The American Journal of Tropical Medicine and Hygiene - Volume 53, Issue 2, August 1995

The priorities of public health planners are often at variance with the community's own environmental sanitation priorities and perspectives. Public opinion about individual, collective, and governmental responsibilities in addressing these issues and priorities is of particular importance when designing community-based programs. In a study conducted in Trinidad and Tobago on knowledge, attitudes, and practice regarding dengue, its prevention and control, a high level of awareness about dengue and its etiology was evident, but there was poor understanding of the symptoms and hence little concern about the health risks associated with it. The most important household pest problem identified by the respondents was related to mosquito nuisance, particularly from night-biting mosquitoes. Rodents were also a major concern perceived as being responsible for economic losses, ruined food, and a health hazard. Unreliable water supply, a factor associated with Aedes aegypti abundance, was an environmental sanitation issue of major importance to householders in rural areas. No correlation was found between knowledge of dengue and levels of Ae. aegypti abundance as measured by larval surveys of the respondents' premises. The study gave a clear indication of the need for broad-based environmental sanitation strategies when planning community-based vector control initiatives for the prevention and control of dengue in Trinidad and Tobago.

A seroprevalence survey of hepatitis B virus (HBV) markers was conducted among health care workers in Belize to help determine the epidemiology of hepatitis B and to determine if screening before immunization might lower vaccine costs. Of the 330 workers tested, 94 (29%) were positive for antibody to HBV core antigen (anti-HBc) and three (1%) had HBV surface antigen. The presence of anti-HBc increased significantly with age from 12% in those 18–24 years old to 52% in those ≥ 50 years old. The rate was 17% of 48 men compared with 30% of 282 women (P = 0.05). Rates increased with years of medical service and were higher among nurses (69 of 228; 30%) and nonprofessional staff (15 of 44; 34%) than among physicians (0 of 20). The presence of anti-HBc also differed significantly among ethnic groups: Mestizo, 4%; Creole, 33% and Garifuna, 57%. Rates differed by district ranging from 3% in a northern district (mostly Mestizo) to 67% in a southern district (mostly Garifuna). Parenteral exposure to hepatitis B through needle stick injuries and blood transfusions was not associated with anti-HBc. Multiple logistic regression analysis confirmed ethnicity, district of residence, and age as the best predictors of anti-HBc in health care workers. Cost analysis suggests that because of regional differences in exposure, testing of health care workers for anti-HBc in the Belize and Stann Creek districts in southern Belize before hepatitis B immunization would result in vaccine program cost savings.

In Virginia, Borrelia burgdorferi was more prevalent in a site along the Atlantic Ocean, near Maryland, than in an inland site near Williamsburg and Yorktown. At the coastal site on Assateague Island, B. burgdorferi was isolated from 4.2% of 475 animals sampled, including four species of small mammals. Serologic tests indicated that 25–37% of the small rodents assayed had been exposed to B. burgdorferi. Immunofluorescence antibody assays specific for B. burgdorferi showed spirochete infection in Ixodes scapularis and Dermacentor variabilis but not in other species of ticks also examined from this site. At another coastal site (Parramore Island), no evidence of Peromyscus leucopus was found, no immature specimens of I. scapularis were collected, and no isolations were made from numerous raccoons or small mammals sampled. Borrelia burgdorferi infection was found in one I. cookei nymph, but not in numerous specimens of I. scapularis or other tick species from this locality. At the inland site between Williamsburg and Yorktown, B. burgdorferi was isolated from two small mammal species and antibodies to B. burgdorferi were found in only 7–10% of the small mammals sampled. Ixodes scapularis were less abundant at this locality than at the Assateague Island site. Borrelia burgdorferi spirochetes were found in I. scapularis and a single nymph of Amblyomma americanum, but not in any of numerous specimens of four other species. Infection with B. burgdorferi was found in 20% of unfed adult I. scapularis from vegetation, but in only 0.2% of numerous adults from hunter-killed deer. Infection in immature ticks was much lower than at Assateague Island. Borrelia burgdorferi may be more prevalent along the Atlantic coast than in inland areas. Isolations, seroprevalence, immature I. scapularis densities, and spirochete infection rates in ticks were higher at the Assateague Island site than the Williamsburg/Yorktown site. Consequently, the risk of human exposure to Lyme disease may be higher in some parts of the coastal area than elsewhere in Virginia. Overall, B. burgdorferi is less intense in Virginia than in the northeastern United States.

