388 Am. J. Trop. Med. Hyg., 83(2), 2010, pp. 388���394 doi:10.4269/ajtmh.2010.09-0688 Copyright �� 2010 by The American Society of Tropical Medicine and Hygiene INTRODUCTION Acute febrile illnesses comprise the majority of the disease burden to most populations in sub-Saharan Africa. 1, 2 Although a considerable percentage of these syndromes is attributable to familiar diseases such as malaria and typhoid fever, recent studies suggest that malaria may be overdiagnosed and that a significant proportion of febrile diseases may be caused by pathogens not frequently considered in most settings in sub- Saharan Africa. 3��� 6 Arboviruses in West Africa include members of the genera Flavivirus [yellow fever (YFV), dengue (DENV), and West Nile (WNV)], Alphavirus [chikungunya (CHIKV) and O���nyong nyong (ONNV)], Phlebovirus [Rift Valley Fever (RVFV)], and Bunyavirus [Tahyna (TAHV)]. Evidence and reason suggest that these and other arboviruses circulate frequently in sub- Saharan Africa: competent mosquito vectors for many arbovi- ruses are found throughout the region, the incidence of other mosquito-borne illnesses is generally high, and outbreaks of YFV, CHIKV, and other arboviruses are periodically noted. Whereas large outbreaks often initiate more active epidemio- logic investigation and attempts at laboratory diagnosis, identi- fication of arboviral syndromes on a daily basis in sub-Saharan Africa is rare. Detection is hampered by the non-specific clini- cal presentation, lack of local laboratory diagnostic facilities, cross-reactions on serologic testing, and passive and inconsis- tent surveillance in most African countries. 7 The reported and confirmed cases of arbovirus infection are thought to repre- sent only a small percentage of the actual cases. 8, 9 Without the possibility for laboratory confirmation, many clinicians��� index of suspicion to diagnosis of arboviral syndromes is low. The Republic of Guinea is located on the Atlantic coast of West Africa, with a population of approximately 9.9 million. Evidence points to the circulation of various arboviruses in Guinea: outbreaks of YFV have occurred almost biannually since 2000 in various parts of the country, including northwest- ern Guinea (2000) and the towns of Conakry and N���Z��r��kor�� (2001), Macenta (2003), and Faranah (2005) ( Figure 1 ). 10 The largest YFV outbreak reported in Africa was in Conakry in 2000���2001. 11 A serosurvey of febrile patients between 1978 and 1989 found that 40% were positive for YFV by hemag- glutination assay. 12 Serological surveys have also showed antibodies to CHIKV in Guinea, although the morbidity asso- ciated with infection has not been thoroughly investigated 12��� 14 (E. Jentes, unpublished data). Furthermore, a host of arbo- viruses have been isolated from arthropods, bats, birds, non- human primates, and rarely, humans in Guinea, although, with the exception of YFV, the incidence of human infection and disease due to arboviruses has not been reported nor sys- tematically sought in that country. 12, 13, 15��� 22 We hypothesized that arboviruses accounted for a considerable proportion of undiagnosed febrile illnesses in Guinea and sought to deter- mine the frequency of human arboviral disease in two hospi- tals in two ecologically distinct regions of Guinea. The original intention was to conduct a matched case-control study over 1 year and establish diagnostics for arboviruses at the Center International de Recherche sur les Infections Tropicale in N���Z��r��kor��, Guinea. However, civil unrest, including nation- wide strikes (including hospital staff), street violence, and gas shortages, forced closure of the study after just a few months. Nevertheless, the limited data that were collected and are reported here provide valuable information on the frequency and diversity of arbovirus infections in Guinea. METHODS AND MATERIALS Study area. Despite its abundant natural resources, including 30% of the world���s bauxite reserves, Guinea remains one of the least developed countries in the world. Similar to many other sub-Saharan African nations, only 5.3% of the gross national product is spent on health. Agriculture, which is approximately 25% of the gross domestic product nationally, includes cultivation of rice, bananas, sweet potatoes, timber, and coffee. The