[Malaria in the central highlands of Madagascar: control strategies]

The purpose of this article is to present data on malaria in the central highland plateaux of Madagascar and strategies to improve the national malaria control program. Use of rapid diagnosis strips, early home-based fever management with pre-packaged chloroquine treatment kits and proposed new therapeutic combination based on artemisinine are discussed for management of patients with high suspicion of malaria attack. Preventive measures including alternated targeted and full-house indoor spraying for vector control, use of insecticide-impregnated bednets, implementation of intermittent preventive treatment in risk groups, optimization of the epidemic early detection and warning system using the Lot Quality Assurance Sampling method for epidemiological investigation if the alert threshold is exceeded, and provision of rapid diagnosis strips are presented.

Efficacy of artemether-lumefantrine treatment in patients with acute uncomplicated Falciparum malaria in Mayotte, a French collectivity of the Comoros Archipelago

Mayotte is a French island located in the Comoros archipelago in the Indian Ocean. Due to the high level of resistance to chloroquine and sulfadoxine-pyrimethamine in this area, new therapeutic strategies are required. The aim was to assess and to document the efficacy of artemether-lumefantrine (AL) combination in four oral dosages. The follow-up was carried out during 21 days to monitor the antimalarial drug efficacy in an open trial in April-May, 2002. Results were obtained from 51 patients, aged from three to 46 years (12% less than five years). No case of therapeutic failure was observed. At day 2 after treatment, all the patients were apyretic and none of them had parasitaemia until day 21. This first therapeutic trial of the AL combination in the Indian Ocean sub-region shows that this association is safe, effective and rapid. AL should be an alternative treatment of uncomplicated malaria attacks in Comoros Archipelago, and will be of help to manage imported chloroquine-resistant falciparum malaria strains in Madagascar.

[Epidemiological data on the plague in Madagascar]

The first case of plague was introduced in Madagascar in 1898 in the east coast by way of boat from India. In 1921, plague reach the highlands and a large epidemic over the next twenty years. Until the beginning of the 80's, only of few case were identified, notified mostly in rural setting. However gradually it has re-emerged as a public health problem. Urban plague is located in the city of Antananarivo (resurgence in 1978 after 28 years of apparent silence) and in Mahajanga port (resurgence in 1991 after 63 years of silence). The reactivation of the Plague National Control Program from 1994 will allow better surveillance. The aim of this analysis is to update the epidemiological data on human plague in Madagascar based on reported cases obtained from the Central Lab of the Pasteur Institute of Madagascar from 1980 to 2001 (16,928 suspected cases of which 3,500 are likely positives or confirmed positives). The Plague season runs from October to March on the central highlands and July to November on the north-western coast. Sex-ratio male/female is 1.3/1, and the age-group of 5 to 25 years is more affected. The case fatality rate was 40% in the beginning of the 1980's, and decreased to 20% by the end of the 1990's. The percentage of case with pulmonary plague decrease from 15% to less than 5%. However, geographical extension is demonstrated: 4 districts in 1980, 30 districts in 1999 and 21 districts in 2001. In 2002, the diffusion of a new rapid test (reagent strip) in the primary health centres (CSB) in 42 endemic districts may help to decrease the morbidity and the letality due to plague and improve its control at the national level.

[Current epidemiological situation of bilharziasis in the Antananarivo plain]

In 1994, prior to the Project of Development of the Plain of Antananarivo, an epidemiological survey was conducted in primary schools to assess the level of schistosomiasis. The results of stool examination by the Kato-Katz method demonstrated a prevalence of 4.3% among 6,169 randomized schoolchildren. The most infected four villages are located close the Mamba river. In 1999, after the realization of the project, another study had been done in the same Public Primary School to evaluate the evolution of schistosomiasis endemicity. 5,222 randomized pupils aged 5 to 16 years old took part in the study. The sex ratio was 1.4/1. The global prevalence was 1.8%. High prevalence are respectively notified in Antanandrano Primary school (23%) close to the Mamba river in the Northern part of the capitol and in Ambohitsoa Primary School (16.3%) located near the Mahazoarivo lac. Thus, between 1994 and 1999, a significant overall decrease in prevalence was observed (p < 10(-1)), although a few areas still have relatively high rates. Many factors may explain this improvement, including sanitary education, urbanization and a lack of conditions needed in the development of intermediate host. In addition, a high prevalence of ascaridiasis (79%) and trichocephalosis (67%) was observed. 4.3% of schoolchildren had been infected by Taenia sp. Schistosomiasis is hypoendemic in the plain of Antananarivo. The strategy against this disease must include an IEC programme, focusing in childhood exposure.