Puumala (PUU) virus is the etiologic agent of nephropathia epidemica (NE) in humans. This disease is highly endemic in Vasterbotten county, Sweden, with an annual incidence of 19.2 (range 3.7–37.4) per 100,000 inhabitants. Voles are considered to be both the main reservoir and the vector of PUU virus. A total of 3,591 rodents (mainly Clethrionomys glareolus, C. rufocanus, and Microtus agrestis) trapped in Vasterbotten between 1979 and 1987 were tested for the presence of PUU virus antibodies by enzyme-linked immunosorbent assay. The predominant species, C. glareolus (71% [2,544 of 3,591]), also had the highest antibody prevalence (19% [483 of 2,544]). In C. glareolus, the antibody prevalence rate increased with weight (age), reaching more than 50% in the heaviest weight group, and suggesting that horizontal infection may be important. The highest frequency (25%) of antibody-positive C. rufocanus was also found in the highest weight groups. Microtus agrestis showed low absolute numbers and a low antibody prevalence rate (5%). In C. glareolus, both antibody prevalence and weight were recurrently higher in the spring than in the previous fall. The antibody prevalence rate in spring was positively correlated with the vole density in the previous fall and spring. The fall antibody prevalence rate was directly dependent on C. glareolus density. The incidence of human NE in the fall was dependent on the concurrent density of C. glareolus, whereas the incidence of NE in the spring was dependent on vole density the previous fall.

Leishmania parasites were isolated from four of seven small rodent species captured in the State of Campeche, Mexico. Signs and/or symptoms of leishmanial infection were restricted to the tail of the mice. Peromyscus yucatanicus endemic to the peninsula of Yucatan and Oryzomys melanotis were added to the list of carriers of Leishmania. Leishmania (L.) mexicana, characterized by an indirect immunofluorescence antibody method using monoclonal antibodies, was identified in one Sigmodon hispidus and one Or. melanotis. The possible role of the different species as primary reservoirs of Leishmania in the peninsula of Yucatan is discussed.

Babesiosis is emerging as an important tick-borne zoonosis in the United States. Most reported cases of this parasitic disease have been acquired in the Northeast. To date, only two clinical cases of Babesia microti infection acquired in the upper Midwest have been described. We report eight more cases. Most if not all of the 10 total cases probably were acquired in northwestern Wisconsin. Three cases (30% of 10) we now report were fatal and occurred in elderly patients (65–75 years old) who died after complicated hospital courses. One patient probably had had a latent Babesia infection that activated because of immunosuppression attributable to high-dose corticosteroid therapy and to splenic infarctions caused by cholesterol emboli. All three fatal cases were diagnosed incidentally and highlight the importance of considering the diagnosis of babesiosis in febrile patients who have been in babesiosis-endemic areas; examining their blood smears carefully; and treating promptly with clindamycin and quinine, and, if indicated, exchange transfusion. Medical personnel should be knowledgeable about this zoonosis, which is not limited to the northeastern United States, and is potentially serious, sometimes fatal.

Quantitative enzyme-linked immunosorbent assays (ELISAs) for the detection of circulating anodic antigen (CAA) and circulating cathodic antigen (CCA) in serum and urine were applied as an epidemiologic tool in a recent, intense focus of Schistosoma mansoni in Senegal. Both CAA and CCA in serum and CCA in urine were found in 94%, 83%, and 95%, respectively, of the population, of which 91% were positive on stool examination. Circulating antigens were also detectable in sera and urines of most egg-negative individuals. The sensitivities of the urine CCA and serum CAA ELISA were substantially higher than that of a single egg count, and increased with egg output. The CAA and CCA levels correlated well with egg counts and with each other. The age-related evolution of antigen levels followed a similar pattern as egg counts, providing supplementary evidence for a genuine reduction of worm burdens in adults in spite of the supposed absence of acquired immunity in this recently exposed community. The antigen:egg ratios decreased in adults, suggesting lower worm fecundity in children. This would be compatible with a density-dependent reduction of fecundity, but not with anti-fecundity immunity in adults that perhaps has not yet developed in this new focus.

