Determining corneal power using Orbscan II videokeratography for intraocular lens calculation after excimer laser surgery for myopia

Purpose

To assess the accuracy of Orbscan II slit-scanning videokeratography for intraocular lens (IOL) calculation in eyes with previous photorefractive surgery for myopia.

Setting

Private practice, St. Louis, Missouri, USA.

Methods

Corneal power (K) was measured by manual keratometry, Placido-based videokeratography (Atlas), slit-scanning videokeratography (Orbscan II), and contact lens overrefraction in 21 post-photoablation eyes having cataract surgery. Postoperative data collected after phacoemulsification were used to back-calculate corneal power (BCK). The BCK values were statistically compared at 3.0 to 6.0 mm central Orbscan II curvature and power measurements, including total axial power, total tangential power, total mean power, and total optical power. Similar comparisons were made to Atlas curvature at the 0.0 to 10.0 mm zones.

Results

The mean corneal power after refractive surgery based on BCK values using the Holladay 2 formula (BCK H2) was 39.35 diopters (D) ± 2.58 (SD). The mean manual value (40.52 ± 1.95 D) and Atlas-based values were statistically higher than BCK H2 values (P<.001). The mean corneal power calculated from historical data was 39.33 ± 2.70 D (P = .83 to BCK H2; n = 19) and from contact lens overrefraction, 41.38 ± 3.11 D (P = .19; n = 5). Orbscan II parameters (n = 21) of the total mean power (3.0 mm, 39.10 ± 2.63 D), total tangential power (3.0 mm, 39.11 ± 2.60), total axial power (5.0 mm, 39.19 ± 2.55 D), and total optical power (3.0 mm, 39.08 ± 2.78 D; 4.0 mm, 39.39 ± 2.76 D) were statistically similar to both the historical and BCK H2 values (P>.11). If used prospectively, 80.9% of eyes would have been within ±0.50 D of the targeted refraction using a 4.0 mm total optical power, 76.2% using a 5.0 mm total axial power, and 42.1% using the historical method.

Conclusion

The Orbscan II 5.0 mm total axial power and 4.0 mm total optical power can be used to more accurately predict true corneal power than the history-based method and may be particularly useful when pre-LASIK data are unavailable.