THE ROLE OF RODENT AND HUMAN FACTORS IN THE PREVALENCE OF ANTIBODY TO LASSA VIRUS IN NIGERIA AND SIERRA LEONE

Abstract

Studies were conducted on distribution of Lassa fever (LF), a viral haemorrhagic disease in humans, in Nigeria and Sierra Leone and presence of Lassa virus (LV) in rodents as it relates to human infection, using the Enzyme Linked Immunosorbent Assay (ELISA) developed by the Centers for Disease Control and Prevention (CDC). Results showed the ELISA was reliable and had high sensitivity and minimal false positive results. Compared to IFA, the ELISA had sensitivity of 78% and specificity of 88%. Using ELISA, prevalence of LV antibody was 4.1% (4512) and 17.2% (5551) in Nigeria and Sierra Leone respectively, with range of 0.3%-28.8% in 15 of 33 (47.5%) locations in Nigeria, which included 7 (46.7%) none disease reporting areas and 2.4-31.1% in all 20 locations in Sierra Leone. Proportion of positive households ranged from 29.4 to 38.8% in Nigeria and 9 to 100% in Sierra Leone. All vegetation zones in this study were affected but the greatest risk was for Guinea Savannah dwellers in Nigeria (OR 16.4, 95% CI 6.0-36.2) and grassland Savannah dwellers in Sierra Leone (OR 1.59, 95% CI 1.3-1.9). Increase in antibody prevalence with age, suggestive of endemic process,was observed in 13 villages and 2 States in Sierra Leone and Nigeria respectively, while 1gM antibody differed with gender in Sierra Leone (P <0.02). The most affected occupations were health workers and housewives in Nigeria and housewives and miners in Sierra Leone. Prevalence of LV antibody among health workers in Nigeria was

1.7-23.7%. Nosocomial transmission probably occurred at a higher rate in private and secondary health facilities than the tertiary facility. Lassa fever was confirmed by ELISA in 44.4% of 53 suspected cases in Nigeria and 0.6% of contacts. Incidence study among febrile patients, showed that 1.5% were accounted for by LF although symptoms were not clearly suggestive of LF. Peak in cases occurred between August and January. In Nigeria, no increased risk was observed with human activities that bring them in contact with rodent or rodent excrement while several risk factors were identified in Sierra Leone. Deafness associated with fever could be attributed to LF in 31.3% of 64 subjects in Sierra Leone. None of the respondents from Nigeria knew about LF or related the disease to rodents. In Sierra Leone, the level of awareness increased with level of education and a decrease in antibody prevalence was observed with increase in level of education. Knowledge of LF also differed with type of occupation, gender, village and domicile in other regions of Sierra Leone. In Sierra Leone, health workers were identified as the most likely source of information about LF and not the schools or village chief, while modified behaviour appeared to have occurred in those who were aware. Rodent studies conducted parallel to the human studies showed that trapped rodents were predominantly Mastomys (75.5%) in Nigeria and Rattus (68.5%) in Sierra Leone. None of the 94 Rodents trapped in Nigeria yielded Lassa virus isolate nor tested positive for Lassa virus antibody or antigen, while 2.2% of 370 rodents from Sierra Leone were positive, of which, 75% Mastomys. Rodent infection was correlated with human infection in houses in Sierra Leone. Sera from 2 human cases from Nigeria yielded Lassa virus isolates.