To examine the role of immaturity in the free radical-mediated rate of lipid peroxidation in premature infants, we studied 27 infants [gestational age, 27.1 (SD 2.4) wk; birth weight, 970 (SD 330) g]. Ethane and pentane were quantitated in expired air during the first 18 d of life. During the first 2 postnatal d ethane [24.1 (SEM 7.8) pmol x kg-1 x min-1] and pentane [24.2 (SEM 4.1) pmol x kg-1 x min-1] were stable but increased during d 5 to maxima of 79.1 (15.8) pmol x kg-1 x min-1 and 62.1 (8.1) pmol x kg-1 x min-1, respectively. Maximum ethane and pentane correlated with gestational age (r = -0.42, p = 0.03 and r = -0.52, p = 0.005, respectively) and birth weight (r = -0.38, p = 0.05 and r = -0.59, p = 0.001, respectively). Infants with high maximum expired ethane and pentane (exceeding 40 pmol x kg-1 x min-1) had higher odds of dying or having bronchopulmonary dysplasia than those with low ethane and pentane (odds ratio, 6.5; 95% confidence interval, 1.1 to 38.5; p < 0.05 for ethane and odds ratio, 5.6; 95% confidence interval, 1.1 to 29.3; p < 0.05 for pentane). We conclude that degree of prematurity is the single most important factor explaining free radical-mediated lipid peroxidation in premature infants. A therapeutic intervention to limit the effects of free radicals should be started during the 1st postnatal d in premature infants to be effective.