Correlation of aberrometry, contrast sensitivity, and subjective symptoms with quality of vision after LASIK.

PURPOSE To compare which parameter category (wavefront data, psychophysical data, or subjective symptoms) predicts best subjective quality of vision after LASIK. METHODS Twenty-eight eyes (15 patients) were included. Twenty-three eyes (12 patients) underwent uneventful LASIK; 5 eyes (3 patients) were symptomatic eyes treated with myopic LASIK elsewhere. Mean preoperative spherical equivalent refraction was -4.79+/-1.92 diopters (D) (range: -1.63 to -7.13 D); mean patient age was 36.6+/-7.4 years (range: 18 to 48 years). All examinations were performed 1 month postoperatively. The wavefront error was described with Zernike polynomials (6-mm pupil). Psychophysical tests included high-contrast visual acuity and contrast sensitivity with and without glare at 167 cd/m(-2), 1.67 cd/m(-2), and 0.167 cd/m(-2) with best spectacle correction. Correspondingly, overall subjective quality of vision and frequency of visual symptoms (glare, halos, starbursts, ghosting, blur) were assessed for three lighting conditions (photopic, high-mesopic, and low-mesopic) using a questionnaire with a visual analog scale. For each parameter category and each lighting condition, a multiple stepwise backwards regression model with the overall quality of vision item value as dependent was applied. RESULTS Under all lighting conditions, subjective symptom scores predicted subjective quality of vision best (adjusted R2=0.83-0.92) with blur as the main predictor throughout all conditions. Psychophysical tests did not significantly predict postoperative subjective quality of vision. The adjusted R2 for the Zernike coefficients was highest for low-mesopic (0.56) and lowest for photopic conditions (0.31). CONCLUSIONS Different parameter categories for the description of optical quality did not predict subjective quality of vision after LASIK equally. Subjective symptom scores had the highest predictability, whereas psychophysical tests with spectacle correction had no predictability. The latter probably do not reflect all dimensions of subjective quality of vision.