The prevention of retinopathy of prematurity (ROP) remains a persistent problem. A previous report has focused on the possible protective effect of bilirubin on the development of ROP. These results still await clinical confirmation by other research groups. Therefore, we undertook a retrospective clinical study trying to confirm this attractive hypothesis. Twelve premature newborns under 32 weeks of gestation with ROP stage 3-4 were matched for gestational age with 12 infants without ROP. Data were collected about the infant's characteristics, medical illnesses, ventilatory settings and treatments. The total serum bilirubin concentrations between the 1st and 8th postnatal day were also gathered. The two matched groups were comparable as to their basic data, clinical characteristics and treatment, except for a slight, but significant longer duration of phototherapy for group ROP 0 (mean, 50.2 h; SD 48,6 vs 31.6 h; SD 42.7 in ROP 3-4; P = 0.02). No statistical difference relative to bilirubin was found between the two groups, neither when expressed as daily mean concentrations, nor as area under the curve (AUC) (mean, ROP 0: 17876.7; SD 6077.3 vs 18888.4; SD 55552.7 in ROP 3-4; P = 0.404) or AUC/h (mean, ROP 0: 135.1; SD 36.3 vs 144.1; SD 23.2 in ROP 3-4; P = 0.515). Our findings do not confirm the hypothesis of a clinically measurable, beneficial role of bilirubin on the development of ROP.