[Validation of questionnaire methods to identify Schistosoma haematobium bilharziasis hyperendemic zones in Madagascar]

Schistosomiasis is a major public health problem in Madagascar. The aim of the national control program is to reduce the morbidity in hyperendemic areas. A prospective study has been conducted in Morombe and Ampanihy to elaborate a simple method to identify Shistosoma haematobium hyperendemic communities. The study included 1,373 children from 5 to 15 years old in 17 primary schools. Moderate sensitivity and Negative Predictive Value, with high specificity and Positive Predictive Value of "blood in urine" and "Schistosomiasis" have been found. Those diagnosis values increase with age. The first symptom should be used in older children.

[Management of breast cancers diagnosed at the Pasteur Institute of Madagascar from 1995 to 2001]

Breast cancer is a great problem of public health all over the world. In developed countries, breast cancer represents the most common cancer in females. Its incidence is also increasing in developing country. In Madagascar, no data is available to estimate the real incidence and prevalence rates of breast cancer. However, the data at the Institut Pasteur de Madagascar can confirm the extent of the problem even if it is not at a national scale. The authors report the results of a retrospective study from histological examination at the Laboratory of pathological anatomy of the IPM, during 7 years. Among 2,337 cases of cancer, 16% (373) were breast cancer. Most of them were a female breast cancer (356 cases). The average age is 48 years old. 30% of the tumors were more than 2 cm in size, corresponding at least to the T2 stade from the International Union Against Cancer anatomoclinical classification. The current histological type is the infiltrating ductal carcinoma (80%), about 2/3 belong to the grade 3 of the Scarff-Bloom-Richardson histopronostical classification. Early diagnosis of the cancer is difficult because of the insufficiency of the sanitary infrastructure, particularly for cervical and breast cancers. A national policy for screening must be set up in order to decrease the rate of these invasive carcinomas. In the meantime, informing women and training all the medical staff is a priority. Recording all the data in Madagascar would be desirable.

Lot quality assurance sampling for screening communities hyperendemic for Schistosoma mansoni

Lot quality assurance sampling (LQAS) was evaluated for rapid low cost identification of communities where Schistosoma mansoni infection was hyperendemic in southern Madagascar. In the study area, S. mansoni infection shows very focused and heterogeneous distribution requiring multifariousness of local surveys. One sampling plan was tested in the field with schoolchildren and several others were simulated in the laboratory. Randomization and stool specimen collection were performed by voluntary teachers under direct supervision of the study staff and no significant problem occurred. As expected from Receiver Operating Characteristic (ROC) curves, all sampling plans allowed correct identification of hyperendemic communities and of most of the hypoendemic ones. Frequent misclassifications occurred for communities with intermediate prevalence and the cheapest plans had very low specificity. The study confirmed that LQAS would be a valuable tool for large scale screening in a country with scarce financial and staff resources. Involving teachers, appeared to be quite feasible and should not lower the reliability of surveys. We recommend that the national schistosomiasis control programme systematically uses LQAS for identification of communities, provided that sample sizes are adapted to the specific epidemiological patterns of S. mansoni infection in the main regions.

[Influenza epidemiologic and virologic surveillance in Antananarivo from 1995 to 2002]

The "Institut Pasteur de Madagascar" virology laboratory is the National WHO Centre for Influenza surveillance in Madagascar. On this surveillance collaborate the Ministry of Health with 9 sentinel centres. In the present article, the authors relate the results of influenza surveillance in Antananarivo between 1995 and 2002. Among 6341 patients with nasal and/or pharyngeal swabs, influenza virus were isolated from 427 patients (6.7%): 307 (68.4%) influenza virus A (H3N2), 124 (27.1%) influenza virus B, 8 (4.0%) influenza virus A (H1N1). The virus had been continually spreading all year long. The weak and the strong points of the influenza sentinel surveillance are also discussed in order to ameliorate the collection processes of influenzal and respiratory morbidity data.

[Epidemics of acute respiratory infections in Madagascar in 2002: from alert to confirmation]

An epidemiological investigation (Ministry of Health/Institut Pasteur de Madagascar (IPM)) was conducted in July 2002, in two districts of a same province (Fianarantsoa: Fianarantsoa II and Ikongo) considering the high frequency of deaths linked with acute respiratory infection (ARI). Morbidity and mortality data was collected in the Centre de Santé de Base (CSB) which gave the alert (village of Sahafata, district Fianarantsoa II). Analysis of monthly activity reports (MAR) allowed calculation of incidence rates of ARI/pneumonia in Fianarantsoa province. Virological data was based on the analysis of nasopharyngeal samples collected during the investigations. Clinical symptoms and homogeneity of laboratory results are consistent with an origin of these epidemics being related to the circulation of an influenza virus A subtype H3N2. Attack rates were very high. CFR was significantly higher in individuals of less than 1 year and more than 65 years. This data was confirmed by posterior investigations of teams from MoH/WHO. Surprisingly, this large epidemic was due to a known influenza virus that previously circulated in countries of northern hemisphere (the year before) and even in Antananarivo weeks before. Different hypothesis could be proposed to explain such phenomenon: great restriction of exchanges between different geographical zones, nutritional status....

