[Quality of caesarean delivery: A cross-sectional study in 12 hospitals in Benin]. / La césarienne de qualité : étude transversale dans 12 hôpitaux au Bénin.

BACKGROUND: Caesarean section (CS) is a major obstetric intervention, widely recognized as an effective means to reduce maternal and perinatal mortality, when appropriately performed. CS numbers and rates are regularly published but quality is rarely taken into account. This study aims to describe the quality of caesarean delivery in selected hospitals in Benin. METHODS: A cross-sectional study was performed among women who had undergone a CS between 18 December 2013 and 8 February 2014 in one randomly selected hospital in each of the 12 administrative districts of Benin. The quality of CS was defined according to the analytical framework of Dujardin and Delvaux (1998) with its four pillars (access, diagnosis, procedure, postoperative care). Data were collected from hospital files and questionnaires from women and hospital directors. Data analysis was performed using Epi Info 3.5.1. RESULTS: Six hundred and thirty-two women delivered by CS during the period and 579 were eligible for the study. They were aged 26.5±6.3 years, 73.2% living more than 5km from the hospital, 63.0% referred to a health facility of whom 46.0% and 21.8% were transported by motorcycle and by ambulance respectively. The median expenditure by family was FCFA 30 000, ranging from 0 to FCFA 200 000. The admission examination was complete in 12.6% of women and the partograph used in 32.6%. The average CS rate was 37.6%, the average response time, 124minutes. Emergency CS was performed in 80.7%, for absolute maternal indications in 48.0% and under spinal anesthesia in 84.2% (98.3% of which were conducted by a nurse or midwife anesthetist). Maternal mortality was 2000 maternal deaths per 100 000 deliveries, while perinatal mortality was 7.4% (88.4% due to stillbirths). CONCLUSION: CS in Benin hospitals partially fulfilled quality criteria. However access to CS remains difficult and errors of diagnosis or excessive delay are too frequent. Quality CS is not yet a reality in Benin hospitals.

[Factors associated with TT2 + coverage in pregnant women in the Zogbodomey- Bohicon-Zakpota Health Zone, Benin in 2013]. / Facteurs associes a la couverture en VAT2+ chez les femmes enceintes dans la Zone Sanitaire de Zogbodomey-Bohicon-Zakpota au Benin en 2013.

AIMS: Our study aimed to investigate factors associated with TT2+ coverage (at least two doses of tetanus-toxoid vaccine) in pregnant women in the Zogbodomey- Bohicon-Zakpota Health Zone, Benin in 2013. MATÉRIALS AND METHODS: A cross-sectional, descriptive and analytical study was conducted between June-July 2013 on mothers of children aged 0-11 months. The sampling method of immunization coverage cluster of WHO has been adapted. Logistic regression was used to identify factors associated to with TT2+. RESULTS: Our study included 210 mothers of children aged 0-11 months. TT2+ coverage of the women surveyed were 61.7% [95% CI 61.4 to 62.0]. The factors associated with TT2+ coverage were: the amount of antenatal care, the use of a private health center, knowledge of the immunization schedule, the use of radio and television, marital status, occupation, waiting time, residence, fear of undesirable reactions, the permanence of immunization services, education level, distance, family support and explanation of the immunization schedule to the women. CONCLUSION: Measures to improve TT2+ coverage should put more emphasis on these identified factors to hope to eliminate maternal and neonatal tetanus.

BUT: Notre étude avait pour objectif d'étudier les facteurs associés à la faible couverture en Vaccin Anti Tétanique deuxième dose (VAT2+) chez les femmes enceintes dans la Zone Sanitaire de Zogbodomey-Zakpota-Bohicon au Benin. MATÉRIELS ET MÉTHODES: Une étude transversale, descriptive et analytique a été menée en juin-juillet 2013 et a porté sur les mères d'enfants de 0­11 mois. La méthode de couverture vaccinale en grappes de l'OMS a été utilisée. Les données ont été analysées avec Epi Info 7 et Stata 11. La régression logistique a été utilisée pour déterminer les facteurs associés à la VAT2+. RÉSULTATS: La couverture en VAT2 + des 210 mères enquêtées était de 61,7 % IC95% =[61,4 ­62,0]. Les facteurs associés à la couverture en VAT2+ étaient : le nombre de CPN, le recours à un centre de santé privé, la connaissance et l'explication du calendrier vaccinal, l'utilisation des médias, le statut matrimonial, la profession, le temps d'attente, la résidence, la peur des réactions secondaires, la permanence des services de vaccination, le niveau d'instruction, la distance, le soutien de la famille. CONCLUSION: Les mesures visant à améliorer la couverture en VAT2+ doivent davantage mettre l'accent sur ces facteurs pour espérer éliminer le tétanos maternel et néonatal.

