Purpose: To compare the efficacy, safety and acceptability of a treatment group (Orthokeratology) to a control group (single vision Spectacles) on slowing axial elongation in children. Methods: We searched studies in MEDLINE, EMBASE and the Cochrane Library up to January 2015 for randomized controlled trials (RCTs) and observational studies. We pooled the mean differences between the Orthokeratology and Control groups for axial elongation and the OR for rates of adverse events and dropout. Results: Three RCTs and six cohort studies with 667 children aged 6-16 years old were included. Two years' mean differences in axial elongation were -0.27mm (95% confidence intervals [CI], -0.32 to -0.23) in all studies, -0.28mm (95% CI, -0.35 to -0.20) in RCTs and -0.27mm (95% CI, -0.32 to -0.22) in cohort studies (p < 0.01). At 6 months, 1 year, 1.5 years and 2 years, mean differences in axial elongation were -0.13mm, -0.19mm, -0.23mm, and -0.27mm (p50.01), respectively. The effect was greater in Asian children than Caucasian (-0.28mm versus -0.22mm) and in children with moderate to high myopia when compared to children with low myopia (-0.35mm versus -0.25 mm). Orthokeratology had more non-significant adverse events (odd ratio [OR], 8.87; 95% CI, 3.79-20.74; p < 0.01) but comparable dropout rates (OR = 0.84, 95% CI, 0.40-1.74, p = 0.64) than control. Conclusion: Orthokeratology has significantly greater efficacy in controlling axial elongation in children compared to Spectacle correction. The safety and acceptability results are good, and there appears to be a greater myopia control effect in Chinese children compared to Caucasians, and in those with higher initial myopia.