Clinical manifestations of Rift Valley fever in humans: Systematic review and meta-analysis

Author summaryRift Valley fever (RVF) is a neglected, arboviral zoonosis that causes severe and diverse disease manifestations in humans. Currently no licenced vaccines exist for use in humans and there is limited surveillance and estimation of disease burden which in part is due to the inability to concisely define the disease. We searched Medline, Embase, Global Health, and Web of Science for published reports on the clinical manifestations of RVF in humans. Studies published in English, reporting frequency of symptoms in humans, and laboratory confirmed RVF were included. We excluded animal studies, studies among asymptomatic volunteers and single case reports for which a proportion could not be estimated. Pooled symptom frequency estimates were calculated using random effects meta-analysis. This review provides a detailed aggregation of the relative frequency of symptoms, and a description of the RVF clinical manifestations in humans. Previous systematic reviews provided a narrative account and it was difficult to identify the most relevant features of RVF disease in areas where other endemic infections present with similar symptoms. This review will refine the clinical diagnosis, improve case detection, and increase public awareness about RVF presentation in humans. BackgroundRift Valley fever (RVF) is an emerging, neglected, mosquito-borne viral zoonosis associated with significant morbidity, mortality and expanding geographical scope. The clinical signs and symptoms in humans are non-specific and case definitions vary. We reviewed and analysed the clinical manifestations of RVF in humans. MethodsIn this systematic review and meta-analysis we searched on different dates, the Embase (from 1947 to 13(th) October 2019), Medline (1946 to 14(th) October 2019), Global Health (1910 to 15(th) October 2019), and Web of Science (1970 to 15(th) October 2019) databases. Studies published in English, reporting frequency of symptoms in humans, and laboratory confirmed RVF were included. Animal studies, studies among asymptomatic volunteers, and single case reports for which a proportion could not be estimated, were excluded. Quality assessment was done using a modified Hoy and Brooks et al tool, data was extracted, and pooled frequency estimates calculated using random effects meta-analysis. ResultsOf the 3765 articles retrieved, less than 1% (32 articles) were included in the systematic review and meta-analysis. Nine RVF clinical syndromes were reported including the general febrile, renal, gastrointestinal, hepatic, haemorrhagic, visual, neurological, cardio-pulmonary, and obstetric syndromes. The most common clinical manifestations included fever (81%; 95% Confidence Interval (CI) 69-91; [26 studies, 1286 patients]), renal failure (41%; 23-59; [4, 327]), nausea (38%; 12-67; [6, 325]), jaundice (26%; 16-36; [15, 393]), haemorrhagic disease (26%; 17-36; [16, 277]), partial blindness (24%; 7-45; [11, 225]), encephalitis (21%; 11-33; [4, 327]), cough (4%; 0-17; [4, 11]), and miscarriage (54%) respectively. Death occurred in 21% (95% CI 14-29; [16 studies, 328 patients]) of cases, most of whom were hospitalised. DiscussionThis study delineates the complex symptomatology of human RVF disease into syndromes. This approach is likely to improve case definitions and detection rates, impact outbreak control, increase public awareness about RVF, and subsequently inform 'one-health' policies. This study provides a pooled estimate of the proportion of RVF clinical manifestations alongside a narrative description of clinical syndromes. However, most studies reviewed were case series with small sample sizes and enrolled mostly in-patients and out-patients, and captured symptoms either sparsely or using broad category terms.