The American Journal of Tropical Medicine and Hygiene - Volume 35, Issue 2, 1986

Three patients who were treated for schistosomiasis mansoni with oxamniquine suffered generalized seizures. We suggest that this side effect may be more common than previously reported. Specific ethnic groups may be particularly at risk.

Experiments were performed to compare the migration and survival of 75Se-labeled schistosomes, introduced by percutaneous cercarial exposure or by intravascular administration of 7-day-old lung stage schistosomula, in control and irradiated cercaria-immunized mice. Schistosomula were intravascularly introduced into the lungs, systemic organs and liver by injection via the femoral vein (FV), left ventricle (LV), and superior mesenteric vein (SMV), respectively. The fate of challenge larvae was examined by autoradiography of host tissues and by recovery of adult worms. It was found that both normal and immune elimination were site-dependent. In control mice 45%–60% of cercarial penetrants and lung schistosomula injected into the FV and LV were recoverable as adult worms, while a significantly greater number (70%–85%) were recoverable when lung schistosomula were injected into the SMV. In immunized mice, parasites introduced as either cercariae or FV-injected schistosomula were both highly sensitive to immune elimination. LV-injected schistosomula were also sensitive but to a slightly lesser degree. In contrast, schistosomula placed directly in the liver by SMV injection were totally insensitive to immune elimination. It was concluded that elimination of schistosomula in irradiated cercaria-immunized mice occurs in the lungs and/or in the systemic organs, but not in the liver. Also, it was concluded that immune elimination is not a rapid process, since more than 7 days were required after intravascular challenge for the development of demonstrable differences between control and immunized mice.

Six dogs reared helminth-free were divided into 2 groups. Four dogs were infected per os with 200,000 protoscoleces each of Echinococcus granulosus and 2 were kept as uninfected controls. All the dogs were kept together until 32 days after infection, when 1 infected dog was killed, its intestine removed and the contents examined to confirm that the infection with E. granulosus had been successful. The remaining 3 infected dogs were transferred to high security housing and their feces inspected daily to establish the time infections became patent. The infected and control dogs were bled every 5 days for 75 days from the time of infection and the sera were stored at -70°C. Sera were tested by the enzyme-linked immunosorbent assay (ELISA) for antibodies to E. granulosus scolex excretory/secretory (ES) antigen, protoscolex antigen and oncosphere antigen. Antibodies to scolex ES antigen and protoscolex antigen were detected in the sera of infected dogs within 2 weeks of infection. Antibody titers rose rapidly and remained at a high level until the dogs were killed 75 days after infection. Antibodies in these sera did not cross react with antigens prepared from Taenia ovis, T. hydatigena, T. pisiformis, Ancylostoma caninum, Trichuris vulpis and Toxocara canis.

A patient with chronic liver disease was treated with large doses of mebendazole for a hepatic hydatid cyst. Eighteen days after beginning treatment he developed marrow aplasia which reverted to normal after the drug was stopped. This is the sixth patient described as developing marrow aplasia when treated with large doses of mebendazole. We suggest that the aplasia is related to the dose of the drug, and that the patient's chronic liver disease was an important factor in its genesis. Patients treated with large doses of mebendazole should have their blood counts monitored during treatment.

In a prospective study of human leptospirosis, thrombocytopenia was demonstrated in 54% of 24 cases. The only additional laboratory evidence suggestive of disseminated intravascular coagulation lay in a mild elevation of fibrinogen degradation products, but this occurred with equal frequency in nonthrombocytopenic patients. There is therefore no causal relationship between disseminated intravascular coagulation and the thrombocytopenia of human leptospirosis.

In order to determine if Peromyscus spp. could become infected with the Lyme disease spirochete (Borrelia burgdorferi) by direct inoculation and to determine the duration of spirochetemia, 4 P. leucopus and 5 P. maniculatus were inoculated by the intramuscular, intraperitoneal, and subcutaneous routes with an isolate of B. burgdorferi obtained from the blood of a trapped wild P. leucopus from Camp McCoy, Wisconsin. All of the mice developed antibodies to B. burgdorferi which reached a peak indirect immunofluorescent (IFA) geometric mean antibody titer of 10 log2 21 days post-inoculation. B. burgdorferi was recovered from the blood of 1 P. maniculatus 21 days post-inoculation. One uninfected Peromyscus of each species was housed in the same cage with the infected Peromyscus as a contact control. Both of the contact controls developed IFA B. burgdorferi antibodies by day 14, indicating contact infection. To determine if B. burgdorferi was being transmitted by direct contact, 5 uninfected P. leucopus and 5 uninfected P. maniculatus were caged with 3 B. burgdorferi infected P. leucopus and 3 infected P. maniculatus, respectively. Each of these contact-exposed P. leucopus and P. maniculatus developed antibodies to B. burgdorferi, and B. burgdorferi was isolated from the blood of 1 contact-exposed P. maniculatus 42 days post-initial contact. These findings show that B. burgdorferi can be transmitted by direct contact without an arthropod vector.

