ePoster #710 - ISHA Annual Scientific Meeting 2016

Is Three-Dimensional Imaging Better Than Plain Radiographs For Defining Pincer Femoroacetabular Impingement Subtypes?

Jeffrey J. Nepple, MD, St. Louis, MO UNITED STATES
James R. Ross, MD, Deerfield Beach, FL UNITED STATES
Asheesh Bedi, MD, Ann Arbor, MI UNITED STATES
John Clohisy, MD, St. Louis, MO UNITED STATES

Washington University School of Medicine, St. Louis, MO, UNITED STATES

FDA Status Cleared

Summary: Three-dimensional imaging allows for more accurate and precise characterization of pincer-type morphologies in patients with FAI, compared to plain radiographs.

Abstract:
Introduction: Pincer-type impingement is thought to commonly occur in combination with cam-type femoroacetabular impingement (FAI). However, recent studies have suggested a high incidence of false-positive findings of retroversion on plain radiographs. The purpose of the current study was to determine if well-positioned plain radiographs appropriately identify subtypes of pincer-type FAI by direct comparison to three-dimensional imaging.

Methods: Twenty-eight consecutive patients undergoing primary surgical treatment of FAI by a single surgeon were included. Low-dose computed tomography (CT) scans were obtained for preoperative planning and allowed for construction of well-positioned, idealized AP pelvis radiographs. Pincer subtypes were defined as focal anterosuperior overcoverage [positive crossover sign (COS), negative posterior wall sign (PWS)], acetabular retroversion (positive COS, positive PWS), or global overcoverage (negative posterior wall sign, lateral center edge angle > 40°). Analysis was performed with the pelvis in its native position on CT. CT-based three-dimensional modeling was utilized to assess acetabular coverage at 48 points circumferentially around the acetabular rim. Overcoverage was defined as three or more consecutive clock face locations of femoral head coverage that was greater than two standard deviations above normative data.

Results: Based on plain radiographs, pincer morphology was noted in 17/28 (60.7%) hips, including 7 focal anterosuperior overcoverage, 5 retroversion, and 5 global subtypes. Re-evaluation based on three-dimensional modeling of the 17 hips with pincer morphology on plain radiographs confirmed pincer type morphology in only 13/17 (76.5%). CT assessment of these hips included 6 with focal anterosuperior overage, 3 with acetabular retroversion, and 4 with global overcoverage. Three hips with a crossover sign on plain radiographs failed to demonstrate pincer-type morphology on three-dimensional imaging due to posterior acetabular undercoverage resulting in the appearance of a crossover sign with normal anterior acetabular coverage. Two additional hips demonstrated a crossover on plain radiographs without anterosuperior overcoverage present on CT secondary to a prominent anterior inferior iliac spine morphology. Five hips characterized as global overcoverage on plain radiographs demonstrated variable findings on CT, including 2 with true global overcoverage, 1 posterolateral overcoverage, 1 anterosuperior overcoverage, and 1 isolated lateral overcoverage. Additionally, one hip with a normal appearance on plain radiographs demonstrated focal anterosuperior overcoverage on CT and four hips demonstrated anterosuperior undercoverage without the characteristic radiographic appearance of acetabular dysplasia on plain radiographs.

Discussion: Three-dimensional imaging allows for more accurate and precise characterization of pincer-type morphologies in patients with FAI, compared to plain radiographs. The presence of a crossover sign on plain radiographs may be due to posterior undercoverage in some patients, rather than anterosuperior overcoverage. Furthermore, absence of a crossover sign on plain radiographs may not preclude the presence of pincer-type or dysplastic acetabular morphology.