Objectives Through comprehensive ophthalmic examination of adult offspring we sought to determine the impact of multiple prenatal ultrasound scans on ocular development. Methods 2743 pregnant women recruited to the Western Australian Pregnancy (Raine) Cohort study during 1989-1991 were randomized to receive at King Edward Memorial Hospital, Western Australia either multiple prenatal ultrasound scans and Doppler flow studies (intensive group) or a single ultrasound scan at 18 weeks' gestation. Neonatal birth weight of the offspring and other physical measurements were collected prospectively. At age 20 years, participants underwent a comprehensive ophthalmic examination including measurement of ocular biometry and visual acuity. Results Complete data were available for 1134 adult offspring participants. The mothers of 563 of these had been randomized to receive multiple prenatal ultrasound scans. The mean age of participants at follow-up was 20.0 years. There was no statistically significant difference between the two groups with regard to ocular biometric or visual outcomes, except for slightly higher intraocular pressure identified in individuals exposed to multiple ultrasound scans (P = 0.034). Although infants in the intensive-ultrasound arm were more likely to have birth weights in the lower quartiles, this was not reflected in adult eye development. Axial length, lens thickness, corneal curvature and thickness and optic cup to disc ratio (a risk factor for glaucomatous optic neuropathy) were not significantly influenced by the more frequent ultrasound protocol. Conclusions Prior to this study, there was a paucity of safety data for ultrasound with regard to eye development. We found that frequent in-utero exposure to ultrasound, including B-mode imaging and the use of spectral Doppler mode from 18 weeks' gestation, had no significant impact on visual outcomes or ocular biometry.