Aflatoxin analysis of 40 percutaneous needle liver biopsies in 27 children with protein-energy malnutrition and 13 children with miscellaneous liver disease in The Sudan is reported. Aflatoxins B1, B2 and aflatoxicol were detected in 5 of the 16 biopsies from kwashiorkor but in none of 11 biopsies from marasmus or marasmic kwashiorkor. Aflatoxins G1, G2 and M2 were detected in 5 of 12 children with chronic liver disease. A very high concentration of aflatoxicol was found in a breast-fed infant with neonatal hepatitis of unknown etiology.

In hyperendemic areas such as Thailand, rapid diagnosis of melioidosis depends upon both bacteriological culture and serological methods. However, interpretation of indirect hemagglutination (IHA) for melioidosis which is the only test available, is seriously hampered by increased IHA titers present in one-third to one-half of the population. In order to get the best results from the available tests, IHA and indirect fluorescent antibody for IgM (IFA-IgM) were evaluated in controls and patients in Thailand. IHA titers of ≥1:40 were considered remote or recent exposure to P. pseudomallei. IHA titers of this level were found in 47.1% of 227 blood donors and 29.5% of 210 sera submitted for other tests, while IFA-IgM was positive in only one donor who had an IHA titer of 1:1,280. IHA was positive in eight out of nine patients with melioidosis with IHA titers of <1:20 to 1:2,560. IFA-IgM was positive in six out of seven melioidosis patients whose sera were available for this test including a serum with IHA titer of <1:20. Six patients were predisposed by diabetes mellitus. Among sera serologically tested for melioidosis, 33 had IHA titers of 1:80–1:1,280, 10 of which were positive for IFA-IgM. This study demonstrates high background IHA titers among Thai people which greatly limits its use for serodiagnosis of melioidosis. In sharp contrast, serodiagnosis by IFA-IgM was more successful. Positive IFA-IgM among healthy Thais did exist indicating that serologic tests for melioidosis at best are only supplementary to bacteriological culture and clinical awareness.

In a search for Yersinia infection in Bangladesh, one isolate of Y. enterocolitica serotype 0:3 was obtained from stools of 1,450 children with fever and diarrhea and one of Y. enterocolitica serotype 0:8 was recovered from intestinal contents of 80 fatal diarrheal cases during postmortem examination. These results suggest that Yersinia infection is an infrequent cause of tropical diarrhea.

Detection of rotavirus by electron microscopy was conducted with fecal specimens from 1,722 infants and young children with acute diarrhea, during a 41-month survey from April 1978 through December 1981 in Guayaquil, Ecuador; 376 of these specimens (21.8%) were positive. The detection rate was higher during the dry season (May to November; 25.2%) than during the rainy season (December to April; 14.7%). When rotaviruses isolated from 59 patients hospitalized with diarrhea (from April 1979 to July 1981) were subjected to genome RNA analysis by polyacrylamide gel electrophoresis, a single dominant electropherotype was found with other less common electropherotypes. An atypical rotavirus with a unique property was also found.

In a sample population of 780 Michigan residents tested for neutralizing antibodies to California serogroup viruses, 216 (27.7%) had specific neutralizing antibody to Jamestown Canyon virus. An additional eight (1.0%) had specific neutralizing antibody to trivittatus virus; none had specific neutralizing antibody to La Crosse virus. Significantly more male residents than female residents of the Lower Peninsula had antibody to Jamestown Canyon virus. The frequency of neutralizing antibody titers fits the Poisson distribution, suggesting that Jamestown Canyon virus infections occur endemically in residents of Michigan. Among 128 sera with specific neutralizing antibody to Jamestown Canyon virus, only two (1.6%) were found to have significant hemagglutination-inhibiting antibody titers with La Crosse virus, while 23 of 44 (52%) had significant titers with Jamestown Canyon virus; a single serum had significant antibody by complement fixation tests with both La Crosse and Jamestown Canyon viruses. This study confirms earlier speculation that complement fixation and hemagglutination-inhibition tests with La Crosse virus (the only tests for California serogroup virus infections performed by most state diagnostic laboratories) fail to detect antibody to Jamestown Canyon virus. ASPEX computer-drawn maps demonstrated that the distribution of persons with antibody to Jamestown Canyon virus and residing in Michigan's Lower Peninsula is closely correlated with the estimated distribution of white-tailed deer in that part of the state, further supporting the hypothesis that white-tailed deer are the primary vertebrate host for Jamestown Canyon virus.

