Low interobserver and intraobserver reliability using the Matta radiographic system for intraoperative assessment of...

The system described by Matta for rating acetabular fracture quality of reduction following ORIF has been used extensively throughout the literature. However, the reliability of this system remains to be validated. We sought to determine the interobserver and intraobserver reliability of this system when used by fellowship-trained pelvic and acetabular surgeons to evaluate intraoperative fluoroscopy.

This is a retrospective evaluation of a prospectively collected acetabular fracture database at an academic level I trauma center. The quality of reduction of all acetabular fractures treated with open reduction internal fixation (ORIF) between May 2013 and December 2015 was assessed using three standard intraoperative fluoroscopic views (anteroposterior and two 45˚ oblique Judets). Displacement of ≤1 mm was considered to be an anatomic reduction, 2-3 mm imperfect, and >3 mm poor according to the system described by Matta. A total of 107 acetabular fractures treated with ORIF with complete intraoperative fluoroscopic images during that time period were available for review. Acetabular fracture reductions were reviewed by the operative surgeon at the time of surgery and subsequently reviewed by two fellowship-trained pelvic and acetabular surgeons. All reduction assessments were performed in a blinded fashion. The primary outcome measure was interobserver reliability for assessing reduction quality. This was evaluated using a weighted kappa (κ_w) statistic between each evaluator and the operative surgeon and a generalized kappa (κ_g) for all 3 surgeons. After a 6-week "washout interval," the surgeons reviewed the images again and intraobserver agreement was calculated using a weighted kappa statistic.

Interobserver reliability based on the initial assessment was low (κ_g = 0.09); however, did slightly improve with the second assessment to fair (κ_g = 0.24). Intraobserver reliability ranged from slight (κ_w = 0.20) to moderate (κ_w = 0.53) among the surgeons.

Low interobserver and intraobserver reliability was found when quality of reduction was assessed with intraoperative fluoroscopic images by the operative and two other pelvic and acetabular surgeons using the Matta system. Given the importance of an anatomic reduction on functional and radiographic outcomes, an accurate and reliable system for assessing intraoperative quality of reduction is essential.