Highlights
-
•
Visual outcomes for deprivation amblyopia remain, on average, disappointing.
-
•
Most of these children also need treatment for strabismus and about one-third for ocular hypertension/glaucoma.
-
•
An average of 5 subsequent surgeries is required per child after congenital cataract removal.
PURPOSE
An aspiration of modern medicine is for disease treatment to improve progressively. Here we examine outcomes of treatment for monocular deprivation amblyopia, caused by a congenital cataract, at an academic medical center over several decades. The question we addressed was whether outcomes have improved substantially, or whether they have remained stubbornly disappointing.
DESIGN
Retrospective, trend study.
METHODS
Patients with deprivation amblyopia due to primary monocular congenital cataract removed before 1 year of age were studied. Preoperative and outcome data were collated from sequential examinations to document corrected distance visual acuity (CDVA), binocular visuomotor behaviors, anterior segment and vitreo-retinal health, adherence to therapies, demographic variables, and the need for subsequent surgical interventions. Biometry measurements were recorded at the time of initial cataract surgery. The primary outcome measure was CDVA. Additional outcomes measures were prevalence of nystagmus, strabismus, ocular hypertension (OHT)/glaucoma, and number of ocular surgeries.
RESULTS
A total of 80 children met inclusion criteria, with a median follow-up age of 12 years. Median CDVA of amblyopic eyes at last follow-up was 1.40 logMAR or 20/500 (interquartile range = 0.75-CF). There were no significant differences in CDVA for patients with cataract extraction between 1990 and 1999, 2000 and 2009, and 2010 and 2022 ( P =.25), with medians of 1.00, 1.44, and 1.40 logMAR, respectively. Better CDVA was related systematically to earlier age at surgery, but the correlation was weak ( P =.06). In all, 96% of the children (77 of 80) had secondary implantation of an intraocular lens. Strabismus developed in 98% of the children (78 of 80), and fusion maldevelopment nystagmus in 73% (58 of 80). A total of 43% (34 of 80) required subsequent surgery for removal of a secondary cataract, and 30% (24 of 80) required surgery to treat aphakic or pseudophakic OHT/glaucoma. The median number of total ophthalmic surgeries was 5 (IQR = 3-6).
CONCLUSIONS
Visual acuity outcomes for monocular deprivation amblyopia remain disappointing. Adherence to patching therapy and contact lens wear is burdensome and often is not achieved. Children with the condition require an average of 4 subsequent surgeries in the first decade of life for intraocular lens implantation, removal of a secondary cataract, strabismus, or glaucoma.
A mblyopia is subnormal acuity caused by early abnormal visual experience. Monocular deprivation is the most severe subtype of amblyopia, with unilateral congenital cataract a common cause. , Over the last several decades, pediatric ophthalmologists have described individual cases or subgroups of children who have been rescued from congenital cataract–induced deprivation amblyopia by treatment initiated in early infancy. In 1981, Beller et al reported 5 children whom they treated who achieved visual acuities (VA) of 20/30 or better, and another 3 children 20/80 or better. Birch and Stager (1988) reported 10 children (of 19 treated) with a VA outcome of 20/80 or better. In 2014, the multicenter, randomized, prospective, National Institutes of Health–sponsored Infant Aphakia Treatment Study (IATS) reported that 19 of the 112 children enrolled achieved a VA of better than 20/40.
These reports verify that some children treated for deprivation amblyopia can emerge with good or reasonably good VA. The question is this: are these children the exception or the rule when predicting VA outcomes? Likewise, have these favorable outcomes become more frequent over the years, as the body of literature demonstrating the benefits of early surgery has grown and techniques for supporting patching compliance have expanded? A cautionary note was sounded by the average outcome reported by the IATS. Despite timely surgery and state-of-the-art closely monitored postsurgical optical and amblyopia therapy, the average corrected distance visual acuity (CDVA) achieved for the 112 children in the IATS was 0.9 logMAR (20/160). The purpose of this study is to re-examine the issue of average visual outcomes in a representative population of children treated for unilateral congenital cataract to determine whether they have improved over time or have remained stagnant.
