A quarter of people presenting with seizures in endemic regions have neurocysticercosis

Design

systematic review and Bayesian meta-analysis of 77 studies, 1990-2023

Population

people presenting with epileptic seizures in endemic regions, all ages and both sexes

Primary outcome

pooled proportion with neurocysticercosis diagnosed by imaging, biopsy or autopsy

Effect

25.2% (95% Bayesian credible interval 19.5-32.2), heterogeneity τ = 1.0 (0.8-1.3)

A systematic review across 20 databases screened 16,202 records and included 77 studies reporting neurocysticercosis frequency among people presenting with epileptic seizures, diagnosed by neuroimaging, biopsy or autopsy. In the 35 studies that used clear case definitions and covered both sexes and all ages, Bayesian pooling gave a proportion of 25.2% (95% credible interval 19.5% to 32.2%). The proportion did not vary by sex or age, and it declined between the 1990-2008 and 2008-2023 periods across Africa, the Americas and Southeast Asia.

Two methodological details carry the clinical message. Studies without clear case definitions reported substantially lower proportions, which the authors read as misclassification - that is, sloppy definitions miss the diagnosis rather than inflate it. And heterogeneity was high (τ = 1.0, 95% credible interval 0.8 to 1.3), so the pooled figure describes a range of very different settings rather than a number to apply to any particular clinic.

For paediatric practice in India this is the background rate behind a common presentation. A child with a first afebrile seizure in an endemic area carries a materially different pre-test probability from one in a non-endemic setting, and that should shape whether imaging is arranged and how quickly - not whether the diagnosis is considered at all, which it always should be.

• Treat a first afebrile seizure in an endemic area as an indication for imaging, not watchful waiting.
• Apply the pooled figure as context, not as a local rate - heterogeneity between settings was high.
• Record the case definition you are using; the review shows loose definitions miss cases.
• The apparent decline may be real or may reflect changed definitions and imaging - the authors do not distinguish them.
• Age and sex did not modify the proportion, so do not discount the diagnosis in younger children.

Why it matters

It puts a defensible number on the pre-test probability behind one of the commonest paediatric neurological presentations in endemic regions.

Don't overread it

High heterogeneity means the pooled estimate is not generalisable to a specific clinic, and the reported decline may be an artefact of changing definitions and imaging.

The statistics, in plain English

A credible interval of 19.5% to 32.2% is the Bayesian equivalent of a confidence interval: the pooled proportion is reasonably well estimated. The heterogeneity statistic matters more. At τ = 1.0 the true proportion varies widely between studies, so the pooled 25.2% is an average across settings that genuinely differ - a figure to reason with, not one to quote as your own population's rate.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.