Ancylostoma caninum, the cosmopolitan canine hookworm prevalent in domestic pets throughout northeastern Australia, has been identified as a cause of obscure abdominal pain. To investigate the influence of climate on putative and confirmed human infections, a retrospective study of patients with eosinophilic enteritis and established infections was undertaken to identify the monthly and seasonal rates. The numbers seen each month were maintained throughout most of the year but decreased in June and remained low during winter. An increased rate of enteritis occurred during spring, which preceded the wet season. It is concluded that climate directly influences the rate of human enteric infection by canine hookworms. The increase in prevalence before the monsoon season was unexpected and suggests other influences, such as extraneous infection promoted by the irrigation of domestic gardens and possibly endogenous reinfection of the gut by hypobiotic larvae.

A randomized, placebo-controlled, double-blind trial was carried out in 1992 in central Tunisia to assess the tolerability and efficacy of paromomycin ointment against zoonotic cutaneous leishmaniasis caused by Leishmania major. One hundred fifteen patients, 2–60 years of age, with a single lesion of parasitologically confirmed cutaneous leishmaniasis, were included in the trial. The ointment was applied twice a day from day 1 through day 14. Clinical and parasitologic evaluations of lesions were done at days 0, 15, 45, and 105. Fifty-seven patients were allocated the treatment and 58 the placebo. Based on local toxicity and laboratory evaluation, there was no difference in tolerability between the two groups. Parasitologic evaluation at day 15 showed that 74.5% of the treated group had negative smears compared with 56.4% among controls (P = 0.06). This difference was no longer apparent at days 45 and 105. Clinical evaluation at days 15, 45, and 105 did not indicate any difference between the two groups. The clinical evaluation at day 15 was a good predictor of the final prognosis of the lesion in the two groups when analyzed separately, suggesting no clinical relapse in either group. These findings suggest that paromomycin ointment should not be used in the present formulation as a treatment for zoonotic cutaneous leishmaniasis in Tunisia.

Schistosoma mansoni was first reported in the area of Richard Toll (Senegal) in 1988 and spread rapidly in the community, after a series of human-engineered ecologic changes. A random population sample (n = 422) from Ndombo, a village near Richard-Toll, was studied in 1991 by stool examination (four Kato slides from two stool samples) and antigen detection in urine and blood. Stool-positive individuals were treated with 40 mg/kg of praziquantel. A house-to-house interview regarding side effects was conducted 24 hr after treatment. Two hundred ninety-eight subjects were re-examined 10 days (antigen detection) and 12 weeks (egg counts, antigen detection) after treatment. Before treatment, positive egg counts were found in 91% of the subjects, with 41% excreting more than 1,000 eggs per gram (epg) of feces. Treatment of 352 individuals caused serious but transient side effects (colic, vomiting, urticaria, and edema), the frequency of which increased with increasing egg counts. The parasitologic cure rate 12 weeks after treatment was only 18%, the frequency of egg counts with more than 1,000 epg decreased to 5%, and the mean egg count of those remaining positive was reduced by 86%. Antigen detection in serum 10 days and 12 weeks after treatment remained positive in 90% of the subjects, although titers decreased sharply. The low cure rates may be due to intense transmission and/or undeveloped immune responses in this recently exposed population. However, reduced drug susceptibility of the parasite strain has now been confirmed in one local isolate.

