Purpose: To determine the range of biometry prediction error of the postoperative refraction and the visual outcome after cataract surgery.
Setting: Postgraduate teaching hospital.
Methods: Refraction and visual outcomes were recorded prospectively for a consecutive case series of 1817 cataract procedures. Complete preoperative and postoperative data were available for 1676 eyes (92.4%). The paired mean error of the difference between the predicted and achieved refraction, the percentages falling within +/-0.50 diopter (D), +/-1.00 D, and +/-2.00 D, and the values that contained 99% of data points were then calculated. The number that achieved a final Snellen visual acuity of 6/12 was determined in all eyes and in eyes without preoperative ocular comorbidity. This analysis was repeated in 1265 eyes that received a single intraocular lens (IOL) design (C10UB/C11UB) after phacoemulsification through a clear corneal incision.
Results: The paired mean error of the difference between the predicted and achieved postoperative refraction in all eyes was -0.32 D +/- 1.05 (SD); 72.3% of eyes were within +/-1.00 D of the planned refraction, with bounds of 99% of observation of +2.92 D and -3.98 D. In the single IOL group, the paired mean error was -0.32 +/- 0.94 D; 74.2% of eyes fell within +/-1.00 D of the planned refraction, with bounds of 99% of observation of +2.54 D and -3.81 D. Axial length did not determine the magnitude of the biometry prediction error. At discharge from the service, 86.9% of all eyes and 95.4% of eyes without preoperative ocular comorbidity achieved a final corrected Snellen acuity of 6/12.
Conclusion: These data provide benchmark information that can be used to monitor clinical practice.