Dr.Rwituja Thomas, R19863, Dr.Mathew Kurian, Dr.Neeraj A Israni, Dr.Sanket Bhatnagar, Dr.Rwituja Thomas
Introduction
Accurate measurement of ocular parameters and appropriate IOL power calculation formulae help choose the correct IOL power to be implanted for attaining the expected refractive outcomes. The accuracy of the standard IOL power calculation formulae is comparable in predicting refractive outcomes in eyes with an average axial length (AL) 1-3. However, beyond average axial lengths these formulae are less accurate in predicting actual refractive outcomes.
The Hill‑Radial basis function (RBF) is a machine‑learned formula based on outcome data of more than 12000 eyes, optimized for Lenstar biometry and Alcon SN60WF IOL implantation, though it also works very well with biometry data from other optical biometry devices and with other biconvex IOL’s from -5 D to +30D.Amongst the fourth generation regression formulae, Barrett has shown to perform most accurately for eyes with longer axial length4. In our study, we compare the accuracy of these two formulae in predicting post-cataract refractive outcomes in myopic eyes.
Purpose
To compare the outcomes of the Hill RBF and Barrett formulae in eyes with high myopia (AL 25 mm to 30 mm) by Lenstar biometry undergoing phacoemulsification with implantation of Alcon IQ SN60WF IOLs.
Methodology
Our study included 28 eyes with axial length between 25– 30 mm by Lenstar biometry (Haag Streit Diagnostics, USA) having senile cataract that were implanted with AcrySof SN60WF(Alcon) monofocalintraocular lenses after uncomplicated phacoemulsification performed by a single surgeon. Adults with traumatic or complicated cataract, ocular comorbidity and missing data were excluded.
For both Barrett and Hill RBF formulae, the predicted refraction for the IOL power used was calculated using AL, K, and the manufacturer’s recommended A constant. Achieved post-operative refraction was performed between 4 to 6 weeks assuming that the majority of patients would have reached refractive stability by this time. Prediction error (PE) = Predicted refraction-Achieved refraction was calculated for each formula.
Statistical Analysis:
Microsoft Excel 2016 spreadsheet software was used to chart the data and Medcalc version 18.2.1 statistical software was used for the purpose of statistical analysis. The Shapiro Wilk test was performed to determine whether the data was normally distributed. Bland Altman plots determined agreement.
Results
Twenty eight eyes of 27 patients were included. Demographic data and biometry parameters of the study population are shown in Table 1
| Parameter | Value |
| Sex
Males Females |
20 (74 %) 7 (26 %) |
| Laterality of eye
Right Left |
17 (60 %) 11(40 %) |
| BIOMETRY PARAMETERS | |
| Mean Axial length (mm) ± SD | 26.56 ± 0.95 |
| Mean K (D) ± SD | 43.37 ± 2.04 |
Table 2 shows the IOL power and the predicted refraction and prediction error by the two formulae.
| Barrett | Hill RBF | |
| IOL power (Mean ± SD) | 12.47 ± 3.43 | -0.068 ± 0.24 |
| Predicted refraction(Mean ± SD) | 12.81 ± 3.55 | -0.032 ± 0.26 |
| Prediction error(Mean ± SD) | -0.08 ± 0.25 | -0.03 ± 0.3 |
| Error >± 0.25 D | n = 10 (35.71%) | n = 10 (35.71%) |
| Error >± 0.5 D | n = 1 (3.57%) | n = 4 (14.29%) |
| Residual refraction> 0.25 D | n = 2 (7.14%) | n = 4 (14.29%) |
| Residual refraction > 0.5 D | n = 0 (0%) | n = 2 (7.14%) |
The preoperative uncorrected distance acuity (UDVA) in logmar was 1.2 ± 0.66 and the Corrected distance vision (CDVA) was 0.62 ± 0.55. The postoperative UDVA was 0.15 ± 0.15 that improved to 0.07 ± 0.12.
DISCUSSION
The Barrett formula mean prediction error was more myopic than the Hill RBF formula. However, it has fewer eyes with error > ± 0.5 D and also the tendency for a hyperopic residual refraction was less than the Hill RBF formula. Both formulae did not give more than 1D residual refractive error in any patient.
Irrespective of the formula used patients can be assured of refractive outcomes close to the predicted refraction suggested by the formula. In the Barrett formula there is a greater possibility of a myopic error which is likely to be advantageous in a pre-existing high myope.
CONCLUSION
The new generation formulae are giving excellent refractive outcomes in challenging clinical situations when performing cataract surgery in eyes with high myopia. In our study population the Barrett formula outperformed the Hill RBF formula which is optimized for the combination of Lenstar biometry and Alcon SN60WF IOL implantation used in this study.
References:
- Sanders DR, Retzlaff J, Kraff MC.Comparison of the SRK II formula and othersecond generation formulas. J Cataract Refract Surg. 1988 Mar;14(2):136-41
- Hoffer, K.J.Clinical results using the Holladay 2 intraocular lens power formula. J Cataract Refract Surg. 2000; 26: 1233–1237
- Hoffer KJ. The Hoffer Q formula: A comparison of theoretic and regression formulas. Journal of Cataract & Refractive Surgery. 1993;19(6):700–12.
- Roberts T, Hodge C, Sutton G, Lawless M. Comparison of Hill-radial basis function, Barrett Universal and current third generation formulas for the calculation of intraocular lens power during cataract surgery. Clinical & Experimental Ophthalmology. 2017;46(3):240-246.
Leave a Comment