Corneal ectasia, the progressive steepening and thinning of the cornea that reduces UCVA and often BCVA, is one of the most feared complications that may occur after a seemingly successful refractive surgery case. Although rare, this complication can have devastating consequences. Identification of high-risk patients in whom surgery should be avoided is critical.
Numerous purported risk factors for ectasia have been proposed, including forme fruste keratoconus, high myopia, residual stromal bed (RSB) thickness less than 250 µm, preoperative corneal thickness less than 500 µm, multiple enhancements, and elevated posterior float values (viewed with the Orbscan II; Bausch & Lomb, Inc., Rochester, New York). Although any of these factors could potentially play a role in determining an individual's risk of developing ectasia after refractive surgery, many have not been validated. Most screening strategies do not account for the full spectrum of combined risk factors, and ectasia cases seemingly without risk factors have continued to be published in the literature.1 Clearly, a more comprehensive screening strategy has been needed.
THE ECTASIA RISK SCORE SYSTEM
Recent studies have sought to (1) evaluate purported ectasia risk factors in a systematic fashion, (2) validate or eliminate individual factors based on scientific analysis, and (3) develop a comprehensive risk-assessment system that assesses individual risk by applying multiple individual factors in a weighted fashion.
We recently conducted a retrospective, comparative, case-controlled study2 of ectasia cases and controls from both the literature and our own referral database, evaluating a variety of pre- and perioperative characteristics (eg, age, gender, spherical equivalent refraction, pachymetry, topographic patterns, type of surgery performed, flap thickness, ablation depth, RSB thickness). We compared 171 ectasia cases that occurred after LASIK (n=164) or PRK (n=7) with a contemporaneous control group of 186 eyes that underwent uneventful LASIK and experienced a normal postoperative course. Patients had at least 1 year follow-up.
Compared with controls, ectasia cases had significantly more abnormal preoperative topographies (0% vs 44%, respectively; P<1.0 x10-15), were significantly younger (40 vs 34.4 years, respectively; P<1.0 x10-7), were more myopic (-5.09 vs -8.53 D, respectively; P<1.0 x10-7), had thinner corneas preoperatively (546.5 vs 521 µm, respectively; P<1.0 x 10-7), and had lower RSB thickness (317.3 vs 256.3 µm, respectively; P<1.0 x 10-10).
Placido-based topographic patterns were categorized into four groups: (1) normal/symmetrical bowtie; (2) asymmetrical bowtie; (3) inferior steepening pattern or skewed radial axis; or (4) abnormal, including keratoconus, pellucid marginal corneal degeneration, and forme fruste keratoconus.
After performing logistic regression multivariate analysis on a subgroup of cases with all necessary information, the most significant risk factors, in order of importance, were: (1) abnormal topography, (2) low RSB thickness, (3) young age, (4) low preoperative corneal thickness, and (5) high myopia.
Taking these recognized risk factors into account in a weighted fashion, we developed the Ectasia Risk Score System (Table 1). This model correctly identified 91% of the ectasia cases in our series as high risk; only 4% of controls were incorrectly identified as high risk (91% sensitivity, 96% specificity). We also recently tested the Ectasia Risk Score System on a novel population of ectasia cases and controls;3 the results were nearly identical to the original study (92% sensitivity and 94% specificity).
Between these two studies, we have confirmed the vaildity of the Ectasia Risk Score System. It represents a significant improvement over previously reported screening strategies. The following are other notable findings in our study: (1) topographic patterns other than forme fruste keratoconus conveyed a risk of ectasia, (2) meticulous topographic analysis remains the most crucial step to avoid performing surgery on high-risk candidates, and (3) patient age was more significant than previously thought. The latter is likely due to a combination of factors, including the inherently lower tensile strength in younger corneas and the predisposition for some younger patients who have not yet manifested significant topographic abnormalities at the time of their screening to develop ectatic corneal disease in the future. Thus, before a younger patient undergoes LASIK, extra attention should be given to all other risk factors. Interestingly, preoperative corneal thickness was a risk factor; however, its impact was less than that of other factors, and therefore preset cutoff limits (eg, 500-µm threshold) do not appear justified at this time.
The Ectasia Risk Score System is a valid and effective method for detecting eyes at risk for ectasia after LASIK. This represents an important improvement over previously available screening options; however, this question still remains: Can we further improve upon current ectasia screening? Newer topographic technologies may provide even more accurate screening that might detect subtle abnormalities in patients who would otherwise qualify as low-risk candidates for surgery.
Further research is needed to test and modify this currently available screening system, which we hope to achieve by collecting more ectasia cases. All surgeons are encouraged to submit ectasia cases to the American Society of Cataract and Refractive Surgery (ASCRS) ectasia registry at www.ectasiaregistry.com.
Claudia E. Perez-Straziota, MD, is beginning her ophthalmology training at the Emory Eye Center, Emory University, Atlanta. Dr. Perez-Straziota states that she has no financial interest in the products or companies mentioned. She may be reached at tel: +1 404 778 4530.
J. Bradley Randleman, MD, is an Assistant Professor of Ophthalmology at the Emory Eye Center and Emory Vision, Emory University, Atlanta. Dr. Randleman states that he has no financial interest in the products or companies mentioned. He may be reached at tel: +1 404 778 2264; e-mail: firstname.lastname@example.org.