Characteristics of amplitude-integrated electroencephalogram in premature infants.

The aim of this study is to evaluate and characterize amplitude-integrated electroencephalogram tracings in preterm infants at varying gestational ages. Eighty-six amplitude-integrated electroencephalogram traces (1.5-9 hours in duration each) were recorded in 32 preterm infants (24-41 weeks of postconception age) during the study period (July 1, 2001, through August 31, 2002). A stepwise rather than gradual progression was detected, with the first step at 24 to 30 weeks, the second at 31 to 34 weeks, and the third at 35 to 41 weeks. The cyclic pattern amplitude-integrated electroencephalogram typical of healthy full-term infants was not detected in the first step, which was marked by a very low baseline and a wide continuous band width of up to 98.5 microV, with a mean of 74.2 microV (SD = 23.3). The first appearance of cycles was in the second group (31-34 weeks), with a mean intercycle band width of 38.3 microV (SD = 26.3) and a cycle band width of 51 microV (SD = 18). The third step (35-41 weeks) differed from the second, having a mean intercycle band width of 16 microV (SD = 3.5; P < or = .001), with quite a similar cycle band width of 43.8 microV (SD = 19.2). The number of cycles per hour increased from a mean of 0.44 (SD = 0.33) per hour to 0.58 (SD = 0.25) per hour in the second and third steps, respectively (P < or = .001). Amplitude-integrated electroencephalogram is a feasible and reliable brain-monitoring tool in healthy preterm infants. Its features are age dependency and stepwise progression. Its real-time clinical significance and prognostic value in healthy and sick preterm infants are yet to be determined.