Abstract
Purpose :
As cataract surgery is nowadays considered a refractive surgery, new formulas are continually being developed in order to achieve optimal refractive outcomes after the surgery. This study compares newer IOL power formulas, Kane and Hill-RBF V 2.0, to other formulas for different axial lengths.
Methods :
This retrospective study included 406 eyes of 406 patients who underwent cataract surgery. Eyes with previous refractive surgery, associated corneal or anterior segment abnormalities, or intraoperative complications were excluded. Biometry was performed using IOL-Master 700. Predicted refraction from 7 formulas (Hill-RBF V 2.0, Kane, Barrett Universal II, Haigis, Hoffer-Q, Holladay 2, and SRK/T) was compared to postoperative refraction at 1-3 months for different axial lengths: <22.5 mm (short), 22.5-25.5 mm (medium), and >25.5 mm (long). Post-hoc analyses and Bonferroni correction were applied for multiple comparisons.
Results :
Overall and within short and medium eyes, all formulas had similar percentages of eyes within ±0.5 D of the target refraction (Table 1). In long eyes, however, the percentages within ±0.5 D were significantly higher for Barrett Universal II and Kane formulas (both 72%) compared to those for Hoffer-Q and Holladay 2 formulas (both 52%) (P=0.004). Mean numerical error (MNE) and mean absolute error (MAE) were similar for all formulas overall as well as within medium and long eyes. However, within short eyes, MNE was significantly lower for Barrett Universal II (0.27 ± 0.38 D) and Haigis (0.28 ± 0.42 D) compared to Holladay 2 (-0.06 ± 0.41 D) (P=0.021 and P=0.014, respectively).
Conclusions :
The accuracy of newer IOL formulas may depend on the axial length. For long eyes, Kane and Barrett Universal II performed better than Hill-RBF V 2.0 and other formulas studied. For short and medium eyes, Kane and Hill-RBF V 2.0 had an accuracy similar to other formulas.
This is a 2021 ARVO Annual Meeting abstract.