Purchase this article with an account.
J. Beverage, J. Schwiegerling, J. Straub; Predicting Refraction Data From Pre-Op and Post-Op LASIK Wavefront Data . Invest. Ophthalmol. Vis. Sci. 2003;44(13):2665.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: We have investigated four techniques for predicting patient refraction based on wavefront data: 1) Zernike analysis with the low order terms, 2) Zernike analysis including higher order terms, 3) Zernike-Gauss analysis with the low order terms, and 4) Zernike-Gauss analysis including higher order terms. The Zernike analysis was performed using the standard set of Zernike coefficients. The Zernike-Gauss analysis modifies the standard coefficients to account for the Stiles-Crawford effect. Methods: Patients’ manifest refraction and wavefront aberrations were measured before and one year after LASIK surgery. Wavefront errors were measured using a Shack-Hartmann wavefront sensor and were expressed as a standard Zernike set. The Zernike coefficients were then converted to Zernike-Gauss coefficients to include the Stiles-Crawford effect. Sphere, cylinder, and axis were computed for both sets of coefficients using only low-order terms (Z3, Z4, Z5), then low and higher order terms (Z3, Z4, Z5, Z11, Z12, Z13). Correlation analysis was then performed to compare the predicted refractions to the manifest refractions for each patient before and after surgery. Results: Correlation with manifest refractions for ten patients shows that all four analysis techniques provide similar results before surgery (slope=0.82, R2=0.80). After surgery however, there is a definite advantage to using only low order Zernike and Zernike-Gauss coefficients. The higher order aberrations induced by refractive surgery also introduce more variability in the predicted refractions. For the low order analysis techniques after surgery, slope=0.92 and R2=0.74, while for the techniques including higher order terms, slope=0.57 and R2=0.33.Conclusions: Because refractive surgery introduces higher-order aberrations, an accurate means of predicting refraction based on wavefront data is needed. All techniques discussed above perform similarly before refractive surgery, but the low order analysis methods predict refractions more accurately in post-op patients. The Stiles-Crawford effect had only a minor effect on the predicted refractions before and after surgery.
This PDF is available to Subscribers Only