The American Journal of Tropical Medicine and Hygiene - Volume 47, Issue 1, 1992

Travel Medicine. The first three papers in this issue deal with a subject not previously addressed as a specific entity in the Journal. German investigators report on a study involving two regimens of halofantrine and 74 cases of acute falciparum malaria acquired in areas of Africa known to have drug-resistant parasites (see page 1). Treatments with drug administration on two days seven days apart was found to be more efficacious than with a single day treatment. This paper emphasizes the problems faced by clinicians dealing with the management of acute malaria in non-immunes who have a history of travel in areas endemic for drug-resistant malaria. Starting on page 6 is an interesting paper from Spain concerned with the diagnosis of three species of schistosomes from a cluster of patients who appeared in a traveller's clinic in Barcelona. All had visited the Dogon country in Mali and had been swimming in one or more rivers and streams.

A multicenter prospective trial was performed to investigate the efficacy and the tolerability of halofantrine in nonimmune patients with malaria imported from areas with drug-resistant falciparum parasites (mainly Africa). Forty-five of the 74 subjects were treated with a one-day regimen (3 × 500 mg) of halofantrine, and the other 29 received the same regimen with an additional treatment on day 7. In the second group, a 100% efficacy rate was demonstrated, but in the group receiving the one-day regimen, four recrudescences were observed in patients with falciparum malaria. Only five mild adverse reactions were seen, which disappeared spontaneously after the end of the treatment. We conclude that halofantrine is highly effective in curing malaria in nonimmune subjects. The treatment scheme for such persons should include an additional treatment on day 7 for nonimmune individuals. This drug was well tolerated in our patients, indicating that halofantrine will be useful in the treatment of multidrug-resistant malaria in nonimmune persons.

A previously unknown area of schistosomiasis transmission is reported based on findings from a travelers' clinic in Barcelona. Three species of Schistosoma (S. haematobium, S. mansoni, and S. intercalatum) were diagnosed in a cluster of 43 patients who had been swimming in the Bandiagara and Bankas districts of Mali, where the Dogon people live. Three villages in the Bankas district appear to harbor these three species. The transmission potential of such a focus in this area is outlined. The travelers involved had little or no information on the risks of contracting schistosomiasis in that area. Obtaining a traveler's history, including accurate geographic data, is shown to be a crucial asset for improving epidemiologic research.

We report a case of intestinal capillariasis in a 32-year-old Italian man. After he made a trip to Indonesia that lasted approximately one month, he developed heartburn, abdominal pain, irregular bowel movements, headache, fatigue, weight loss, low-grade fever, and severe itching. The diagnosis was provided by the recovery of Capillaria philippinensis eggs in the stool. Treatment with oral albendazole, 200 mg twice a day for 21 days, resulted in clinical and parasitologic cure. This is the first report of C. philippinensis infection acquired in Indonesia.

Giardia lamblia is the most commonly reported enteric pathogen in Wisconsin. Since giardiasis became a notifiable disease, the annual number of cases reported to the Wisconsin Division of Health has increased more than 20-fold, from 2.2 cases per 100,000 population in 1981 to 49.1 cases per 100,000 population in 1988. To better understand the nature of this increasing trend, we reviewed records of G. lamblia infections reported to the Wisconsin Division of Health from 1981 to 1988. Although the increase in reported cases was a general phenomenon that was not limited to a few high-risk groups, the highest annual incidence and greatest increase occurred in children 1–4 years old; 34% of the cases in this age group occurred in children who attended day care centers. A remarkably consistent late summer (August) increase was observed across all demographic and risk groups, suggesting that G. lamblia may be more common in the environment during late summer, or that risk factors for transmission may differ during these months. Additional studies are needed to further explain the increasing incidence and seasonal nature of reported giardiasis and to identify opportunities for prevention.

