Non-Operative Versus Operative Treatment of Z-Type Comminuted Clavicle Fractures in Adolescents: A Prospective Sub-Stratified Cohort Analysis

Objectives: To compare the clinical, radiographic, and patient-reported outcomes of non-operative and operative treatment of adolescents with comminuted ‘Z-type’ mid-shaft clavicle fractures. Methods: Design: Prospective observational cohort Setting: Eight tertiary-care pediatric centers Patient Selection Criteria: Patients 10 to 18 years-old who were treated either operatively or non-operatively for a diaphyseal clavicle fracture between 2013 and 2017 were screened/enrolled at the time of injury. The current sub-cohort analysis was derived from a larger adolescent clavicle study population of 907 patients. Outcome Measurements and Comparisons: Complications and validated patient-reported outcome measures (PROs):(ASES, QuickDASH, Marx Shoulder Activity Score, EQ-5D, EQ-VAS, and patient satisfaction score) were compared between operative and non-operative cohorts. Results: Eighty-one patients (69 male (85.2%), 12 female; average age 15 years (11.1-18.7); 78 with sports participation (96.2%)), were followed through bony healing and return to sports, while 59 patients (73%) completed two-year follow-up with patient-reported outcomes (PROs), 26 of whom were treated non-operatively and 33 treated operatively. All demographic and fracture characteristics were similar (p>0.05) between the two-year follow-up cohorts except for fracture shortening, which was greater in the operative cohort (23 mm vs 29 mm, p=0.01). After controlling for this potential confounder through both regression and propensity matched sub-group analysis, non-operative versus operative cohorts showed no difference in rates of nonunion (0%), delayed union (0% vs. 2.3%, p=1.0), symptomatic malunion (2.7% vs 0%, p=0.4), refracture (2.7% vs 2.2%, p=1.0), unexpected subsequent surgery (5.4% vs. 11.4%, p=0.45), or clinically significant complications (5.4% vs. 16%, p=0.17). There were no differences in any PROs between cohorts, both before and after controlling for the difference in fracture shortening (all p-values >0.05). Conclusions: In this prospective comparison of complications and two-year PROs in adolescents with comminuted Z-type clavicle fractures, non-operative and operative treatment yielded similar outcomes. Level of Evidence: Therapeutic Level II. See Instructions for Authors for a complete description of levels of evidence.