We obtained archived CSF specimens from public sector hospitals across Gauteng Province, South Africa, through the National Health Laboratory Service, Tshwane Academic division, from patients who had neurologic signs and symptoms during January–May 2017. We grouped the CSF specimens into 4 categories based on age: age group 1 was children (<1–12 years of age); age group 2, adolescents (13–18 years of age); age group 3, adults (19–59 years of age); and age group 4, senior adults (>60 years of age). SHUV-positive cases were determined by an Orthobunyavirus genus-specific RT-PCR and confirmed using Sanger sequencing and phylogenetic analysis.

Figure

Figure. A) Phylogenetic confirmation that the orthobunyavirus small (S) segment specific reverse transcription PCR (14) positive products identified in this study clustered with SHUV strains. The 155-bp sequence of...

We extracted RNA from the CSF samples using the QIAamp Viral RNA Kit (QIAGEN, https://www.qiagen.com). We performed an Orthobunyavirus genus-specific RT-PCR using the Agpath-ID One Step RT-PCR (Thermo Fisher Scientific, https://www.thermofisher.com) with primers designed to amplify a 155-bp fragment of the nucleocapsid gene of the small (S) segment of orthobunyaviruses (14). We analyzed the sequences using the BioEdit DNA sequence alignment editor v7.0 5.3 software (15) and Blast search analysis (https://blast.ncbi.nlm.nih.gov/Blast.cgi). We performed phylogenetic analysis using maximum-likelihood analysis (MEGA X, http://www.megasoftware.net) as confirmation that the amplicons represent SHUV (Figure, panel A). A larger region of the S segment (≈460 bp) could be sequenced for only 1 of the positive samples (ZRUH131/17, GenBank accession no. MN937197) (Figure, panel B) because of low viral RNA concentration and sample volume in the other CSF samples.

A total of 7 of 130 (5.4%) CSF samples tested positive with an Orthobunyavirus rRT-PCR targeting the S segment and were confirmed by DNA sequencing to represent SHUV (Figure, panel A). A longer region was obtained for a CSF sample taken from a patient who was confirmed to have had neurologic diseases (meningitis, encephalitis, and seizures) with a clinical diagnosis of TB meningitis. Apart from neurological signs, additional clinical diagnoses in other patients included respiratory diseases (tuberculosis, upper respiratory tract infection, and pneumonia), gastrointestinal diseases, vomiting, and hydrops fetalis (Table 1). Only 3 patients’ HIV status was recorded, of which 1 patient’s mother was confirmed to be HIV positive and undergoing treatment. The baby of the positive mother subsequently received nevirapine. The other 2 patients were HIV negative; however, 1 of the children was given nevirapine for reasons not stated. No apparent travel history was recorded for any of these patients.

Most specimens screened were from children (63.1%). Groups with the lowest number of patients were adolescents (1.5%) and the elderly (4.6%) (Table 2). There was only a slight difference in the percentage of males and female patients tested (46.2% male and 51.7% female). A total of 6 (85.7%) of 7 positive cases were in children and 1 of 7 (14.3%) was in an adult. Three of the children with positive test results were <6 months of age. One of these positive children was a newborn admitted to the intensive care unit at 13 days of age who had not left the hospital since birth. Aside from neurologic signs that were present in all patients, the most common recorded symptoms were vomiting, diarrhea, seizures, and fever.

SHUV was reported in horses with severe neurologic signs in South Africa during 2009–2012 (11), which prompted us to also investigate its occurrence in human cases. Screening of CSF specimens from hospitalized patients with neurologic signs around Gauteng Province in South Africa, where some of the equine cases were detected, suggests that up to 5.4% of unidentified neurologic human cases during the arbovirus season may be caused by SHUV. Six of the 7 patients who tested positive for SHUV were children <5 years of age, with 1 being a newborn 13 days of age; only 1 case was identified in a woman. Three of the 7 patients were discharged after being found to be stable; the outcomes of the other 4 are unknown.

These patients were also tested for other viral and bacterial infections, such as influenza, Neisseria meningitidis, pneumonia, herpes simplex virus (HSV), rubella, and cytomegalovirus (Table 1). All 7 patients showed negative results for all requested diagnostic assays except for the 13-day-old infant, who received a diagnosis of hydrops fetalis. He was IgG positive for rubella and cytomegalovirus but IgM negative for both, suggesting maternal antibody transmission. The patient was positive for HSV by PCR and was subsequently placed on treatment for 12 days postnatal until the HSV PCR yielded a negative result. Although the diagnosis of a HSV co-infection cannot rule out HSV as the cause of the hydrops fetalis, the fact that he had not left the hospital since birth suggests a likely vertical transmission of both HSV and SHUV. The patient was stable at discharge after 21; with no death has been reported. None of the patients had any travel history, indicating that they may have been infected in or around their area of residence. Equine cases had previously been identified in these areas, suggesting possible similar vector exposure (11).

A limitation of this study was that all other potential causes of neurologic signs were not exhaustively investigated. Previous detection of SHUV in Cullicoides midges and Culex theileri mosquitoes (McIntosh BM, Epidemiology of arthropod-borne viruses in southern Africa. Unpublished thesis, University of Pretoria, 1980) suggests that SHUV has the potential to expand its geographic range and potentially emerge in new regions. The reservoir host for SHUV is not known but is thought to be ruminants and wildlife, from which transmission to humans would likely be accomplished through susceptible mosquitoes.

Detection of SHUV RNA in the CSF is highly suggestive of SHUV contributing to neurologic signs and likely crossing of the blood–brain barrier. However, further investigations with larger cohorts are needed to determine the disease burden of SHUV in humans across all age groups. These investigations can also include determining the geographic range, clinical presentation, potential vectors, and reservoir hosts in Africa. Improved diagnoses that include IgM serology and early PCR detection of SHUV will aid in defining the true incidence and epidemiology of SHUV.