Refractive surgery screening and corneal safety

Corneal screening before refractive surgery is a risk assessment, not a single pass-or-fail number. A glasses prescription describes the correction needed; it does not show the cornea’s shape, thickness or structural behaviour. Eye history, ocular surface, refraction and procedure-specific tissue changes all inform the discussion. No scan or score can guarantee that ectasia or another complication will not occur. [1]

Why screening uses several kinds of information

Topography maps surface curvature; tomography adds structural information, including the posterior corneal surface; pachymetry measures thickness. These are considered as patterns alongside refraction, contact-lens and eye history, tear-film status and the proposed procedure. The AAO Preferred Practice Pattern is included as a relevant professional guideline, but its full text was not accessible for this review. No specific screening checklist or threshold is attributed to the inaccessible full text. The 2026 international consensus on keratoconus and ectatic diseases is a newer expert update on diagnosis and management; it informs corneal-disease context, not an individual refractive-surgery clearance decision. [8] [1]

The concern in much of the research is corneal ectasia: progressive thinning or shape change that can induce irregular astigmatism and affect vision. It is uncommon, but reported cases show that screening cannot eliminate risk. Studies can identify patterns associated with ectasia; they cannot certify an individual outcome. [3] [7]

How the historical LASIK risk score was developed

Randleman and colleagues retrospectively assembled 171 ectasia cases (158 from published reports and 13 institutional cases) and 186 controls with uneventful LASIK. Of the ectasia cases, 164 followed LASIK and seven followed PRK. Their Ectasia Risk Score System combined topographic pattern, predicted residual stromal bed (RSB), age, preoperative corneal thickness and refractive error. Reported sensitivity was 96% and specificity 91% in the assembled comparison. [3]

Because the sample was deliberately enriched with cases and controls, those figures are classification performance, not the probability that a candidate with a particular score will develop ectasia. Retrospective case collection, historical surgical practice and the chosen control group limit how directly the results apply to today’s general screening population.

A separate retrospective validation compared 50 post-LASIK ectasia eyes with 50 uncomplicated LASIK controls from Emory University procedures performed between June and October 2004; the controls had at least one year of postoperative follow-up. The score labelled 46/50 ectasia eyes high risk (92% sensitivity in that sample) and also labelled 3/50 controls high risk (6%). It detected more cases than the paper’s historical comparison criteria of abnormal topography or RSB below 250 µm (92% versus 50%; P<.00001). The selected 50:50 design cannot estimate absolute risk, and the LASIK score is not thereby validated for PRK or SMILE. [4]

Why tissue metrics depend on the procedure

Percent tissue altered (PTA) was studied as a LASIK measure: flap thickness plus ablation depth, divided by preoperative central corneal thickness. Santhiago and colleagues compared 30 ectatic eyes from 16 patients whose preoperative Placido topography had appeared normal with 174 uncomplicated LASIK control eyes from 88 patients; controls had at least three years’ follow-up. PTA of at least 40% occurred in 97% of ectatic eyes, and the reported odds ratio was 223. The accessible abstract did not give a confidence interval for that estimate. [5]

This was a retrospective case-control association, not a prospective test of a universal threshold. PTA includes a LASIK flap measurement; it should not be carried over as a validated cut-off for flapless SMILE or surface-ablation PRK. RSB and other scores also need interpretation with the maps, prescription, history and intended technique. None is a self-test or a guarantee.

What tomography and biomechanics add

Ferreira-Mendes and colleagues evaluated the Tomographic and Biomechanical Index (TBI) in a retrospective diagnostic study: 312 normal eyes, 118 keratoconic eyes and 57 very-asymmetric-ectasia eyes and the 57 fellow eyes with normal topography from those same patients. For the selected normal-versus-ectatic comparison, a TBI cut-off of 0.335 had 94.4% sensitivity and 94.9% specificity; area under the ROC curve was 0.988 (95% CI 0.982–0.995). In the normal-topography fellow-eye subgroup, an optimised cut-off of 0.295 had 89.5% sensitivity and 91.0% specificity; AUC was 0.960 (95% CI 0.937–0.983). [6]

Biomechanics here refers to how the cornea deforms and recovers in response to a brief air puff, measured alongside tomographic shape data. Sensitivity is the proportion of the selected disease group detected; specificity is the proportion of the selected comparison group correctly classified. AUC summarises how well a test separates those groups across thresholds. These are diagnostic classification measures, not prediction of post-surgical outcomes in an unselected cohort or a patient's absolute probability. The cut-off changed with the comparison, so the index is not a universal clearance threshold.

