Visual and motor outcomes following surgery in children with congenital cataracts: a study of a series of 143 cases Authors: Dilara Pirhan, Sevgi Subasi, Enes Kesim, Veysel Levent Karabas, Özgül Altıntaş Published: 2025 ABSTRACT BackgroundData on long-term visual and ocular motility outcomes after congenital cataract surgery remain limited. We retrospectively reviewed all surgical cases in our database to evaluate these outcomes and to identify prognostic factors using multivariate analysis.MethodsA retrospective analysis of patients who had congenital cataract surgery and secondary intraocular lens (IOL) implantation was conducted. Whenever possible, visual and motor outcomes were collected at each visit. In total, 233 eyes from 143 patients were included, with 90 bilateral and 53 unilateral cases. Statistical analyses included t-tests, Mann–Whitney U-tests, chi-square tests, and multivariate linear regression to evaluate associations and identify independent predictors of postoperative visual outcomes.ResultsPatients were followed for a median of 6 years (range 1–15 years). At the final visit, the mean best corrected visual acuity (BCVA) in logarithm of the minimum angle of resolution (logMAR) units was 0.22 logMAR (≈ 20/32; range 20/20–20/66) in bilateral cases and 1.0 logMAR (≈ 20/200; range 20/50 − 20/400) in unilateral cases. Strabismus was present before surgery in 21.3% of bilateral and 30% of unilateral patients, rising postoperatively to 45.3% and 62%. Nystagmus was initially seen in 14.6% of patients and remained postoperatively in 20% of bilateral and 5.7% of unilateral cases. Myopic shift (MS) occurred in 37.2% of eyes; 62.8% remained hyperopic. Spectacle/contact lens adherence was 86.9% in bilateral and 72.7% in unilateral cases. Eyes with normal interocular axial length difference (IALD, defined as < 0.5 mm) showed better BCVA than those with similar or different IALD (p = 0.004). Final BCVA was significantly associated with adherence to treatment, cataract laterality, presence of nystagmus, and amblyopia therapy compliance (F = 13.789, p < 0.001).ConclusionsThese results show that timely surgery alone is not enough to get good visual and motor outcomes in congenital cataracts. Adherence to visual rehabilitation, cataract laterality, and the presence of nystagmus were the three most critical things that affected the final visual acuity in our sample. It is very crucial to keep up with regular follow-ups and make sure that treatment is followed in unilateral cases because these people are more likely to get amblyopia and interocular imbalance. Early diagnosis, well-planned surgery, and long-term care that is tailored to each person’s needs are all necessary for the best results. Introduction Congenital cataracts are one of the most common causes of preventable childhood blindness, particularly in underdeveloped regions [1]. Although the true global incidence remains uncertain due to underreporting, estimates suggest that 20,000 to 40,000 infants are born with cataracts each year. Beyond visual loss, early-onset visual impairment may negatively impact neurodevelopment, mobility, and long-term access to education and healthcare services. Infants with congenital cataracts may have functional vision if the cataract is identified and treated early with surgery, optical rehabilitation, and amblyopia therapy. Postoperative follow-up, duration of amblyopia, and cataract severity also influence the final visual prognosis. Some problems may not appear until later in life, making continuous monitoring of the child’s vision and ocular development essential. Despite these challenges, advances in surgical techniques and postoperative care have improved outcomes, reduced unpredictability, and enhanced visual prospects in pediatric cataract surgery [2–4]. Early surgery is critical to minimize deprivational amblyopia, and postoperative rehabilitation with aphakic contact lenses, glasses, or intraocular lens (IOL) implantation remains essential [3]. However, there is limited data on the long-term visual and ocular motility outcomes in children with congenital cataracts [5–7]. Understanding these outcomes is important for evaluating current treatment strategies and guiding improvements. Only a few studies have assessed visual acuity alongside ocular motor performance while also considering multiple prognostic factors. To address this gap, long-term follow-up studies in large pediatric cohorts are needed. Therefore, the purpose of this study was to investigate the incidence of congenital cataracts and associated visual outcomes, including strabismus, nystagmus, myopic shift (MS), interocular axial length difference (IALD), and adherence to amblyopia treatment. Materials and methods The database for children with congenital cataracts treated at the Department of Ophthalmology, Kocaeli University Medical Faculty, was reviewed retrospectively from January 2004 to December 2020. Patients who dropped out of follow-up and had missing data were not included in the study. Missing data was defined as follows: patients with a follow-up period of less than 12 months, patients with missing preoperative records, or those with gaps of more than 12 months between follow-up visits due to incomplete follow-up or lost records. These cases were excluded from the analysis (n = 12). Baseline features of excluded patients could not be systematically compared due to incomplete records. The study included 233 eyes from 143 patients. For patients with bilateral cataracts, only the first operated eye was included in eye-based analyses to avoid intra-subject correlation. The first operated eye was determined based on cataract density and visual significance, with denser or more visually significant eyes typically operated first. Laterality (unilateral vs. bilateral) was also recorded for each patient and included as a categorical variable in multivariate analyses. The purpose of this study and the methods described adhered to the tenets of the Helsinki Declaration, and the study was approved by the local institutional review