JBJS - 2026-09-15 - Journal Article
Classification of Forearm Involvement in Multiple Hereditary Exostoses: Consensus from the CoULD Study Group.
Vuillermin C, Canizares MF, Miller PE, Wall LB, Steinman SE, Van Heest A, CoULD Study Group
Topics
Key Takeaway
The CoULD MHE Classification achieved substantial interrater reliability (kappa=0.64) on refined testing, significantly outperforming the Masada original (kappa=0.21), Masada modified (kappa=0.35), Gottschalk (kappa=0.43), and Jo (kappa=0.43) classifications for forearm deformity in multiple hereditary exostoses.
Summary Depth
Choose how much analysis to show on this article page.
Summary
A 7-round Delphi consensus process among 15 pediatric hand surgeons from the CoULD Study Group was used to develop a new classification for MHE-related forearm deformity, incorporating ulnar variance, radial articular angle, radial head alignment, and osteochondroma size into a 3-tier system with 93% closing consensus. Initial interrater reliability was moderate (kappa=0.55), improving to substantial (kappa=0.64, concordance=0.82) after refinement adding osteochondroma size and radial head subluxation/dislocation subtypes (3A/3B). All reliability metrics significantly exceeded those of the Masada, Gottschalk, and Jo classifications.
Key Limitation
Reliability was assessed on a single cross-sectional radiograph set by only 4 raters; longitudinal reproducibility and correlation with functional outcomes or surgical thresholds remain unvalidated.
Original Abstract
BACKGROUND
Existing classifications for forearm deformity in multiple hereditary exostoses (MHE) have demonstrated low interrater reliability and limited clinical applicability. Our primary research question was whether a Delphi consensus process could generate a more reliable and clinically useful classification for longitudinal surveillance of MHE-related forearm deformities.
METHODS
Fifteen pediatric hand surgeons from the Congenital Upper Limb Difference (CoULD) Study Group participated in 7 rounds of online Delphi surveys (June 2022 to September 2023). An investigator team facilitated the process with anonymized feedback. Round 1 identified key clinical and radiographic features. Subsequent rounds refined these features, determined categorical structure, and piloted the system with representative radiographs. Consensus was defined as ≥70% agreement. After final consensus, interrater reliability was tested by 4 independent surgeons who classified 101 radiographs from 67 patients with forearm MHE. Reliability was assessed with kappa and concordance coefficients and compared with the Masada, Gottschalk, and Jo classifications.
RESULTS
The Delphi process identified radiographic predictors that included ulnar variance, the radial articular angle (RAA), radial head alignment, and osteochondroma size. Consensus favored a 3-tier classification: Type 1, mild; Type 2, moderate; and Type 3, severe. Closing consensus was achieved, with 93% agreement. Reliability testing showed moderate agreement on the first reading (kappa = 0.55), significantly higher than for the Masada (kappa = 0.21 [original classification], kappa = 0.35 [modified version]), Gottschalk (kappa = 0.43), and Jo (kappa = 0.43) classifications. After refinement that incorporated osteochondroma size and the radial head subtypes (3A, subluxation, and 3B, dislocation), agreement improved to substantial (kappa = 0.64) (p < 0.001), with high concordance (0.82).
CONCLUSIONS
The CoULD MHE Classification, developed through expert consensus, integrates validated radiographic predictors with defined thresholds. It demonstrated significantly improved reliability over prior systems and provides a practical framework for longitudinal monitoring of forearm deformities in patients with MHE.
LEVEL OF EVIDENCE
Diagnostic Level V. See Instructions for Authors for a complete description of levels of evidence.