METHODS
PATIENT INCLUSION AND EXCLUSION CRITERIA
In this trend study, patient records were identified for perusal via a search of Current Procedural Terminology (CPT) procedural codes and International Classification of Diseases, Ninth Revision/Tenth Revision ( ICD-9/10 ) diagnostic codes. The code list culled children with encounters at St. Louis Children’s Hospital between January 2000 and May 2024 associated with billing codes for a unilateral congenital cataract or associated complications. The final set of patients included those with surgery dates ranging from April 1990 to May 2022. Retrospective chart review was conducted to verify that the child met the following inclusion criteria: (1) observation of an optical axis opacity by the parent, pediatrician, or examiner in the first 3 months of life; (2) diagnosis of monocular deprivation amblyopia after clinical examination by an eye care provider; (3) cataract surgery prior to 12 months of age; (4) postoperative initiation of patching therapy; and (5) follow-up of at least 2 years after surgery. Lensectomy, anterior vitrectomy, and posterior capsulectomy were performed. Infants were left aphakic and corrected optically using a soft contact lens. After 1 year of age, they were treated by implantation of a posterior chamber intraocular lens (IOL). Patients were excluded if a cataract could be attributed to any of the following: congenital glaucoma; infantile globe trauma or non-accidental trauma; intraocular hemorrhage; uveitis; congenital infection; retinopathy of prematurity; chronic corticosteroid administration; or radiation treatment. Amblyopia was defined as CDVA 2 lines worse than age norms (eg, worse than 20/32 at age 5 years or older). The study protocol complied with the Association for Research in Vision and Ophthalmology (ARVO) resolution on the use of human subjects in research and was approved by the Washington University Human Research Protection Office (IRB #202406158) prior to the start of the project.
OPHTHALMIC AND DEMOGRAPHIC FEATURES
Office examination measures were supplemented by examination under anesthesia (EUA) at the time of the primary cataract procedure and any subsequent surgeries. The examinations included the following: age- and developmentally appropriate testing (ETDRS, HOTV, Allen figure acuity, Cardiff figure acuity, or Teller grating acuity) of corrected distance visual acuity (CDVA) in each eye ; pupillary examination for afferent defects and diameter; sensorimotor examination of eye alignment/eye movement/binocular function; and manual and, when feasible, automated cycloplegic refractions. Additional measurements obtained under anesthesia, before the surgical procedure, included the following: intraocular pressure, pachymetry, keratometry, corneal diameter, slitlamp biomicroscope evaluation of the anterior segment, A-scan ultrasonographic axial length, B-scan echography, anterior chamber depth, lens thickness measurement, indirect ophthalmoscopy using scleral depression and, when indicated, Retcam digital imaging of the fundi. Annotation was also made of sex (as assigned at birth), race, ethnicity, and insurance type.
ADHERENCE TO THERAPY
For each patient, adherence to patching therapy and corrective lens wear was rated as good, fair, or poor based on queries of the child’s caregiver annotated in the medical record. Patching therapy was recommended through ages 1 to 3 years. Because VA tended to plateau by age 3 years or non-adherence by this age made further recommendations futile, adherence was not quantified further. Good adherence was defined as an estimated time of 75% to 100% of that prescribed, fair as 25% to 75%, and poor as 0% to 25%.
VISUAL ACUITY RANKING
Optotype and grating VAs were converted to logMAR units for analysis. Because 19 patients had acuities too poor to quantify with a numeric acuity, nonparametric ranks of acuity were used for analysis. In the ranking system used, NLP acuities were given the poorest rank, followed by LP, then HM, then CF, and finally all logMAR acuities in descending order of their value. Another 3 patients had visual acuities charted using the fix–follow–maintain system due to an inability to cooperate with quantitative VA testing because of young age or neurodevelopmental disability. Because of the difficulty of accurately placing these acuities on a ranked scale, these 3 patients were excluded from visual acuity analyses.
STATISTICAL ANALYSIS
Differences in visual acuity were analyzed via a nonparametric Mann–Whitney test, or a Kruskal–Wallis test in cases where more than 2 categories were compared. When a Kruskal–Wallis test indicated a significant difference among 3 or more categories, a post hoc Dunn z test was conducted with adjustment for multiple comparisons. A nonparametric Spearman rank order correlation was used to determine the correlation between age at surgery and visual acuity. t Tests were conducted in a paired fashion for comparison of EUA measurements between amblyopic and fellow eyes. Significance was defined as P <.05. Statistical analysis was conducted using SPSS Statistics (IBM, Inc), Microsoft Excel (Microsoft, Inc), and Prism (GraphPad Software, LLC). Values reported wherever appropriate are median and interquartile range (IQR).