board (GOKAEK-2022/04.19 2022/63). The requirement for informed consent was waived in accordance with institutional and national ethical guidelines, as the study involved a retrospective review of anonymized patient records and posed minimal risk to participants. All patient data were coded and de-identified prior to analysis to ensure confidentiality and privacy. No identifiable information was stored or shared outside the research team. Before cataract surgery, all children were thoroughly examined. Slit-lamp biomicroscopy of the anterior segment, intraocular pressure (IOP) measurement, and cycloplegic refraction were performed whenever possible. IOP was measured using a handheld tonometer (Tono-Pen AVIA, Reichert Technologies, Buffalo, NY, USA). Preoperative axial length (AL) was measured with an ultrasonic A-scan (Echoscan US 800, Nidek Co., Gamagori, Japan). Postoperatively, axial length was measured with the same ultrasonic A-scan at follow-up visits when feasible. A dilated fundus examination was performed on each child if possible; if not, the examination was conducted under general anesthesia. B-scan ultrasonography (Echoscan US 4000, Nidek Co., Gamagori, Japan) was used to rule out any posterior segment abnormalities, and visual evoked potentials were used to assess the prognosis of children with cataract-obscured media. All operations were performed under general anesthesia by the same experienced surgeon using a limbal approach. A 2.4 mm limbal incision was created, and an infusion cannula was placed to maintain balanced salt solution in the anterior chamber throughout the procedure. An ophthalmic viscosurgical device (OVD, sodium hyaluronate 1.4%, Healon GV; Johnson & Johnson Surgical Vision Inc., Santa Ana, CA, USA) was injected into the anterior chamber to stabilize the chamber before performing the anterior continuous curvilinear capsulorhexis (ACCC). After completion of the ACCC, lens material was aspirated and the capsular bag was inflated with OVD. A posterior continuous curvilinear capsulorhexis (PCCC) of approximately 4 mm in diameter, slightly smaller than the ACCC, was then created. Limited central anterior vitrectomy was preferred in patients younger than 8 years of age; in children older than 8 years the posterior capsule was left intact. A 23-gauge cutter (Oertli® OS3, Berneck, Switzerland) was utilized to perform a limited anterior vitrectomy through the limbal route at 1,500 cuts/min and 150 mmHg aspiration. The goal was to maintain vitreous traction at a minimum. After thorough removal/exchange of viscoelastic material, the limbal incision was closed with a single 10 − 0 nylon suture. Primary IOL implantation was routinely performed in children older than 2 years with bilateral cataracts and in those older than 1 year with unilateral cataracts; ciliary sulcus placement was used when capsular support was insufficient. These procedures were routinely performed up to 8 years of age. For primary implantation, we considered both age and ocular parameters, including corneal diameter (≥ 10.5 mm), white-to-white distance (≥ 10 mm), normal intraocular pressure, adequate capsular support, and sufficient pupillary dilation. Eyes that did not meet these criteria were left aphakic and corrected optically until secondary IOL implantation became possible. Secondary implantation was typically performed after 4 years of age, once ocular growth and biometric parameters stabilized. Preoperative assessment included measurement of corneal diameter, white-to-white distance, and axial length, and confirmation of adequate capsular support, normal intraocular pressure, and a clear cornea. All implanted lenses were three-piece hydrophobic acrylic IOLs of the Sensar AR40e model (Johnson & Johnson Surgical Vision, Inc., USA), placed in the capsular bag or in the sulcus according to capsule status. Secondary IOL implantation in aphakic patients was preferred after 4 years of age when the eye was suitable for surgery (normal IOP, adequate pupillary dilation, and sufficient residual capsule to support an IOL). Surgery was also considered for prolonged uncorrected aphakia when contact lens correction was not feasible. Keratometry was obtained preoperatively under general anesthesia using a handheld autokeratometer (Retinomax K-Plus 3, Righton, Tokyo, Japan); the mean of three consecutive readings was used for IOL calculation. When reliable keratometry could not be obtained (e.g., due to corneal pathology or equipment unavailability), a theoretical keratometry value of 43.0 D was applied. IOL power calculation was performed using the Holladay I formula for shorter eyes and SRK/T for average eyes. Undercorrection was deliberately applied to induce mild postoperative hyperopia in anticipation of the progressive myopic shift associated with ocular growth. The approximate target refractions were + 4.00 D at age 2 years, + 2.00 D at age 3 years, and + 1.00 D at age 4 years. This approach is supported by previous studies on pediatric cataract surgery outcomes [8–10]. The postoperative regimen included moxifloxacin ophthalmic solution (Vigamox 0.5%, Alcon Laboratories, Fort Worth, TX, USA), instilled as one drop six times per day for 2 weeks. Prednisolone acetate (Pred-Forte 1%, Allergan Inc., Irvine, CA, USA) was instilled one drop every hour for the first 5 days, then tapered over the subsequent 5 weeks according to clinical response. Cycloplegic drops (tropicamide 1%, Alcon Laboratories, Fort Worth, TX, USA) were prescribed, one drop twice daily for 2 weeks. After cataract surgery, all children had a full eye exam. Refractive error was assessed by retinoscopy every 3 months until the age of 4 1/2 years, then every 6 months until the last visit. At each follow-up visit, BCVA in relation to patient age and compliance was reported. Visual acuity was assessed according to the child’s age and level of cooperation by the same pediatric ophthalmologist (D.P.) throughout follow-up. In preverbal or noncooperative children, fixation behavior, following response, and objection to occlusion were used to