Human immunodeficiency virus (HIV)-positive children may present with a wide range of neurological disorders. Among these, epilepsy is of key concern because of its lifelong impact and potential for damage to the central nervous system (CNS). Few studies in developing regions have investigated the prevalence and aetiology of epilepsy in HIV-infected children as a key population.
We describe the prevalence of epilepsy, associated neurological disabilities, immunological status, clinical stage and history of CNS infection at epilepsy diagnosis in a cohort of HIV-infected children receiving antiretroviral therapy (ART) in the Eastern Cape of South Africa.
We conducted a retrospective study (2004–2014) at two major referral sites for HIV-infected children diagnosed with epilepsy aged 0–16 years. Eligible subjects were extracted from the electronic medicine bridging access to care in excellence (EMBRACE) Paediatric Cohort using the Paediatric ART Data Management Tool (PADMT). Fixed data fields were interrogated for exposures to antiepileptic drugs. Unstructured ‘comments’ fields were searched for the terms: epilepsy, seizures, fits and szs, as well as abbreviated versions of common antiepileptic drug names. Eligible subject folders were then retrieved to validate the digital data.
From 2139 children enrolled in the two sites, 53 children were diagnosed with epilepsy (2.48%). In these, the median CD4 count was 591 cells/mm 3, and the mean viral load was 4.9 log copies/mL, with undetectable viral loads in only seven children (14.0%). World Health Organization (WHO) clinical HIV stage was available for 46 patients of the sample, with 3, 6, 26 and 11 children graded at stages 1, 2, 3 and 4, respectively. Forty percent children had a history of CNS infection prior to the epilepsy diagnosis, and 55% children were reported to have school problems.
In this descriptive study, the prevalence of epilepsy among children with HIV was 2.48%, mostly diagnosed in advanced HIV-disease stages. Our findings support the usefulness of early detection and initiation of ART in HIV-infected children in order to reduce the risk of epilepsy. In addition, our study demonstrates that novel techniques are effective in accessing cohort-level data that allow interrogation of both structured and unstructured clinical data.