The American Journal of Tropical Medicine and Hygiene - Volume 46, Issue 6, June 1992

The Boston meeting of the Society was notable for many reasons. The Presidential Address by Dr. Scott Halstead will be remembered for its stimulous for a reevaluation of the mission of the Society and the importance of our potential contribution to the improvement of the health of people living in the developing tropics. This issue of the Journal opens with another highlight of the meeting — the Soper Lecture. Dr. Donald Hopkins provided yet another extension of the disease prevention and control concept through an intriguing discussion that reintroduced the concept of eradication as a potentially rational goal for some diseases. His lecture provided a detailed examination of three initiatives to reduce or eradicate helminthic diseases: the Onchocerciasis Control Program, the Dracunculiasis Eradication Program and a new effort to control schistosomiasis in school children in Ghana through mass chemotherapy. The text of Dr. Hopkins' comments is an interesting and exciting look at the possible in terms of progress in disease control, and is recommended reading for everyone with an interest in tropical medicine.

Fred Soper dedicated his life to the eradication of disease. Soper's legendary efforts in the battle against yellow fever and his leadership of the successful campaigns to eradicate Anopheles gambiae from Brazil and later from Egypt are emblazoned on the records of public health achievements of the 20th century. In Ventures in World Health, the Memoirs of Fred L. Soper, the medical historian John Duffy noted, “Fred Soper can best be described as a secular medical missionary. Appalled by sickness and disease, he was determined to do all in his power to eliminate its cause. In many respects he epitomizes the best of those American qualities of practicality, forthrightness, moral certainty, and an impelling urge to reform.”

As a tribute to this remarkable public health leader, the Gorgas Memorial Institute established the Soper Lectureship shortly after Fred Soper's death in 1977 and has recently made provision for its endowment.

I am honored to give this lecture dedicated to the memory of Dr. Fred Soper, whose pioneering efforts against the insect vectors of malaria and yellow fever are legendary. Soper and his colleagues also played key roles in restoring the concept of eradication to respectability after the collapse of earlier campaigns to eradicate hookworm and yellow fever. When Soper was himself invited to give the first of five Fred Soper Lectures to mark his retirement as regional director of the Pan American Health Organization in 1959, he entitled his lecture Rehabilitation of the Eradication Concept in Prevention of Communicable Diseases. Ten years later, The World Health Organization (WHO)'s abandonment of the declared goal of malaria eradication marked another nadir in the acceptability of the concept of eradication, from which we are only now reemerging.

One measure of the change is that the Director-General of WHO in 1980 wrote that “Important lessons can be learned from smallpox eradication, but the idea that we should single out other diseases for worldwide eradication is not among them.”

This report describes the results of a community-based Aedes aegypti control program in Merida, Yucatan, Mexico. Baseline surveys concerning knowledge, beliefs, and practices of 577 women and a larval survey of 616 house lots were carried out in October 1989. Following development of a public health communication intervention from this data, the program was implemented in six communities. Evaluation used an untreated control group design with pretest and two post-tests, one at completion and one six months later. Significant changes in knowledge and behavior were seen in the treatment group in both post-tests. Women in the intervention group were able to identify the Ae. aegypti mosquito, the larval production sites of the mosquito, and appropriate control methods. A behavior change proxy was measured by examining changes in the Breteau (number of positive containers/100 houses surveyed) and container indices. The Breteau index remained the same in the intervention group while it increased significantly in the comparison group. Changes were also seen with respect to individual containers. This project demonstrated that a community-based communication program aimed at larval production site elimination or control can be effective in changing behavior and reducing larval production sites.

The Thai-Cambodian border is a difficult area in which to provide adequate malaria chemoprophylaxis because of multiple drug-resistant Plasmodium falciparum. In 1990–1991, Thai soldiers were randomly selected to receive proguanil (200 mg/day) combined with dapsone (4 mg or 12.5 mg/day) (n = 184) or pyrimethamine/dapsone (12.5 mg and 100 mg/week) (n = 177). Doxycycline (100 mg/day) was given to men with glucose-6-phosphate dehydrogenase deficiency (n = 77). Falciparum malaria attack rates were the same whether proguanil/dapsone (10.3%) or pyrimethamine/dapsone (11.3%) was used. However, proguanil/dapsone was more effective than pyrimethamine/dapsone in preventing vivax malaria (1.6% versus 12.4%). Men receiving doxycycline had falciparum malaria (3.9%) and vivax malaria (1.3%) at low rates. Adjusting the dapsone component from 4 mg to 12.5 mg did not improve the prophylactic effectiveness. Hematologic toxicity was not observed with the proguanil/dapsone combination. We conclude that proguanil/dapsone is not a useful alternative for malaria chemoprophylaxis on the Thai-Cambodian border.