estimate visual preference, consistent with established pediatric cataract protocols. In cooperative preschool-aged children, age-appropriate Lea symbols were used, and in older or literate children, best-corrected visual acuity (BCVA) was measured using Snellen charts. According to the IRIS Registry recommendations, children with impaired fixation or vision had their approximate acuity levels converted to logMAR equivalents. This includes finger counting (2.0), hand movement (2.3), light perception (2.7), and absence of light perception (4.0) tests. All tests were performed by the same person to ensure consistent results and minimal differences between observers. Since this was a retrospective study, masking was not possible, but standard testing methods were used to minimize measurement bias. Ocular motility and alignment were assessed by the same doctor. Follow-up with Hirschberg, Krimsky, and the alternate cover tests revealed strabismus, which was diagnosed if the deviation exceeded 10 prism diopters. Postoperative occlusion therapy was suggested for amblyopic patients. Amblyopia management was initiated with full optical correction (spectacles or contact lenses). Occlusion therapy was prescribed in unilateral cases and in eyes with a significant interocular visual acuity difference. The patching schedule was age- and severity-adjusted: 2–4 h/day in younger children and up to 6 h/day in older children or in those with more severe amblyopia, consistent with previous PEDIG protocols [11–13]. Near activities such as drawing or reading were encouraged during occlusion. Patching duration was modified over time according to visual acuity outcomes at follow-up visits. Adherence to amblyopia therapy and visual rehabilitation was assessed retrospectively from examiner notes and parental reports. Adherence was classified as “good” when ≥ 50% of the prescribed daily duration and frequency of therapy was achieved, meaning that the child used the occlusion patch and spectacles for at least half of the recommended hours per day. For contact lens wear, adherence was considered “good” when lenses were worn for ≥ 50% of waking hours on most days. In cases using spectacles only, compliance was similarly based on the proportion of time the correction was worn during waking hours. “Poor” adherence indicated use below these thresholds. An investigator masked to preoperative IALD independently reviewed each chart to confirm adherence grading. The information collected from patient charts included age at operation and last visit, type of cataract, laterality, and strabismus or nystagmus. Postoperative complications and ocular pathology were noted, and the secondary IOL implantation time, final BCVA, refraction, adherence to spectacles or contact lenses, and adherence to occlusion therapy were determined. IOL implantation is considered a part of the overall visual rehabilitation strategy but does not require active adherence. In contrast, the consistent use of contact lenses, spectacles, and patching therapy depends on patient cooperation and is thus included in our operational definition of “adherence to visual rehabilitation”. MS was determined by differences in the postoperative and last-visit spherical equivalent refraction. Methods for MS included MS that were designated as “No shift” (≤ 2.0 diopters), “Mild shift” (2.25–4.0 diopters), and “Large shift” (> 4.25 diopters) that corresponded to clinically relevant refractive change ranges described in previous pediatric pseudophakia studies. IALD (preoperative and postoperative) was analyzed. IALD was defined as the absolute axial length difference between the operated and fellow eyes. Patients were classified as “similar” (IALD < 0.5 mm) or “different” (IALD ≥ 0.5 mm). Strabismus was defined as a constant horizontal or vertical deviation greater than 10 prism diopters detected by alternate cover testing or Krimsky/Hirschberg methods. All statistical analyses were performed using IBM SPSS for Windows, version 25.0 (SPSS, Chicago, IL, USA). A Kolmogorov–Smirnov test indicated that BCVA (logMAR) and other continuous variables were not normally distributed (p < 0.05). Accordingly, nonparametric tests were preferred for these analyses. For the regression model, residuals were examined and confirmed to meet assumptions of linearity and homoscedasticity. BCVA values were converted to logarithm of the minimum angle of resolution (logMAR) units prior to analysis. Both the Kolmogorov–Smirnov (p < 0.001) and Shapiro–Wilk (p < 0.001) tests indicated a deviation from perfect normality; however, visual inspection of the Q–Q plot demonstrated an approximately linear trend with only mild tail deviations. Given this near-normal distribution and the symmetric residual pattern observed in model diagnostics, a linear regression model was considered appropriate, as logMAR BCVA is a quasi-continuous variable that approximates linearity in clinical datasets. Normally distributed continuous variables were expressed as mean ± standard deviation, while those that were not normally distributed were expressed as median (25th–75th percentile). Categorical variables were summarized as counts (percentages). Comparisons between normally distributed continuous variables were performed using the student’s t-test, while non-normally distributed data were analyzed using the Mann–Whitney U-test. Associations between categorical variables were determined using the chi-square test, with a two-sided p-value < 0.05 considered statistically significant. Multiple comparisons were corrected using the Bonferroni method. Additionally, a multivariate linear regression analysis was performed to identify the independent effects of potential predictors on final BCVA. Fifteen clinical variables were initially evaluated as potential predictors. Variables showing an association with postoperative BCVA in univariate analyses (p < 0.20) or considered clinically relevant (e.g., laterality, presence of nystagmus, adherence to visual rehabilitation) were included in the multivariate linear regression model. SPSS