Predicting the refractive outcome and accuracy of IOL power calculation after phacoemulsification using the SRK/T formula with ultrasound biometry in medium axial lengths

Abstract

PurposeTo evaluate the accuracy of the SRK/T formula using ultrasound (US) biometry in predicting a target postoperative refraction of ±1.00D in eyes with medium axial length (AL) that underwent phacoemulsification.MethodsThe present study was a retrospective analysis, which included 538 eyes with an AL from 22.0 to 24.60 mm that underwent phacoemulsification and foldable intraocular lens (IOL) implantation (six different IOLs) in the bag. Preoperative AL was measured by US biometry and IOL power (IOLp) was calculated with the SRK/T formula. Patients had a complete ophthalmic examination, preoperatively and 1, 7, and 30 days after surgery. The achieved spherical equivalent (SE) and the prediction error (PE) were calculated. The prediction error was defined as the difference between attempted predicted target refraction and the achieved postoperative SE refraction. Statistical analysis was performed with SPSS V21.ResultsThe mean age of the patients was 66.96±9.67 years, the mean AL was 23.29±0.62 mm, the mean K1 was 43.62±1.49D, the mean K2 was 43.69±1.53D, the mean IOL power was 21.066±1.464D, the mean attempted (predicted) SE was −0.178±0.266D, and the mean achieved SE was −0.252±0.562D. The mean PE (difference between predicted and achieved SE) showed a relatively hyperopic shift (mean ± standard deviation: 0.074±0.542D, ranging from −1.855 to 2.170D, P=0.001). A total of 93.87% of eyes were within ±1.00D of the PE and 92.75% of eyes within ±1.00D of achieved postoperative refraction. A total of 39 eyes (7.25%) had a refractive surprise. A total of 32 of 39 eyes were more myopic than −1.00D and 7 of them were more hypermetropic than +1.00D. There was no correlation between the mean PE and IOL type, AL, K1, K2, and IOLp. There were a positive statistically significant correlation between PE and age (r=0.095; P=0.028) and a negative statistically significant correlation between achieved SE and AL (Spearman’s r=−0.125; P=0.04), and age (r=−0.141; P=0.01).ConclusionThe IOLp calculation using the SRK/T formula with US biometry may demonstrate very good postoperative refractive outcomes in medium eyes with a few refractive surprises.

Accuracy of the SRK/T Formula Using Partial Coherence Interferometer, Al-Scan after Phacoemulsification

“…In this study by using the SRK/T formula, the mean numeric error ranged from −1.38 D to +1.34 D. The mean error was -0.08 ± 0.61 D. The majority of patients (90%) were within 1 D of the predicted refractive error (Table 2). These results were similar to the study done by Karabela et al (3) with a prediction accuracy of 38.66% for refractive errors of 0.25D, 69.51% for refractive errors of 0.50D, and 93.87% for refractive errors of 1.00D. Lagrasta et al (9) showed a prediction accuracy of 55% for refractive error within 0.50D, and 91% for refractive error within 1.00D using SRK/T formula with US biometry in 33 eyes of 33 patients with medium ALs (22.2-24.5 mm).…”

“…Accurate biometric data are essential for achieving good surgical outcomes and patient satisfaction after cataract and refractive surgery. Third-generation formulas such as the Holladay 1, the Hoffer Q, and the SRK/T; fourth-generation formulas such as the Holladay 2, the Haigis, and the Olsen; and newer formulas are the most commonly used in all eyes (3) . Different studies to evaluate the predictive accuracy of various IOL power calculation formulas; Wang et al (6) showed similar performance of the Haigis, Hoffer Q, Holladay1, and SRK/T in medium eye length, in another study SRK/T and Haigis performed equally well and outperformed the Hoffer Q and Holladay 1 in 34 eyes between 25 and 28mm (7) .…”

“…Third-and fourthgeneration formulas are now the most preferred formulas. The SRK/T is one of the third-generation formulas that was representing a combination of linear regression method with a theoretical eye model (3) . Ultrasound (US) biometry (A-Scan) had been considered the gold standard for axial length and ACD measurement.…”

Cataract surgery refractive outcomes: representative standards in a National Health Service setting

“…In a multicentre, prospective cohort study involving 1553 eyes, Hahn et al reported that 95% achieved a 1-month postoperative refraction within 1 D of target 9. Karabela et al reported a relatively high refractive success of 93% of patients being within 1 D of target refraction 10. However, these results were in selected populations.…”

“…Hahn et al excluded patients with any ocular comorbidity or previous eye surgery; lenses used were only in the range of +16 to+26 to avoid possible pitfalls in predicting refraction at the low and high ends of the refractive spectrum, all patients were suitable for optical biometry, and all of the operating surgeons were highly experienced in cataract surgery having previously performed over 10 000 cases 9. Similarly, Karabela et al only included patients with medium AL ranging from 22.00 to 24.60 mm in their study 10. The Swedish National Cataract Register Study, which included 17 056 operations, reported 92.70% of cases to be within 1 D of target refraction 1–2 months postoperatively 12.…”