HIGHLIGHTS
- •
Younger gestation-adjusted age better predicts glaucoma outcome.
- •
Younger chronological age better predicts visual acuity.
- •
Age at the time of cataract surgery is a risk factor for glaucoma and visual acuity.
OBJECTIVE
To determine the association among age, gestational age, glaucoma-related adverse events, and visual acuity following unilateral cataract surgery.
DESIGN
Post hoc follow-up of a multicenter randomized clinical trial.
METHODS
An analysis of 112 infants who had unilateral congenital cataract surgery between 1-6 months of chronological age. Data on long-term glaucoma-related status and visual acuity were collected when children were 10.5 years old. Intergroup comparison and correlation tests were performed, and associations of adjusted and chronological age with glaucoma and visual acuity at 10.5 years of age were assessed by logistic regression.
RESULTS
Chronological and adjusted age were highly correlated (Pearson correlation, r = 0.98, P < .001). Gestational age was not significantly associated with glaucoma ( P = .08) nor visual acuity ( P = .6). Younger adjusted age had a significant association with glaucoma ( P = .04) and with better visual acuity ( P = .03). Younger chronological age was not significantly associated with glaucoma ( P = .06) but had a significant association with better visual acuity ( P = .02).
CONCLUSIONS
Age adjusted for gestation and chronological age are important risk factors for glaucoma and visual acuity in patients who have undergone cataract removal in infancy. Among the IATS participants, age adjusted for gestation significantly predicted glaucoma-related adverse events at 10.5 years. Conversely, chronological age was significantly associated with visual acuity at 10.5 years. Adjusted age should be considered in addition to chronological age when determining age of cataract surgery to minimize the risk of glaucoma and maximize visual acuity.
G laucoma is the most frequent vision- threatening adverse event following congenital cataract surgery, with a reported incidence ranging from 6.6% to 50% and increased incidence over longer follow-up duration. Published incidence rates depend on inclusion criteria of the study. For example, inclusion of infants having surgery in the first 4 weeks of life or after 6 months of life have vastly different risk profiles for glaucoma. The Infant Aphakia Treatment Study (IATS), a multicenter, randomized, controlled clinical trial, compared the outcomes of unilateral cataract surgery either with or without an intraocular lens implant in infants between 1 and 6 months of age. The IATS reported a 10.5-year glaucoma incidence of 20%.
Magnusson and associates published the early discovery of increased glaucoma risk following congenital cataract surgery. The IATS and other studies further reinforced our understanding that younger age at the time of surgery is a key risk factor for glaucoma-related adverse events (including glaucoma and glaucoma suspect). , Younger age at surgery is also associated with better visual acuity, which complicates clinician decisions about the optimal age for unilateral congenital cataract surgery. , General consensus suggests that chronological age of <3-4 weeks presents unacceptable glaucoma risk, whereas surgery at an age >7-9 weeks will sacrifice visual acuity. , The association of gestational age with glaucoma-related adverse events in patients who have undergone cataract removal in infancy remains unknown.
In this study, we investigated the association between younger gestational age and the development of glaucoma-related adverse events and better visual acuity in IATS participants at 10.5 years. We tested our a priori hypothesis that younger age, when adjusted for gestation, is more strongly associated with (1) increased risk of glaucoma-related adverse event (including glaucoma or glaucoma suspect) at 10.5 years and (2) better visual acuity at 10.5 years, compared with chronological age.
METHODS
The IATS study design, criteria, surgical technique, vision, and glaucoma evaluation protocols have been previously reported. , , Definitions of glaucoma, glaucoma suspect, and glaucoma-related adverse event were detailed in previous IATS publications and are available in the Supplementary Table. Gestational age refers to the duration of pregnancy measured from the date of last menstrual period to the date of birth. Estimates for gestational age are refined by early ultrasonographic measurements, when available. The IATS was approved by each institutional review board (IRB) at its 12 institutional sites and written informed consent was obtained for each participant. This study’s post hoc analysis protocol was reviewed by the University of Maryland IRB and determined to be exempt. The IATS is registered at ClinicalTrials.gov, identifier: NCT00212134.
