Visual Acuity and Ophthalmic Outcomes 5 Years After Cataract Surgery Among Children Younger Than 13 Years

Abstract

Importance: Cataract is an important cause of visual impairment in children. Data from a large pediatric cataract surgery registry can provide real-world estimates of visual outcomes and the 5-year cumulative incidence of adverse events.

Objective: To assess visual acuity (VA), incidence of complications and additional eye operations, and refractive error outcomes 5 years after pediatric lensectomy among children younger than 13 years.

Design, setting, and participants: This prospective cohort study used data from the Pediatric Eye Disease Investigator Group clinical research registry. From June 2012 to July 2015, 61 eye care practices in the US, Canada, and the UK enrolled children from birth to less than 13 years of age who had undergone lensectomy for any reason during the preceding 45 days. Data were collected from medical record reviews annually thereafter for 5 years until September 28, 2020.

Exposures: Lensectomy with or without implantation of an intraocular lens (IOL).

Main outcomes and measures: Best-corrected VA and refractive error were measured from 4 to 6 years after the initial lensectomy. Cox proportional hazards regression was used to assess the 5-year incidence of glaucoma or glaucoma suspect and additional eye operations. Factors were evaluated separately for unilateral and bilateral aphakia and pseudophakia.

Results: A total of 994 children (1268 eyes) undergoing bilateral or unilateral lensectomy were included (504 [51%] male; median age, 3.6 years; range, 2 weeks to 12.9 years). Five years after the initial lensectomy, the median VA among 701 eyes with available VA data (55%) was 20/63 (range, 20/40 to 20/100) in 182 of 316 bilateral aphakic eyes (58%), 20/32 (range, 20/25 to 20/50) in 209 of 386 bilateral pseudophakic eyes (54%), 20/200 (range, 20/50 to 20/618) in 124 of 202 unilateral aphakic eyes (61%), and 20/65 (range, 20/32 to 20/230) in 186 of 364 unilateral pseudophakic eyes (51%). The 5-year cumulative incidence of glaucoma or glaucoma suspect was 46% (95% CI, 28%-59%) in participants with bilateral aphakia, 7% (95% CI, 1%-12%) in those with bilateral pseudophakia, 25% (95% CI, 15%-34%) in those with unilateral aphakia, and 17% (95% CI, 5%-28%) in those with unilateral pseudophakia. The most common additional eye surgery was clearing the visual axis, with a 5-year cumulative incidence of 13% (95% CI, 8%-17%) in participants with bilateral aphakia, 33% (95% CI, 26%-39%) in those with bilateral pseudophakia, 11% (95% CI, 6%-15%) in those with unilateral aphakia, and 34% (95% CI, 28%-39%) in those with unilateral pseudophakia. The median 5-year change in spherical equivalent refractive error was -8.38 D (IQR, -11.38 D to -2.75 D) among 89 bilateral aphakic eyes, -1.63 D (IQR, -3.13 D to -0.25 D) among 130 bilateral pseudophakic eyes, -10.75 D (IQR, -20.50 D to -4.50 D) among 43 unilateral aphakic eyes, and -1.94 D (IQR, -3.25 D to -0.69 D) among 112 unilateral pseudophakic eyes.

Conclusions and relevance: In this cohort study, development of glaucoma or glaucoma suspect was common in children 5 years after lensectomy. Myopic shift was modest during the 5 years after placement of an intraocular lens, which should be factored into implant power selection. These results support frequent monitoring after pediatric cataract surgery to detect glaucoma, visual axis obscuration causing reduced vision, and refractive error.

Figure 1.. Visual Acuity 5 Years After Lensectomy by Laterality of Surgery and Primary Intraocular Lens Implantation

Monocular visual acuity was converted to logMAR. Three ordinal categories were defined for visual acuities beyond the range of the testing method, and each was assigned an incremental logMAR value, starting 0.1 logMAR increment worse than 20/800: (1) worse than 20/800 = 1.7 logMAR, (2) counting fingers = 1.8 logMAR (20/1262), and (3) hand motions, light perception, no light perception, or prosthesis = 1.9 logMAR (20/1589). The top of each box represents the 75th percentile of the data and the bottom, the 25th percentile. Horizontal lines indicate group medians; filled circles, group means; and open circles, statistical outliers. The bars extending above and below each box represent 1.5 times the IQR or the maximum and minimum observed values within the range if not as extreme as the calculated value.

Figure 2.. Cumulative Incidence of Glaucoma-Related Adverse Events

A Cox proportional hazards regression model was used to estimate the 5-year cumulative incidence of glaucoma-related adverse events. The robust sandwich variance estimator was used to adjust for the correlation between eyes of participants who contributed data from both eyes to the analysis. The date of onset for glaucoma-related adverse events was not collected; thus, the date of the most recent office visit at which they were reported was used to estimate the 5-year cumulative incidence. Glaucoma-related adverse events reported up to 5.5 years after lensectomy were included in the analysis.

Figure 3.. Cumulative Incidence of Surgery to Clear the Visual Axis

A Cox proportional hazards regression model was used to estimate the 5-year cumulative incidence of surgery to clear the visual axis. The robust sandwich variance estimator was used to adjust for the correlation between outcomes collected for participants who underwent bilateral lensectomy by the last documented office visit. The actual date of surgery was known and was used to document the timing of the surgery.

Figure 4.. Refractive Error at 5 Years by Laterality of Surgery and Primary Intraocular Lens Implantation

The top of each box represents the 75th percentile of the data and the bottom, the 25th percentile. Horizontal lines indicate group medians; filled circles, group means; and open circles, statistical outliers. The bars extending above and below each box represent 1.5 times the IQR, or the maximum or minimum observed value within the range if not as extreme as the calculated value.