Diagnostic accuracy of interictal source imaging in presurgical epilepsy evaluation: A systematic review from the E-PILEPSY consortium

•Interictal source imaging studies are biased and show heterogeneity for population and test method.•Source imaging sensitivity and specificity was 82% (95% CI: 75–88%) and 53% (37–68%) respectively.•Diagnostic accuracy for (extra)temporal and (non)-lesional patients show no statistical differences. Interictal high resolution (HR-) electric source imaging (ESI) and magnetic source imaging (MSI) are non-invasive tools to aid epileptogenic zone localization in epilepsy surgery candidates. We carried out a systematic review on the diagnostic accuracy and quality of evidence of these modalities. Embase, Pubmed and the Cochrane database were searched on 13 February 2017. Diagnostic accuracy studies taking post-surgical seizure outcome as reference standard were selected. Quality appraisal was based on the QUADAS-2 framework. Eleven studies were included: eight MSI (n = 267), three HR-ESI (n = 127) studies. None was free from bias. This mostly involved: selection of operated patients only, interference of source imaging with surgical decision, and exclusion of indeterminate results. Summary sensitivity and specificity estimates were 82% (95% CI: 75–88%) and 53% (95% CI: 37–68%) for overall source imaging, with no statistical difference between MSI and HR-ESI. Specificity is higher when partially concordant results were included as non-concordant (p