Brucellosis seroprevalence and risk factors for seroconversion among febrile attendants of urban health care facilities in Mali.

Transcription

1 Revue Africaine de Santé et de Productions Animales 2006 E.I.S.M.V. de Dakar ARTICLETICLE ORIGINAL Brucellosis seroprevalence and risk factors for seroconversion among febrile attendants of urban health care facilities in Mali. Abstract P. STEINMANN 1, B. BONFOH 1, 2, M. TRAORE 3, A. FANE, E. SCHELLING 1, 4, M. NIANG 5 et J. ZINSSTAG 1 1 Public health and Epidemiology, Swiss Tropical Institute, Socinstrasse. 57, P.o. box, CH-4002 Basel, Switzerland. 2 Centre Suisse de Recherche Scientifique en Côte d Ivoire, 01 BP 1303 Abidjan 01, Côte d Ivoire. 3 Clinique "Dr. Yamadou Sidibé", BP E 11, Bamako, Mali. 4 International Livestock Research Institute (ILRI) PO Box Nairobi 00100, Kenya. 5 Laboratoire Central Vétérinaire, BP 2295, Bamako, Mali. Correspondance et tirés à part, Bassirou.bonfoh@csrs.ci The objectives of the present study were to assess the endemicity of human brucellosis among febrile attendants of urban health care facilities in Mali, and to investigate possible risk factors for seroconversion. Serum samples from 168 febrile individuals were collected and tested for Brucella spp specific antibodies by the Rose Bengal agglutination test. An overall seroprevalence of 7.7% was found among all study participants. Seropositive individuals were only identified in Bamako, with a local prevalence of 16.9%. Livestock owners, farmers, helpers during animal parturition or slaughtering, and consumers of unpasteurized milk were significantly more likely to be seropositive. Self-reported signs or symptoms were not related to serostatus but brucellosis seropositives were significantly more likely to be diagnosed with toxoplasmosis. Human brucellosis is endemic in the urban Malian population, and risk factors for infection include direct contacts with animals and consumption of fresh milk. The high seroprevalence rate suggests that efforts for differential diagnosis and treatment of febrile diseases in humans, transmission prevention, and control in livestock are warranted. The study suggests that there is a need for an intersectoral approach (human and animal health) for brucellosis control in Mali. (RASPA, 4 (3-4) : ). Key-Words: Brucellosis - Febrile - Serology - Risk factor - Milk - Mali Résumé La seroprévalence de la brucellose et les facteurs de risque de séroconversion chez des patients fébriles des centres de santé urbains au Mali. Les objectifs de la présente étude étaient d évaluer l endémicité de la brucellose humaine chez les patients fébriles des centres de santé urbains au Mali et d identifier les facteurs de risques potentiels de séroconversion. Les échantillons de sérum de 168 individus fébriles sont prélevés et testés pour les anticorps spécifiques de Brucella spp. avec le test d agglutination Rose Bengal. Une séroprévalence de 7,7% a été trouvée parmi tous les patients. Les individus séropositifs ont été identifiés seulement à Bamako, avec une prévalence locale de 16,9%. Les propriétaires d animaux, les fermiers, les assistants pendant la mise base et l abattage et les consommateurs de lait non pasteurisé sont significativement séropositifs. Les signes et les symptômes décrits par les patients eux-mêmes n étaient pas liés au statut sérologique mais les séropositifs à la brucellose était significativement liés au diagnostic de la toxoplasmose du médecin. La brucellose humaine est endémique dans la population humaine en zone périurbaine du Mali et les facteurs de risques d infection comprennent le contact direct avec les animaux et la consommation de lait cru. Le taux de séroprévalence élevé milite pour la garantie des efforts de diagnostic différentiel et de traitement des maladies fébrilisantes chez l homme ; de prévention de la transmission et de la lutte chez les animaux. L étude démontre enfin le besoin d une approche intersectorielle (homme et animale) pour la lutte contre la brucellose au Mali. Mots-clés : Brucellose - Fébrile - Sérologie - Facteurs de risque - Lait - Mali Introduction Brucellosis is an animal and human zoonotic disease caused by several species of the gram-negative genus Brucella [22]. Infections can be mild in young animals but cause abortions in females, and orchitis and epididymitis in males. The bacteria are shed in high