CONCLUSION

The epidemic episodes of acute respiratory infections in Madagascar in July 2002 were due to an influenza virus A subtype H3N2 without any genotypic or phenotypic features. Various factors, could explain the importance of the epidemic and particular high lethality found in some age groups. This epidemic illustrates the relative incapacity for a developing country, to face and manage a flu epidemic caused by a classical influenza virus.

[The current epidemiological situation of cysticercosis in Madagascar]

Being associated to fecal-oral transmission, cysticercosis is contracted either by auto-infection or by ingestion of food contaminated with eggs from the pork tape worm (Taenia solium). In the stomach, the larvae named cysticercus (Cysticercus cellulosae) hatches from the eggs and invades the host through the mucosa membrane. Human cysticercosis occurs in highly prevalent proportions in many developing countries including Madagascar where hygiene conditions are deplicable. Serology tests applicable to epidemiological surveillance of cysticercosis and associated pathology in the Malagasy population have been developed: an enzyme-linked immunosorbent assay (ELISA) for screening purpose, and an enzyme-linked immunoelectrotransfer blot assay (EITB) for confirmative testing. Two specific bands (13 and 14 kDa) have been identified as significant markers of the cysticercus in an active (vesicle) stage of the infection when cestocidal treatment is strongly indicated. The same bands may on the other hand be absent at early (cyste) as well as late (calcified) stages of the infection. Series of studies, including 4,375 serum samples, have been undertaken from 1994 until 1999 aiming at determinating the cysticercosis sero-prevalence in different provinces of Madagascar. It was confirmed that cysticercosis is highly frequent on the island, and that there exists a marked variation in the prevalence from 7 to 21% between the different provinces: less than 10% in coastal regions (Mahajanga and Toamasina) increasing to 20% in central regions (Ihosy, Ambositra and Mahasolo). It has also been observed that cysticercosis may occur in individuals at any age, and that it is equally distributed in urban as in rural areas. However, it is more frequently detected in women than in men. Madagascar is an endemic country for cysticercosis, which causes major and severe disease with implications in the public health sector. A national control program is, therefore, urgently warranted.

[Cervical cancers diagnosed at the Pasteur Institute of Madagascar from 1992 to 2002]

In Madagascar, the epidemiological data actualized concerning the cancer of the collus of uterus are not available because of the absence of register of cancer. The objective of this study is to achieve a first assessment of the problem, to complete the epidemiological knowledge, to point out the tool of precoce detection of the precancerous lesions, to propose the measures aiming to improve the management of the patients and to contribute to the institution of a register of cancer. This is a retrospective survey on the frequency of the cancer of the cervix observed from 1992 to 2002 about 23,908 withdrawals addressed to the Institut Pasteur de Madagascar for anatomopathological exam and 12,605 cervical smears for cytological exam. In pathological anatomy, 2,621 (63.4%) of 4,136 cases of diagnosed cancer, have been observed in women. 687 cases (26.2%) of them were localized in the collus. The 3/4 of the cancers of the cervix is invasive and the mean age is 48.2 years old at the time of diagnosis. The cytology detects only 74 cases of invasive cancer of which most don't have an histological confirmation. 274 pre-lesions of cervix cancer were diagnosed for this period, the majority lesions are cytological diagnosis. In spite of a non representative recruitment of the general population, and by the number of withdrawals considered, these results may represent indicators of the epidemiological situation and justify the institution of program to detect the precancerous lesions in a national scale.

[Persistence of an endemic circulation of the West Nile virus in Madagascar]

The wide geographic distribution of the West Nile virus and the increase in virulence observed since 1994 in the Mediterranean basin, central Europe and North America, with several outbreaks of lethal encephalitis, demonstrate the importance of regular surveillance of the epidemiological data regarding this virus in the world. The Institut Pasteur de Madagascar has shown between 1975 and 1990 that this arbovirus was most abundant in Madagascar, where it had an endemic circulation. There has been no further study since that time. In order to evaluate the level of circulation, the seroprevalence of anti-West Nile antibodies in children that are 15 or less was measured on two different collections of sera. These collections came from population studies realised respectively in the region of Ambositra in the Highlands in 1996 and in the city of Mahajanga on the north west coast in 1999. The seroprevalence were 2.1% and 10.6% respectively, these results indicate that the circulation of this climatic dependent virus is still significant.