[Factors associated with low coverage of tetanus-toxoid vaccine in pregnant women in the Health Zone Zogbodomey- Bohicon-Zakpota, Benin]. / Facteurs associes a la faible couverture en vaccin antitetanique chez les femmes enceintes dans la Zone Sanitaire de Zogbodomey-Bohicon-Zakpota au Benin.

AIMS: Our study aimed to investigate associated factors with TT2 + coverage (at least two doses of tetanus-toxoid vaccine) in pregnant women in the Health Zone Zogbodomey- Bohicon-Zakpota, Benin in 2013. MATÉRIALS AND METHODS: A cross-sectional, descriptive and analytical study was conducted in june-July 2013 on mothers of children aged 0-11 months. The sampling method of immunization coverage cluster of WHO has been adapted. Logistic regression was used to identify factors associated to with TT2 +. RESULTS: Our study included 210 mothers of children aged 0-11 months. TT2 + coverage of the women surveyed were 61.7% [95% CI 61.4 to 62.0]. The final model, factors associated with TT2 + coverage were: the number of antenatal care, the use of a private health center, knowledge of the immunization schedule, the use of radio and television, marital status, occupation, waiting time, residence, fear of undesirable reactions, the permanence of immunization services, education level, distance, family support and explanation of the immunization schedule the woman. Conclusion: Measures to improve TT2 + coverage should put more emphasis on these identified factors to hope to eliminate maternal and neonatal tetanus.

BUT: Notre étude avait pour objectif d'étudier les facteurs associés à la faible couverture en Vaccin Anti Tétanique deuxième dose (VAT2+) chez les femmes enceintes dans la Zone Sanitaire de Zogbodomey-Zakpota-Bohicon au Benin. MATÉRIELS ET MÉTHODES: Une étude transversale, descriptive et analytique a été menée en juin-juillet 2013 et a porté sur les mères d'enfants de 0­11 mois. La méthode de couverture vaccinale en grappes de l'OMS a été utilisée. Les données ont été analysées avec Epi Info 7 et Stata 11. La régression logistique a été utilisée pour déterminer les facteurs associés à la VAT2+. RÉSULTATS: La couverture en VAT2 + des 210 mères enquêtées était de 61,7 % IC95% =[61,4­62,0]. Les facteurs associés à la couverture en VAT2+ étaient: le nombre de CPN, le recours à un centre de santé privé, la connaissance et l'explication du calendrier vaccinal, l'utilisation des médias, le statut matrimonial, la profession, le temps d'attente, la résidence, la peur des réactions secondaires, la permanence des services de vaccination, le niveau d'instruction, la distance, le soutien de la famille. CONCLUSION: Les mesures visant à améliorer la couverture en VAT2+ doivent davantage mettre l'accent sur ces facteurs pour espérer éliminer le tétanos maternel et néonatal.

[Distribution of Buruli ulcer in the Zè district of Benin]. / Distribution spatiale de l'ulcère de Buruli dans la commune de Zê (Bénin).

The goals of this cross-sectional study conducted in the Zè district of Benin were to determine the overall distribution and prevalence of Buruli ulcer (BU) and to identify environmental and behavioral risk factors. A total of 425 current or previous BU patients from the study district were included. Data was obtained by direct observation, semi-structured interviews, and document review. The main findings can be summarized as follows. The overall prevalence of BU in the Zè district in 2006 was 52 cases per 10000 inhabitants. The prevalence of current and previous cases was 28.1 and 23.9 per 10 000 inhabitants respectively. The distribution of BU within the district was highly variable from one subdistrict to another and from one village to another within the same subdistrict. The subdistricts showing the highest and lowest endemicity were Djigbé with 265 cases per 10 000 inhabitants and Koundokpoé with 3 cases per 10 000 inhabitants respectively. Proximity of the hamlets to water bodies was a risk factor for the disease.