RESULTS
CLINICAL AND DEMOGRAPHIC CHARACTERISTICS
Of the pool of children (634 in total) whose records were culled, 80 met all inclusion/exclusion criteria. Table 1 describes their baseline demographic and clinical characteristics. The cohort was majority male (54%), White (88%), and privately insured (81%). The median age at initial cataract surgery was 2 months (IQR = 1-4), and the median age at last follow-up was 12 years (IQR = 7-18; mean 13). The median delay between initial presentation and cataract extraction was 15 days (IQR = 6-29) days. Cataract surgery was accomplished by 1 month of age in 25% of the infants, by 3 months of age in over 50%, and by 5 months of age in over 75%.
TABLE 1
Characteristics of Studied Patients.
| Characteristics | Patients | % of Patients |
|---|---|---|
| Sex, n | ||
| Male | 43 | 54 |
| Female | 37 | 46 |
| Race, n | ||
| White | 70 | 88 |
| Black | 7 | 9 |
| Asian | 1 | 1 |
| Unknown/other | 2 | 3 |
| Insurance type, n | ||
| Medicaid | 14 | 18 |
| Private | 65 | 81 |
| Uninsured/other | 1 | 1 |
| Neurodevelopmental disorder, n | 6 | 8 |
| Age at surgery, mo, median (IQR) | 2 (1-4) | N/A |
| Age at last follow-up, y, median (IQR) | 12 (7-18) | N/A |
IQR = interquartile range (reported as 25th percentile to 75th percentile); N/A = not applicable.
VISUAL ACUITY OUTCOMES OVER THE LAST 3 DECADES
The median CDVA of amblyopic eyes at last follow-up was 1.40 logMAR, or 20/500 (IQR = 0.75-CF; n = 77 eyes). Median acuities were 1.00, 1.44, and 1.40 logMAR for patients with cataract extraction from 1990 to 1999, 2000 to 2009, and 2010 to 2022, respectively ( Figure 1 ). A Kruskal–Wallis test showed no significant differences in CDVA at last follow-up between these surgical era groups ( P =.25).
Box and whisker plot of visual acuity (VA) of patients, sorted by era of cataract surgery. No significant difference by Kruskal–Wallis testing ( P =.25). CF = count fingers; HM = hand motion; LP = light perception; NLP = no light perception.
Median CDVA for children operated on at ≤2 months of age was 1.18 logMAR, compared to 1.48 logMAR for those operated on after 2 months. When separated into 4 groups based on age at initial cataract surgery ( Figure 2 , A), Kruskal–Wallis testing indicated a significant difference in median VA between groups ( P =.03). Post hoc testing revealed a significant difference in CDVA only between those with surgery between 0 and 2 months (median CDVA = 1.18 logMAR) vs 2 to 4 months (median CDVA = 1.48 logMAR). When graphed on a scatterplot and analyzed via Spearman rank order correlation ( Figure 2 , B), there was a nonsignificant trend associating better CDVA at last follow-up with younger age at cataract surgery; however, the correlation was nonsignificant ( r s = 0.22; P =.06). Marked variation in CDVA outcome was apparent even among patients who had surgery at a similar age.
A. Box and whisker plot of visual acuity (VA) of patients, sorted by age at cataract surgery. Median VAs were significantly different ( P =.03; Kruskal–Wallis test). Post hoc Dunn z -test results indicated with an asterisk (* P <.05). B. Scatterplot of age at cataract surgery vs VA at last follow-up. Spearman rank correlation coefficient indicated as r s .
BASELINE KERATOMETRIC AND BIOMETRIC FEATURES
Exam under anesthesia (EUA) measurements (consisting of axial length, corneal diameter, average corneal curvature, and corneal thickness) were recorded at the time of cataract extraction and compared between the cataractous eye and the normal, fellow eye. As shown in Table 2 , amblyopic eyes had shorter axial lengths ( P =.02), smaller corneas ( P <.001), and steeper corneal curvature ( P =.004). There were no significant differences in corneal thickness between the eyes with cataract and fellow eyes ( P =.24). These findings demonstrate that our study cohort was representative of baseline measures reported in previous major studies of monocular congenital cataract. ,
TABLE 2
Baseline Examination Under Anesthesia Meaurements.
| Eyes Undergoing Lensectomy | Fellow Eyes | P Value a | No. of Patients | |||
|---|---|---|---|---|---|---|
| Mean | SEM | Mean | SEM | |||
| Axial length, mm | 18.04 | 0.39 | 18.36 | 0.34 | .02 | 49 |
| Corneal diameter, mm | 10.51 | 0.12 | 10.85 | 0.10 | <.001 | 55 |
| Corneal power, D | 47.40 | 0.60 | 45.61 | 0.35 | .004 | 33 |
| Corneal thickness, µm | 576 | 14 | 560 | 8 | .24 | 37 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