numbers during birth and abortions but can also be found in milk, urine and faeces. They remain infective for up to 72 days in dust at C and up to 180 days in goats cheese at 4-8 C [22]. Even asymptomatic animals can excrete bacteria [15]. Human infections usually originate from livestock or livestock products. Infections occur through skin lesions, the mucous membranes, inhalation of contaminated dust and consumption of dairy products. Especially consumers of unpasteurized dairy products and veterinarians as well as slaughterhouse workers are at risk of infection. Human clinical disease is characterized by a wide range of symptoms and can develop into a chronic status. The main symptoms are intermittent or constant fever, chills, sweating, aches and pains, anorexia and fatigue. Diagnosis is usually based on serology or bacteria culture [13], [15], [22]. The brucellosis prevalence in animals is not well known, and the global number of human cases is unknown. It is generally assumed that the prevalence of brucellosis is underestimated [13]. Brucella spp. does not only cause disease in animals and humans, but also has massive economic implications [1], [5], [9], [14], [19]. Mali is a landlocked Sahelian country where nomadic or transhumant livestock production systems prevail. The population has a long tradition of dairy product consumption. Livestock is a major asset and has social, cultural and economic value, especially with the Fulani ethnic groups. 117

2 P. STEINMANN et al. In recent years, more intensive livestock production systems have been developed in urban and peri-urban areas to serve the strong demand of urban centres for meat and dairy products. This development is accompanied by uncontrolled attempts for genetic improvements through cross-breeding with foreign breeds (Montbeliard, Holstein, Rouge des steppes) [7]. Animal brucellosis has repeatedly been found in the Malian livestock and human population [14], [21], [22], [24]. It is estimated that the fertility of Brucella-infected herds is reduced by 20%. Milk production of infected cows may be down by 15%. Brucella-specific antibodies were recently detected in Malian cow milk at a level of 30% in bulk milk samples from cattle farms around Bamako, and 5-10% in other cities [5], [6]. These findings probably reflect the production system transformations with high prevalences in cattle herds; the individual prevalence in cows is about 20% in Bamako [5], [24]. The epidemiology of human brucellosis in Mali and throughout West Africa is not well known and recent reports are scarce. The local medical community is rarely aware of the disease and the unspecific symptoms of brucellosis could be confounded with other febrile diseases like malaria [3]. The aims of the present study were to assess the endemicity of human brucellosis among febrile attendants of urban health care facilities in Mali by a conventional serological screening test, and to investigate possible risk factors for seroconversion. Materials and Methods 1. STUDY SITES, FIELD PROCEDURES AND QUESTIONNAIRE The study sites and field procedures including the questionnaire and blood sampling were described previously [20]. In brief, the study participants were recruited among attendants of two private clinics and two health centres in Bamako and Mopti between December 2002 and February The inclusion criteria were an age over 5 years and recent symptoms of a febrile illness of unknown origin. Severely ill patients were not considered. Potential participants were invited by the treating physician to join the study. Upon informed consent, the contact details were recorded by the physician and the participant was assigned an identification number. Subsequently, local collaborators administered a questionnaire in the native languages (Bambara, Peulh) that captured demographic details and information on livestock exposure, dairy product consumption and currently experienced symptoms. After completion of the questionnaire, a nurse collected a venous blood sample in an evacuated tube (Becton Dickinson ) following local standards and procedures. Blood samples were stored at ambient temperature for one hour and later cooled on ice for transfer to the laboratory where they were centrifuged (250g, 10 min.). The collected serum was frozen (-18 C). 2. SEROLOGICAL TEST The undiluted serum samples were screened for Brucella spp.