[Medical waste incinerator constructed with locally produced materials: experience during the immunization campaign 2002 against measles in Douala, Cameroon]. / Incinérateur de déchets médicaux à base de matériaux locaux: expérience de la campagne 2002 de vaccination anti-rougeole à Douala, Cameroun.

UNLABELLED: Incineration is still the method of choice for disposal of biomedical waste. Combustion at 800+ degrees C eliminates all risk of injury by reducing the waste to ashes and destroying all microorganisms. The purpose of his report is to present a modular double-chambered incinerator in conformity with recommendations for developing countries. The incinerator described here provides the following advantages: suitability for low-cost artisanal construction, excess-air combustion, low risk of accidents and contamination, after-burner for smoke, energy efficiency, easy operation/maintenance, and safety. It is made from locally produced clay bricks reinforced with a steel frame and features two chambers for combustion and post-combustion. It functions on a batch basis with a 2-hour cycle. The immunization campaign against measles that was carried out in 5 districts of Douala in December 2002 produced 800,000 self-disabling syringes with needles. These by-products were immediately discarded in cardboard safety boxes (5-liter capacity, 1 kg) that were taken to the incinerator and burned at the rate of two batches per day. RESULTS: The construction cost of the incinerator was 3,500,000 FCFA (5,300 euros). A total of 5,816 boxes were incinerated, i.e., 29,080 litres, 872,400 syringes, 6,281 kg. Incineration required 126 batches over a period of 11 weeks. The ashes (240 litres, i.e. 0.8% of the total incinerated volume) were virtually free from solid residue. Fuel consumption was negligible. Smoke was never excessive, thick or odorous. Temperatures measured during preliminary trials were always above 800 degrees C. According to the WHO, toxic emissions are negligible if incineration is carried out at temperatures higher than 700 degrees C for less than 2 hours a day. Further study will be necessary to confirm this preliminary experience with regard to temperature control and smoke/ash composition and to test the incinerator in various climatic and operating conditions.

[Impact of parental attitudes on infant vaccinal coverage in Benin]. / Influence de certaines caractéristiques parentales sur la couverture vaccinale des nourrissons au Bénin.

Few studies have addressed the impact of parental attitudes on vaccinal coverage in early childhood. The purpose of this descriptive-analytical transverse study was to assess this problem in a cohort of parents with a 12- to 23-month-old child randomly identified by cluster analysis in five communities in the Oueme department. Data were collected using a questionnaire and tabulation sheet. Statistical analysis was performed by logistic regression using the stepwise digression method. Most of the 438 parents in the study cohort (74.2%) were between 21 and 35 years of age. More than half had not attended school and less than 20% were unemployed. The proportion refusing vaccination for their children was 35% among parents who had to walk more than 30 minutes back and forth to the health care facility and 38% among parents who had poor knowledge about vaccination. The refusal rate was 1.4 times higher for parents with no education than for parents who had attended school (P=0.005). Poor parental knowledge about vaccination was significantly correlated with refusal of vaccination (p<0.001). This study suggests that communication strategies aimed at enhancing parental knowledge and understanding about vaccination should be promoted at health care facilities as well as through other channels, e.g., news media and public events such as social and religious gatherings.

[Underweight prevention and treatment of the young child in Mono (Benin)]. / Prévention et traitement de l'insuffisance pondérale chez le jeune enfant du Mono (Bénin).

BACKGROUND: This study aimed to assess the prevention and treatment effectiveness of underweight in less than two years old children admitted in 2001 to community nutrition programme of Mono area in Benin. METHODS: A longitudinal retrospective survey took place in 30 villages chosen at random. It concerned a sample of 1799 children. Inside of those children, during the 2001 year, when they were admitted to the programme, 1545 were without underweight, have had more than one month following visit and about them the underweight effectiveness prevention had been analyzed; the others 111 were underweighted, have had more than one month following visit and about them the underweight treatment had been assessed. The 143 other children have been followed for only one month. The prevention effectiveness had been explored by the evolution of the cumulative incidence of underweight according to the number of following visit and the underweight incidence density. The treatment effectiveness has been analysed by calculating the recovery rate and incidence density. RESULTS: During 2001, there have been observed 10% for the underweight cumulative incidence and 51.4% for the cumulative incidence of underweight recovery. The underweight and recovery incidence density were respectively 14 cases for 1000 persons-months of following and 118 cases for 1000 persons-months of following. The children with serious underweight at their last following visit have had the number of following visits lower than those who were with moderate underweight. The recovery was arrived after two to twelve months. There was a significant increase of underweight prevalence with the increase of following visit number inside the children with a normal weight at their admission into the programme. CONCLUSION: During the 2001 following, the underweight prevention and treatment activities of community nutrition workers have been little effectiveness. The low level of instruction of community nutrition workers, their low level of knowledge of essential community nutrition activities, their multiple tasks and the not used appropriate foods for the treatment of underweight are the possible causes of the poor performance observed.