Modeler’s Automatic Data Preparation (ADP) module automatically handled missing data and excluded variables with excessive missingness or minimal variance, leaving ten predictors in the final analysis. A forward stepwise (likelihood ratio) linear regression approach was used. Multicollinearity was assessed (VIF < 3 for all variables), and residuals were confirmed to be approximately normally distributed. Results We reviewed the records of 233 eyes from 143 children, including 90 bilateral and 53 unilateral cases. Table 1 provides an overview of demographic data, cataract types, surgical timing, IOL implantation ages, follow-up durations, and postoperative outcomes in congenital cataract cases. No significant differences were found between the bilateral and unilateral groups in terms of median age at cataract extraction, age at secondary IOL implantation, or age at final follow-up (p > 0.05). Table 1Overview of demographic data, cataract types, and postoperative outcomes in congenital cataract casesCharacteristicBilateral (n = 90)Unilateral (n = 53)Total (n = 143)Sex Female32 (35.6%)25 (47.2%)57 (39.9%) Male58 (64.4%)28 (52.8%)86 (60.1%)Cataract type Lamellar77 (42.8%)20 (37.7%)97 (41.6%) Nuclear21 (11.7%)8 (15.1%)29 (12.4%) Total28 (15.6%)4 (7.5%)32 (13.7%) Anterior polar18 (10%)5 (9.4%)23 (9.8%) Posterior subcapsular13 (7.2%)4 (7.5%)17 (7.2%) Cortical3 (1.7%)4 (7.5%)7 (3.0%)Posterior polar16 (8.9%)7 (13.2%)23 (9.8%) Unrecorded4 (2.2%)1 (1.9%)5 (2.1%)Median age at surgery (months)*7.5 (3–34)6 (2.25–22.5)-Median age at last follow-up (years)*6.75 (4–10)5 (2-8.5)-Secondary IOL implantation time (months)*46 (33–64)37 (30–54)-Preoperative IALD < 0.5 mm73 (74.5%)10 (40%)-Preoperative IALD > 0.5 mm25 (25.5%)15 (60%)Postoperative IALD < 0.5 mm71 (78%)10 (43.5%)-Postoperative IALD > 0.5 mm20 (22%)13 (56.5%)Postoperative nystagmus18 (20%)3 (5.7%)-Postoperative strabismus38 (48.1%)30 (61.2%)-Final BCVA (LogMAR)0.22 (0-0.52)1.0 (0.4–1.3)-Myopic shift No shift (0–2 D)96 (67.6%)23 (56.1%)- Mild shift (2.25-4 D)32 (22.5%)11 (26.8%)- Large shift (> 4.25 D)14 (9.9%)7 (17.1%)-Data are presented as median (interquartile range, IQR)Abbreviations: IOL = intraocular lens; IALD = interocular axial length difference; BCVA = best corrected visual acuity; D = diopter Table 1Overview of demographic data, cataract types, and postoperative outcomes in congenital cataract casesCharacteristicBilateral (n = 90)Unilateral (n = 53)Total (n = 143)Sex Female32 (35.6%)25 (47.2%)57 (39.9%) Male58 (64.4%)28 (52.8%)86 (60.1%)Cataract type Lamellar77 (42.8%)20 (37.7%)97 (41.6%) Nuclear21 (11.7%)8 (15.1%)29 (12.4%) Total28 (15.6%)4 (7.5%)32 (13.7%) Anterior polar18 (10%)5 (9.4%)23 (9.8%) Posterior subcapsular13 (7.2%)4 (7.5%)17 (7.2%) Cortical3 (1.7%)4 (7.5%)7 (3.0%)Posterior polar16 (8.9%)7 (13.2%)23 (9.8%) Unrecorded4 (2.2%)1 (1.9%)5 (2.1%)Median age at surgery (months)*7.5 (3–34)6 (2.25–22.5)-Median age at last follow-up (years)*6.75 (4–10)5 (2-8.5)-Secondary IOL implantation time (months)*46 (33–64)37 (30–54)-Preoperative IALD < 0.5 mm73 (74.5%)10 (40%)-Preoperative IALD > 0.5 mm25 (25.5%)15 (60%)Postoperative IALD < 0.5 mm71 (78%)10 (43.5%)-Postoperative IALD > 0.5 mm20 (22%)13 (56.5%)Postoperative nystagmus18 (20%)3 (5.7%)-Postoperative strabismus38 (48.1%)30 (61.2%)-Final BCVA (LogMAR)0.22 (0-0.52)1.0 (0.4–1.3)-Myopic shift No shift (0–2 D)96 (67.6%)23 (56.1%)- Mild shift (2.25-4 D)32 (22.5%)11 (26.8%)- Large shift (> 4.25 D)14 (9.9%)7 (17.1%)-Data are presented as median (interquartile range, IQR)Abbreviations: IOL = intraocular lens; IALD = interocular axial length difference; BCVA = best corrected visual acuity; D = diopter Overview of demographic data, cataract types, and postoperative outcomes in congenital cataract cases Data are presented as median (interquartile range, IQR) Abbreviations: IOL = intraocular lens; IALD = interocular axial length difference; BCVA = best corrected visual acuity; D = diopter In bilateral cases, 92.2% (n = 166) of eyes had no additional pathology, 7.2% (n = 13) had microphthalmia, and 0.6% (n = 1) had persistent hyperplastic primary vitreous (PHPV); in unilateral cases, the rates were 73.6% (n = 39), 7.5% (n = 4), and 18.9% (n = 10), respectively. Multiple subgroup comparisons were performed for visual and treatment outcomes (Tables 2 and 3). To control for the family-wise type I error, Bonferroni correction was applied. For Table 2, six independent comparisons were performed (corrected α = 0.05 / 6 = 0.0083), and for Table 3, seven independent comparisons (corrected α = 0.05 / 7 = 0.0071). Only p-values below these thresholds were considered statistically significant after correction. Effect sizes and 95% confidence intervals were calculated to describe the strength of associations beyond statistical significance. Visual outcomes Postoperative BCVA outcomes are summarized in Tables 1 and 2. Final visual acuity was significantly better in bilateral cases compared to unilateral ones (p < 0.001). The influence of visual rehabilitation, amblyopia treatment, and strabismus on BCVA is further explored in Tables 2 and 3, and Fig. 1. Table 2Comparison of visual and treatment outcomes by subgroups in congenital cataract casesBilateral BCVA (n = 90)Unilateral BCVA(n = 53)Effect Size (r)95% CIp-ValueGood patient adherence to amblyopia treatment0.13 (0-0.4)0.46 (0.22-1)0.250.09–0.42 0.003 Poor patient adherence to amblyopia treatment0.4 (0.1–0.7)1.3 (1.3–1.7)0.270.10–0.44 < 0.001 Good patient adherence to visual rehabilitation0.26 (0.05–0.52)0.7 (0.4–1.3)0.270.10–0.44 0.001 Poor patient adherence to visual rehabilitation1 (0.22–1.3)1.3 (1.15–1.3)0.15–0.02–0.320.106Presence of nystagmus0.52 (0.22-1.0)1.7 (1.3-NA)‡0.220.06–0.39 0.002 No nystagmus0.1 (0-0.4)0.7 (0.4–1.3)0.340.18–0.51 < 0.001 Presence of postoperative strabismus0.4 (0.05–0.7)1.3 (1-1.3)0.350.19–0.52 < 0.001 No postoperative strabismus0.19 (0-0.4)0.4 (0.04–0.65)0.10–0.07–0.270.225Preoperative IALD < 0.5 mm0.3 (0.1–0.7)1 (0.12–1.4)0.180.01–0.350.032Preoperative IALD > 0.5 mm0.22 (0-0.61)0.7 (0.4–1.3)0.250.09–0.42 0.003 Postoperative IALD < 