الفهرس 
Axial LengthOnly 14 pages are availabe for public view
 |  | from | 161 |  |  |
| | | from | 161 | | |
Abstract
Cataract surgeons dedicate a significant portion of their preoperative workflow towards performing IOL calculations, picking a formula, and selecting the most appropriate lens power to implant into a given patient’s eye. This may be a challenging and time-consuming process for many surgeons given the number of formulae and factors to consider with each eye. Since there has not been a single perfect IOL formula or solution which can help to simplify this complex process, surgeons have always relied on using multiple IOL formulae and picking one which they feel will suit their needs.
Expectations from cataract surgery remain high from both patients and surgeons. Current IOL calculation formulae generally perform well for average eyes. However, due to the mathematical and biometric imperfections, these formulae remain suboptimal in eyes with atypical values of axial length, keratometry and anterior chamber depth. Several new IOL formulae have been developed in the past decade which have sought to improve post-operative outcomes by introducing additional variables and incorporated advanced
mathematical techniques. However, the data on these new formulae is still lacking.
This study aims to compare the accuracy of intraocular lens calculation formulae (Holladay I, Haigis, SRK/T, Hoffer Q, Holladay II, and Barrett Universal II) in the prediction of postoperative refraction for eyes with AL 26.0 mm or more.
This prospective study was carried on Forty-seven eyes of Forty-seven patients. The mean age of patients is 56.9 years with SD ± 10.00. Females represent 53.2 % of the studied group of the total number of patients, while males represent 46.8 %.
Mean absolute error was calculated for each formula showed that the least mean absolute error is for Barrett Universal II formula. SRK/T formula is the 2nd lower mean absolute error followed by Haigis formula and the last is Holladay II formula.
The scatter plots between axial length and absolute error from each of the six calculations formulae was done and showed that Barrett universal II has negative correlation with axial length which means when axial length increase the prediction error decrease. As for SRK/T and Haigis formulae
there is a linear correlation which means the prediction error is constant with the variation of axial length.
Numerical Predictive Error of different estimating methods was calculated as the difference between the actual postoperative refractive outcome and the predicted refraction estimated by IOL master (actual postoperative refraction − predicted refraction) for each formula. The mean numerical error was done for postoperative spherical equivalent by Friedman’s test statistic = 112.026, df = 5, p < 0.005. As mean, SD, Range and Median showed that SRK/T formula had the lowest mean numerical error followed by Haigis formula then Barrett Universal II and last came Holladay II formula. The Hoffer Q & Holladay I formulae had the highest mean numerical error.
Pairwise comparison between prediction error of different formulae and adjusted significance by Bonferroni correction showed no statistical difference between the four formulae (Barrett Universal II, SRK/T, Haigis and Holladay II). However, it is highly significant between these four formulae and the other two formulae (Hoffer Q and Holladay
I) with a p-value = zero.
For high axially myopic eyes, the Barrett Universal II formula is the best in this group of patients and provide the lowest mean absolute error, highest percentage of eyes within
≤ ± 0.50 D and the lowest risk of refractive surprise compared to other IOL power calculation formulae. Followed by SRK/T and Haigis formulae which performed similarly and provide a good predictable outcome then at last came Holladay II formula. However, Holladay I and Hoffer Q formulae are not suitable for high axial patients.
More comparison studies will need to be performed to compare these formulae as they continue to improve with time to help provide a more conclusive calculation solution.