To compare axial length applanation ultrasonography (A-scan) (CineScan B-Scan; Quantel Medical, Bozeman, MT, USA) and low-coherence reflectometry (Lenstar LS900®; Haag-Streit, Bern, Switzerland), the accuracy of the predictive postoperative refraction of both instruments, and the intraocular lens (IOL) calculators.

A total of 250 eyes of 191 patients who received cataract surgery were included in the study. The axial length was measured by the A-scan and Lenstar LS900®. The SRK-2, SRK/T, and Olsen formulas were used to calculate the IOL power, and the difference between the predictive and actual postoperative refractions after 6 weeks and the probability that they were within 0.25 diopters (D) and 0.5 D were compared.

The mean axial lengths measured by the A-scan and Lenstar LS900® were 23.42 ± 0.94 mm and 23.55 ± 0.95 mm, respectively, which showed a statistically significant difference (paired t -test, p = 0.000). When comparing the difference between the predictive and actual postoperative refractions, the results of the A-scan using the SRK-2 and SRK/T formulas were significant toward the hyperopia, and the results of the Lenstar LS900® using the SRK-2, SRK/T, and Olsen formulas were significant toward the myopia (paired t -test, p = 0.001 and p < 0.001, respectively). When comparing the mean absolute difference between the two refractions and the probability that they were within 0.25 D and 0.5 D, the Lenstar LS900® using the Olsen formula significantly showed the highest accuracy (McNemar test, p = 0.045 and p = 0.002; p = 0.010 and p = 0.002, respectively).

The A-scan using the SRK-2 and SRK/T formulas showed that the actual postoperative refraction was more hyperopic than the predicted refraction, whereas the Lenstar LS900® resulted in more myopic findings. The accuracy of predictive postoperative refraction was highest with the Lenstar LS900® using the Olsen formula.