An outbreak of classical dengue fever occurred from March to August 1988 in the city of Taxco, Guerrero State, Mexico. Taxco is at an elevation of 1,700 meters above sea level, and this study represents the highest altitude at which an outbreak of dengue has been documented. An investigation was conducted to obtain serologic confirmation of dengue infection, determine the extent of the outbreak, and identify risk factors for dengue illness. Toxorhynchites cell lines were used for viral isolation, and hemagglutination inhibition was used to measure anti-dengue antibody titers. The case definition used in the investigation was any person with fever, headache, myalgias, and arthralgias, or rash or retroocular pain. Dengue virus type 1 was isolated from five acute cases. Of 1,686 persons living in the affected area, 42% (715) met the case definition. Large (200-liter) water containers were significantly associated with infection (relative risk = 1.7, 95% confidence interval 1.5–1.9). The effect of altitude on epidemic transmission is most likely modulated by seasonal temperatures. The epidemiologic and serologic confirmation of a dengue outbreak at 1,700 meters above sea level represents the capability of Aedes aegypti to adapt to new environments, and the potential for epidemic spread in cities at comparable altitudes or higher.

Determinations were made of the source of 16, 330 bloodmeals from 10 species of Culex mosquitoes, including recognized vectors of Japanese encephalitis (JE) virus, in two epidemiologically distinct areas in southern India. In Madurai, where cases occurred sporadically and pigs were reared only in some villages, Cx. tritaeniorhynchus, Cx. pseudovishnui, and Cx. vishnui had fed predominantly on cattle (89.2–91%), but less frequently on humans (2.1–6.2%) and on pigs and ardeid birds (0–0.1%). In Nallur, which was endemic for JE and had a large pig population, 4.4–5.4% of the feedings were on these hosts. Cattle feedings accounted for 84.6–88% of the total feedings, human feedings for 2.4–6.2%, but there were no ardeid-positive feedings. Culex tritaeniorhynchus and Cx. vishnui showed a marked increase in the proportion of human feedings during the hot season, due to increased availability of humans sleeping outdoors to mainly exophagic mosquitoes. Feeding indices were corrected for spatial and temporal concurrence of hosts in each season, but these factors were found to require further elucidation. Discrepancies in the relative abundance of vectors as monitored by two different methods are discussed in the light of these observations.

Following the detection of an Ebola-like virus in cynomolgus macaques recently imported into the United States from The Philippines, studies were initiated to document transmission at export facilities located in the latter country. At one export facility, 52.8% of 161 monkeys that died over a 2.5-month period were shown to be infected with this virus using an enzyme-linked immunosorbent assay to detect antigen in liver homogenates. A case fatality rate of 82.4% was documented for the infected monkeys. The initial anti-viral antibody prevalence among the captive macaques at this facility was 25.9% (indirect fluorescent antibody titer ≥ 1:16). Followup documented infection of 24.4% of initially seronegative animals and 8.7% of initially seropositive monkeys. Being held in a gang cage versus a single cage was found to be a significant risk factor for subsequent virus infection, and the presence of IFA antibody was shown to predict protection. This study documents unequivocally for the first time the presence of an Ebola-related filovirus in Asia.

A followup of Loa loa and Mansonella perstans microfilaremia was carried out in an adult population living in a highly endemic area of the Congo. Infection rates and parasite loads were found to be stable in the general population, both in the short-term (two months) and long-term (3–4 years) followup. The microfilarial status of most of the subjects examined did not change between tests. At the individual level, the microfilarial densities of L. loa and M. perstans also remained remarkably constant over time. This results in a qualitative and quantitative stability of the parasitic material available for the vectors.

