Yield of sphenoidal recording in sleep-deprived outpatients.

We retrospectively reviewed the results of sphenoidal recordings performed after sleep deprivation in 101 consecutive outpatients whose prior scalp EEGs failed to show epileptiform discharges (EDs). These patients were suspected of having seizures of temporal lobe origin. Blinded reviews of the recordings by two electroencephalographers showed a tendency for sphenoidal recording to detect EDs at a higher rate than scalp recording, but the difference did not reach statistical significance (p = 0.06). However, EDs appeared exclusively at sphenoidal electrodes in 11 patients and exclusively at scalp electrodes in 3. The probability of detecting EDs with combined scalp and sphenoidal recordings was greater when spells were characterized by focal symptoms or signs. Except in three patients, initial EDs occurred during the first 30 min of recording. We recommend that post-sleep-deprivation recording be performed initially with scalp and other noninvasive electrodes (such as anterior temporal, ear, or cheek electrodes). If no EDs occur after 30 min, sphenoidal electrode insertion and recording may then be considered, especially in patients with spells characterized by focal signs or symptoms.