average yearly temperature is 25.4��C, with average yearly rainfall of around 1,700 mm. There are single rainy and dry seasons extending from June to November and December to May, respectively. Guinea is divided roughly into four distinct geographic and administrative regions: Basse C��te (lowlands near the oceanic Acute Arboviral Infections in Guinea, West Africa, 2006 Emily S. Jentes ,* Jaimie Robinson , Barbara W. Johnson , Ibrahima Conde , Yosse Sakouvougui, Jennifer Iverson , Shanna Beecher , M. Alpha Bah , Fousseny Diakite , Mamadi Coulibaly , and Daniel G. Bausch Department of Tropical Medicine, Tulane School of Public Health and Tropical Medicine, New Orleans, Louisiana Division of Vector-Borne Infectious Diseases, Centers for Disease Control and Prevention, Fort Collins, Colorado Centre International de Recherche sur le Infections Tropicales, N���Z��r��kor��, Guinea Abstract. Acute febrile illnesses comprise the majority of the human disease burden in sub-Saharan Africa. We hypothesized that arboviruses comprised a considerable proportion of undiagnosed febrile illnesses in Guinea and sought to determine the frequency of arboviral disease in two hospitals there. Using a standard case definition, 47 suspected cases were detected in approximately 4 months. Immunoglobulin M antibody capture enzyme-linked immunosorbent assays and plaque-reduction neutralization assays revealed that 63% (30/47) of patients were infected with arboviruses, includ- ing 11 West Nile, 2 yellow fever, 1 dengue, 8 chikungunya, and 5 Tahyna infections. Except for yellow fever, these are the first reported cases of human disease from these viruses in Guinea and the first reported cases of symptomatic Tahyna infection in Africa. These results strongly suggest that arboviruses circulate and are common causes of disease in Guinea. Improving surveillance and laboratory capacity for arbovirus diagnoses will be integral to understanding the burden posed by these agents in the region. * Address correspondence to Emily S. Jentes, Epidemic Intelligence Service, Office of Workforce and Career Development, and Division of Global Migration and Quarantine, Centers for Disease Control and Prevention, Atlanta, GA 30333. E-mail: ejentes@cdc.gov

389 ACUTE ARBOVIRAL INFECTIONS IN GUINEA coast), Fouta Djalon (mountainous middle region), Haute- Guin��e (northeast savannah), and Guin��e Foresti��re (south- eastern rainforest). This study was conducted in the towns of N���Z��r��kor�� and Faranah ( Figure 1 ). Faranah (population of 88,000) is located in the savannah region (Haute-Guin��e) along the Niger River basin, whereas N���Z��r��kor�� (population of 120,000) is in an area of secondary tropical forest (Guin��e Foresti��re) with interspersed agricultural plots. As in sur- rounding Liberia, C��te d���Ivoire, and Sierra Leone, Guin��e Foresti��re has now been significantly deforested, leaving mainly palm and banana trees in areas that were previously dense with primary tropical forest. Human subject selection. The study was approved by the Tulane University Internal Review Board and the Ethics Committee of the Guinean Ministry of Health. Patients who met a broad case definition for suspected acute febrile viral diseases were recruited from N���Z��r��kor�� Regional Hospital (NZRH) and Faranah Regional Hospital (FRH) ( Table 1 ). Patients were informed about the study, and oral consent was obtained before inclusion in the study. Because these hospitals charge for services on a sliding economic scale and are the reference centers for their respective regions, they are visited by people of all socio-economic levels from the town and surrounding prefecture and thus, have large catchment areas. NZRH has 175 beds, and FRH has 100 beds. Both hospitals provide services in general medicine as well as pediatric, maternity, and surgical wards and ambulatory services. Subjects were enrolled from August 2006 until civil unrest forced early termination of the study in January 2007. The study was initiated in August to correspond with the middle of the rainy season in Guinea, because the majority of cases of YFV in West Africa are noted from the middle of the rainy season to the middle of the dry season. 