0.5 mm0.22 (0.05–0.52)0.7 (0.12–1.3)0.14–0.02–0.310.087Postoperative IALD >0.5 mm0.4 (0.02-1)1 (0.4–1.3)0.170.00–0.340.048Data are presented as median (interquartile range, IQR). Statistical comparisons were made using the Mann-Whitney U test. To control for multiple comparisons, Bonferroni correction was applied, yielding a corrected significance threshold of p < 0.0083. For bilateral cases, only the first operated eye was included in the analysis to avoid duplication‡IQR could not be calculated due to limited sample sizeAbbreviations: BCVA = best-corrected visual acuity; IALD = interocular axial length difference; CI = confidence interval Table 2Comparison of visual and treatment outcomes by subgroups in congenital cataract casesBilateral BCVA (n = 90)Unilateral BCVA(n = 53)Effect Size (r)95% CIp-ValueGood patient adherence to amblyopia treatment0.13 (0-0.4)0.46 (0.22-1)0.250.09–0.42 0.003 Poor patient adherence to amblyopia treatment0.4 (0.1–0.7)1.3 (1.3–1.7)0.270.10–0.44 < 0.001 Good patient adherence to visual rehabilitation0.26 (0.05–0.52)0.7 (0.4–1.3)0.270.10–0.44 0.001 Poor patient adherence to visual rehabilitation1 (0.22–1.3)1.3 (1.15–1.3)0.15–0.02–0.320.106Presence of nystagmus0.52 (0.22-1.0)1.7 (1.3-NA)‡0.220.06–0.39 0.002 No nystagmus0.1 (0-0.4)0.7 (0.4–1.3)0.340.18–0.51 < 0.001 Presence of postoperative strabismus0.4 (0.05–0.7)1.3 (1-1.3)0.350.19–0.52 < 0.001 No postoperative strabismus0.19 (0-0.4)0.4 (0.04–0.65)0.10–0.07–0.270.225Preoperative IALD < 0.5 mm0.3 (0.1–0.7)1 (0.12–1.4)0.180.01–0.350.032Preoperative IALD > 0.5 mm0.22 (0-0.61)0.7 (0.4–1.3)0.250.09–0.42 0.003 Postoperative IALD < 0.5 mm0.22 (0.05–0.52)0.7 (0.12–1.3)0.14–0.02–0.310.087Postoperative IALD >0.5 mm0.4 (0.02-1)1 (0.4–1.3)0.170.00–0.340.048Data are presented as median (interquartile range, IQR). Statistical comparisons were made using the Mann-Whitney U test. To control for multiple comparisons, Bonferroni correction was applied, yielding a corrected significance threshold of p < 0.0083. For bilateral cases, only the first operated eye was included in the analysis to avoid duplication‡IQR could not be calculated due to limited sample sizeAbbreviations: BCVA = best-corrected visual acuity; IALD = interocular axial length difference; CI = confidence interval Comparison of visual and treatment outcomes by subgroups in congenital cataract cases Data are presented as median (interquartile range, IQR). Statistical comparisons were made using the Mann-Whitney U test. To control for multiple comparisons, Bonferroni correction was applied, yielding a corrected significance threshold of p < 0.0083. For bilateral cases, only the first operated eye was included in the analysis to avoid duplication ‡IQR could not be calculated due to limited sample size Abbreviations: BCVA = best-corrected visual acuity; IALD = interocular axial length difference; CI = confidence interval Fig. 1Comparison of median BCVA (logMAR) between bilateral and unilateral congenital cataract cases according to clinical subgroups Comparison of median BCVA (logMAR) between bilateral and unilateral congenital cataract cases according to clinical subgroups Subgroups include postoperative and preoperative IALD, presence or absence of nystagmus and strabismus, and adherence to amblyopia or visual rehabilitation treatment. Bars represent median BCVA (logMAR) values for bilateral (blue) and unilateral (orange) cases within each subgroup. Data presented are descriptive and not adjusted for potential confounders. Error bars represent interquartile ranges (IQR). Abbreviations: BCVA = best-corrected visual acuity; IALD = interocular axial length difference. Baseline severity differed between age groups, which helps to explain the finding regarding surgical timing. Infants who underwent surgery before 6 months of age more often presented with severe baseline characteristics, including a higher frequency of total cataracts (21.6% vs. 8.3%), different preoperative IALD > 0.5 mm (50.5% vs. 24.8%), and poorer adherence to amblyopia therapy (64.2% vs. 41%) compared with those operated later. Strabismus and nystagmus Preoperatively, the most common strabismus types were esotropia (11.2%) and exotropia (9%) in bilateral cases and 24% and 6%, respectively, in unilateral cases. A significant association between postoperative strabismus and worse BCVA was observed in unilateral cases (p = 0.003), while no such correlation was found in bilateral cases (p = 0.102). The most frequent postoperative deviation was esotropia in both groups (22.1% in bilateral and 54% in unilateral cases), followed by exotropia (20.9% and 2%, respectively). Strabismus surgery was performed in 22 patients (15.9%). Patients with orthotropia had significantly better BCVA than those with heterotropia (p = 0.002). Nystagmus, observed in 21 patients (14.6%), was associated with a poorer visual prognosis (p < 0.001). Additional details, including postoperative nystagmus rates and adherence to optical correction, are provided in Tables 2 and 3. Table 3Association of clinical factors with visual acuity outcomes in children with congenital cataractsBCVA (n = 143)Effect Size (r)95% CIp-ValueTime of cataract surgery0.340.18–0.51 < 0.001 Before 6 months0.7 (0.22-1) After 6 months0.19 (0-0.49)Preoperative IALD0.08–0.09–0.250.344 Less than 0.5 mm0.40 (0.1–0.7) More than 0.5 mm0.40 (0.06–0.92)Postoperative IALD0.240.07–0.41 0.004 Less than 0.5 mm0.22 (0.05–0.7) More than 0.5 mm0.7 (0.28-1)Nystagmus0.310.15–0.48 < 0.001 Presence0.7 (0.22–1.08) Absence0.26 (0-0.7)Patient adherence to amblyopia treatment0.180.01–0.350.033 Good0.22 (0-0.65) Poor0.40 (0.1–0.96)Patient adherence to visual rehabilitation0.310.15–0.48 0.001 Good0.40 (0.05–0.7) Poor1.15 (0.77–1.3)Strabismus0.260.09–0.43 0.002 Orthotropia0.22 (0-0.4) Heterotropia0.46 (0.14-1)*Data are presented as median (interquartile range, IQR). Statistical comparisons were made using the Mann-Whitney U test. To control for multiple comparisons (n = 7), Bonferroni correction was applied, yielding an adjusted significance threshold of p < 0.0071Abbreviations: BCVA = best