Responsiveness of three measurements in cone-beam computed tomography transverse analyses during both tooth-supported and mini-screw-assisted rapid maxillary expansion

Abstract

ABSTRACT Objective To evaluate the responsiveness of three cone-beam computed tomography (CBCT) transverse analyses (University of the Pennsylvania [UPenn] analysis, Boston University analysis and Yonsei University [YU] analysis). Materials and Methods A consecutive cohort sample of patients was retrospectively reviewed for eligibility. CBCT records before treatment (T0) and immediately after maxillary expansion (T1) of 71 patients receiving tooth-supported rapid maxillary expansion (RME) and 57 patients receiving mini-screw-assisted RME (MARME) were finally analyzed. Responsiveness was assessed by comparing changes of measures (T1-T0) to mid-palatal suture opening distance (MSOD) at T1. Correlational responsiveness was assessed by Pearson correlation coefficient (r). Absolute agreement responsiveness was assessed by Bland-Altman analysis. A specialized intraclass correlation coefficient (ICC) was selected to assess responsiveness combining correlation and absolute agreement. Results Changes of all three measures were moderately to strongly correlated to MSOD (r > 0.5). The highest correlation coefficient (0.79) was found between the YU analysis and MSOD. When exploring absolute agreement responsiveness, the smallest deviation (0.14 mm) was observed in the UPenn analysis. For ICC, the highest ICC value (0.63) was observed when the YU analysis was used. In addition, all three measurements were more responsive to MSOD in the MARME group than to those in RME group. Conclusions All three transverse measurements responded well to true changes of maxillary transverse deficiency during both tooth-supported and mini-screw-assisted RME. Deviations of responsive properties of these measurements from true skeletal changes were below a clinically meaningful level (1 mm).