Predictors of Unsuccessful Nonoperative Management of Capitellar Osteochondritis Dissecans. Issue 11 (September 2019)

• Record Type: Journal Article
• Authors: Funakoshi, Tadanao; Furushima, Kozo; Miyamoto, Azusa; Kusano, Hiroshi; Horiuchi, Yukio; Itoh, Yoshiyasu
• Is Part Of: American journal of sports medicine. Volume 47:Issue 11(2019)
• Page Start: 2691
• Page End: 2698
• Publication Date: 2019-09
• Subjects: elbow -- osteochondritis dissecans -- overhead athletes -- baseball; Sports medicine -- Periodicals; Sports injuries -- Periodicals; Orthopedic surgery -- Periodicals; 617.102705
• DOI: 10.1177/0363546519863349
• Languages: English
• ISSNs: 0363-5465

Abstract

Background: Osteochondritis dissecans of the humeral capitellum (capitellar OCD) is a common injury among adolescent throwing athletes. Some younger patients with incomplete maturity of the epiphysis and early-stage capitellar OCD are good candidates for nonoperative treatment. However, during initial examination, predicting the need for surgical treatment in patients with capitellar OCD is difficult. Purpose: To perform multivariate ordered logistic regression analysis of data obtained from patients' medical records and images on initial examination and identify the predictors of unsuccessful nonoperative management of capitellar OCD. Study Design: Cohort study (prognosis); Level of evidence, 2. Methods: This study included 245 elbows with capitellar OCD (all male; mean age, 14 years [range, 10-27 years]). Patients were divided into 3 groups comprising 148 elbows requiring an immediate operation on initial examination, 48 requiring nonsurgical treatment, and 49 requiring an operation after nonoperative management. Baseline data and radiographic parameters, such as lesion location, lesion size, modified Minami classification, radial head size, skeletal age difference between both elbows on initial examination, lesion size on computed tomography, and staging on magnetic resonance imaging, were retrospectively reviewed. Univariate and multivariate ordered logistic regression analyses of spontaneous healing of the lesion were conducted. Results: Univariate logistic regression analysis showed that radial head enlargement and skeletal age difference were significantly associated with spontaneous healing. In multivariate ordered logistic regression analysis, radial head enlargement (anteroposterior and lateral) and skeletal age difference were significant predictors of lack of spontaneous healing (odds ratio [OR], 2.76, P =.025; OR, 7.92, P =.026; and OR, 1.84, P =.0089, respectively). Conclusion: To predict spontaneous healing in the moderate stage, plain radiographs would be important to evaluate radiocapitellar congruity and skeletal age. This study showed that preoperative radiographic findings of radial head enlargement and advanced skeletal age of the throwing side compared with that of the nonthrowing side were predictors of advanced-stage capitellar OCD. Despite several limitations, the statistical significance and correlations herein provide important information on preoperative surgical planning to surgeons.