In a Mexican village in which Taenia solium infection was known to be endemic, we selected a cluster sample of 368 households (21% of the total) for demographic, environmental, and diagnostic surveys, and medical histories for taeniasis and cysticercosis. Coproparasitologic studies of 1, 531 participants revealed infection by Taenia sp. in four (0.3%) individuals; however, 5.8% of the respondents reported a history of having passed tapeworm proglottids in feces. Of 1, 552 human serum specimens, 10.8% tested positive in the cysticercosis immunoblot assay. Seropositivity increased with age and reached a maximum in subjects ages 46–55 years. Risk factors associated with seropositivity included a history of passing tapeworm proglottids, frequent consumption of pork, and poor personal and household hygiene (P < 0.05). A history of seizures was also significantly associated with seropositivity (P < 0.05); approximately one-third of persons with such histories were seropositive. Of 571 pigs examined by tongue inspection, 23 (4.0%) had cysticerci; infection rates increased with the age of pigs, and were higher in pigs that habitually ran loose or were fed human feces (P < 0.05). Goodness of fit analysis confirmed that seropositive persons (but not infected pigs) were significantly clustered within households, particularly, in households in which a member reported a history of having passed tapeworm proglottids. The results of this study have identified community behavioral and environmental practices that must be modified to prevent continued transmission of cysticercosis and taeniasis.

Mortality trends of missionary staff serving in sub-Saharan Africa were tracked for the period 1945–1985. For 1945–1970, when more complete incidence data were available, the missionary death rate was approximately 40% lower, after adjustment, than would be expected in a comparable US population. This trend persisted through 1985. Between 1945 and 1970, the largest number of fatalities was attributable to malignancy, atherosclerosis, accidents, and infectious disease, and the greatest mortality risks, compared with the US experience, were from homicides, the complications of pregnancy, and infections, notably malaria, hepatitis, and polio. Beginning in the late 1950s, motor vehicle accidents became the leading cause of death. Since the 1960s, accidental causes of death have been approximately 50% higher than in the US, and homicides have been four times higher. During this same period, the infectious disease death rate decreased to approximately that within the US. Currently, the leading causes of mortality are motor vehicle accidents, malignancy, and atherosclerosis, followed by other accidental causes, notably aircraft mishaps and drownings. Viral hepatitis is presently the leading infectious disease cause of death. Other contemporary lethal infections include malaria, rabies, typhoid, Lassa fever, and retroviral infection. It was concluded that missionaries in sub-Saharan Africa had a death rate approximately half that expected in a comparable domestic control population. Preventive strategies, particularly relative to accident and infectious disease prevention, could effectively reduce mortality risk further.

We tested the ability of a recombinant DNA-encoded fragment (C7Ag) of a Plasmodium falciparum merozoite protein (p75) and of two carrier-free peptide models (28-mer and 76-mer) to stimulate boostable antibody responses in Aotus nancymai monkeys. In addition, we evaluated protection against challenge with the Uganda Palo Alto (FUP) strain of this parasite. The data indicate that C7Ag elicited a strong and boostable IgG antibody response in all the monkeys immunized. However, studies with the peptide models demonstrated that various animals produce antibodies to different portions of this structure. When the post-boost sera from monkeys immunized with C7Ag were analyzed for reactivity against two major portions of C7Ag, most of the antibody response was observed against the disulfide-bonded 76-residue region that forms a conformational immunogenic epitope. In the same sera, antibody levels against the charged helical region modeled with a 28-mer were generally low. Immunization with synthetic peptides revealed that the 76-mer stimulated an antibody response almost as strong as C7Ag, with substantial cross-reactivity against the parasite antigen. The 28-mer evoked a response that was not efficient or uniform, and showed little reactivity with the authentic parasite antigen. Aotus nancymai was shown to be susceptible to infection with the Uganda Palo Alto strain of P. falciparum; however, maximum parasitemia varied markedly in both immunized and control monkeys. Statistical analysis failed to recognize differences in maximum parasitemia between the vaccine and control groups. The variation in maximum parasitemia suggests that the FUP strain in this species of Aotus is a poor model for the detection of differences in efficacy based on maximum parasitemia. This initial study with structures based on parts of the 75-kD merozoite surface antigen of P. falciparum indicated that both the recombinant-produced protein C7 and the 76-mer synthetic peptide, when combined with a Syntex adjuvant formulation, were safe and immunogenic in A. nancymai monkeys. However, the data emphasize the problems of using animal models to evaluate the potential effects of immunogens in humans.