[Incidence of meticillin-resistant Staphylococcus aureus in neonatal care unit of departmental hospital centre of Zou Collines in Benin]. / Incidence de Staphylococcus aureus résistant à la méticilline dans le service de néonatologie du centre hospitalier départemental du Zou et des Collines au Bénin.

UNLABELLED: Meticillin resistant Staphylococcus aureus (MRSA) is among the most important causes of nosocomial infections. It possesses a particular ability to spread in hospitals worldwide. OBJECTIVE: To analyze the proportion of MRSA among S. aureus isolated from specimen taken for diagnosis purposes. To make the medical staff aware of the problem of MRSA infections and to take a better care of patients. PATIENTS AND METHODS: During 3 months, a prospective study was carried out in the neonatal unit of centre hospitalier départemental du Zou et Collines in Benin. We identified newborn carriers of SA, particularly MRSA and factors associated with the carriage. Two hundred and ninety patients were admitted in the 3 divisions of the neonatal unit. From 195 specimens examined for diagnosis purposes 48 h after hospitalization, 112 patients were detected by nose swabbing. Concurrently, swabbing of environment was achieved. RESULTS: Among patients'specimens, 141 isolations of S. aureus were observed. The proportion of MRSA was 36% amongst diagnostic specimens. MRSA represented 39% of the environment specimens. None of the isolated MRSA produces Panton Valentine leukocidin. CONCLUSION: Our survey revealed a high level of MRSA among S. aureus isolated from diagnostic specimens. Consecutive to such findings and for decreasing nosocomial infection, an appropriate prevention program was installed, including intensive promotion of hands hygiene, correct sterilization and disinfection of materials and patients.

[Distribution of Mycobacterium ulcerans (Buruli ulcer) in the district of Lalo in Benin]. / Distribution de l'infection a Mycobacterium ulcerans (Ulcere de Buruli dans la commune de Lalo au Benin.

This study aimed to determine the distribution of Buruli ulcer (BU) in Lalo, one of the endemic districts of the Couffo department in Bénin. A total of 752 BU patients were detected in this district with 160 active and 592 inactive cases. The overall prevalence of BU in this district is 86.6 per 10,000 inhabitants, varying from 0 to 249/10,000 between sub-districts. At village level the prevalence varies between 0 and 561 cases per 10,000 inhabitants. Our findings confirm the large variation of distribution of the disease at the village level in endemic area. Children under 15 years are frequently affected. We also found a significant association between age and location of Buruli lesions. Further epidemiological and environmental studies are needed to identify the reasons for the extraordinary variation in BU distribution between villages from the same sub-district, and to confirm if it is associated with temporal variations.

[Traditional treatment for Buruli ulcer in Benin]. / Le traitement traditionnel de l'ulcère de Buruli au Bénin.

The purpose of this transversve qualitative study on traditional treatment for Buruli ulcer in Benin was to track the treatment itinerary of patients, the main phases of traditional treatment, cost and efficacy of such treatment, and the knowledge and skills of traditional practitioners. A total of 20 traditional practitioners, 35 patients treated by traditional therapy, and 35 patients treated by surgery were included. Findings showed that both traditional and surgical treatment was sought at a late stage. Reasons determining the type of treatment chosen included religion, access to adequate care facilities, constraints involved in surgical treatment, duration of hospitalization, and fear of scarring. The four main steps in traditional treatment were diagnosis, removal of necrotic tissue, wound care, and exorcism. The cost of traditional treatment was high not only in currency but also by payment in kind (eg., livestock and land). Although it is performed with patient consent, traditional treatment presents a number of risks. Information campaigns are necessary to inform populations about available treatments and the possible risks associated with each modality. Care centers must do more to lessen the constraints involved in surgical treatment both in terms of duration of hospitalization and cosmetic outcome.