The dynamics of vectorial transmission of Trypanosoma cruzi and the level of host (guinea pigs) protection after immunization with attenuated parasites (TCC strain) was studied under natural climatic conditions in an endemic region of northern Argentina. The experimental design included two guinea pig corrals isolated by mosquito netting. One (controls) had 17 healthy and susceptible adult guinea pigs. The other had 19 guinea pigs immunized with attenuated T. cruzi TCC strain. Each corral was colonized in April 1988 with equal-sized populations of Triatoma infestans naturally infected by T. cruzi. To evaluate relevant variables in the natural transmission of Chagas' disease, corrals were sampled in both winter and late spring to assess vector populations, and to carry out parasitologic studies on both vertebrate and invertebrate hosts. In both corrals, vector density decreased in winter and reached a maximum in the hot season. The vector infection rate was very high (> 50%) throughout the experiment. Vector infectivity increased with temperature and vector age, but did not differ between the experimental and control corrals. The vector-host contact rate showed a close relationship with temperature, although a very high vector density decreased this rate, even with high ambient temperatures. Initial infections by T. cruzi occurred among guinea pigs only during the hot season. Vectorial transmission risk was estimated from the total number of bug bites per day, the proportion of infected bugs, and the daily incidence in the guinea pig population. During the hot season, this risk was 6.84 × 10-4 in the control group and 1.82 × 10-4 in the immunized group.

Four hundred eighty house mice (Mus musculus) were trapped primarily from urban sites in Baltimore, Maryland from 1984 to 1989 and tested for antibody to lymphocytic choriomeningitis virus (LCMV). The majority of mice (95%) were trapped in residences at two city locations (n = 260), or in an urban park (n = 196); five additional sites were sampled. Overall, 9.0% of the mice were LCMV antibody positive and infected animals were obtained from six of eight sites, including all three of the primary city sites, where the prevalence varied significantly (3.9–13.4%). The location with the highest prevalence was an inner city residential site where positive mice were found significantly clustered within blocks and households. In this location, LCMV antibody prevalence was also significantly correlated with estimates of mouse density within individual blocks. The focal nature of LCMV infection in house mice may result from contact or vertical transmission of virus in conjunction with the highly structured social system of mice, which promotes inbreeding and limited dispersal.

An outbreak of Q fever occurred among patients and staff of a psychiatric institution in southern France. Some of the patients and staff left the institution daily to work on a farm where goats were raised for raw milk and cheese production. The goats had all been vaccinated annually with a commercial vaccine containing phase II Coxiella burnetii antigen. A serologic survey revealed that 40 (66%) of the 61 patients and staff had elevated titers to C. burnetii. Seropositive persons were more likely to report an acute illness (P = 0.001), fever (P = 0.04), weakness (P = 0.04), arthralgia (P = 0.04), and headaches (P = 0.06) in the preceding year than were seronegative persons. Seropositivity rates were significantly higher among persons who worked on the farm and consumed unpasteurized milk products (69% [22 of 32]; P = 0.007), those who only had worked on the farm (75% [9 of 12]; P = 0.009), and those who only had consumed unpasteurized milk products (75% [9 of 12]; P = 0.009), compared with those who had not worked with the goats or consumed unpasteurized milk products (0 of 5). Despite vaccination against Q fever, no antibodies to C. burnetii were detectable in 17 (59%) of 29 goats. All 12 seropositive goats had antibodies to both phase I and phase II antigens, indicating that they were naturally infected, and two of three goats examined were shedding C. burnetii in their milk. Vaccination of this herd did not prevent the outbreak and might have increased shedding of C. burnetii in the dairy products.

Anopheles gambiae s.l. and An. funestus were sampled for Plasmodium spp. ookinetes in two P. falciparum-endemic sites in western Kenya. Since the ookinete is a transitional stage of short duration, occurring after fertilization and before oocyst development, only females in the half-gravid and gravid stages of blood digestion were examined. Preparations of homogenized midguts were spotted onto microslides and examined microscopically after staining with Giemsa. Overall, ookinetes were detected in 4.4% of 1,079 anophelines examined over an eight-month period. Anopheles funestus had higher ookinete rates than An. gambiae s.l., and ookinete rates were higher in half-gravid than in gravid An. gambiae s.l. Geometric mean numbers of ookinetes per infected female were less than five for each species at the two sites, and the maximum number observed was only 12. The low frequencies and numbers of ookinetes were sufficient to produce sporozoite rates of 4–18% in the vector populations. The intense transmission of P. falciparum in these two sites is maintained by anthropophilic vectors where only one in 23 blood meals initiates an infection of generally less than five ookinetes. Relationships between human malaria infectiousness and vector infectivity are dependent upon the high efficiency of the developmental transition from the ookinete to the subsequent oocyst and sporozoite stages.

