This retrospective clinical cohort study included pediatric patients with congenital or developmental cataracts who underwent cataract extraction and intraocular lens (IOL) implantation at the Eye Hospital of Wenzhou Medical University between 2003 and 2018.Eligibility criteria required: (1) a definitive diagnosis of congenital or developmental cataract with primary IOL implantation performed at our institution; (2) comprehensive preoperative biometric data, including age at surgery, axial length (AL), corneal curvature (K), implanted IOL power; and any complications occurring preoperatively or postoperatively, (3) longitudinal postoperative follow-up with stable refraction data at the final visit and a minimum follow-up of 5 years. Patients were excluded if they presented with: (1) systemic comorbidities precluding normal ocular development; (2) postoperative complications that could confound ocular growth or refractive assessment, such as secondary glaucoma, strabismus, corneal opacification, or significant posterior capsular opacification (PCO); or (3) insufficient clinical records. No patients in the final cohort received any myopia control interventions (e.g., low-dose atropine, orthokeratology, or myopia control spectacles) during the follow-up period. One patient who underwent IOL exchange surgery due to high myopia was included only up to the time point before the exchange procedure. Another patient with IOL decentration was excluded from the study, as this condition may have affected normal visual experience and refractive development. For patients with bilateral involvement, Given the high correlation between eyes in pediatric patients, only the first operated eye was included in the final dataset to avoid unit-of-analysis error. The study ultimately included a total of 106 eyes. The study protocol conformed to the tenets of the Declaration of Helsinki and received formal approval from the Institutional Review Board of the Eye Hospital of Wenzhou Medical University (Approval No. 2026-029-K-021).

The following data were extracted by trained chart abstractors: (1) preoperative ocular parameters including axial length, corneal curvature, intraoperatively implanted IOL power, and intraoperative complications. (2) postoperative follow-up time points, refractive outcomes from each visit, and slit-lamp examination findings. Preoperative axial length measurements were obtained using either the Axis Nano ultrasound biometer (Quantel Medical Inc., France) or the IOL Master 500 (Carl Zeiss Meditec, Jena, Germany). Postoperative refraction was performed by experienced pediatric optometrists under cycloplegia. Cycloplegia was achieved with 1% cyclopentolate hydrochloride, instilled as one drop every 5 minutes for 3 doses. Retinoscopy was carried out 30 to 40 minutes after the last drop, once the pupillary light reflex was abolished. Subjective refraction was added when feasible in older children.

To evaluate the generalizability of the predictive model, we performed external validation using an independent cohort from Shanghai Eye and ENT Hospital of Fudan University. This cohort consisted of 72 pediatric patients who underwent congenital cataract surgery with primary IOL implantation between 2013 and 2020. The same inclusion and exclusion criteria as the primary cohort were applied. All preoperative and postoperative variables were collected using identical definitions and measurement protocols. The final model derived from the development cohort was applied directly to this external dataset without refitting. Model performance was assessed using the same metrics: Harrell’s C-index and time-dependent AUC at 5 and 10 years. In addition, calibration was evaluated using calibration plots.

Data normality was evaluated using the Shapiro-Wilk test. Continuous variables are presented as mean ± SD (SD) for normally distributed data. Categorical variables are expressed as frequencies and percentages. Continuous variables with a normal distribution were compared using the Student’s t-test. Categorical data were compared using the chi-square test or Fisher’s exact test, as appropriate. High myopia was defined as a spherical equivalent (SE) of −6.00 diopters (D) or less at the final follow-up visit. The myopic shift rate was defined as the annualized magnitude of refractive change. It was calculated by taking the absolute difference between the spherical equivalent at the final visit and the first postoperative refraction, divided by the total follow-up duration in years. To identify independent predictors of postoperative high myopia, univariate and multivariate Cox proportional hazards regression models were employed. In the univariate analysis, the following 9 variables were evaluated: sex, ocular laterality, age at surgery, axial length, IOL power, radius of corneal curvature, axial length to corneal radius ratio (AL/CR ratio), IOL Power to Axial length Ratio (IOL/AL ratio), and initial postoperative spherical equivalent (Postop SE). Variables demonstrating a significance level of P <.05 in the univariate analysis were entered into the multivariate stepwise model. Significant factors identified in the final multivariate analysis were then utilized to develop a predictive model. The model’s predictive performance was rigorously assessed through discrimination and calibration. Discriminative ability was quantified using the Harrell’s C-index and the area under the receiver operating characteristic curve (AUC). Model calibration was evaluated via calibration plots to cf predicted vs observed probabilities. Internal validation was conducted using the bootstrap resampling method (500 iterations) to ensure the stability of the model and to adjust for potential overfitting.