Two virus strains were isolated by mouse inoculation from blood of Rousettus aegyptiacus fruit-eating bats collected from Kasokero Cave in Uganda. Shortly after these strains were introduced in the laboratory, four additional strains were recovered from laboratory workers who had developed mild to severe illnesses presumably as a result of laboratory infection. Serological studies established that these six isolates are strains of the same virus. Serological tests showed also that this virus is related to Yogue, an unclassified virus.

Twenty-one dengue (DEN) viruses isolated from the Caribbean (Dominica and Jamaica) during the 1981–1982 epidemic year were distinct serological and genetic variants of DEN-4 virus. These isolates were clearly identified as DEN-4 viruses using type-specific monoclonal antibodies in indirect immunofluorescence assays. However, they either were not neutralized, or were neutralized poorly using hyperimmune mouse ascitic fluids (HMAF) or rhesus monkey serum directed against the H-241 prototype strain of DEN-4 virus isolated in the Philippines in 1956. HMAF prepared against a representative Caribbean isolate, however, neutralized with similar effectiveness the homologous virus, the H-241 prototype strain, and virus strains isolated from the Pacific and Southeast Asian areas from 1973 to 1984. The Caribbean isolate exhibited no more than 30% and 16% oligonucleotide spot homology with the H-241 and Bangkok viruses, respectively, by RNA fingerprint analysis, while demonstrating 82% and 89% homology with the Gilbert and Niue Island isolates, respectively. The isolation of dengue viruses which are serologically and genetically distinct from the prototype virus emphasizes the need for continued dengue virus surveillance. The recognition of unique dengue isolates should allow the selection of reference strains and vaccine candidate strains which will induce antibodies that are equally effective in neutralizing viruses from all geographic areas.

In order to explore the significance of a previous observation that the most important pathologic changes in fatal Lassa fever are hepatic, we have studied postmortem liver biopsies from 19 patients with fatal Lassa fever. We observed a vigorous macrophage response to cellular damage, but we found no evidence of lymphocyte infiltration in infected hepatic tissues. Using semi-quantitative estimates of liver cell damage, we found a wide range in the severity and progression of Lassa virus hepatitis in our fatal cases. We have classified for descriptive purposes three general nosopoeitic phases: active hepatocellular injury (<20% necrosis), continued damage and early recovery, and mitotic activity representing hepatic recovery. We conclude that the liver goes through cellular injury, necrosis and regeneration and any or all may be present at death. In no instance was the degree of hepatic damage sufficient to implicate hepatic failure, and all three phases were represented among our cases. We conclude that the hepatitis of Lassa fever in humans is not the primary cause of death.

An Aedes albopictus line originally selected for efficient transovarial transmission (TOT) of San Angelo (SA) virus displayed a progressive decline in filial infection rate (FIR) whenever artificial selection pressure was relaxed. This observation brought into question whether the efficiency of TOT in this vector-virus model was sufficiently high to permit persistent vertical transmission over numerous freely reproducing generations. By tracing descendants of individual transovarially infected females for several generations, it was found in the present study that this decline in FIR resulted not only from the spontaneous appearance of uninfected females, but also from an accumulation of transovarially infected females which were themselves inefficient transovarial transmitters. Some efficient transovarial transmitters continued to be produced even when FIR had declined to low levels, assuring transmission of virus to subsequent generations, irrespective of the overall infection rate. Outcrossing experiments suggested that a “refractory” genetic factor or factors had been selected out of the TOT-efficient line during its long history of inbreeding.

New TOT-efficient lines were more readily established from parenterally infected Ae. albopictus of a Taiwan strain than had been the original TOT-efficient line from a Hawaiian mosquito strain. One of the newly selected lines differed from the others in producing progenies with FIRs no higher than 36% in the third and fourth generations of transovarial passage. The familial and strain variation revealed by these studies suggests a genetic influence on both the establishment and maintenance of persistent oogonial infections.

The antigenic and biological characteristics of a new Orbivirus, designated Netivot virus, are described. This agent was originally recovered in cultures of the C6/36 clone of Aedes albopictus cells from a pool of Culex pipiens captured in Israel. Netivot virus is not pathogenic for newborn mice, nor did it initially produce detectable cytopathic effect (CPE) in Vero cells. It is closely related antigenically to Umatilla and Llano Seco viruses; these 3 agents appear to constitute a new serogroup within the genus Orbivirus. Netivot virus is also more distantly related to a number of other orbiviruses in the blue-tongue, epizootic hemorrhagic disease of deer, and Eubenangee serogroups. Netivot virus replicated to high titer and produced CPE in a variety of mosquito cell cultures, but it did not grow in 2 sand fly cell lines. Inoculation of Ae. aegypti and Ae. albopictus with Netivot virus resulted in almost 100% mortality in both species within 15 days after infection. The recovery of this and a number of other yet unidentified viral agents from field-collected mosquitoes in cultures of C6/36 cells, but not in the conventional vertebrate assay systems, suggests the existence in nature of many yet unrecognized mosquito-associated viruses. It also demonstrates the value of using new isolation methods in arbovirus studies.