Among the initial cohort of 114 participants in the IATS, 112 subjects had gestational age data available. Subjects were 36-42 weeks’ gestational age by IATS inclusion criteria. Gestational age was documented in initial enrollment form by the site investigator from medical record review or parent report. Gestational age was defined as the number of completed weeks since conception as estimated by the obstetrician.
For determination of age adjustment, the number of weeks relative to 40-week term birth was calculated (eg, –4 weeks for 36 weeks’ gestational age, 0 weeks for 40 weeks’ gestational age, and +2 weeks for 42 weeks’ gestational age). The age adjustment was then converted from weeks to days and added to the chronological age in days to determine the adjusted age. For example, a subject born at 36 weeks who underwent cataract surgery at a chronological age of 6 weeks and 4 days would have and an age adjustment of –4 weeks = –28 days, a chronological age at surgery of 46 days, and an adjusted age at surgery of 46 – 28 = 18 days. Of note, the contact lens arm and intraocular lens arm of the IATS were identical with regard to glaucoma and age at surgery, which allowed all study subjects to be included and analyzed together.
Visual acuity and glaucoma-related adverse events (glaucoma + glaucoma suspects) were previously collected as part of the IATS 1-, 4.5-, and 10.5-year follow-up visit. Visual acuity was evaluated at the 10.5-year visit using the electronic Early Treatment Diabetic Retinopathy Study (e-ETDRS) protocol. Visual acuities were converted to logMAR, with extrapolation of low vision acuities to stepwise logMAR values greater than 1.5 for statistical analysis (logMAR 2.64 for hand motion, 2.78 for light perception, and 2.93 for no light perception). Missing data for gestational age (n=2) and visual acuity (n=9) was considered random and was excluded.
Primary outcomes of interest, selected prior to analysis, were glaucoma-related adverse event at 10.5 years and visual acuity. Glaucoma-related adverse event was selected, rather than glaucoma, because of the high conversion rate of glaucoma suspect to glaucoma in the 1- to 4.5-year interval and the 4.5- to 10.5-year interval of the IATS. Secondary outcomes included all definitions of glaucoma at all time points and time to glaucoma.
STATISTICAL ANALYSIS
The independent variables evaluated were gestational age, gestational age less than 38 weeks, adjusted age at the time of cataract surgery, and chronological age at the time of cataract surgery. The dependent variables were glaucoma-related adverse event (glaucoma or glaucoma suspect diagnosis) at 10.5 years, time to first glaucoma event (glaucoma or glaucoma suspect diagnosis), logMAR visual acuity, vision better than 20/40 (logMAR <0.3), and vision better than 20/100 (logMAR <0.7) in the treated eye at 10.5 years. Normally distributed univariate analysis was conducted using Student t test. Nonparametric tests of significance were conducted in univariate analysis with Wilcoxon test. Multivariable logistic and linear regression were used to identify coefficients and strength of association.
Models included sex, race, chronological age, adjusted age, glaucoma status, and visual acuity. Sex was categorized as male or female as designated at birth. Nonsignificant variables were then eliminated from the model to create a parsimonious model. Colinear variables were also eliminated from the model. All statistical analyses were conducted using SAS, version 9.4 (SAS Institute Inc). In all analysis, 2-sided P <.05 was considered statistically significant. Exact p -values and odds ratios were reported as a measure of strength of association for models with multiple variables demonstrating P <.05. Data were analyzed from September 2023 to September 2024.
RESULTS
Patient characteristics in the IATS have been previously published. Of the 114 participants in the original study, 2 participants were excluded in the adverse glaucoma events analysis because of missing gestational age data (n=112). Nine participants were excluded in the visual acuity analysis because of missing visual acuity data (n=103) ( Figure 1 ).