A new approach for direct identification and characterization of Trypanosoma cruzi stocks in biological samples was tested for field applicability on an extensive sample of feces collected from triatomine vectors from four different species found in Bolivia. The first step of the technique is polymerase chain reaction (PCR) amplification of the hypervariable region of kinetoplast DNA minicircles of T. cruzi parasites. In this report, 345 fecal samples were analyzed and the PCR results were compared with microscopic examination. For Triatoma infestans, the principal Bolivian vector, both techniques were in concordance 85.3% of the time. For the three other species, Rhodnius pictipes, Eratyrus mucronatus, and Triatoma sordida, the fecal samples were all negative by microscopic examination whereas PCR results showed several T. cruzi-infected insects in each species. The second step of the procedure is the characterization of the T. cruzi clones by means of hybridization of the PCR products with clone-specific probes generated by the PCR. We used two probes corresponding to major clones circulating in high frequency in Bolivia (as shown by previous population genetic studies using isoenzyme characterization). We obtained four primary results: 1) we confirm the importance of two major clones in Bolivia in two distinct regions; 2) we report high rates of mixed infections (multiple clones in a single vector) in Triatoma infestans, up to 22% and 35% in Cochabamba and La Paz departments, respectively; 3) the results favor the absence of interaction between different clones; and 4) we find, for the first time, evidence of the major clones circulating in three species of triatomines that are known as mainly sylvatic species. The origin of these clones, sylvatic or domestic, is also discussed.

A human case of infection by Trichinella pseudospiralis has recently been described. Some morphologic anomalies of the muscle larvae, however, raise the possibility of an incorrect taxonomic attribution. A molecular taxonomic approach has therefore been applied for the identification of the parasite. Random amplified polymorphic DNAs were obtained from a single larva extracted from a muscle biopsy of the suspected case of T. pseudospiralis infection, and compared with those derived from 27 reference strains of Trichinella spp. Nearly identical amplification patterns were obtained from the suspected larva and from reference strains of T. pseudospiralis, thus supporting the original morphology-based identification. An enzyme-linked immunosorbent assay and Western blots carried out on pretreatment and post-treatment sera provided further confirmation.

Antigen-specific cellular immunity in paracoccidioidomycosis (PCM) has been poorly studied due to lack of standard in vitro lymphocyte proliferation assays. To standardize such an assay, we studied T and B cell responses to a Paracoccidioides brasiliensis cell wall extract (PbAg) in healthy subjects sensitized to either P. brasiliensis [Pb(+)Hc(-)] or to Histoplasma capsulatum [Hc(+)Pb(-)], and in nonsensitized persons. All subjects showed, as expected, a vigorous proliferative response to a control fungal antigen obtained from Candida albicans. Lymphocytes from Pb(+)Hc(-) donors, but not from Pb(-)Hc(-) donors, reacted to PbAg by proliferating in a dose-dependent manner with a maximum reaction after 6–9 days, suggesting a secondary specific immune response. Most activated cells were CD4+ lymphocytes. However, Hc(+)Pb(-) donors' cells reacted with PbAg. Cross-reactivity with H. capsulatum was not unexpected, since both fungi, but not C. albicans, share cell wall immunogenic compounds. An enzyme-linked immunosorbent assay to detect human immunoglobulins (Ig) demonstrated that B cells from Pb(+)Hc(-) donors, but not from Pb(-)Hc(-) ones, reacted with PbAg by secreting high levels of IgG and IgM in 12-day culture supernatants. This secretion was possibly mediated by PbAg-activated CD4+ cells. We believe that analysis of T and B lymphocyte responses to PbAg will be useful in the investigation of the infection-associated immune impairment seen in some PCM patients.

Forty-three Brazilians were immunized against American tegumentary leishmaniasis using a vaccine made of whole antigens from killed promastigotes of five American dermotropic Leishmania strains. None of the immunized subjects had a positive reaction in the Montenegro skin test (leishmanin) before vaccination, and 74% developed positive reactions in the skin test after vaccination. The proliferative responses of peripheral blood mononuclear cells (PBMC) induced by antigens from dermotropic Leishmania species were significantly higher after vaccination than before vaccination. However, with antigens from L. chagasi (a causative agent of American visceral leishmaniasis), there was no significant difference between the proliferative responses obtained before and after vaccination. Interferon-γ was detected in the supernatants of L. braziliensis antigen-stimulated PBMC cultures after vaccination (but not before vaccination). One year after vaccination, PBMC were obtained from eight of the immunized individuals and stimulated with L. braziliensis antigens in proliferative response assays. In all cases, the majority of the responding cells were CD8+ T cells, in contrast to the results of a group of patients with active lesions of tegumentary leishmaniasis, whose L. braziliensis-reactive cells were mainly of the CD4+ T cell phenotype.