[Effect of a participation in a community-based nutrition program on children's outcomes]. / Effet de la participation à un programme de nutrition communautaire sur l'état nutritionnel des enfants.

BACKGROUND: This study aimed to assess the effectiveness of a protein-energetic malnutrition prevention and treatment community program on the nutritional status of 18 to 36-month-old children in the poor communities of the Mono Region in Benin. METHODS: Two cross sectional anthropometric surveys were conducted. The first survey took place in 1998, before the implementation of the program; the second survey was conducted three years later. At each survey, cluster samples proportional to the population size was done; the sizes of the samples were 617 in 1998 and 648 in 2001. The nutritional status of 18 to 36 month-old-children was assessed by z-score of the indexes height-for-age, weight-for-age and weight-for-height and the cut-off point of -2 standard deviation was used to identify children who had stunting, wasting or underweight; children with serious stunting, wasting or underweight were identified with the cut-off point of -3 standard deviation. RESULTS: From 1998 to 2001, there was a significant reduction in the prevalence of growth retardation (-13.1%) and underweight (-13.0%). The reductions observed for the three types of protein-energetic malnutrition were more pronounced in the 30 to 36 months age group than in the others age groups. CONCLUSION: Substantial improvement was observed during the three years activities. This improvement cannot be attributed to the program unconditionally, because there was no difference between the nutritional status of participant and non participant children.

[Assessment of partogram utilisation in Benin]. / Evaluation de l'utilisation du partogramme au Bénin.

A safe motherhood program was launched in Benin in 1989. One of the methods used to decrease maternal mortality and morbidity was the partogram. We recently conducted a survey in maternity facilities in urban and rural Benin to assess its utilisation rate and the quality of its use. In this cross-sectional survey, all facilities were asked to respond to a questionnaire. In addition, we retrospectively studied files in half the facilities, based on a stratified randomisation. Partograms were used in 98% of all cases; in 13.3% of files (all in rural areas), partogram completion stopped before delivery. Overall completion was less good: of the 984 partograms examined, administrative data were complete on only 20% and medical delivery data on 50%. Action taken before the alert line was crossed was incorrect in 48% of cases (particularly oxytocin use). The alert line was crossed in 13.5% of the cases, but correct action always followed (artificial rupture of membranes, oxytocin administration). The patient transfer rate was 13% and the cesarean rate 5.2%. This alert line was crossed only once. These results thus show very high coverage of partogram use, but inadequate quality and thus demonstrate the need for refresher training for maternity staff about partogram use.

[Preliminary evaluation of morbidity due to S. haematobium and S. mansoni in the area of the future Adjarala Dam in Benin]. / Evaluation préliminaire de la morbidité liée à S. haematobium et à S. mansoni dans la zone du futur barrage d'Adjarala au Bénin.

The aim of this study was to provide a preliminary assessment of morbidity due to schistosomiasis in the area of the future Adjarala Dam in Benin. We included 412 schoolchildren from five villages in the study. The mean age of the children was 9.4 +/- 2.3 years and the sex ratio (F/M) was 0.94. These children underwent parasitological examination. The children provided urine samples, which were examined by eye, filtered and tested for blood and protein. Stool samples were examined using Kato's technique and an ultrasound scan of the urinary tract was performed. We carried out a survey of snails at the possible transmission sites of all the villages, with a view to identifying the intermediate hosts. We detected S. mansoni and S. haematobium in the area. Urinary schistosomiasis was mesoendemic (prevalence of 19.7%) whereas schistosomiasis due to S. mansoni was hypoendemic (prevalence of 3. 9%). Ultrasound scans showed that 28% of the children had bladder lesions and that 2.5% were carriers of hydronephrosis. We compared the diagnostic performances of the various indirect indicators of morbidity. We found that a history of hematuria was the most sensitive indicator (88%) but that the sensitivity of this indicator differed significantly according to the sex of the child. Hematuria >= 1+ was the most effective indicator. Snail surveys showed that Biomphalaria pfeifferi and Bulinus globosus were present in the area.

[Infant malnutrition and associated maternal factors in a secondary city south of Benin, Ouidah]. / Malnutrition infantile et facteurs maternels associés dans une ville secondaire au sud du Bénin, Ouidah.