corrected visual acuity; IALD = interocular axial length difference; CI = confidence interval Table 3Association of clinical factors with visual acuity outcomes in children with congenital cataractsBCVA (n = 143)Effect Size (r)95% CIp-ValueTime of cataract surgery0.340.18–0.51 < 0.001 Before 6 months0.7 (0.22-1) After 6 months0.19 (0-0.49)Preoperative IALD0.08–0.09–0.250.344 Less than 0.5 mm0.40 (0.1–0.7) More than 0.5 mm0.40 (0.06–0.92)Postoperative IALD0.240.07–0.41 0.004 Less than 0.5 mm0.22 (0.05–0.7) More than 0.5 mm0.7 (0.28-1)Nystagmus0.310.15–0.48 < 0.001 Presence0.7 (0.22–1.08) Absence0.26 (0-0.7)Patient adherence to amblyopia treatment0.180.01–0.350.033 Good0.22 (0-0.65) Poor0.40 (0.1–0.96)Patient adherence to visual rehabilitation0.310.15–0.48 0.001 Good0.40 (0.05–0.7) Poor1.15 (0.77–1.3)Strabismus0.260.09–0.43 0.002 Orthotropia0.22 (0-0.4) Heterotropia0.46 (0.14-1)*Data are presented as median (interquartile range, IQR). Statistical comparisons were made using the Mann-Whitney U test. To control for multiple comparisons (n = 7), Bonferroni correction was applied, yielding an adjusted significance threshold of p < 0.0071Abbreviations: BCVA = best corrected visual acuity; IALD = interocular axial length difference; CI = confidence interval Association of clinical factors with visual acuity outcomes in children with congenital cataracts *Data are presented as median (interquartile range, IQR). Statistical comparisons were made using the Mann-Whitney U test. To control for multiple comparisons (n = 7), Bonferroni correction was applied, yielding an adjusted significance threshold of p < 0.0071 Abbreviations: BCVA = best corrected visual acuity; IALD = interocular axial length difference; CI = confidence interval Myopic shift Data on the degree of MS in relation to IOL implantation timing are presented in Table 4. Importantly, earlier secondary IOL implantation was significantly associated with greater degrees of MS in both bilateral and unilateral cataract patients (p = 0.029 and p = 0.014, respectively). Table 4Comparison of secondary IOL implantation timing according to degree of myopic shift (MS) in congenital cataract casesMyopic Shift (-)Myopic Shift (+)p-ValueSecondary IOL implantation time (months)Bilateral42.29 ± 16.4557.04 ± 28.360.029Unilateral35.20 ± 11.7152.56 ± 23.020.014*Data are presented as mean ± standard deviation or median (interquartile range), where appropriate. Statistical comparisons were performed using the Mann-Whitney U test. Statistical significance was defined as p < 0.05. Abbreviations: IOL = intraocular lens; MS = myopic shift Table 4Comparison of secondary IOL implantation timing according to degree of myopic shift (MS) in congenital cataract casesMyopic Shift (-)Myopic Shift (+)p-ValueSecondary IOL implantation time (months)Bilateral42.29 ± 16.4557.04 ± 28.360.029Unilateral35.20 ± 11.7152.56 ± 23.020.014*Data are presented as mean ± standard deviation or median (interquartile range), where appropriate. Statistical comparisons were performed using the Mann-Whitney U test. Statistical significance was defined as p < 0.05. Abbreviations: IOL = intraocular lens; MS = myopic shift Comparison of secondary IOL implantation timing according to degree of myopic shift (MS) in congenital cataract cases *Data are presented as mean ± standard deviation or median (interquartile range), where appropriate. Statistical comparisons were performed using the Mann-Whitney U test. Statistical significance was defined as p < 0.05. Abbreviations: IOL = intraocular lens; MS = myopic shift Interocular axial length difference At the time of surgery, axial length (AL) values were measured for both eyes. For bilateral cases, the first operated eye had an IALD of ≤ 0.5 mm [14]; values exceeding this threshold were classified as “similar” or “different” IALD. Eyes with normal IALD showed better BCVA outcomes compared to eyes with asymmetric ALs in both preoperative and postoperative visits. These differences were statistically significant (p < 0.001). Of note, in the unilateral group, only one patient had a preoperative IALD < 0.5 mm. Postoperative IALD was significantly correlated with final BCVA, as also demonstrated in Table 3. Adherence to amblyopia and visual treatment Amblyopia therapy adherence reflects patients’ efforts to wear patches and utilize glasses or contact lenses following cataract surgery. Adherence was “good” when the patient met more than 80% of the attending physician’s amblyopia and visual therapy requirements. An investigator, masked to preoperative IALD, reviewed each chart and rated patch adherence as good (patch wearing and glasses use as prescribed, with either complete adherence or only brief durations of departure) or poor (refusing to wear the patch and glasses, or documented use for less than 50% of the prescribed duration as noted by the examiner) based on the examiner’s notes. The visual acuities of children who had cataract surgery and were well adjusted to amblyopia treatment and visual rehabilitation were much better in terms of BCVA than those who did not (p = 0.033 and p = 0.001, respectively) (Table 3). In bilateral cases, patient adherence to both amblyopia and visual rehabilitation was better, by means of BCVA, than in unilateral cases (Table 2). Postoperative glaucoma and other complications Postoperative complications were observed in 23.8% of patients. Secondary glaucoma developed in 13.3% of bilateral and 11.3% of unilateral cases; all were under control at the last follow-up, with eight bilateral cases managed medically and six unilateral cases requiring glaucoma surgery. Visual axis opacification occurred in 7.8% of bilateral and 5.7% of unilateral cases. Less common complications included retinal detachment (0.7%), endophthalmitis (2.8%), and intraocular lens decentration (2.8%), the latter managed with lens exchange. No cases of corneal decompensation were recorded. Multivariate regression analysis According to the multiple linear regression model, the overall fit was statistically significant (F = 13.789, p < 0.001), explaining 