The Agreement Between Preoperative Digital Radiographic Measurements and Clinically Fitted Stainless Steel Crown Size in Primary Molars: A Retrospective Study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background/Objectives: Accurate selection of stainless steel crown (SSC) size is essential for successful restoration of primary molars. Conventional crown selection relies on a trial-and-error approach, which may increase chair time and reduce clinical efficiency. Preoperative digital radiographic measurements may provide an objective adjunct for predicting SSC size; however, evidence supporting their agreement with clinically selected crown sizes remains limited. This study aimed to evaluate the agreement between preoperative digital radiographic measurements and clinically fitted SSC size in primary molars, and to assess whether this agreement is influenced by dental arch or tooth type. Methods: This retrospective cross-sectional study analyzed archived patient records and preoperative digital radiographs of pediatric patients who received SSCs on primary molars. Mesiodistal tooth dimensions were measured and converted to SSC sizes according to manufacturer guidelines. The clinically fitted and cemented crown size, as recorded in the patient chart, served as the reference standard against which radiographically predicted size was compared. Intra-examiner reliability was assessed using the intraclass correlation coefficient. Associations were analyzed using chi-square tests and binomial logistic regression (p ≤ 0.05). Given repeated teeth within patients, the primary logistic regression model was additionally re-fit using generalized estimating equations (GEEs) with an exchangeable working correlation structure and robust standard errors, and prediction error was further characterized using near-miss accuracy, direction of bias, weighted kappa, and Bland–Altman analysis. Results: A total of 152 primary molars from 71 patients were included. Intra-examiner reliability was excellent (ICC = 0.99). Radiographic prediction agreed with the clinically fitted crown size in 33.6% of cases, and 69.7% of predictions fell within one crown size of the cemented crown. Agreement was significantly higher for mandibular molars than maxillary molars (p = 0.010) and varied by tooth type (p < 0.001). Mandibular molars were more likely to show accurate prediction (OR = 2.52; 95% CI: 1.23–5.17); this association remained statistically significant after accounting for within-patient clustering using GEEs with an exchangeable correlation structure (OR = 2.46; 95% CI: 1.10–5.49; p = 0.029). Maxillary first molars showed a systematic under-prediction of exactly two crown sizes in 97% of cases (mean bias −2.00), identifying a specific, correctable calibration problem. Conclusions: Preoperative digital radiographic measurements show moderate agreement with clinically fitted crown size and may serve as a useful adjunct to clinical judgment for SSC size selection, particularly for mandibular molars. Because the reference standard reflects clinical judgment itself rather than an independent anatomical gold standard, these findings should be interpreted as a concordance analysis, and further prospective work is needed before efficiency benefits can be confirmed.