Revised modified oxford bone score: A simpler systemfor prediction of contralateral involvement in slipped capital femoral epiphysis

Background: The decision to prophylactically treat the uninvolved hip in slipped capital femoral epiphysis (SCFE) remains controversial. The modified Oxford bone score (mOBS) is predictive of future contralateral involvement in slipped capital femoral epiphysis. The scoring method for this system is challenging to remember because out-of-context irrational number sequences and total score range (16 to 26) are used. This study was performed to evaluate intraobserver and interobserver reliability of the mOBS and to determine whether revising the scoring sequence to 0 to 2 for all 5 categories (total score range: 0 to 10) would be easier for orthopedic surgeons to remember. Methods: Six orthopedic surgeons scored 30 normal pelvis radiographs using the mOBS (original or revised scoring system) on 2 separate occasions, at least 2 weeks apart, with the aid of reference diagrams and an explanatory key. At a later date, the observers were asked to complete blank reference diagrams from memory for both scoring systems (16 to 26 and 0 to 10). Results: Intraobserver reliability was analyzed for each parameter independently and as a total score. Overall, intraobserver reliability was excellent, with total scores being within 1 and 2 points of each other 80.5% and 94.9% of the time, respectively. Interobserver reliability was very good, with total scores within 1 and 2 points of each other 69.6% and 87% of the time, respectively. None of the 6 observers were able to complete the blank mOBS key correctly from memory, despite being reminded of the 16 to 26-point range. Five of the 6 were able to correctly complete the revised key using the 0 to 10 point range system. Conclusions: The mOBS is a useful method to estimate risk of contralateral slip, with excellent intraobserver and very good interobserver reliability. Difficulty in remembering the original scoring scheme because of its illogic sequences in the modified method limits its clinical applicability. Revision of the mOBS to a consistent 0 to 2 (range: 0 to 10) system greatly enhanced the observers ability to recall the scoring system. Level of evidence: Level II - development of diagnostic criteria with reference to "gold" standard.