51.4% of the variance in postoperative BCVA (R² = 0.514). The most influential factors were adherence to visual rehabilitation (β = 0.39, p < 0.001; importance = 0.247) and laterality (β = 0.38, p < 0.001; importance = 0.247). Nystagmus (β = 0.21, p = 0.005; importance = 0.118) and amblyopia treatment adherence (β = 0.16, p = 0.025; importance = 0.077) also showed significant effects, while postoperative strabismus demonstrated borderline significance (β = 0.12, p = 0.055). After adjustment, preoperative strabismus, cataract type, time of surgery (< 3 years vs. ≥ 3 years), secondary intraocular surgery, and associated ocular pathology showed no independent associations (p > 0.05). Model residuals were normally distributed, and no influential outliers were detected (Cook’s distance < 0.20) (Fig. 2; Table 5). Fig. 2The relative importance and statistical significance of predictors The relative importance and statistical significance of predictors Table 5Multiple linear regression analysis for predictors of logMAR BCVAVariableβ (Standardized)SE95% CIP-valueVIFAdherence to visual rehabilitation0.390.090.21–0.56< 0.0011.32Laterality0.380.090.20–0.55< 0.0011.28Presence of nystagmus0.210.070.06–0.350.0051.14Postoperative strabismus0.120.06-0.01-0.250.0551.19Adherence to amblyopia treatment0.160.070.02–0.300.0251.10Preoperative strabismus0.080.06-0.03-0.190.2101.16Cataract type0.070.05-0.04-0.180.2701.12Time of surgery (< 3 years)-0.100.05-0.22-0.020.0891.08Secondary intraocular surgery0.060.05-0.04-0.160.3201.17Associated ocular pathology (microphthalmia, PHPV, etc.)0.050.06-0.07-0.170.4501.20Abbreviations: β: standardized coefficient; SE = standard error; CI = confidence interval; VIF = variance inflation factor; PHPV = persistent hyperplastic primary vitreous Table 5Multiple linear regression analysis for predictors of logMAR BCVAVariableβ (Standardized)SE95% CIP-valueVIFAdherence to visual rehabilitation0.390.090.21–0.56< 0.0011.32Laterality0.380.090.20–0.55< 0.0011.28Presence of nystagmus0.210.070.06–0.350.0051.14Postoperative strabismus0.120.06-0.01-0.250.0551.19Adherence to amblyopia treatment0.160.070.02–0.300.0251.10Preoperative strabismus0.080.06-0.03-0.190.2101.16Cataract type0.070.05-0.04-0.180.2701.12Time of surgery (< 3 years)-0.100.05-0.22-0.020.0891.08Secondary intraocular surgery0.060.05-0.04-0.160.3201.17Associated ocular pathology (microphthalmia, PHPV, etc.)0.050.06-0.07-0.170.4501.20Abbreviations: β: standardized coefficient; SE = standard error; CI = confidence interval; VIF = variance inflation factor; PHPV = persistent hyperplastic primary vitreous Multiple linear regression analysis for predictors of logMAR BCVA Abbreviations: β: standardized coefficient; SE = standard error; CI = confidence interval; VIF = variance inflation factor; PHPV = persistent hyperplastic primary vitreous Discussion Although many studies have focused on the long-term visual results of surgery [15], only a few have examined both ocular motility and visual outcomes together, along with contributing factors to poor vision after surgery in children [16–21]. Final BCVA was significantly correlated with age at surgery. Infants who underwent surgery before 6 months had worse BCVA (p < 0.001), whereas later surgery has been associated with better acuity [22], possibly due to milder opacification and some preserved binocularity [23, 24]. Moreover, visually significant cataract types are more common in children under 6 months. Although our analysis indicated that surgery performed after 6 months was associated with better BCVA, this observation should be viewed with caution. In our cohort, children who underwent surgery before 6 months more frequently had severe baseline characteristics, including a higher proportion of total cataracts, larger preoperative IALD, and poorer adherence to amblyopia therapy. These factors likely account for the less favorable outcomes observed in the early surgery group. Therefore, our findings should not be interpreted as contradicting the established recommendation for early intervention in visually significant congenital cataracts, but rather as evidence that disease severity, ocular asymmetry, and treatment adherence confound outcome comparisons across age groups. While this relationship demonstrates an association between surgical timing and visual outcome, it should not be interpreted as evidence of causality, as baseline severity and adherence factors likely account for these differences. As previous studies have shown, bilateral cataracts led to better vision results than unilateral cases [10, 17, 22–28]. Louison et al. reported a median BCVA of 0.1 logMAR in bilateral cases compared to 0.5 logMAR in unilateral cases [25]. This difference likely results from the absence of a dominant fellow eye in bilateral cases. Even with early surgery, unilateral cataracts are still hard to treat because of amblyopia and competition between the eyes. Outcomes strongly depend on adherence to patching and spectacle use [9, 28], as shown in our cohort. Cross-study variability likely reflects differences in therapy protocols and compliance. Bilateral cases may benefit from some residual input, whereas unilateral cases had more frequent strabismus, nystagmus, and anisometropia. Optical correction is a key part of visual rehabilitation. All patients received secondary IOL implantation after long-term use of contact lenses or spectacles. Aphakia in infants disrupts emmetropization, leading to MS, which may exceed 10 D and follows a logarithmic pattern. In our cohort, bilateral cases had a mean MS of 1.9 ± 2.1 D, and unilateral cases had 2.1 ± 2.2 D. MS was slightly lower in bilateral cases, who also had better VA, suggesting more balanced ocular growth. In contrast, unilateral cases are more prone to anisometropia, amblyopia, and impaired binocularity. Aphakic or pseudophakic eyes often showed greater MS than their unoperated fellow eyes [29]. MS tended to be higher in younger children undergoing secondary IOL implantation, consistent with previous findings [9, 29–34]. Lu et al. reported an