Closing the temporal blind spot: Six additional electrodes increase patient-level EEG abnormality detection by ~12.

Abstract

Objective: To quantify the additional detection yield of the 25-channel electrode array recommended by the International Federation of Clinical Neurophysiology (IFCN), which includes six inferior-temporal electrodes beyond the conventional 10-20 International System (IS) in identifying interictal epileptiform discharges (IEDs) and focal slow activity (SA).

Methods: We analyzed 258 consecutive adult EEGs (routine or sleep-deprived; mean age, 57 ± 19 years; 55% female; 61% on anti-seizure medication). Indications were suspected epilepsy (33%) and seizure follow-up (28%). Two blinded neurophysiologists harmonized IED/SA terminology and applied the validated three-item subset of the IFCN's operational IED criteria. Each EEG was reviewed twice: first with the standard 19-electrode 10-20 system and, after a ≥ 4-week washout, with the 25-channel IFCN array. Reading lists were randomized between reviewers so that the sequence of cases differed, thereby reducing recall bias. Abnormalities were classified by type and lobe. Agreement was measured using Cohen's κ; yield gain was the detection-rate difference.

Results: Agreement ranged from substantial to almost perfect (κ = .65-.88; 95% CI: .48-1.00), with disagreement rates of 1.2-7.8%. Temporal IEDs were detected more frequently with IFCN (72/71 patients) than with 10-20 (63 each), corresponding to gains of 12.5% and 11.3%, respectively. Temporal SA was also higher (98 vs. 86; +12.2% for both). Gains in extratemporal SA were smaller (10.7% and 9.5%) and differences in extratemporal IEDs were minimal (≤3.6%). Pooled across all abnormalities, IFCN detected 183 and 182 patients versus 172 and 170 with 10-20, respectively, an overall gain of 11.5%. The largest relative increase was seen in temporal IEDs, with an additional 11-12% of patients identified using the IFCN array compared with 10-20.

Significance: The addition of six inferior-temporal electrodes increased the relative detection yield of EEG abnormalities by 11.5%, with the largest gains for temporal IEDs, and required only minimal extra setup time. Although the true diagnostic accuracy of these additional detections cannot be determined without an external gold standard, the findings support the routine use of the IFCN 25-electrode array to address a major blind spot of the 10-20 system in adult EEG practice.