Early Radiographic Healing and Functional Results After Autologous Osteochondral Grafting for Osteochondritis Dissecans of the Capitellum: Introduction of a New Magnetic Resonance Imaging–Based Scoring System

Abstract

Background: Autologous osteochondral grafting (OG) is an option in the treatment of capitellar osteochondritis dissecans (COCD). However, radiographic healing after this procedure has not been well documented. Purpose: To develop a magnetic resonance imaging (MRI)–based scoring system specific for evaluating healing after single-plug OG in COCD and to evaluate correlation between radiographic healing and early clinical outcomes. Study Design: Cohort study (diagnosis); Level of evidence, 3. Methods: Between 2014 and 2017, 183 elbows with COCD were enrolled in a prospective registry. A total of 61 elbows in 59 patients underwent single-plug OG. Of these, 52 elbows in 50 patients had pre- and postoperative MRI scans. Postoperative MRI and clinical outcome data from this group were used to develop the novel BOGIE score (Boston Osteochondral Graft Incorporation in the Elbow), with a possible range of 4 to 12. Results: Median age at surgery was 14.2 years (interquartile range, 13.1-15.0 years). Median clinical follow-up after OG was 12.4 months (interquartile range, 9.5-16.9 months; range, 6-53 months). Compared with before surgery, elbow function at 6 months after surgery and at latest follow-up was significantly improved as measured by the Timmerman and Andrews score (TAS; median: 145 before surgery, 185 at 6 months after surgery, 190 at latest follow-up; P < .001, before vs after surgery), as well as the short version of Disabilities of the Arm, Shoulder and Hand score; median: 21 before surgery, 7 at 6 months after surgery, and 0 at latest follow-up; P < .001 before surgery vs after surgery). Median BOGIE score at 6 months after surgery was 10 (range, 4-12). BOGIE score intraobserver reliability was 0.90 (95% CI, 0.82-0.94) for reader 1 and 0.91 (95% CI, 0.86-0.95) for reader 2. Interobserver reliability between the readers was 0.86 (95% CI, 0.78-0.92). Correlation was observed between the 6-month BOGIE score and the concurrent postoperative objective TAS ( P < .001) as well as total TAS ( P = .01) but not the subjective TAS ( P = .08). Patients who underwent subsequent secondary surgery for persistent symptoms had a significantly lower postoperative BOGIE score at 6 months than those who did not (median, 7.8 vs 10.3; P = .016) Conclusion: Quantitative evaluation for radiologic healing after single-plug OG in COCD is possible. The MRI-based BOGIE score appears to correlate with early clinical function and may be useful as an adjunct tool in decision making on activity progression. The use of a standardized MRI score may improve comparability of outcomes after OG in the literature.