BACKGROUND: The aim of this study was to find the needs and priorities for interventions to improve children's nutritional state in a secondary city in Bénin. It addressed the issues of the magnitude and distribution of infant malnutrition and related maternal factors. It also aimed to identify an easy to use and low cost, but valid, technique to diagnose malnutrition in children. METHOD: First of all, the prevalence of infant malnutrition was assessed with a representative sample of 492 children aged less than 36 months, in all four communities of the city. Then the sensitivity and specificity of arm circumference were studied and the associations between the children's anthropometric indices were assessed. Secondly in a sub-group of 200 couples of mothers and children, an analysis was conducted to show the links between the indices of mothers' nutritional status, some of their social and economical variables, and the children's anthropometric indices. RESULTS: The prevalences of wasting and stunting and all other forms confounded among children aged 0-3 years were respectively 5.7%, 22% and 25.9%. They represented 44.7% for the arm circumference. Wasting was more prevalent among children aged 6-23 months (9.6%) than those aged less than 6 months (1.1%) and those of 24 to 36 months (5.2%). The boys had a higher prevalence of stunting (25.1%; p = 0.049) than the girls (18.1%). The correlation between children arm circumference and their indices weight/height, weight/age and height/age were all significant (p < 0.001), but they were higher for weight/age (r = 0.48) and weight/height (r = 0.36) than for height/age (r = 0.30). Low, but significant correlation (r ranged from 0.17 to 0.25) were observed between anthropometric indices of mothers and children. Mothers' instruction level had a tendency to be associated positively and significantly with children z-score weight/height. The effect of socio-economic level on children's nutritional status was significant only at p < 0.10. Unlike the condition observed in the big cities of under-developed countries in general, the central area of Ouidah was more affected by infant malnutrition than peripheral area recently urbanized. CONCLUSIONS: Infant malnutrition appears to be a really public health problem in this town and children at weaning age are more affected. The interventions to improve children's nutritional status must concern, not only children with malnutrition, but also their mothers. Those interventions must also improve mother's knowledge and practices about weaning foods and their instruction and socioeconomic levels. The cut-off-point 12.5 cm of arm circumference seems to be more appropriate to diagnose wasting among children aged less than 12 months; 13.5 cm is better for 12-36 months aged children.

[Work schedule management in the health institutions of Abomey-Calavi (Benin)]. / Gestion du temps de travail dans les formations sanitaires de la sous-préfecture d'Abomey-Calavi (Bénin).

Work time management of the health personnel of the health institutions in the sub-prefecture of Abomey-Calavi (Benin) This study was carried out with a sample size of 130, comprising health personnel, patients, members of management committees and twelve health institutions of the subprefecture of Abomey-Calavi. The survey was carried out through observation, one-on-one interviews and consultation of documents. The results reveal that work time is poorly managed in almost all of the programmes surveyed. There exist long waiting lines in almost all the health programmes. On average, the traffic report per service is 82 and 10 out of 12 of the programmes investigated do not have an agenda of activities.

[Antibiotic sensitivity of gram negative bacilli isolated from urinary tract infections at NUHC of Cotonou (Benin) from March to December 1992]. / Sensibilité aux antibiotiques de bacilles a gram négatif isolés d'infections urinaires au CHNU de Cotonou (Bénin) de mars à Décembre 1992.

Eleven antibiotics were tested against 1,194 Gram negative bacilli isolated from urinary tract infections at the National University Hospital Center at Cotonou. Among the betalactams tested, only cefotaxime remained active against most of the bacteria tested: 90% of the strains of Escherichia coli and 75% of the strains of Enterobacter cloacae were sensitive. Ampicilline, on the other hand, had lost its activity even on strains which are usually the most susceptible. Thirteen percent of the E. coli strains were sensitive. This reduction in antibiotic activity against bacterial strains in Cotonou, which concerned to various degrees the tetracyclines, chloramphenicol, cotrimoxazole, is less pronounced for the amino-sides (gentamicine and netilmicine), and the quinolones of which nalidixique acid was active against 83.9% of the strains of E. coli. The low frequency of isolation of wild type strains (sensitive to betalactams) is probably the consequence of strong selection pressure due to a massive, and uncontrolled use of antibiotics in Cotonou.