3 cases were reported from 1994 to 1997; none were seroconvertors. Two of the 3 cases occurred in households with previous cases of visceral leishmaniasis. All were successfully treated with meglumine antimoniate. The incidence of visceral leishmaniasis decreased in the study area over the decade. During the 4 years before the study, there were an average of 6 new cases per year, whereas during the first 3 years of the project only 12 cases of visceral leishmaniasis were reported. The explanation for the decrease in the incidence of visceral leishmaniasis in the area is uncertain. The number of cases in Ceará and other areas of Brazil has been observed to fluctuate in ∼10-year cycles since visceral leishmaniasis was first identified there in the late 1930s. Differences in rainfall, vector density, and herd immunity have been postulated to be important factors [8]. Several additional changes were observed during the period of study. In general, economic conditions improved in this region as they had elsewhere in Brazil, and public health measures were strengthened. Although not quantified, the number of dogs, which are a reservoir for L. chagasi, seemed to have decreased. Finally, some of the families moved from rural farming sites to small towns. Whether these factors affected the transmission of L. chagasi is uncertain. In summary, none of the 95 asymptomatic seroconvertors followed longitudinally over a decade developed visceral leishmaniasis. The data suggest that if L. chagasi infection does not progress to visceral leishmaniasis within the first or second year after infection (the longest incubation period in this study being 14 months), it is unlikely to do so in otherwise healthy individuals. However, although late recrudescences were not observed, they might occur in the future if L. chagasi-infected persons become immunocompromised by concurrent HIV infection, chemotherapy, or other factors.