Objective Recent experimental studies have shown that early brain development is exquisitely sensitive to activity, and that early brain activity may be very sensitive to pharmacological manipulations. Such activity in humans consists of endogenous events, which we have called ‘spontaneous activity transients’ (SAT). We studied whether and how some routinely used drugs affect SAT activity.

Methods We collected retrospectively eighteen EEG recordings from fifteen infants at 26–33 weeks of conceptional age without brain lesions detectable with ultrasound. The recordings were divided into five drug-treatment groups: phenobarbital (N = 3), fentanyl (N =  2), ketamine (N = 1), theophylline (N =  9), and controls (N = 3). First, SAT frequency was counted in every recording. Second, SAT events from each recording was collected, and taken to a detailed time-frequency analysis by wavelets.

Results The electric activity between SATs was decreased by ketamine. Fentanyl and phenobarbital reduced the length of the SATs and enhanced oscillations at higher frequencies. Theophylline reduced the activity at these frequencies. No hemispheric difference was seen in SATs in any group.

Conclusions All of the drugs examined had an effect on the SAT activity. Some affected certain frequency bands or the length of SATs, others reduced SAT occurrence. In animal experiments, SAT-like activity has been suggested to play a crucial role in shaping neuronal connectivity during brain development. Our observations support the idea that centrally acting drugs might have effects on brain development which escape detection unless sensitive measures of brain activity are used.