-specific antibodies by the Rose Bengal test (Bio-Rad, Hercules, CA, USA). The Rose Bengal test consists of Rose Bengal-stained Brucella spp. antigen and agglutinates Brucella spp-specific antibodies in human or animal serum. Equal amounts of serum and Rose Bengal test solution (30 µl each) were mixed on a slide and agglutination was evaluated after 3 minutes. Positive samples were re-tested at increasing dilutions (1:2 and 1:4). Agglutination was graded from 0 (no agglutination) to 4 (strong agglutination) and participants were considered seropositive if a sample was rated 2 at a dilution of 1:2. The cut-offs for true seropositivity are disputed [16]. 3. DATA MANAGEMENT AND STATISTICAL ANALYSIS All data were double-entered in Epi Info v. 6.4 (Center for Diseae Control, Atlanta, GA, USA) and analyzed with Stata TM 8.0 (Stata Corp LP; College Station, USA). Answers to open questions like symptoms and diagnoses were categorized. The Fisher's exact test was used to explore associations between seropositivity and different risk factors or symptoms among the whole study population, and the participants from Bamako only. 4. ETHICAL ASPECTS Ethical clearance to conduct the study was obtained from the local health authorities (Direction Régionale de la Santé). Potential participants were asked for informed consent prior to inclusion in the study. Seropositive participants were offered a 3-week course of tetracycline treatment. Results 1. STUDY POPULATION The final cohort included 168 patients, 77 from Bamako and 91 from Mopti. Females represented 56.6% of the study participants and the age structure was as follows: 5-14 years (5.4%), years (75.6%) and over 45 years (19.0%). Significantly more participants from Mopti reported direct contacts to animals (60.4% versus 42.9%, p=0.023), but they were less often present during birth or slaughter of an animal (11.1% versus 26.3%, p=0.011). The prevalence of reported consumption of milk, unpasteurized milk and dairy products was not significantly different between the two cities (p>0.05). 2. SEROPREVALENCE OF BRUCELLOSIS The results of the Rose Bengal agglutination test are displayed in table 1. The overall seroprevalence of brucellosis was 7.7% (13 participants). Table 1: Results of the serum antibody agglutination test using Rose Bengal stained Brucella spp. antigen Rating Undiluted Dilution 1:2 Dilution 1:4 0 - Negative Slightly positive Positive Strong positive Very strong positive Total Positive Table 2 shows the associations between the serostatus and the different risk factors. All seropositive participants originated from Bamako where the local prevalence was 16.9%. The prevalence was higher among males and older age groups, albeit not significantly (p>0.05). A significantly higher seroprevalence was found among farmers (both livestock and crop breeders; all: p=0.004/in Bamako: p=0.014), animal owners (all: p=0.056/in Bamako: p=0.042) and those who reported past attendance of animal parturition or slaughtering (all: p=<0.001/in Bamako: p=0.004). The reported consumption of unpasteurized milk was another significant risk factor for seroconversion (all: p=0.003/in Bamako: p=0.010). 118

3 Brucellosis seroprevalence and risk factors for seroconversion among febrile attendants of urban health care facilities in Mali Table 2: Seroprevalence of brucellosis and risk factors for seroconversion among all study participants (n=168) and among participants from Bamako only (n=77) Variable Total Seropositive p-value Seropositive p-value (n) (Fisher s exact test) in Bamako (Fisher s exact test) n % n % Total <0.001 Sex: Age: Male Female years years >44 years Occupation: Farmer (Livestock, crop) Other Ethnic group: Fulani Other Direct contacts with animals Animal owner Presence during animal parturition or slaughter < Consumption of milk Consumption of unpasteurized milk Consumption of other dairy products SELF-REPORTED SYMPTOMS AND SYMPTOMS OF SEROPOSITIVES Table 3 shows the self-reported signs and symptoms mentioned more than 10 times and the differences in their frequency between Bamako and Mopti. More than 20% of all participants complained about fever, headache, stomach ache and constipation. In Bamako, significantly less participants mentioned fever, headache, stomach ache and constipation at the time of the survey than in Mopti (all