Purpose: The Wilkins-modified Gartland classification of pediatric supracondylar humerus fractures does not consider coronal or sagittal obliquity. The purposes of our study were (1) to identify and describe fracture characteristics with unique properties and (2) to propose a fracture classification system that can be reproduced reliably. Methods: We retrospectively studied 203 consecutive displaced pediatric extension-type supracondylar humerus fractures treated operatively from January 1998 to January 2003. Fracture characteristics (eg, coronal and sagittal obliquity, postoperative alignment), type of surgical treatment, outcome, and complications were assessed and analyzed statistically with Student t test and a receiver operating characteristic curve. Significance was defined as P < 0.05. We incorporated significant cutoff values for fracture obliquity into our classification scheme and tested the classification's interobserver and intraobserver reliability. Results: We identified 4 coronal (typical transverse, medial oblique, lateral oblique, and high fractures) and 2 sagittal (low sagittal and high sagittal) subtypes with significantly different characteristics and outcome. Compared with fractures with coronal obliquity of less than 10 degrees, fractures with coronal obliquity of 10 degrees or greater were associated with significantly more comminution and rotational malunion. Compared with fractures with sagittal obliquity of less than 20 degrees, fractures with sagittal obliquity of 20 degrees or greater were associated with a significantly higher incidence of additional injuries and were more likely to result in extension malunion. Analysis of the interobserver and intraobserver reliability for our system identified correlation coefficients ranging from 0.772 to 0.907 and 0.860 to 0.899, respectively. Conclusions: Because pediatric extension-type supracondylar humerus fractures vary significantly in terms of characteristics, identification of sagittal oblique and coronal oblique angles may have an important role in surgical decision making and may impact outcomes. Level of Evidence: Level 3 (retrospective study).
