Journal of Cataract & Refractive Surgery
Published online April 20, 2018
In a large multicenter study, Lundström et al. documented risk factors for refractive error after cataract surgery. In addition to previously reported risk factors, they identified several new indicators, including poor preoperative visual acuity, corneal opacities, and surgical complications such as vitreous loss and capsular break.
The authors gathered data from consecutive cases of cataract extraction reported to the European Registry of Quality Outcomes for Cataract and Refractive Surgery (EUREQUO) in 2014 and 2015. All told, 100 clinics and 12 countries were represented. Collected information included demographics, preoperative corrected distance visual acuity (CDVA), target refraction, coexisting eye disease, previous eye surgery, type of surgery and any surgical difficulties, and type of intraocular lens (IOL) implanted.
Of the 548,392 reported cases, follow-up data were available for 282,811 (mean age of patients, 74 years). The absolute mean biometry prediction error was 0.42 D. The prediction error was within 1.0 D for 93% of eyes and within 0.50 D for 72%. Strong indicators of poor refractive outcome were target refraction (negative or absolute), poorer preoperative CDVA, coexisting eye disease, and surgical difficulty and complications. The odds ratios of refractive error in the presence of a surgical complication were 2.55, 5.57, and 13.8 for >0.5 D, >1.0 D, and >2.0 D, respectively.
The authors found that older age (>60 years) was associated with biometry prediction errors >0.5 D, while younger age was linked to prediction errors >2.0 D. There were no significant differences in refractive outcomes between men and women. The absolute mean biometry prediction error was 0.43 ± 0.55 D in 2014 and 0.41 ± 0.48 D in 2015 (p < .001).
The number of risk factors for refractive error is larger than expected. Results of this study may aid in updating evidence-based guidelines. The authors suggest lowering the absolute biometry prediction error from ≤0.6 D (as stated in 2012 guidelines based on the EUREQUO data) to ≤0.45 D, to more closely resemble their findings. They also propose increasing the benchmark percentage of error within 1.0 D from ≥87% (per the 2012 guidelines) to at least 90%. Moreover, the authors recommend that all risk factors be considered during preoperative planning, including selection of the most appropriate IOL.
The original article can be found here.