[Identification of communities endemic for urinary bilharziosis by the "Lot Quality Assurance Sampling" method in Madagascar]

Reduction of morbidity is the main component in the National Schistosomiasis Control Program in Madagascar. The lot quality assurance sampling (LQAS) method has previously been shown as a useful tool in assessment of immunization coverage. A study was carried in the western part of Madagascar aiming to evaluate the applicability of the method in measuring the level of Schistosoma haematobium endemic level in different communities. Parasitological examination of urine samples from 1,124 children aged 5 to 19 years from 12 different schools by use of filtration technique constituted the reference in determining the prevalence. Three schools were found hyper-endemic (prevalence more than 60%), 5 schools were intermediate-endemic (prevalence between 30 to 59%), and 4 were hypo-endemic (prevalence less than 30%). Those figures indicate a heterogeneous distribution of S. haematobium in the study area. A sampling plan (16.6) was then tested in the same area while other sampling plans were simulated in the laboratory. School teachers randomized under supervision the children to participate in this study and collected urine samples. All sampling plans (16.6), (14.5), (12.4), (10.3), (8.2), (6.1) et (4.0) allowed correct identification of hyper-endemic and hypo-endemic areas. Misclassifications occurred frequently for intermediate-endemic areas. The study confirms that the LQAS method by use of a (16.6) sampling plan constitute a valuable tool for large scale screening of hyper-endemic areas for therapeutic intervention as part of the control program. The study has also shown that school teachers may offer a potential source of manpower locally in such screening operations.

In vitro sensitivity of Plasmodium falciparum to amodiaquine compared with other major antimalarials in Madagascar

Chloroquine has been used in Madagascar since 1945 and remains the first-line treatment for uncomplicated cases of malaria. Low-grades of resistance type R1 and R2 have been reported. Thus, in vitro tests were performed in order to monitor the drug sensitivity of Plasmodium falciparum from different study sites, with the aim of identifying alternatives to chloroquine. Chloroquine IC50 values ranged from 0.2 nM to 283.4 nM (n = 190, mean IC50 = 52.6 nM; 95% CI = 46.1-59.1 nM). Fifteen isolates (7.9%) were chloroquine-resistant. One mefloquine-resistant isolate was detected (1/139). The test isolates were sensitive to amodiaquine (n = 118), quinine (n = 212), pyrimethamine (n = 86) and cycloguanil (n = 79). The median IC50 for amodiaquine was 12.3 nM (mean IC50 = 15.3 nM, 95% CI = 13.3-17.3 nM). Amodiaquine was 3.4 times as active as chloroquine in vitro and 7 times as active as quinine against P. falciparum. These results indicate that amodiaquine may be a potent alternative to chloroquine in Madagascar. There was positive correlation between tested quinoline-containing drugs activities, which suggests in vitro cross-susceptibility.

[Profile of the patients lost to treatment in the National Anti-Tuberculosis Program in Madagascar]

A national anti-tuberculosis program has been in operation in Madagascar since 1991. Despite the significant efforts made in the management, education and surveillance of patients, the number of patients lost to treatment remains high (18.8%). Noncompliance with treatment is the principal cause of treatment failure and of the development of resistance to tuberculosis drugs. This study investigated the profile of the patients who discontinue treatment. The study population consisted of the patients withdrawing from a clinical trial carried out between August 1994 and September 1996. The aim of the trial was to compare several treatments in routine practice conditions in Madagascar. The treatments tested were streptomycin (S) or ethambutol (E) associated with isoniazid (H), rifampicin (R) or pyrazinamide (Z) for the first two months, followed by 6 months of treatment H and thiacetazone (T). The trial involved four public and three private diagnosis and treatment centers in Antananarivo, Fianarantsoa and Mahajanga. A total of 1, 023 patients were included in the trial and 192 (18.8%) withdrew during the eight-month treatment period. We tracked down 109 of these patients (56.8%) and 19 patients came back to the treatment center on their own initiative for a checkup. The rate of response to a recall letter sent by mail was low. The 106 patients interviewed mostly gave professional, financial or family reasons for discontinuing treatment. Many patients stopped the treatment as soon as they began to feel better. To reduce the number of patients discontinuing treatment, the National Anti-Tuberculosis Program should improve the education of patients and their families.