p<0.05). No significant relationship was observed between the reported symptoms and brucellosis serostatus (all p>0.05, Table 4). The observations "repeated abortion" (n=1) or "would like to become pregnant" (n=2) concerned seronegative women. Table 3: Self-reported signs and symptoms of all study participants, and differences in symptom frequency between Bamako and Mopti (symptoms reported by <10 participants not reported, 5 symptoms/participant) Symptom All participants Bamako Mopti p-value (n = 165) (n = 75) (n = 90) (F= Fisher's exact test) n % n % n % Fever* Headache Stomach ache Constipation Fatigue Vertigo Joint ache Anorexia Sleeping problems Cough Back ache General fever (n=90), nocturnal fever (n=15), intermittent fever (n=7) 4. DIAGNOSES OF PARTICIPANTS TDiagnoses were based on the clinical skills of the treating physician and basic laboratory testing. Specific diagnostic tools were available for the diagnosis of malaria, typhoid fever, several sexually transmitted diseases and toxoplasmosis. The most common diagnoses are displayed in Table 5. Malaria was diagnosed in 33.8% of the study participants who disclosed their diagnosis. Other frequently given diagnoses included typhoid fever (33.1%) and hypocalcaemia (15.8%). 119

4 P. STEINMANN et al. In Bamako, a positive Rose Bengal test result was significantly associated with the diagnosis of toxoplasmosis (p=0.024). No other significant associations between serostatus and diagnosis were observed (data not shown). Table 4: Self-reported signs and symptoms of study participants in Bamako, stratified by serostatus. Symptom Seronegative participants, Seropositive participants, p-value Bamako (n=62) Bamako (n=13) (F= Fisher's exact test) n % n % Fever Intermittent fever Headache Stomach ache Joint aches Vertigo Fatigue Constipation Sleeping problems Vomit Anorexia Cough Backache Table 5: Frequent (n>3) medical diagnoses among 139 study participants in Bamako and Mopti who disclosed their diagnosis (participants could get multiple diagnoses) 120 Diagnosis Frequency n % Malaria Typhoid fever Hypocalcaemia High arterial blood pressure Toxoplasmosis Headache Urinary infection Gastritis Discussion We used the Rose Bengal agglutination test to check serum samples of febrile patients attending urban health care facilities in Bamako and Mopti for Brucella spp.-specific antibodies and investigated risk factors for seroconversion. The observed seroprevalence of 7.7% confirms the endemicity of human brucellosis in Mali. Interestingly, seropositive participants were only identified in Bamako where BONFOH and colleagues [5] [7] reported an increasing seroprevalence in livestock compared to rural area. The observed risk factors for seroconversion obviously included the city, but also activities involving animal contacts and the consumption of unpasteurized milk. Consumption of other dairy products or boiled milk was not associated with seropositivity. The observed seroprevalence among febrile patients is in line with findings from other areas. Among febrile Bedouins in Israel, 8% were found to be seropositive [17] and in Kenya, a seroprevalence of 12% was found among patients with influenza-like symptoms in a health care facility [14]. The detection of seropositive humans in Bamako but not in Mopti correlates well with the available results of studies on animal brucellosis. A Brucella spp. infection prevalence of 43% and 7.7% was found among cattle in Koulikoro (in the vicinity of Bamako) and Mopti, respectively [24]. The prevalence of human brucellosis usually reflects the prevalence among livestock, modified by the effective contact rate [14]. Many urban residents have contacts with livestock in Mali, be it as animal owner, farmer, on livestock markets or when butchering animals. We found an elevated seroprevalence among study participants who reported respective exposures. The consumption of unpasteurized milk and other dairy products is the main risk factor for human brucellosis in urban areas with limited direct contacts to livestock [13], [15]. We identified the consumption of unpasteurized milk as a risk factor for seroconversion. This result is corroborated by findings from Yemen [2] and Saudi Arabia [8]. In Malian cities, brucellosis probably is a consumer hazard, be it through consumption of unpasteurized milk, home butchering of animals or exposure