Risk assessment for ectasia after corneal refractive surgery.

PURPOSE To analyze the epidemiologic features of ectasia after excimer laser corneal refractive surgery, to identify risk factors for its development, and to devise a screening strategy to minimize its occurrence. DESIGN Retrospective comparative and case-control study. PARTICIPANTS All cases of ectasia after excimer laser corneal refractive surgery published in the English language with adequate information available through December 2005, unpublished cases seeking treatment at the authors' institution from 1998 through 2005, and a contemporaneous control group who underwent uneventful LASIK and experienced a normal postoperative course. METHODS Evaluation of preoperative characteristics, including patient age, gender, spherical equivalent refraction, pachymetry, and topographic patterns; perioperative characteristics, including type of surgery performed, flap thickness, ablation depth, and residual stromal bed (RSB) thickness; and postoperative characteristics including time to onset of ectasia. MAIN OUTCOME MEASURES Development of postoperative corneal ectasia. RESULTS There were 171 ectasia cases, including 158 published cases and 13 unpublished cases evaluated at the authors' institution. Ectasia occurred after LASIK in 164 cases (95.9%) and after photorefractive keratectomy (PRK) in 7 cases (4.1%). Compared with controls, more ectasia cases had abnormal preoperative topographies (35.7% vs. 0%; P<1.0x10(-15)), were significantly younger (34.4 vs. 40.0 years; P<1.0x10(-7)), were more myopic (-8.53 vs. -5.09 diopters; P<1.0x10(-7)), had thinner corneas before surgery (521.0 vs. 546.5 microm; P<1.0x10(-7)), and had less RSB thickness (256.3 vs. 317.3 microm; P<1.0x10(-10)). Based on subgroup logistic regression analysis, abnormal topography was the most significant factor that discriminated cases from controls, followed by RSB thickness, age, and preoperative corneal thickness, in that order. A risk factor stratification scale was created, taking all recognized risk factors into account in a weighted fashion. This model had a specificity of 91% and a sensitivity of 96% in this series. CONCLUSIONS A quantitative method can be used to identify eyes at risk for developing ectasia after LASIK that, if validated, represents a significant improvement over current screening strategies.