An outbreak of cutaneous leishmaniasis occurred in a unit of 608 Puerto Rican national guardsmen conducting jungle warfare training in the Panama Canal Area in July 1984. An epidemiologic investigation of reported nonhealing, ulcerating skin lesions was conducted among 540 (89%) unit members in November and December 1984. Fifteen (88%) of 17 individuals with chronic, ulcerating skin lesions were confirmed as cases of cutaneous leishmaniasis by culture or histopathology. Twelve cases yielded positive Leishmania cultures, identified as L. braziliensis panamensis by cellulose acetate electrophoresis. Evaluation of different diagnostic techniques revealed that direct examination of tissues by Giemsa-stained histological examination was the most sensitive test (87% sensitivity), with an indirect immunofluorescent antibody test being rather insensitive (67%). All but one of the confirmed cases operated in small units that trained and slept overnight at a mortar firing site for a period of three days, yielding a site-specific attack rate of 22% (14 of 64). This contrasted with a much lower attack rate of 0.2% (1 of 476), experienced by unit members who trained at other locations during the same time frame (P < 0.001). The median incubation period calculated from day of arrival at the mortar firing site was 17 days (range 2–78) for the 15 confirmed cases. Available personal protection methods, such as the use of insect repellents, were not appropriately implemented by unit personnel and thus, were not found to effectively protect against Leishmania infection. This is the largest reported outbreak of cutaneous leishmaniasis in military personnel associated with a single geographic focus of infection and contrasts with the usual sporadic disease experience in Panama.

In areas where the agent of Lyme disease is intensely enzootic, the mouse reservoirs may be universally infected. Because a large proportion of the vector tick population appears to feed upon these hosts, the prevalence of infection in the vectors should approach 100%. However, infection in host-seeking nymphal ticks in nature rarely exceeds 40%. To help reconcile this apparent paradox, we examined whether estimates of prevalence might differ if we did not assume that infected ticks are randomly or uniformly distributed within a site. Nymphal Ixodes dammini were collected by dragging a series of 10-meter replicates within an intensely enzootic site. Estimates of the prevalence of spirochetal infection, based upon the monthly means of individual 10-meter collections, were then compared with estimates derived by pooling all samples. Host-seeking ticks tended to cluster. The Lyme disease spirochete was present in 15.6% of 469 pooled ticks. When the prevalence estimate was based solely on ticks in clusters that contained one or more infected ticks, however, at least 50% of the ticks were infected. We conclude that nymphal deer ticks infected by Lyme disease spirochetes tend to aggregate spatially in nature, and that prevalence estimates based upon a mean value for pools may be misleading.

Primer-extension sequencing of the RNA template of polio, dengue, Rift Valley fever, and Japanese encephalitis (JE) viruses has provided new information on their geographic distribution, origin, and evolution. In a previous study of 46 diverse JE virus strains, we demonstrated the existence of three distinct JE genotypes in Asia. We now report the occurrence of a fourth genotype. In the present study, 19 JE virus isolates, representing various geographic regions of Asia and a 50-year time span, were compared with each other and with Murray Valley encephalitis, West Nile, and Kunjin viruses. Twelve of the JE strains from the Indonesian Archipelago and the Philippines had not been previously examined; the remainder were representatives of the three previously identified genotypes. Two hundred forty nucleotides from the pre-M gene region of the virus were used in these comparisons. Using 12% divergence as a cut-off point, the 19 JE strains fell into four distinct genotypic groups; maximum divergence across the comparison region was 21%. The newly recognized fourth genotype was comprised of five Indonesian isolates that were 7% divergent from the rest of the JE viruses.

The heat-shock response induced in Trichomonas vaginalis by exposure to various incubation temperatures was traced by metabolic labeling and monitored by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and fluorography. Increasing the incubation temperature from 37°C to 43°C depressed normal protein production and enhanced synthesis of heat-shock proteins (hsp). Smaller increases in incubation temperature resulted in little change in protein synthesis, whereas larger temperature increases inhibited protein synthesis. The hsp produced by T. vaginalis included molecules with approximate molecular masses of 85, 78, 66, 61, 35–31, 20–15, and 12 kD. Trichomonas vaginalis switched to hsp synthesis gradually. Full conversion to hsp production occurred within 60–90 min after the initiation of the 43°C stress. The period of synthesis was different for individual hsp, suggesting independent regulation of hsp production. Four strains of freshly reinitiated and culture-adapted (extended in vitro culture) T. vaginalis synthesized the 85-, 78-, and 66-kD hsp, but varied in the synthesis of the 61-, 35-, 34-, 32-, and 31-kD molecules. Culture adaptation affected the heat-shock response of two of the four strains tested.