Seroepidemiologic studies using the indirect immunofluorescent antibody test (IFAT) are valuable in malaria control programs in identifying local foci of malaria, in diagnosing malaria in asymptomatic, low-parasitemia blood donors in nonendemic countries, in detecting imported malaria and preventing its introduction into new areas, and in excluding recurrent fever from causes other than malaria. Because other diseases may occur in areas where malaria is prevalent, the aim of this work, using the IFAT, was to determine the frequency of cross-reactions between blood-stage antigens of Plasmodium falciparum and antibodies present in the serum of individuals with leishmaniasis, toxoplasmosis and Chagas' disease. Since malaria transmission does not occur in the study area (State of Minas Gerais, Brazil) where these other diseases are present, we studied sera from individuals living in this area who had never been in the areas endemic for malaria in the Amazon region. Positive reactivity of sera with blood malaria antigens evaluated by IFAT at dilutions ≥ 1:40 was detected in 19 (38%) of 50 individuals with cutaneous leishmaniasis, five (62%) of eight individuals with visceral leishmaniasis, 14 (32%) of 44 individuals with Chagas' disease, four (11%) of 36 individuals with toxoplasmosis, and in none of the 14 uninfected controls. All 23 of the control malaria sera from the endemic area (State of Mato Grosso, Brazil) were positive at high dilutions. We found no correlation between titers of the IFAT with malaria and the specific antigens used for serodiagnosis of the other protozoan infections studied. At dilutions of 1:20 and 1:40, the sensitivity of the IFAT test was 100% and specificity was 52% and 72%, respectively. At a dilution of 1:80, the sensitivity was 86% and the specificity was 90%. Thus, the choice of the final dilution of serum for use for the diagnosis of malaria should take into account the study population and the type of parasites prevalent in the area. At the present time, leishmaniasis is endemic in many areas of the Amazon region where malaria is endemic. Thus, we recommend the use of sera diluted ≥ 1:80 to exclude cross-reactive antibodies to this parasite in seroepidemiologic studies of malaria.

Experimental primary infection with Plasmodium berghei in rats is known to be influenced by several cytokines. Dietary supplementation of n-3 fatty acids has been shown to influence cytokine production capacity and to protect mice from cerebral malaria. We investigated the effect of dietary fish oil (FO) supplementation on cytokine and nitric oxide production and liver schizont development in male brown Norway rats. Control groups were fed either a corn oil-supplemented diet (CO) or standard lab chow (LC). After six weeks on either diet, rats given supplementary FO had a significantly lower production of interleukin-1 (IL-1) and IL-6 after stimulation with lipopolysaccharide, and also had significantly lower production of interleukin-1 (IL-1) and IL-6 after stimulation with lipopolysaccharide, and also had significantly lower numbers of liver schizonts compared with CO- or LC-fed animals. We conclude that in rats, an FO-supplemented diet reduces the production capacity of IL-1 and IL-6 and inhibits schizont development after intravenous inoculation of P. berghei sporozoites. Fish oil did not influence nitric oxide production by peritoneal macrophages.

In spite of being separated by more than 20 million years of evolution, the murine and human immune systems share extensive similarities. Thus, experimental results obtained with the murine model may have predictive value for human Chagas' disease. Challenge of the H-2 congenic mouse strains A.SW (H-2s) and A.CA (H-2f) with Trypanosoma cruzi yields different results. The A.CA animals die approximately 12 days postinfection, while A.SW mice survive indefinitely. A 45-kD protein (Tc45), an antigen differentially recognized by the A.SW strain, is present in cultured epimastigotes and blood trypomastigotes. We describe here its purification from epimastigotes. The presence of Tc45 was monitored and a single band was detected. Since the molecular weights of Tc45, cruzipain, cruzain, and a 46-kD parasite polypeptide are similar, it was important to determine if these molecules are related. A complete lack of homology was observed when the sequences of cruzain, cruzipain, and the 46-kD polypeptide were compared with the preliminary sequence of Tc45.