at animal markets. It remains to be seen which way of transmission prevails in rural settings. Most participants who reported livestock exposure or consumption of unpasteurized milk indicated contacts with multiple livestock species. Therefore, our data do not allow the identification of the livestock species which posed the greatest risk of infection in our study setting. It is also not possible to distinguish the different Brucella species by the Rose Bengal agglutination test. In other settings, dairy products from and contact to sheep and goats posed greater infection risks than cattle [8], [16]. Brucellosis is a febrile illness and differential diagnosis is difficult in the absence of specialized laboratory facilities. Supposedly, the disease is often confounded with malaria in countries like Mali where malaria is endemic. Specific symptoms can only be linked to brucellosis and guide differential diagnosis in regions

5 Brucellosis seroprevalence and risk factors for seroconversion among febrile attendants of urban health care facilities in Mali where malaria is not endemic [16]. Our data further suggest that brucellosis and toxoplasmosis could be confounded. The Rose Bengal agglutination test is commonly applied for veterinary purposes. The sensitive test is suitable for brucellosis screening in humans as well [14]. Serological diagnosis of active human brucellosis is possible if 2 serum samples are screened over a certain time to observe changes in antibody titres [12]. For routine use, the Rose Bengal test needs standardization and evaluation under local conditions to define appropriate cut-off criteria for seropositivity, and investigate cross-reactions with other locally endemic diseases. Cross-reactions with antibodies against Francisella tularensis, Vibrio cholerae, Yersinia enterocolitica and Salmonella urbana have been reported [13]. For diagnostic purposes, positive results should be confirmed by other tests such as ELISA or Complement Fixation. In our study, we observed a close correlation between serostatus and known risk factors for brucellosis, indicating a reasonable protocol and good performance of the Rose Bengal test under the local conditions. The high brucellosis seroprevalence should be further investigated and, if confirmed, it could be worthwhile both economically and from a public health point of view to introduce routine testing and treatment in health facilities, and to step up brucellosis control in animals [18], [19]. In a related study, we found a Q-fever seroprevalence of 40% in the same study population [20]. Therefore, it is possible that both brucellosis and Q-fever contribute to the high number of febrile diseases that are often summarily diagnosed (and treated) as malaria in Mali and other sub-saharan countries. Efforts for diagnosis and treatment in humans should be complemented by the promotion of milk pasteurization [15] and increased efforts for livestock vaccination [18], [19]. The local dairy industry could disproportionately profit from increased milk production through routine vaccination and marketing of pasteurized milk because the demand for locally produced dairy products exceeds local production [10]. The difficult economic conditions, the fragmentation of the milk marketing system [4] and a tradition that favours fresh, unboiled milk [11], [23] and close contacts to livestock are serious challenges for these preventive measures. However, prevention could also have positive effects on the prevalence of other diseases, e.g. tuberculosis due to Mycobacterium bovis and Q-fever due to Coxiella burnetii. The epidemiology of human and animal brucellosis in Mali should be further explored. The seroprevalence of human brucellosis and the incidence of active disease in the urban and rural population should be assessed, and the locally prevalent Brucella species and their reservoirs and main transmission routes should be identified. Acknowledgements The authors acknowledge support from the Swiss National Centre of Competence in Research (NCCR) North South: Research Partnerships for Mitigating Syndromes of Global Change, co-funded by the Swiss National Science Foundation (SNF) and the Swiss Agency for Development and Cooperation (SDC). We would like to thank the study participants, and the field workers and staff of the health care facilities in Bamako and