The 2015 consensus used three questionnaire rounds and a meeting with 36 ophthalmologists, with a two-thirds agreement threshold. Its 2026 second edition expanded the process to four questionnaire rounds plus a meeting, involving 128 ophthalmologists across seven panels, again with a two-thirds threshold. These are expert consensus processes, not prospective trials validating a scan, a screening threshold or an individual’s eligibility. [2] [8]

What reported ectasia rates can and cannot show

A systematic review of reports published from 1984 to 2021 counted ectasia cases without identified preoperative risk factors: 57 after PRK, 1,453 after LASIK and 11 after SMILE. Including cases with risk factors, counts were 70, 1,681 and 19 respectively. The authors estimated annual global procedure volumes of 283,920 PRK, 1,608,880 LASIK and 96,750 SMILE; dividing reports by these estimated denominators produced approximate rates of 20, 90 and 11 per 100,000 for cases without identified risk factors. [7]

The review cautions that these are not robust, directly comparable incidence rates: case reporting, procedure volumes, follow-up and denominators vary, and under-reporting may occur. SMILE also had a shorter history and far fewer reported cases. The figures do not establish that one procedure is universally safer or validate LASIK-specific scores for other techniques. They describe the limits of available case-based evidence.

Using screening information in a clinical discussion

Procedure choice and suitability require an individual assessment by an appropriately qualified eye-care professional. This article summarises research and guideline material; it cannot assess an individual or provide personal medical advice. If you are planning a consultation, you can review general LASIK and PRK prices in Turkey and the described LASIK and PRK laser eye surgery package as planning information; these pages do not establish suitability or clinical recommendation. Just Clinic Istanbul is an independent marketing and lead-generation platform, not a healthcare provider. Enquiries may be shared with authorised healthcare partners only with your consent.

Questions to ask at a screening appointment

• Which maps and measurements were used, and what pattern mattered?
• Could contact-lens wear or ocular-surface issues affect a reading, and should it be repeated?
• Which evidence applies to the proposed procedure?
• If findings are borderline, what alternatives or further assessment will be discussed?
• What follow-up and contact arrangements apply if vision changes?

References

1. Jacobs DS, Lee JK, Shen TT, Afshari NA, Bishop RJ, Keenan JD, Vitale S Refractive Surgery Preferred Practice Pattern®. Ophthalmology. 2023;130(3):P61–P135. Published online 19 December 2022. doi: 10.1016/j.ophtha.2022.10.032. PMID: 36543604.
2. Gomes JAP, Tan D, Rapuano CJ, Belin MW, Ambrósio R Jr, Güell JL, Malecaze F, Nishida K, Sangwan V Global consensus on keratoconus and ectatic diseases. Cornea. 2015;34(4):359–369. doi: 10.1097/ICO.0000000000000408. PMID: 25738235.
3. Randleman JB, Woodward M, Lynn MJ, Stulting RD. Risk assessment for ectasia after corneal refractive surgery. Ophthalmology. 2008;115(1):37–50.e4.e4. doi: 10.1016/j.ophtha.2007.03.073. PMID: 17624434.
4. Randleman JB, Trattler WB, Stulting RD. Validation of the Ectasia Risk Score System for preoperative laser in situ keratomileusis screening. American Journal of Ophthalmology. 2008;145(5):813–818. doi: 10.1016/j.ajo.2007.12.033. PMID: 18328998. PMCID: PMC3748728.
5. Santhiago MR, Smadja D, Gomes BF, Mello GR, Monteiro MLR, Wilson SE, Randleman JB Association between the percent tissue altered and post-laser in situ keratomileusis ectasia in eyes with normal preoperative topography. American Journal of Ophthalmology. 2014;158(1):87–95.e1. doi: 10.1016/j.ajo.2014.04.002. PMID: 24727263.
6. Ferreira-Mendes J, Lopes BT, Faria-Correia F, Salomão MQ, Rodrigues-Barros S, Ambrósio R Jr Enhanced ectasia detection using corneal tomography and biomechanics. American Journal of Ophthalmology. 2019;197:7–16. doi: 10.1016/j.ajo.2018.08.054. PMID: 30201341.
7. Moshirfar M, Tukan AN, Bundogji N, Liu HY, McCabe SE, Ronquillo YC, Hoopes PC Ectasia after corneal refractive surgery: a systematic review. Ophthalmology and Therapy. 2021;10(4):753–776. doi: 10.1007/s40123-021-00383-w. PMID: 34417707. PMCID: PMC8589911.
8. Gomes JAP, Hafezi F, Ambrósio R Jr, Henriquez MA, Campos M, Kinoshita S, Nkurikiye J, Koh S, Watson S, Mazzotta C, Torres-Netto E, Belin MW, Güell JL, Izquierdo L, Mahjoub S, Nubile M, Sharma N, Sarnicola V, Scorsetti DH, Mehta JS, Vinciguerra R, Cochener-Lamard B, Fogla R, Lee WB, Rapuano CJ; Panelists of the Global Consensus on Keratoconus and Ectatic Diseases—Edition 2; Hillen M; Intrials Scientific Team. Global Consensus on Keratoconus and Ectatic Diseases—Edition 2. Cornea. 2026;45(7):888–908. doi: 10.1097/ICO.0000000000004170. PMID: 42228627. Published 1 July 2026.