Individuals living in a malaria-endemic area in northern Peru were found to have antibodies to the variant repeat sequence of the circumsporozoite (CS) protein of Plasmodium vivax. The presence of IgG antibody to the predominant repeat sequence GDRAA/DGQPA represented by the recombinant protein NS181 V20 (V20), and the variant repeat sequence ANGAGNQPG contained in the synthetic peptide Pvk247, was determined by enzyme-linked immunosorbent assay. IgG antibodies to the repeats were present in 78 (26%) of 298 serum samples; 56% of the positive serum samples had antibodies to V20 and 60% had antibodies to Pvk247. These findings stress the importance of considering the variant epitope in designing a vaccine based on the repeat region of the vivax CS protein. In a malaria-endemic area such as the one in this study, in which exposure to the variant repeat epitope may be as frequent as exposure to the predominant repeat, a vaccine based solely on the predominant repeat epitope may be ineffective against the variant form.

Antibodies in human sera recognizing epitopes I, IIa, III, and IV on the Plasmodium falciparum gametocyte antigen Pfs 48/45 have been investigated by competitive enzyme-linked immunosorbent assay. More than one-third of the residents of three villages in Madang, Papua New Guinea responded to epitopes I, IIa and III, with little variation by village or with time. There was a bimodal distribution of positive sera by age, with the highest proportion of responders in the 5–9- and > 20-year-old age groups. The data suggest a lower prevalence of antibodies against epitopes IIa and III in P. falciparum gametocyte carriers than in non-carriers. Enhancement of binding of monoclonal antibodies to epitopes IIa and III was also observed more frequently with sera from gametocyte carriers. Sera from gametocyte carriers in Papua New Guinea and Thailand, whose infectivity to mosquitoes had been tested, were used to examine the relationship between recognition of particular epitopes and infectivity. There was a significant association between lack of infectivity of P. falciparum gametocyte carriers and recognition of epitope IIa on Pfs 48/45 by antibodies in their sera.

In areas studied in the Rondonia State of Brazil, a high rate of malaria transmission and a low prevalence of anti-(NANP)4 antibodies are reported. The entomologic data are comparable to those observed in some malaria-endemic areas of Africa and Asia. However, the frequency of individuals with antibodies to the immunodominant epitope of the circumsporozoite protein of Plasmodium falciparum recorded in the four localities of Rondonia state was very low when compared with the frequencies recorded in other African and Asian endemic areas. Most of the studies performed in Africa and Asia concerned the native population of hyperendemic areas, whereas we studied a migrant population who were mostly from malaria-free areas of Brazil and living in Rondonia State for 2–4 years. In positive individuals the antibody production was influenced by previous malaria experience, suggesting that infective bites must occur in cumulative numbers before anti-(NANP)4 antibodies are detected. Therefore, it is possible that the individuals described in this report have not been exposed long enough to malaria infection to develop detectable levels of anti-(NANP)4 antibodies.

To improve diagnostic capabilities, an ultrasound unit was installed at a major hospital in Wad Medani, Sudan. During the implementation period (October 1986 to March 1987) of ultrasound service, 863 patients were examined cooperatively by Sudanese and German physicians. The service covered internal medicine (47.5%), obstetrics and gynecology (31.6%), surgery (12.5%), and pediatrics (8.4%). Pathologic findings were seen in 75% of the patients. In obstetrics, the rate of pathologic findings was 40%. Five hundred eighty-three pathologic findings were detected in organ systems that constituted the primary indication for ultrasonography (of a total of 1,009 indications). The clinical benefit of ultrasonography was evaluated in 289 randomly selected patients in a standardized questionnaire completed by the attending physicians. The final clinical diagnosis was established or substantially revised in 21.5% of the patients, based on the ultrasonographic report. The previously reported diagnosis was supported in 69.5% of the patients. Medical management was directly influenced in 26% of the patients. We conclude that the clinical benefit of ultrasonography at a district hospital in the tropics is substantial, providing cost-effective, immediate therapeutic benefits in 25% of the patients examined. The technique thereby contributes to better and more rational patient management in institutions with limited resources.

Using Hansel's stain, eosinophiluria greater than 5% of total urinary white blood cells was found in 59% of a randomly selected population sample in an area endemic for schistosomiasis hematobium. The prevalence and mean level of eosinophiluria were significantly higher in infected subjects than in noninfected subjects (P < 0.05). The sensitivity (80%), specificity (86%), and positive predictive value (82%) of eosinophiluria as a diagnostic index for schistosomiasis hematobium were significantly higher (Youdin index 0.66; P < 0.05) than those of proteinuria, hematuria, and leukocyturia taken singly or in combination. However, unlike the latter three measurements, this method involves microscopy. There is a great need for a chemical method for measuring eosinophiluria.

