Corneal asphericity affects refractive outcomes of IOL formulas

Prolate corneas were associated with a myopic outcome, while oblate corneas produced a hyperopic result.

Issue: June 25, 2015

Anterior surface corneal asphericity had a significant impact on refractive outcomes after IOL implantation, according to a study.

Perspective from Jack T. Holladay, MD, MSEE, FACS

The influence of corneal asphericity on postoperative refractive outcomes has been largely overlooked compared with effective lens position, refraction and axial length, the study authors said.

“Corneal asphericity has received little to no attention when calculating the IOL power, especially if standard theoretical thin lens formulas are used,” Giacomo Savini, MD, the corresponding author, told Ocular Surgery News.

Giacomo Savini

Methods and measurements

The retrospective study, published in the Journal of Cataract and Refractive Surgery, included 115 eyes of 115 patients who underwent phacoemulsification and implantation of a one-piece AcrySof SA60AT posterior chamber IOL (Alcon).

The Haigis, Hoffer Q, Holladay 1 and SRK/T IOL formulas were used to calculate IOL power. A Placido disc corneal topographer (Keratron, Optikon), a rotating Scheimpflug camera (Pentacam, Oculus) and a rotating Scheimpflug camera combined with Placido disc corneal topography (Sirius, CSO) were used to measure anterior corneal asphericity Q-values at a diameter of 8 mm.

Savini and colleagues used linear regression analysis to assess the relationship between error in refraction prediction and Q-values.

“All formulas were equally influenced. Rather, there were some differences when evaluating the Q-value measurements by the three devices, as the measurements by the Placido disc corneal topographer showed a slightly stronger correlation with the refractive outcome,” Savini said.

Results and conclusions

Mean axial length was 23.98 mm, and mean IOL power was 20.48 D.

Mean Q-values were –0.18 for the Keratron, –0.27 for the Pentacam and –0.32 for the Sirius.

There was a statistically significant correlation between the error in refraction prediction and Q-values with all calculation formulas and all measurement devices.

Errors in refraction prediction corresponded most strongly with Q-values in eyes measured with the Hoffer Q formula and the Keratron. The association between corneal asphericity and refraction prediction error was strongest with the Hoffer Q and Holladay 1 formulas, Savini said.

More negative Q-values were associated with a negative refraction prediction such as myopia, while more positive Q-values correlated with positive refraction errors such as hyperopia.

“As expected, prolate corneas were associated with a myopic outcome, due to the fact that the corneal curvature inside the ring along which K readings are taken is steeper than in the 3 mm diameter. The opposite holds true for oblate corneas, where a hyperopic outcome may be achieved,” Savini said. – by Matt Hasson

Reference:

Savini G, et al. J Cataract Refract Surg. 2015;doi:10.1016/j.jcrs.2014.07.035.

For more information:

Giacomo Savini, MD, can be reached at Fondazione G.B. Bietti IRCCS, via Livenza 3, Rome, Italy; email: giacomo.savini@alice.it.
Disclosure: Savini reports no relevant financial disclosures.

Perspective

Jack T. Holladay, MD, MSEE, FACS

The lens used in the Savini study was a spherical Alcon AcrySof SA60AT. Today most surgeons realize that the best visual outcome in normal and post-refractive surgery corneas is with an aspheric IOL because it compensates for the positive spherical aberration (SA) in the cornea, results in less total ocular SA and eliminates the prediction error associated with residual SA. The positive SA of the cornea may be measured with each of the topography or tomography devices used in their study using the Z(4,0) Zernike term over a 6-mm zone for which the human average is +0.27 µm. Abbott Medical Optics (–0.27 µm), Alcon (–0.18 µm) and Bausch + Lomb (0.0 µm) all have aspheric IOLs that can be matched to the patient’s measured corneal spherical aberration to achieve the minimum ocular SA.
Unfortunately, the Q-value is only the minor component of the two variables needed to calculate the SA … the most important variable is the corneal radius of curvature (r). Calossi, de Ortueta and Pérez-Escudero have all shown and agree that the formula for converting to SA is proportional to the Q-value and inversely proportional to the third power of the corneal radius (1/r3). Using the Q-value alone is why their findings only explain 15% to 26% (R2) of their data in Table 2 [in the study]. Also, the 8-mm zone used in the study is far too large and does not correspond to the 6-mm zone used for labeling IOLs. The same zone must be used for the cornea as for the IOL so the values match. If they had used the Z(4,0) value determined over a 6-mm zone from each of the instruments used, their correlations and R2 values would have been much higher. In short, the Q-value alone is a poor indicator of corneal SA and therefore the residual ocular SA.
Most importantly, the recommendation should be to use aspheric IOLs. Matching the patient’s corneal SA to the closest available aspheric IOL will not only avoid the prediction error they found in the final refraction, but also significantly improve the quality of vision for visual acuity, contrast sensitivity and nighttime driving as well as reducing complaints of halos and glare.

Jack T. Holladay, MD, MSEE, FACS

OSN Optics Section Editor

Disclosures: Holladay reports no relevant financial disclosures.