We attempted to tabulate all Bunyamwera serogroup (family Bunyaviridae, genus Bunyavirus) isolates from North America. By summarizing information from the laboratories of the Centers for Disease Control, data generously shared by other laboratories, and the published literature, we were able to accumulate data regarding 1,372 Bunyamwera serogroup viruses. These were: Tensaw (664, including 8 from vertebrates), Cache Valley (396, including 6 from vertebrates), Main Drain (160, including 14 from vertebrates), Lokern (69, including 8 from vertebrates), Northway (13, including 5 from vertebrates), Tlacotalpan (7), Santa Rosa (2), Santa Cruz (1 from a horse), and 60 of undetermined serotype. Virus isolation rates by month of collection were correlated with collection efforts, but associations of viruses and arthropod vectors varied by location, vertebrate host, and arthropod distribution. Tensaw virus was isolated principally from Anopheles crucians mosquitoes (466/656 isolates from arthropods) in the southeastern United States; Cache Valley virus principally from An. quadrimaculatus (94), Coquillettidia perturbans (59), Culiseta inornata (45), Aedes sollicitans (30), Psorophora columbiae (23), An. punctipennis (18), and Ae. vexans and trivittatus (18 each) mosquitoes (total = 305/382 isolates from arthropods from all of the United States and Canada, except the southeastern United States); Main Drain virus from Culicoides variipennis (31), Culicoides (Selfia) sp. (65), and Psorophora (23) and Aedes (21) species mosquitoes in the western United States; Lokern virus from Culicoides species (55/61 isolates from arthropods) in the western United States. Relationships between vector and vertebrate host distributions are discussed briefly in regard to geographic distribution of the Bunyamwera serogroup viruses.

My first surprise on receiving this second (1984) edition for review was that I had never heard of the first (1976). Surprise number two was the ambitiousness of the undertaking: to create an authoritative textbook with special relevance to the practitioner in Africa. The third surprise was the degree to which the book succeeds.

The contributions of the well selected array of authors are based upon their first-hand experience in practicing medicine in Africa. As a multi-authored book, however, it is readable and uniform in style—attributable apparently to the editor's wife (alas, we are not told her name).

The scope of the book is impressive. The first four chapters, which deal with the relationship between disease and climate, people, food and agriculture, and animals, are well written and exemplify the fact that this book, in teaching principles of medicine in Africa, goes far beyond the narrow confines of the purely clinical arena.

This book is a printing of papers contributed by colleagues of Professor Bernard G. Greenberg (Kenan Professor of Biostatistics and former Dean, School of Public Health, University of North Carolina, Chapel Hill) and dedicated to him for his many contributions to both theoretical and applied biostatistics. The twenty-seven essays in this volume cover a wide range of biostatistical topics, grouped into six major areas: clinical trials (three papers), demography and family planning (five), epidemiology and environmental biostatistics (three), general methodology (thirteen), human genetics (two) and neurophysiology (one). The topics are of current interest from both a theoretical and applications point of view. Most papers are methodologic in flavor, with numerous examples in a variety of biomedical areas and, with minor exceptions, include up-to-date references.

Topics in the area of clinical trials include the use of prognostic factors in the planning of trials and in the analysis of trial results.

This is the second of two volumes. The first was subtitled Volume 1. Experience and Components from Conventional Chemical Control. This volume is subtitled Volume 2. Biocontrol and Other Innovative Components and Future Directions. The title Integrated Mosquito Control Methodologies implies coverage of all methods of mosquito control currently or potentially available for use in operational mosquito control programs, but only two are covered. In the preface to Volume 1, the editors state: “We have considered Integrated Mosquito Control Methodologies as comprising chemical, biological and environmental procedures used conjointly or sequentially against the background of an exhaustive ecological understanding of the selected target pest or vector so as to maximize efficacy and be fully acceptable from health and environmental standpoints.”

Also in the foreword of Volume 1 Dr. Norman G. Gratz, Director, Division of Vector Biology and Control, World Health Organization states: “Whenever possible every effort should be made to utilize environmental methods which will prevent vector mosquito breeding.”