The effectiveness of praziquantel in treating schistosomiasis is most commonly assessed by quantitating egg production or anti-schistosome antibodies in serum. We have used a monoclonal antibody (MAb)-based antigen-capture enzyme-linked immunosorbent assay (ELISA) for the serologic diagnosis of schistosomiasis, and to monitor the efficacy of praziquantel therapy in 49 individuals with parasitologically proven schistosomiasis. The MAb used, 128C3/3/21, recognizes a repeating carbohydrate epitope expressed at all stages of parasite development, and antibodies recognizing this epitope are found in the serum of infected humans. The overall sensitivity of the ELISA was 78%, with a sensitivity of 100% for patients excreting > 100 eggs/g of feces and 72% for those excreting < 100 eggs/g of feces. The positivity of the ELISA was directly related to the fecal egg counts obtained on days -3, -2, and -1 before treatment with praziquantel, but there was no correlation between antigen levels and the clinical stage of the disease. After praziquantel treatment, we observed a highly significant correlation (P < 0.0001) between the time elapsed since treatment and the decrease in antigenemia. Furthermore, although no eggs were detected in any of the stool specimens at week 12 after treatment, the antigen was detected in 21% of the treated patients (seven of 33 ELISA-positive patients). Antigen levels decreased over the 12-week period in six of these patients, whereas the antigen level increased with time in one individual. The persistence of antigenemia suggests that these individuals are either still clearing antigen or remain infected. These data indicate that quantitation of circulating schistosome antigens may be of value for monitoring the fate of the parasite after praziquantel treatment, and particularly, in identifying patients that remain infected after drug treatment.

To determine the frequency of renal abnormalities occurring with Bancroftian filarial infections and to assess the effects of treatment on such abnormalities, we initiated a prospective, hospital-based study of 20 microfilaremic and five amicrofilaremic patients with Wuchereria bancrofti infections. Thorough clinical evaluations and detailed renal assessments were made prior to treatment and at multiple time points for 60 days following a standard twelve-day course of treatment with diethylcarbamazine (DEC). There were two important findings. First, even prior to DEC treatment, almost half of the microfilaremic patients had hematuria and/or proteinuria. Second, treatment with DEC induced these same abnormalities in almost all of the remaining microfilaremic patients. However, this DEC-induced hematuria and/or proteinuria was transient, and the long-term response to DEC in all of the microfilaremic patients was resolution of the abnormal renal findings during the two-month followup period. In the amicrofilaremic study patients, no hematuria or proteinuria was detected before, during, or after treatment with DEC.

In 1971 we estimated that Schistosoma mansoni eggs in the tissues of mice were destroyed with an approximate half-life of four weeks. Our present results of five experiments suggest that egg destruction is not as rapid, and no significant destruction of eggs was detected for up to 26 weeks after treatment. However, in these experiments, a mean of 60% of the eggs in intestinal tissues were found in the feces at the time of treatment. In previously reported experiments, only 15% of gut eggs were passed in the feces. We now believe that underestimation of the number of eggs passed in the feces led to an overestimation of the number of eggs destroyed in the tissues. We analyzed liver eggs separately because eggs lost from this site are unaffected by eggs passed in the feces. No significant decrease in liver eggs occurred in the present experiments, but reanalysis of previously published data showed significant egg destruction in the liver in several experiments, although at a much slower rate than previously estimated. However, inspection of the data in the previously published and present experiments does not show a convincing difference in the number of eggs in the liver after treatment. The persistence of egg shells is probably not important in the pathogenesis of disease, but is of concern in calculating worm fecundity. Hepatic collagen levels increased markedly two weeks after treatment and subsequently decreased significantly in some, but not all, experiments.

Two black African women and one black American man had carcinomas of cervix, perineum, and sigmoid colon, respectively. In each of these patients, trophozoites of Entamoeba histolytica had invaded the surface of the tumor, and in some areas had invaded more deeply into the stroma between the tumor cells. Although it is well known that cutaneous amebiasis of anus, penis, vulva, and cervix can mimic squamous cell carcinoma, it may be, perhaps, less well known that carcinomas at these sites may be colonized by trophozoites of E. histolytica. In patients with amebiasis but without an associated carcinoma, a correct diagnosis of amebiasis spares the patient unnecessary and sometimes mutilating surgery. But a diagnosis of amebiasis, when there is an unrecognized underlying carcinoma, delays effective treatment of the carcinoma. A smear that establishes a diagnosis of cutaneous amebiasis, therefore, should be followed by biopsy to exclude or confirm an underlying carcinoma.