Absolute external apical root resorption after self-ligating bracket treatment: a systematic review and meta-analysis of 2D and 3D imaging studies

To synthesize the available evidence on the absolute magnitude of external apical root resorption (EARR) in patients treated with self-ligating brackets (SLBs) using linear and volumetric measurements, and to explore whether imaging modality contributes to variability in reported outcomes, presented as an exploratory quantitative analysis rather than a source of definitive clinical benchmarks. A systematic search was conducted across five databases up to January 2026. Randomized controlled trials and cohort studies reporting absolute pre- to post-treatment EARR changes in SLB patients were included. Meta-analysis was performed using a random-effects model to calculate pooled mean resorption for maxillary incisors. Subgroup and sensitivity analyses were conducted based on tooth type and imaging modality (2D vs. 3D). Thirteen studies met the inclusion criteria for the qualitative synthesis, of which nine studies were included in the quantitative meta-analysis. For maxillary central incisors (U1), the pooled mean linear resorption was 0.74 mm (95% CI: 0.40, 1.07). Subgroup analysis by imaging modality showed pooled means of 0.59 mm (95% CI: -0.02, 1.20) for 2D studies and 0.84 mm (95% CI: 0.29, 1.39) for 3D CBCT studies, with no statistically significant difference between modalities ( P = 0.55). For maxillary lateral incisors (U2), the pooled mean resorption was 0.47 mm (95% CI: 0.06, 0.87), with 2D and 3D subgroups also showing no significant difference ( P = 0.53). Volumetric analysis revealed a mean loss of 27.83 mm³ (95% CI: 25.48, 30.18). Heterogeneity was consistently high for the linear root length outcomes (I² = 87–94%) but low for the volumetric outcome (I² = 0%), although the latter was derived from only three studies. The overall certainty of evidence for all outcomes, assessed using GRADE, was very low. Treatment with SLBs is associated with a measurable degree of root resorption of approximately 0.74 mm linear shortening and 27.83 mm³ volume loss in maxillary central incisors. However, these pooled estimates should not be interpreted as stable clinical benchmark values, as certainty of evidence was very low, heterogeneity was high, subgroup sizes were small, and some variance estimates depended on imputed change-score standard deviations and methodological limitations including overall high risk of bias in all RCTs and serious risk of bias in all non-randomized studies. While 3D CBCT studies yielded numerically higher pooled estimates than 2D studies in maxillary central incisors, this pattern was not consistent across tooth types, and the difference between imaging modalities did not reach statistical significance in either outcome, reflecting the limited number of CBCT studies available and persistent within-subgroup heterogeneity. The current evidence base remains insufficient to establish definitive 3D benchmarks for orthodontically induced root resorption. Future research should prioritize standardized CBCT protocols to enable more precise quantification of the true magnitude of root loss.