Prediction of Radial Head Subluxation and Dislocation in Patients with Multiple Hereditary Exostoses

David S Feldman; Troy J Rand; Jaroslaw Deszczynski; Tomasz Albrewczynski; Dror Paley; Aaron J Huser

doi:10.2106/JBJS.20.02205

Prediction of Radial Head Subluxation and Dislocation in Patients with Multiple Hereditary Exostoses

J Bone Joint Surg Am. 2021 Aug 25. doi: 10.2106/JBJS.20.02205. Online ahead of print.

Authors

David S Feldman^{1

2}, Troy J Rand¹, Jaroslaw Deszczynski³, Tomasz Albrewczynski², Dror Paley^{1

2}, Aaron J Huser¹

Affiliations

¹ Paley Advanced Limb Lengthening Institute, St. Mary's Hospital, West Palm Beach, Florida.
² Paley European Institute, Medicover Hospital, Warsaw, Poland.
³ Department of Orthopedics and Rehabilitation, Warsaw Medical University, Warsaw, Poland.

PMID: 34432742
DOI: 10.2106/JBJS.20.02205

Abstract

Background: Multiple hereditary exostoses (MHE) is a rare bone disease that results in growth of benign cartilage-capped tumors and a number of skeletal deformities. Forearm deformities are present in up to 60% of patients with MHE, and radial head subluxation or dislocation occurs in 20% to 30%. Radial head subluxation/dislocation results in a shortened forearm and loss of motion. The purpose of this study was to identify radiographic variables that are most predictive of radial head subluxation/dislocation in an effort to determine the need for prophylactic treatment.

Methods: We retrospectively reviewed the cases of consecutive patients with MHE treated in our center between April 2007 and December 2019. Radiographic measurements included the presence or absence of distal ulnar osteochondromas, total ulnar bow, total radial bow, and percent ulnar length. Participants were separated into 3 groups based on the status of the radial head: located, subluxated, and dislocated. Radiographic measurements were compared using a Kruskal-Wallis H test with Dunn post-hoc analysis. A prediction model was run using a binomial logistic regression, and a prediction matrix was created.

Results: A total of 88 patients were included in the study. There were significant differences in the located group compared with the dislocated group in terms of pronation, supination, and extension. The percent ulnar length, total ulnar bow, and total radial bow differed significantly between the located and dislocated groups (p < 0.0001); however, in the binomial regression analysis, only the percent ulnar length and total ulnar bow could be used to distinguish between the located group and the subluxated/dislocated group. Both of these measurements were significant predictors of subluxation/dislocation. There was no radial head subluxation/dislocation in patients with an ulnar bow of <17°.

Conclusions: The data indicate that total ulnar bow and percent ulnar length are good predictors of radial head subluxation/dislocation. These 2 parameters can be utilized to monitor forearm deformity and guide timing for prophylactic treatment.

Level of evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.