To compare the effects of secondary in-the-bag versus ciliary sulcus intraocular lens (IOLs) implantation on the accuracy of IOL power calculation in pediatric eyes.Prospective non-randomized interventional study.Pediatric aphakic eyes underwent either in-the-bag or ciliary sulcus secondary IOL implantation were included. The mean prediction error (PE), mean absolute error (MAE), median absolute error (MedAE) and percentages of eyes with PE within ±0.25 diopter (D), ±0.50D, ± 0.75 D, and ± 1.00 D were calculated and compared using SRK/T formula.One hundred fourteen eyes (38.26%) received in-the-bag IOL implantation and 184 (61.74%) underwent ciliary sulcus IOL implantation. Compared with the sulcus group, the capsular group displayed statistically significant lower MAE and higher percentage of eyes within ±0.50 D of PE (MAE: 0.90 vs. 1.56D; ±0.50 D: 40.40% vs. 14.29%, both P0.001). The eyes receiving in-the-bag IOL implantation (Sulcus IOL implantation β: -1.060, 95CI: -1.415∼-0.705; P0.001), unilateral (β: 0.647, 95CI:0.144∼1.150; P=0.012) or with deeper anterior chamber depth (β: 0.362, 95CI:0.068∼0.656; P=0.016) were prone to maintain hyperopia (PE0). To reduce PE, when the predicted capsular IOL power was between 11.50-30.00D, the power of a sulcus implanted IOL should be reduced by 0.50-2.50D accordingly (the exact amount of reduction is positively related to the predicted power).In-the-bag implantation yielded smaller PE in pediatric eyes undergoing secondary IOL implantation. Adjustment of IOL power for ciliary sulcus implantation is required to reduce PE, and the amount of adjustment is positively correlated with the IOL power predicted by SRK/T formula.