For the initial postoperative spherical equivalent (Postop SE), the maximally selected rank statistics method was employed to identify an optimal cutoff of +0.75 D. This allowed for the dichotomization of the variable to enhance its practical utility in clinical settings.

To evaluate potential heterogeneity between unilateral and bilateral cases, we performed stratified sensitivity analyses. Each subgroup model included the 3 key predictors from the main analysis: age at surgery, IOL/AL ratio, and initial postoperative spherical equivalent (Postop SE). Predictive performance in each subgroup was assessed using time-dependent area under the receiver operating characteristic curve (AUC) at 5 and 10 years of follow-up, as well as the C-index for overall discrimination. All statistical analyses were performed using R software version 4.5.1, and a two-tailed P <.05 was considered statistically significant.

Ultimately, 106 eligible patients were included in this study ( Figure 1 ), comprising 49 females and 57 males. The mean age at the time of surgery was 3.73 years. The cohort was followed for a mean duration of 8.19 years, with a mean age of 11.9 years (range: 3.38 to 21.22 years) at the final follow-up. At the last visit, the mean spherical equivalent (SE) was −4.85 D, with 63 patients exhibiting an SE of −6.00 D or less. For external validation, an independent cohort of 72 patients who underwent congenital cataract surgery with primary IOL implantation at another tertiary center (2013-2020) was included, using the same inclusion and exclusion criteria. The mean follow-up duration in this external cohort was 7.83 years. Baseline characteristics of both the primary and external validation cohorts are summarized in Table 1 . The intergroup comparisons of the primary cohorts based on the development of high myopia at the final follow-up are summarized in Table 2 .

Flowchart of patient evaluation.

In the univariate cox regression analysis, the following factors were significantly associated with the development of postoperative high myopia: implanted IOL power (Hazard Ratio [HR], 1.06; 95% CI [CI], 1.01-1.12; P =.015), IOL-to-axial length (IOL/AL) ratio (HR, 2.64; 95% CI, 1.17-5.96; P =.02), and an initial postoperative spherical equivalent (SE) ≤ 0.75 D (HR, 2.89; 95% CI, 1.71-4.87; P <.001).In the multivariate cox regression model, only the IOL/AL ratio (HR, 3.04; 95% CI, 1.38-6.70; P =.006) and an initial postoperative SE ≤ 0.75 D (HR, 3.06; 95% CI, 1.82-5.14; P <.001) remained independent predictors of high myopia( Table 3 ).

Table 3

Factors Associated with the Development of Postoperative High Myopia: Univariate and Multivariate Cox Regression Analysis

Variables	Univariate Analysis						Multivariate Analysis
	β	S.E	Z	P	HR (95%CI)		β	S.E	Z	P	HR (95%CI)
Age	0.00	0.08	0.06	0.950	1.00 (0.86 ∼ 1.17)
IoL Power	0.06	0.03	2.42	.015	1.06 (1.01 ∼ 1.12)
Axial length	−0.15	0.09	−1.78	.075	0.86 (0.73 ∼ 1.02)
Radius of Corneal Curvature	−0.42	0.42	−0.99	.323	0.66 (0.29 ∼ 1.51)
AL/CR Ratio *	−0.76	0.62	−1.22	.222	0.47 (0.14 ∼ 1.58)
IOL/AL Ratio ⁎⁎	0.97	0.42	2.33	.020	2.64 (1.17 ∼ 5.96)		1.11	0.40	2.75	.006	3.04 (1.38 ∼ 6.70)
Sex
Female					1.00 (Reference)
Male	−0.31	0.25	−1.24	.216	0.73 (0.44 ∼ 1.20)
Ocular Laterality
Monocular					1.00 (Reference)
Binocular	0.09	0.26	0.33	.738	1.09 (0.65 ∼ 1.83)
Postop SE
SE > 0.75 D					1.00 (Reference)						1.00 (Reference)
SE ≤ 0.75 D	1.06	0.27	3.97	<.001	2.89 (1.71 ∼ 4.87)		1.12	0.27	4.21	<.001	3.06 (1.82 ∼ 5.14)

Only gold members can continue reading. Log In or Register to continue

Share this:

• X
• Facebook

Related

Related posts:

1. Microperimetry as an Outcome Measure Improves Patient Eligibility and Efficacy Detection in RPGR Gene Therapy Trials
2. Posterior Segment Risk Factors for Penetrating Keratoplasty Failure
3. Clinical Utility of Ultra-Widefield Swept-Source OCT for Intraocular Tumors: Comparison With Ultrasonography, SD-OCT, and MRI
4. Comment on: Visual Field Progression in Patients With Glaucoma With Delayed Follow-Up

Stay updated, free articles. Join our Telegram channel

Join