The average chronological age at surgery was 10.7 ± 7.0 weeks with the range of age between 4 and 30 weeks. The average gestational age was 38.9 ± 1.3 weeks with the range of age between 36 and 42 weeks. After the gestational age was accounted for, the mean age adjustment was –8 days (±9 days). The mean adjusted age was less than chronological (9.5 ± 7.3 weeks vs 10.7 ± 7.0 weeks). Chronological and adjusted age had a strong positive correlation (Pearson r = 0.98, P < .001; Kendall r =0.83, P < .001) ( Table 1 ).
| Mean ± SD | Range | |
|---|---|---|
| Chronological age at surgery, wk | 10.7 ± 7.0 | 4-30 |
| Adjusted age at surgery, wk | 9.5 ± 7.3 | 9-29 |
SURGICAL TIMING AND GESTATIONAL AGE
There was no significant correlation between gestational age and chronological age at time of surgery ( r = 0.2, P = .1). Surgeons performed surgery earlier in the younger gestational age subgroup (36-39 weeks’ gestation) at 70 days, compared to at 87 days for subjects born late (41-42 weeks’ gestation); however, this difference was not significant ( P = .1).
GLAUCOMA OUTCOME AT 10.5 YEARS
At 10.5 years, 46 of 112 subjects had glaucoma or glaucoma suspect status in the operated eye, or 41% of the cohort; 25 subjects (22%) had glaucoma; 21 subjects (19%) were glaucoma suspects; 66 subjects (59%) did not have glaucoma. The mean gestational age was lower at 38.6 ± 1.2 weeks, in the glaucoma group, compared with 39.0 ± 1.3 weeks in the no glaucoma group, but the difference was small and not significant ( P = .07). Both measures of younger age, chronological and adjusted, were significantly associated with glaucoma ( P = .01 and P = .005, respectively). We found younger mean chronological and adjusted age at the time of surgery among glaucoma eyes compared with eyes that did not develop glaucoma ( Table 2 ).
| Glaucoma or Glaucoma Suspect at 10 y | No Glaucoma nor Glaucoma Suspect at 10 y | P Value | |
|---|---|---|---|
| Gestational age, wk * | 38.6 ± 1.2 | 39.0 ± 1.3 | .07 * |
| Chronological age at surgery, wk | 9.2 ± 6.7 | 11.7 ± 7.0 | .010 ⁎⁎ |
| Adjusted age at surgery, wk | 7.7 ± 7.0 | 10.7 ± 7.3 | .005 ⁎⁎ |
| Vision, logMAR | 0.854 ± 0.522 | 0.850 ± 0.518 | .90 ⁎⁎ |
⁎ Normally distributed, Student t test
The final logistic regression model included only adjusted age and glaucoma outcome. We found that chronological age and corneal diameter were colinear with adjusted age. Both chronological age and corneal diameter had relatively weaker association with glaucoma outcome and thus were removed from the model. We found no association of sex ( P = .6) and race ( P = .3) with glaucoma event at 10 years. Logistic regression model coefficient demonstrates that each week surgery is delayed after birth, accounting for gestation, decreases the 10.5-year risk of glaucoma by 6%. Results were similar for glaucoma outcomes at 1 and 5 years compared with 10 years. Survival analysis was significant for the association between adjusted age and time to glaucoma event ( P = .028) and between chronological age and time to glaucoma event ( P = .048).
VISUAL ACUITY
Mean vision was logMAR 0.85 ± 0.52, range 0-1.7. In addition, visual acuity was not associated with gestational age ( P = .6) ( Table 2 ).
GLAUCOMA AND VISUAL ACUITY AT 10.5 YEARS
At 10.5 years, subjects with lower chronological age less than 48 days at the time of surgery were 4.71 (95% CI 2.02, 11.00; P = .0003) times more likely to have vision better than 20/100 and 3.31 (95% CI 1.50, 7.28; P = .04) times more likely to have glaucoma compared with subjects >48 days. Subjects with adjusted age less than 48 days at the time of surgery were 2.57 (95% CI 1.14, 5.78; P = .02) times more likely to have a vision better than 20/100 and 3.28 (95% CI 1.42, 7.6; P = .02) times more likely to have glaucoma compared with subjects >48 days’ adjusted age at the time of surgery ( Table 3 ). Improved visual acuity and earlier onset of glaucoma were both associated with younger chronological age and younger adjusted age at surgery in our evaluation of vision and glaucoma as continuous variables (logMAR vision and age at diagnosis of glaucoma or glaucoma suspect) ( Table 4 ).