7 The original intention of obtaining matched controls for each subject to investigate and compare risk factors for arbovirus infection in the community was not possible given the civil unrest. Specimen collection. After explaining the objectives of the study and obtaining informed consent, 5 mL of blood were taken through venipuncture in red-top Vacutainer tubes (Becton, Dickinson, and Company, Franklin Lakes, NJ) from each subject at enrollment, which was usually on the first day of hospitalization (acute), and again on days 7 (late acute) and 28 after hospitalization (convalescent). However, for reasons of both subject adherence and civil unrest, we were often unable to obtain a complete set of three blood samples from each subject (see below). Because Lassa virus, a biosafety level-4 agent, is endemic in the area of study, all samples were inactivated by heating (56��C for 30 minutes) before further manipulation. 23 This protocol has since been changed to 60��C for 60 minutes. The serum was separated from the clot by centrifugation and stored in labeled cryovials in a solar- powered freezer (approximately -11��C) until testing. Laboratory analysis. All serum samples were first tested in Guinea by YFV immunoglobulin M (IgM) antibody capture (MAC) enzyme-linked immunosorbent assay (ELISA) and PanBio DEN IgM ELISA (PanBio Ltd, Brisbane, Australia ) following the Centers of Disease Control and Prevention (CDC) Clinical Laboratory Improvement Amendments (CLIA)- approved protocol and manufacturer���s instructions, respectively. 24, 25 When the deteriorating civil situation made it clear that the study must be terminated, the serum samples were transported on cold packs over land to Sierra Leone and subsequently, to C��te d���Ivoire, where they were shipped on dry ice to Tulane University in New Orleans, LA. Finally, they were shipped overnight to the Arboviral Diseases Branch of the CDC in Fort Collins, CO. At CDC, the samples were tested by MAC ELISA against a standard African antigen panel that includes WNV, YFV, DENV (serotype 2), CHIKV, and TAHV. 25 Samples with positive or equivocal IgM results were confirmed by virus-specific neutralizing antibody titer 90% endpoint plaque-reduction neutralization assay (PRNT) in Vero cells using a 0.5% agarose double overlay visualized with neutral red staining in the second overlay. 26, 27 Neutralizing antibody titers were determined using PRNT 90% cutoff values for confirmation of IgM positive and equivocal samples. 28, 29 Neutralizing antibody titer is expressed as the reciprocal of the endpoint serum dilution that reduced the challenge virus plaque count by 90% based on the back titration. Samples were confirmed as positive with a positive or equivocal IgM and ��� 4-fold neutralizing titer difference between paired Figure 1 . Guinea and surrounding countries in West Africa. The four defined topographical regions of Guinea are demarcated. Epicenters of YFV outbreaks in Guinea since 2000 are signified by stars. The study described here took place in N���Z��r��kor�� (Forest Region) and Faranah (Savannah Region). Table 1 Case definition used to detect patients with suspected acute febrile viral diseases at N���Z��r��kor�� and Faranah Regional Hospitals * Major signs Minor signs Abnormal bleeding (from the mouth, nose, rectum, and/or vagina) General malaise Headache Retrosternal pain Muscle or joint pain Vomiting Cough Sore throat Abdominal pain Diarrhea Proteinuria Leucopenia 4,000/��L Edema of the neck and/or face Conjunctival or sub-conjunctival hemorrhage Jaundice Spontaneous abortion Buzzing in the ears or acute deafness Persistent hypotension Elevated liver transaminases (serum glutamic oxaloacetic transaminase [SGOT]/aspartate aminotransferase [AST] ) * For inclusion in the study, the patient must present with fever 38��C for less than 3 weeks, absence of signs of local inflammation (i.e., the illness is systemic), negative thick smear for malaria, absence of a clinical response after 48 hours of antimalaria treatment and/or broad- spectrum antibiotics, and two major signs or one major sign and two minor signs. Common antimalarial drugs used in the area include chloroquine, quinine, sulfadoxine/pyrimethamine, and artemisinin compounds. Common antibiotic regimens include combinations of vari- ous beta-lactams, including penicillin derivatives and cephalosporins, aminoglycosides, sulfa drugs, macrolides, and chloramphenicol.