Analysis of seven strains designated as Rickettsia conorii for reactivity with a panel of 12 monoclonal antibodies to surface-protein epitopes of spotted fever group rickettsiae and by Western immunoblotting with standard serotyping sera revealed remarkable antigenic diversity. Rickettsial strains from France, Morocco, Ethiopia, Kenya, South Africa, India, and the USSR differed from one another in reactivity with at least one and as many as five monoclonal antibodies. Simko and Indian strains were similar to one another and differed substantially from other R. conorii strains. All seven strains reacted with three R. conorii-specific monoclonal antibodies. Western immunoblotting demonstrated a major 120-kD protein and a major 135-kD protein in all strains. The principal differences were the presence of a major undenatured 130-kD protein in all strains except Indian and Simko, which had an analogous protein of 124 kD. Immunodominant antigenically related, heat-denatured protein bands of 170 kD (Malish 7 strain), 175 kD (Manuel strain), and 190 kD (Kenya tick typhus, Indian, and Simko strains) were not detected in the M-1 and Moroccan strains. This antigenic diversity is greater than that previously reported for other spotted fever group rickettsial species, suggesting that R. conorii is an older species than R. rickettsii with a longer period of time for evolutionary divergence.

Cytotoxic nucleoside derivatives may become useful in the treatment of parasitic infections. As part of our drug development studies, the effect of a number of nucleosides (100 µM) on the cellular transport of 3H-adenosine and 3H-inosine (each at 1 µM) in promastigotes from four Leishmania major strains was investigated. When 3H-inosine was used as permeant, all strains exhibited essentially the same inhibition profile, with unlabeled inosine, guanosine, formycin B, and 3′-deoxyinosine being strongly inhibitory, and adenosine-related compounds such as 2′-deoxyadenosine and tubercidin being inactive. However, when 3H-adenosine was used as permeant, considerable differences in the inhibition profiles were noted among strains. Thus, both inosine transporter-selective nucleosides such as inosine and guanosine and adenosine transporter-selective nucleosides such as 2′-deoxyadenosine and tubercidin showed variable activity as inhibitors of 3H-adenosine transport in different strains. These observations indicated that an adenosine transporter was variably expressed in different strains, and that inhibition profiles for adenosine transport indicated cellular entry via both the inosine and adenosine transporters. The existence of different types of adenosine transporters as an alternative explanation could not be ruled out. The apparent uniform expression of an inosine transporter among different species and strains of Leishmania suggests that inosine derivatives may be useful as anti-leishmanial drugs.

We evaluated the Yersinia pestis capsular (fraction 1 [F1]) antigen as a potential oral immunogen in mice. We found that single doses of as much as 0.4 mg of F1, administered by intragastric (ig) intubation, were unprotective and did not stimulate production of detectable levels of specific antibody. Three weekly ig doses resulted in low serum antibody levels that also did not provide protection against challenge with virulent Y. pestis. Assays of type-specific antibody following intubation and subsequent challenge with a subcutaneous inoculation of F1 revealed that the quantity of antigen intubated and the secondary IgG2a antibody levels were inversely related, suggesting the induction of tolerance to intragastrically administered F1 antigen. Transfer of spleen cells from intubated mice to F1 immune recipients failed to demonstrate suppression of specific antibody, indicating that the immune tolerance observed in intubated mice was not due to a T suppressor cell-mediated effect. The results of this study indicate that Y. pestis F1 antigen is not likely to be an efficacious immunogen for oral vaccination of mice against plague.

Studies were conducted to determine the potential for transmission of a live, attenuated chikungunya (CHIK) virus vaccine by orally exposed or virus-inoculated mosquitoes. The vaccine (CHIK 181/clone 25) replicated in and was transmitted by female Aedes albopictus and Ae. aegypti after intrathoracic inoculation. Mosquitoes also became infected with the vaccine after ingesting virus from either a blood-soaked cotton pledget or a viremic monkey. However, because of the low viremias produced in inoculated humans, it is unlikely that mosquitoes would become infected by feeding on a person inoculated with the live, attenuated CHIK vaccine. Although the vaccine was transmitted by mosquitoes after intrathoracic inoculation, there was no evidence of reversion to a virulent phenotype.

