Interventions for myopia control in children: a living systematic review and network meta-analysis.

Abstract

Rationale: The increasing prevalence of myopia is a growing global public health problem, in terms of rates of uncorrected refractive error and significantly, an increased risk of visual impairment due to myopia-related ocular morbidity. Interventions to slow its progression are needed in childhood, when myopia progression is most rapid. This is a review update, conducted as part of a living systematic review.

Objectives: To assess the comparative efficacy and safety of interventions for slowing myopia progression in children using network meta-analysis (NMA). To generate a relative ranking of interventions according to their efficacy. To produce a brief economic commentary, summarising economic evaluations.

Search methods: We searched CENTRAL, MEDLINE, Embase, and three trial registers. The latest search date was 19 February 2024.

Eligibility criteria: We included randomised controlled trials (RCTs) of optical, pharmacological, light therapy and behavioural interventions for slowing myopia progression in children, up to 18 years old.

Outcomes: Critical outcomes were progression of myopia (mean difference (MD) in the change in spherical equivalent refraction (SER, dioptres (D)), and axial length (AL, mm) in the intervention and control groups at one year or longer), and difference in the change in SER and AL following cessation of treatment (rebound).

Risk of bias: We assessed the risk of bias (RoB) for SER and AL using the Cochrane RoB 2 tool.

Synthesis methods: We followed standard Cochrane methods. We rated the certainty of evidence using the GRADE approach for change in SER and AL at one and two years. We used the surface under the cumulative ranking curve (SUCRA) to rank the interventions for all available outcomes.

Included studies: We included 104 studies (40 new for this update) that randomised 17,509 children, aged 4 years to 18 years. Most studies were conducted in China or other Asian countries (66.3%), and North America (14.4%). Eighty-four studies (80.8%) compared myopia control interventions against inactive controls. Study durations ranged from 12 months to 48 months.

Synthesis of results: Since most of the networks in the NMA were poorly connected, our estimates are based on direct (pairwise) comparisons, unless stated otherwise. The median change in SER for controls was -0.65 D (55 studies, 4888 participants; one-year follow-up). These interventions may reduce SER progression compared to controls: repeated low intensity red light (RLRL: MD 0.80 D, 95% confidence interval (CI) 0.71 to 0.89; SUCRA = 93.8%; very low-certainty evidence); high-dose atropine (HDA (≥ 0.5%): MD 0.90 D, 95% CI 0.62 to 1.18; SUCRA = 93.3%; moderate-certainty evidence); medium-dose atropine (MDA (0.1% to < 0.5%): MD 0.55 D, 95% CI 0.17 to 0.93; NMA estimate SUCRA = 75.5%; low-certainty evidence); low dose atropine (LDA (< 0.1%): MD 0.25 D, 95% CI 0.16 to 0.35; SUCRA = 53.2%; very low-certainty evidence); peripheral plus spectacle lenses (PPSL: MD 0.45 D, 95% CI 0.16 to 0.74; SUCRA = 50.2%; very low-certainty evidence); multifocal soft contact lenses (MFSCL: MD 0.27 D, 95% CI 0.18 to 0.35; SUCRA = 49.9%; very low-certainty evidence); and multifocal spectacle lenses (MFSL: MD 0.14 D, 95% CI 0.08 to 0.21; SUCRA = 30.8%; low-certainty evidence). The median change in AL for controls was 0.33 mm (58 studies, 9085 participants; one-year follow-up). These interventions may reduce axial elongation compared to controls: RLRL (MD -0.33 mm, 95% CI -0.37 to -0.29; SUCRA = 98.6%; very low-certainty evidence); HDA (MD -0.33 mm, 95% CI -0.35 to -0.30; SUCRA = 88.4%; moderate-certainty evidence); MDA (MD -0.24 mm, 95% CI -0.34 to -0.15; NMA estimate SUCRA = 75.8%; low-certainty evidence); LDA (MD -0.10 mm, 95% CI -0.13 to -0.07; SUCRA = 36.1%; very low-certainty evidence); orthokeratology (ortho-K: MD -0.18 mm, 95% CI -0.21 to -0.14; SUCRA = 79%; moderate-certainty evidence); PPSL (MD -0.13 mm, 95% CI -0.21 to -0.05; SUCRA = 52.6%; very low-certainty evidence); MFSCL (MD -0.11 mm, 95% CI -0.13 to -0.09; SUCRA = 45.6%; low-certainty evidence); and MFSL (MD -0.06 mm, 95% CI -0.09 to -0.04; SUCRA = 26.3%; low-certainty evidence). Ortho-K plus LDA probably reduces axial elongation more than ortho-K monotherapy (MD -0.12 mm, 95% CI -0.15 to -0.09; SUCRA = 81.8%; moderate-certainty evidence). At two-year follow-up, change in SER was reported in 34 studies (3556 participants). The median change in SER for controls was -1.01 D. The ranking of interventions to reduce SER progression was close to that observed at one year; there were insufficient data to draw conclusions on cumulative effects. The highest-ranking interventions were: HAD (SUCRA = 97%); MDA (NMA estimate SUCRA = 69.8%); and PPSL (SUCRA = 69.1%). At two-year follow-up, change in AL was reported in 33 studies (3334 participants). The median change in AL for controls was 0.56 mm. The ranking of interventions to reduce axial elongation was similar to that observed at one year; there were insufficient data to draw conclusions on cumulative effects. The highest-ranking interventions were: ortho-K plus LDA (SUCRA = 94.2%); HAD (SUCRA = 96.8%); and MDA (NMA estimate SUCRA = 88.4%). There was limited evidence on whether cessation of myopia control therapy increases progression beyond the expected rate of progression with age. Adverse events and treatment adherence were not consistently reported. Two studies reported quality of life, showing little to no difference between intervention and control groups. We were unable to draw firm conclusions regarding the relative costs or efficiency of different myopia control strategies in children.

Authors' conclusions: Most studies compared pharmacological and optical treatments to slow the progression of myopia with an inactive comparator. These interventions may slow refractive change and reduce axial elongation, although results were often heterogeneous. Less evidence is available for two years and beyond; uncertainty remains about the sustained effect of these interventions. Longer term and better quality studies comparing myopia control interventions alone or in combination are needed, with improved methods for monitoring and reporting adverse effects.

Funding: Cochrane Eyes and Vision US Project is supported by grant UG1EY020522, National Eye Institute, National Institutes of Health.

Registration: The previous version of this living systematic review is available at doi: 10.1002/14651858.CD014758.pub2.