Summary Background:  Infants are noted to frequently sleep during spinal anesthesia, with a concomitant fall in Bispectral Index. However, there are suggestions that EEG derived anesthesia depth monitors have inferior performance in infants. The aim of this study was to quantify the degree of sedation during spinal anesthesia in infants using another EEG derived measure of anesthesia effect – the Cerebral State Index (CSI). Methods:  Twelve infants, <52 weeks postconceptual menstrual age, scheduled for bilateral inguinal hernia repair under spinal anesthesia were enrolled. Patients received a standard anesthetic protocol with a subarachnoid dose of 1 mg·kg−1 of levobupivacaine 0.5%. No premedication, sedatives, opioids or anticholinergics were administrated during the perioperative period and patients were left undisturbed during the surgical time, without tactile stimulation or loud auditory stimuli. CSI score (0–100) and bust suppression (BS) (0–100%) were continuously recorded during the surgical time and then statistically re‐evaluated. Results:  In all patients the CSI fell during the procedure and there were significant levels of BS recorded by the CSI monitor. The BS occurred between 12 and 34 min after spinal anesthesia with the peak being at 30 min and mean onset time being 15 (2.6) min after spinal block. A statistical significant difference was found between the lowest mean CSI as well as the highest BS if compared with their baseline values. A negative correlation was found between CSI and BS. Conclusions:  The degree of burst suppression detected by the CSI in our study supports the hypothesis that infants may have discontinuous patterns of EEG during spinal anesthesia similar to those seen during emergence from general anesthesia. Moreover, the limitations in the application of the adult algorithms to infant EEG may lead to an overestimation of the degree of sedation.