The five-domain framework
The original ERSS organizes five preoperative or planned-procedure domains: anterior topographic pattern, predicted residual stromal bed, age, preoperative corneal thickness, and manifest refractive spherical equivalent. The components are weighted rather than treated as equivalent observations.
Development and validation context
The system was derived from retrospective post-refractive-surgery ectasia cases and controls and was subsequently evaluated in a separate LASIK screening study. The original publications should be consulted for cohort definitions, scoring tables, operating characteristics, and exclusions.
What ERSS contributes
ERSS makes established clinical risk factors explicit and auditable. It also illustrates why a planned procedure cannot be assessed from corneal shape alone: tissue removal and flap-related geometry contribute information not contained in a preoperative topography map.
What ERSS cannot establish
ERSS is not a universal lifetime probability calculator. Its reported performance belongs to the studied cohorts and methodology. Modern tomography and biomechanics can reveal information that the original anterior-topography-era framework did not directly encode.
CER-AI ERSS component scoring
| Component | Current operational rule | Points |
|---|---|---|
| Topography | Normal/symmetric; asymmetric bow-tie; inferior steepening/SRAX; abnormal/ectatic | 0; 1; 3; 4 |
| LASIK RSB or PRK RST | >=300; 280 to <300; 260 to <280; 240 to <260; <240 µm | 0; 1; 2; 3; 4 |
| Age | >=21; 19 to <21; 18 to <19 years | 0; 2; 3 |
| Thinnest pachymetry | >=510; 500 to <510; 480 to <500 µm | 0; 1; 2 |
| Manifest MRSE | >=-8; <-8 to >=-10; <-10 to >=-12; <-12 to >=-14; <-14 D | 0; 1; 2; 3; 4 |
Age <18 is incomplete. Pachymetry <480 µm is not assigned an ERSS row score because it is an independent CER-AI STOP-DEFER. PRK uses RST for the tissue row; its independent RST hard stop is <310 µm.
From ERSS total to CER-AI pathway result
| ERSS total | CER-AI result | Report behavior |
|---|---|---|
| 0–2 | PASS | Five rows, total, topography category, and disposition are reported. |
| 3 | CAUTION | Requires explicit surgeon review; does not alone impose the global final result. |
| >=4 | STOP-DEFER | Becomes a stop driver in the final report. |
| Critical input missing | ASSESSMENT INCOMPLETE | A complete report token is not issued until the required field is resolved. |
CER-AI does not convert the ordinal total into an individual lifetime probability.
Selected sources
- Randleman et al., Ophthalmology 2008
- Randleman et al., American Journal of Ophthalmology 2008
- Chan et al., Clinical & Experimental Ophthalmology 2010
See the complete CER-AI medical reference registry for broader context.