Mopti for their dedication and support. We are also grateful to the personnel of the Institut du Sahel and the Central Veterinary Laboratory (LCV) in Bamako for logistic support. References 1. AKAKPO, A.J., Brucelloses animales en Afrique tropicale. Particularités épidémiologiques, cliniques et bactériologiques. Revue Elev. Méd. Vét. Pays trop., 40 (4): AL-SHAMAHY, H.A.; WHITTY, C.J.; WRIGHT, S.G., Risk factors for human brucellosis in Yemen: a case control study. Epidemiol Infect 125 (2): BABA M.M.; SARKINDARED S.E.; BRISIBE F., Serological evidence of brucellosis among predisposed patients with pyrexia of unknown origin in the north eastern Nigeria. Cent Eur J Public Health 9 (3): BONFOH B.; FANE A.; NETOYO L.; MBAYE A.; SIMBE C.F.; ALFAROUKH O.I., 2003a.- Collection and distribution of locally produced milk in urban Bamako, Mali. Sahelian Studies and Research (8-9): BONFOH B.; FANE A.; STEINMANN P.; HETZEL M.W.; TRAORE A.N.; SIMBE C.F.; ALFAROUKH O.I.; NICOLET J.; AKAKPO A.J.; FARAH Z.; ZINSSTAG J., 2003b- Qualité microbiologique du lait et des produits laitiers vendus au Mali et leurs implications en santé publique. Etudes et recherches sahéliennes (8-9): BONFOH B., FANE A., TRAORE A.P., TOUNKARA K., SIMBE C.F., ALFAROUKH O.I., SCHALCH L., FARAH Z., NICOLET J., ZINSSTAG J., Use of an indirect enzyme immunoassay for detection of antibody to Brucella abortus in fermented cow milk. Milk Sci Int 57 (7). 7. BONFOH B., SALL A., DIABATE M., DIARRA A., NETOYO L., SIMBE C.F., ALFAROUKH O.I., FARAH Z., ZINSSTAG J., 2003c-. Viabilité techno-économique du système extensif de production et de collecte de lait à Bamako. Etudes et recherches sahéliennes (8-9) COOPER C.W Risk factors in transmission of brucellosis from animals to humans in Saudi Arabia. Trans R Soc Trop Med Hyg 86 (2): COULIBALY A Contribution à l'étude de la brucellose bovine au Mali, cercle de Youvarou, Thèse Médecine vétérinaire, Ecole Nationale Vétérinaire de Lyon. 10. DEBRAH S., SISSOKO K., SOUMARÉ S Etude économique de la production laitière dans la zone périurbaine de Bamako au Mali. Revue Elev. Méd. Vét. Pays trop., 48 (1): HETZEL M.W., BONFOH B., FARAH Z., SIMBE C.F., ALFAROUKH O.I., TANNER M., ZINSSTAG J Milk consumption patterns in an area with traditional milk production: data from a case-control study in peri-urban Bamako, Mali. RASPA 3 (3-4): KRAUSS H., WEBER A., APPEL M Zoonosen, von Tier zu Mensch übertragbare Infektionskrankheiten. Köln. Deutscher Ärzte-Verlag. 13. MADKOUR M.M Madkour's brucellosis. Berlin, Heidelberg, New York, Springer. 14. MCDERMOTT J.J., ARIMI S.M Brucellosis in sub-saharan Africa: epidemiology, control and impact. Vet Microbiol 90 (1-4): NICOLET J Kompendium der veterinärmedizinischen Bakteriologie. Berlin, Hamburg, Parey. 16. OMER M.K., ASSEFAW T., SKJERVE E., TEKLEGHIORGHIS T., WOLDEHIWET Z Prevalence of antibodies to Brucella spp. and risk factors related to high-risk occupational groups in Eritrea. Epidemiol Infect 129 (1): PEREZ-AVRAHAM G., YAGUPSKY P., SCHLAEFFER F., BORER A., CAISERMAN S., RIESENBERG K Zoonotic infections as causes of hospitalization among febrile Bedouin patients in southern Israel. Trans R Soc Trop Med Hyg 95 (3): ROTH F., ZINSSTAG J Economic analysis of the brucellosis control in Mongolia. Swiss Tropical Institute, Basel: ROTH F., ZINSSTAG J., ORKHON D., CHIMED-OCHIR G., HUTTON G., COSIVI O., CARRIN G., OTTE J Human health benefits from livestock vaccination for brucellosis: case study. Bull World Health Organ 81 (12): STEINMANN P., BONFOH B., PÉTER O., SCHELLING E., TRAORÉ M., ZINSSTAG J Seroprevalence of Q-fever in febrile individuals in Mali. Trop Med Int Health 10 (6): TASEI J.P., RANQUE P., BALIQUE H., TRAORE A.M., QUILICI M La brucellose humaine au Mali. Acta Trop 39 (3): THIMM B.M Brucellosis, distribution in man, domestic and wild animals. Berlin, Heidelberg, New York, Springer. 23. THOMAS L L'approvisionnement et la distribution des laits et produits laitiers dans les villes d'afrique francophone vu à travers l'exemple du Mali. FAO: TOUNKARA, K., MAIGA, S., TRAORE, A., SECK, B.M., AKAKPO, A.J Epidémiologie de la brucellose bovine au Mali: enquête sérologique et isolement des premières souches de Brucella abortus. Rev Sci Tech 13 (3): WHO The control of neglected zoonotic diseases: a route to poverty alleviation. Report of a joint WHO/DFID-AHP meeting. Geneva sept p. 121