average MS of ~ 5.15 D in the first postoperative year in infants with primary IOLs, attributed to rapid axial elongation [9]. Pseudophakic eyes continued to elongate similarly to fellow eyes. These findings support a hyperopic target in infancy to accommodate future myopization. Some studies also report greater MS in unilateral eyes compared to bilateral ones [35]. We also found that MS decreased with increasing age at IOL implantation. This may partly reflect better adherence to amblyopia therapy in younger children during the aphakic phase. Preoperatively, AL differences were more common in unilateral cases but decreased postoperatively, possibly due to improved visual input [36]. Although IALDs were associated with poorer visual outcomes, this was not statistically significant. Still, IALD may serve as a clinical marker of visual prognosis. The multivariate linear regression model explained approximately half of the variance in postoperative visual acuity (R² = 0.51), implying a moderate-to-strong overall fit. This confirms the stability of observed predictor factors (visual rehabilitation adherence) and laterality as significant long-term predictors of surgery for congenital cataract. Similarly, nystagmus and adherence to amblyopia treatment were important, and borderline significance was observed for postoperative strabismus. Findings underscore the overlapping impact of anatomical and behavioral factors. These findings are similar to Takashima et al., who presented better visual acuity in bilateral cases and pseudophakic eyes, and worse outcomes in individuals with systemic comorbidities or secondary glaucoma [37]. Birch et al. demonstrated that unilateral cataracts and prolonged visual deprivation (>6 weeks) were associated with a higher risk of nystagmus and strabismus [38], highlighting the need for monitoring of a patient as a means that allows for timely and consistent intervention. In summary, the implications of our findings are that tailored, multidisciplinary care addressing surgical and rehabilitative aspects has to be tailored to long-term care. This study has several limitations. As this was a retrospective, single-center study, the possibility of selection bias cannot be excluded. Furthermore, as all surgeries were performed by a single experienced surgeon, procedural consistency was maximized; however, this also limits the external validity of our findings. Amblyopia and the compliance with visual rehabilitation were evaluated mainly through the notes taken by the examiner and the reports from parents, which may introduce a level of subjectivity and contribute to misclassification bias. Although adherence grading was verified by a blinded reviewer, formal inter-rater reliability statistics could not be calculated because each chart was reviewed only once and did not undergo independent reviews by multiple assessors. A sensitivity analysis applying a more stringent adherence criterion yielded consistent results, indicating that the observed association with visual outcomes was consistent across analyses. Using final BCVA rather than early postoperative values may also have introduced bias. Small subgroup sizes reduced statistical power, and some data were missing due to follow-up loss. Cataract severity or density could not be included in the multivariate analysis because these parameters were not consistently recorded in the medical files. This limitation may have influenced the interpretation of factors associated with final visual outcomes. Although the multivariate regression model explained more than half of the variance in postoperative BCVA, some residual confounding may still exist. Certain factors such as cataract density, surgical complexity, intraoperative complications, or systemic comorbidities (e.g., prematurity, metabolic disorders) were not available for inclusion or showed insufficient variability for statistical modeling. These unmeasured parameters may partly account for the remaining unexplained variance. Nevertheless, the consistency of our findings with established predictors of visual outcome in congenital cataract supports the robustness of the model. In addition, we included patients with elevated IOP or corneal edema to reflect real-world clinical conditions; however, subgroup analysis was not feasible due to the small number of such cases. Although the multivariate regression model explained roughly half of the variance in postoperative BCVA, some unmeasured confounding is likely to persist. Factors such as cataract density, surgical complexity, intraoperative challenges, and systemic comorbidities were either unavailable for analysis or exhibited insufficient variability for statistical assessment. These unmeasured variables may account for the observed unexplained variation. Nevertheless, our findings consistently align with established predictors of visual outcome in congenital cataract, supporting the robustness of the model. Despite these limitations, to our knowledge, this is the first study to evaluate anatomical factors (IALD, nystagmus, strabismus) and behavioral factors (adherence to amblyopia therapy and visual rehabilitation) together within a single multivariate model in a large pediatric cohort. This combined approach provides a more comprehensive view of long-term outcomes and underscores the importance of individualized strategies, particularly in unilateral cases where prognosis remains most challenging. Future studies should use prospective, multicenter designs with standardized follow up to validate IALD and adherence as predictors and to explore their relationship with surgical timing and IOL planning. Incorporating more objective methods to assess adherence, such as digital patch monitors, structured parental diaries, or electronic tracking systems, may provide more reliable measures of compliance. In addition, the integration of AI tools and public health initiatives, including early detection, family education, and long-term rehabilitation access, will be essential to optimize outcomes in children with congenital cataracts.