Evaluation of Fracture and Osteotomy Union in the Setting of Osteogenesis Imperfecta: Multicenter Reliability of the Modified Radiographic Union Score for Tibial Fractures (RUST).

BACKGROUND

A critical component of patient care for osteogenesis imperfecta patients is understanding the development of union following fracture and osteotomy surgery. Studies evaluating the criteria of bony union in this population have varied. An objective standardized method for fracture healing was developed as the radiographic union score for tibial fractures (RUST), and subsequently revised to the modified RUST (mRUST). The RUST score has been shown to have excellent interobserver and intraobserver reliability in a single-center investigation in an OI population. The purpose of the study was to evaluate the mRUST reliability in a multicenter investigation.

METHODS

Thirty OI patients with tibial fractures or osteotomy radiographs were identified. Nine observers from 3 institutions measured on 2 separate occasions. Four cortices were scored with the following classification: 1=no callus; 2=callus present; 3=bridging callus; and 4=remodeled, fracture not visible. The modified RUST score ranges from 4 to 16. Interclass and intraclass coefficients (ICC) of 95% CIs were used for interobserver and intraobserver reliabilities.

RESULTS

Interobserver reliability ICC representing the interobserver reliability for the first and second scores was the following, respectively, 0.926 (0.864-0.962) and 0.915 (0.845-0.957). Three reviewers' intraobserver reliability for series 1 and series 2 measurements were 0.860 (0.707-0.934), 0.994 (0.986-0.997), and 0.974 (0.946-0.988).

CONCLUSIONS

Excellent interobserver and intraobserver reliability was demonstrated for the mRUST in the setting of OI bone, indicating the precision of the mRUST for OI tibia healing. Modified RUST application and routine use may help standardize the evaluation of osteotomy and fracture healing in the OI population with further work necessary for it to help make clinical decisions.

LEVEL OF EVIDENCE: Level III-retrospective study of nonconsecutive patients.