Incidence rate of secondary glaucoma following congenital cataract surgery: an in-depth systematic review and meta-analysis

Author: Li L, Wang X, Liu C, Wang S, Wang X.

Geographical coverage: China, India, Saudi Arabia, Turkey, Iran, South Korea, USA, Germany, Israel, Italy, Jordan, Spain, and the United Kingdom

Sector: Burden of disease

Sub–sector: Incidence

Equity focus: Not explicitly stated

Study population: Children who underwent congenital cataract surgery

Review type: Effectiveness review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background:

Congenital cataract is a leading cause of childhood blindness, affecting roughly one in 3 000 births worldwide. Early surgical extraction is essential for visual rehabilitation. Children older than two years usually receive a primary intra‑ocular lens (IOL) at the time of extraction, whereas younger infants are left aphakic and corrected with contact lenses or spectacles until a secondary IOL can be implanted. Surgery, however, disrupts the anterior segment and increases the lifetime risk of secondary glaucoma. This review examines how frequently glaucoma develops after congenital cataract surgery (referred to as glaucoma following congenital cataract surgery, GFCS) and how risk varies by IOL status, age at surgery and region.

Objectives

To determine the incidence of secondary glaucoma after congenital cataract surgery in children and to explore demographic and surgical factors that modify this risk.

Main findings

Thirty‑six cohort studies (prospective and retrospective) comprising 3 151 participants (4 717 eyes) published between 2001 and 2023 were eligible. All were rated good quality on the Newcastle–Ottawa Scale.

• Overall incidence: 6.6 % (95 % confidence interval [CI] 3.9 % to 9.9 %).
• Aphakia versus pseudophakia: Incidence was higher in aphakic eyes (13.5 %; 95 % CI 7.7 % to 20.6 %) than in eyes with a primary IOL (3.3 %; 95 % CI 1.5 % to 5.8 %) or a secondary IOL (3.5 %; 95 % CI 0 % to 11.4 %) (p < 0.01).
• Regional variation: Children from Asian studies had a higher incidence (6.9 %; 95 % CI 4.1 % to 10.4 %) than those from European studies (0.9 %; 95 % CI 0 % to 3.0 %) (p < 0.01).
• Age at surgery: Meta‑regression showed younger age at cataract extraction predicted a higher risk of glaucoma (p < 0.01).
• Laterality: Incidence did not differ between unilateral (10.0 %; 95 % CI 1.7 % to 22.7 %) and bilateral cataracts (11.8 %; 95 % CI 3.6 % to 23.5 %) (p = 0.90).

Sensitivity analyses, subgroup analyses and a leave‑one‑out approach produced similar estimates. Funnel‑plot symmetry and Egger’s/Begg’s tests suggested no significant publication bias.

Methodology

The authors searched PubMed, Embase, Cochrane Library and Web of Science to 16 March 2023 and performed a supplementary search after extraction to capture newly indexed records. Eligible English‑language studies included ≥ 20 eyes, ≥ 12 months’ follow‑up and clear diagnostic criteria for postoperative glaucoma. Two reviewers independently screened articles, extracted data and assessed risk of bias with the Newcastle–Ottawa Scale. Heterogeneity was quantified using Cochran’s Q and the I² statistic; random‑effects models were applied when I² > 50 %. When heterogeneity was substantial, subgroup analyses, sensitivity analyses and meta‑regression were undertaken to identify its sources.

Applicability/external validity

Although the review incorporated studies from multiple continents, definitions of glaucoma varied and data were sparse from Africa and South America. Most cohorts came from tertiary centres, so incidence may differ in community settings. Nevertheless, the findings give a reasonable benchmark for clinicians counselling families on the long‑term risks after congenital cataract surgery, particularly in similar healthcare systems.

Geographic focus

Included studies were conducted in China, India, Saudi Arabia, Turkey, Iran, South Korea, USA, Germany, Israel, Italy, Jordan, Spain and the United Kingdom.

Summary of quality assessment

We have medium confidence in the review’s conclusions. The search was limited to English‑language literature, reference lists were not searched systematically and the authors did not report contacting study investigators or providing a list of excluded papers. Although individual‑study risk of bias was assessed, pooled estimates were not stratified by this risk, and heterogeneity remained high despite subgroup analyses. These factors temper certainty in the incidence figures reported.