Prediction Of Residual Astigmatism Using Intraoperative Wavefront Aberrometry Versus Multiple Toric Iol Calculators

The Barrett Toric IOL calculator offers excellent performance in the prediction of residual astigmatism following toric IOL implantation by including estimation of posterior corneal astigmatism and effective lens position. The option of including measured posterior corneal astigmatism is also available. The Kane Toric formula, EVO 2.0 formula and the Naeser/Savini toric calculator are other recently introduced methods. Intraoperative Wavefront Aberrometry (IWA) is an alternative method for intraoperative IOL cylindrical power selection and axis refinement. In this study we compared the prediction of residual astigmatism by each of these methods.

Department of Ophthalmology, Centro Hospitalar e Universitário de Coimbra - CHUC, Coimbra, Portugal. Unidade de Oftalmologia de Coimbra (UOC), Coimbra, Portugal.

Prospective study with 92 eyes implanted with a toric IOL (SN6ATx, SND1Tx or TFNTx, Alcon) using IWA (ORA, Alcon) for intraoperative toric power selection and residual astigmatism prediction. Residual astigmatism for the same toric IOL power was back-simulated using the Barrett Toric calculator (with estimated (ePCA) and measured (mPCA) posterior corneal astigmatism), Kane formula, EVO 2.0 formula and Naeser/Savini Toric Calculator, using Oculus Pentacam keratometry data. Each suggested toric power by the calculators was noted. Subjective refraction was obtained 3 months postoperatively. The postoperative refractive astigmatism prediction error in the spectacle plane was evaluated by the centroid and the mean absolute error for each method.

The centroid prediction error was 0.11D@177±0.54 for EVO, 0.14D@2±0.54 for Barrett ePCA, 0.13D@1±0.54 for Barret mPCA, 0.17D@180±0.55 for Naeser/Savini, 0.17D@8±0.85 for IWA and 0.24D@1±0.55 for Kane. The Barrett ePCA and EVO formulas showed the highest proportion of eyes with a prediction error within 0.50D, 0.75D and 1D with 66%, 89% and 92% for Barrett (no improvement with mPCA) and 68%, 88% and 92% for EVO. They were followed by Naeser/Savini (61%, 86%, 90%), Kane (57%, 82%, 92%) and lastly IWA (40%, 64%, 80%). The chosen IOL cylindrical power (T) using IWA differed from the suggested T power using Barrett ePCA in 48/92 eyes (52%) and EVO in 55/92 eyes (60%), with tendency lower T values using IWA compared to Barrett and EVO formulas.

The EVO 2.0 formula had the lowest centroid and the lowest mean absolute prediction error, followed closely by Barrett. Measured posterior corneal astigmatism did not further enhance Barrett formula’s predictions.