In seven patients with paragonimiasis westermani, parasite-specific IgE and IgG levels in sera and pleural effusion were determined by an enzyme-linked immunosorbent assay (ELISA). The sensitivity to adult excretory-secretory (E-S) antigen was compared with the sensitivity to whole worm extract antigen, and the former was more sensitive in both an IgE-ELISA and IgG-ELISA. Both parasite-specific IgE and IgG could be detected by ELISA at levels much higher than those in control subjects using E-S antigen. When specific IgE and IgG levels in sera and pleural effusion of individual patients were compared, the latter had higher values. The difference between levels of specific IgE in pleural effusion and serum did not correlate with that of specific IgG. These results indicate that specific IgE and IgG antibodies form locally, i.e., in the lung, and that pleural effusions from patients with paragonimiasis are more suitable than serum for immunodiagnosis.

Reports of deteriorating quinine efficacy prompted us to investigate the ability of quinine-tetracycline to clear parasites and fever from patients with multiple drug-resistant Plasmodium falciparum infections. Past and present treatment results were compared at two study sites along the Thai-Cambodian border. In northeastern Thailand, quinine-tetracycline cleared parasites more quickly in 1990 than in 1987 (mean 3.4 and 4.0 days, respectively; P = 0.006). In southeastern Thailand, there were no significant differences between 1990 (n = 26) and 1981–1983 (n = 42) in the time taken to clear either parasites (median 96 and 93 hr, respectively; P = 0.35) or fever (mean 74 and 66 hr, respectively; P = 0.30). In vitro drug sensitivity testing revealed a two-fold decrease in susceptibility to quinine between 1983 and 1990 in isolates from the southeastern Thai-Cambodian border (mean inhibitory concentration 166 ng/ml and 320 ng/ml, respectively; P < 0.001). We conclude that oral quinine-tetracycline continues to reliably clear parasites and fever from falciparum malaria patients infected in eastern Thailand. Periodic re-evaluations are warranted, however, since the decrease in in vitro susceptibility to quinine may be followed by an in vivo decay in the treatment response.

Mefloquine was introduced into Thailand in 1985 for the treatment of Plasmodium falciparum infection. Recently, clinical failure of mefloquine was observed in southeastern Thailand, where an epidemic of falciparum malaria occurred. Beginning in 1984 and continuing until 1989, in vitro monitoring of P. falciparum isolates from Borai, a border district in the southeastern part of the country, showed a progressive decrease in mefloquine sensitivity until 1989; in 1990, the degree and prevalence of resistance accelerated. A similar pattern of resistance was observed for halofantrine, an antimalarial drug not yet commercially available in Thailand. In vitro sensitivity patterns of mefloquine and halofantrine elsewhere in the country remained relatively unchanged. These observations suggest a serious deterioration in available drugs for the treatment of falciparum malaria in southeastern Thailand that is predicted to spread throughout the country and Southeast Asia.

We report that in vitro sensitivity to pentavalent antimony (Sb5) of 35 Leishmania isolates as determined by the semiautomated microdilution technique (SAMT) showed an 89% and 86% correlation with clinical outcome after Pentostam and Glucantime treatment, respectively. These results suggest that in over 85% of the cases, the clinical outcome of treatment (cure or failure) could have been predicted by using the SAMT technique. Furthermore, the results clearly indicate that drug resistance is a problem, and that at least in some instances, failure to respond to treatment is due to the parasite as well as patient factors. Strains from Sb5-treated patients with American cutaneous and mucocutaneous disease who fail at least one complete course of Pentostam are as highly nonresponsive to this drug as laboratory-proven drug-resistant Leishmania strains. It was determined that some Leishmania isolates are innately less susceptible to Sb5 than others, and that moderate resistance to Sb5 exists in nature. A 10- and 17-fold increase was detected in the 50% inhibitory concentration (IC50) of Sb5 for L. mexicana and L. braziliensis isolates after subcurative treatment of the patients, when compared with the mean IC50 of seven and six isolates from the same endemic areas in Guatemala and Peru, respectively. Thus, we have correlated subcurative treatment to a decrease in drug sensitivity in at least these two cases. Collectively, these results indicate that under Sb5 pressure from undermedication, the parasites inherently most drug resistant are favored. The degree of resistance of a strain to antimony in association with host-specific factors will determine whether the clinical response to treatment with this drug is a total cure or a partial response followed by relapse(s), and possibly secondary unresponsiveness resulting in total resistance to antimony. It is evident from our in vitro test data that the SAMT is an extremely powerful and highly accurate technique for the prediction and determination of drug sensitivity of leishmanial isolates, as well as a means to screen for anti-leishmanial agents.