Abstract

To quantify the incidence of lumbopelvic instability in the setting of unilateral and bilateral sacral fractures and assess whether the presence of bilateral sacral fractures on axial imaging is a useful screening test for lumbopelvic instability.Retrospective case series.Level I trauma center at an academic medical center.A hospital database was used to identify patients diagnosed with a sacral fracture by The International Classification of Diseases, Ninth Revision (ICD-9) code from 2000 to 2014.Axial cross-sectional imaging was reviewed to confirm the presence of unilateral or bilateral sacral ala fractures. Sagittal reconstructions were scrutinized for a transverse fracture line separating the lumbar spine from the pelvis, which was used to define lumbopelvic instability.The Roy-Camille classification system was applied to all identified cases of lumbopelvic instability.One thousand five hundred twenty-six patients were diagnosed with sacral fractures by the ICD-9 code. Four hundred ninety had adequate axial and sagittal cross-sectional imaging. Four hundred forty-three of these patients had unilateral sacral ala fractures, and none of these were associated with lumbopelvic instability. Forty-seven patients had bilateral sacral ala fractures, and 41 of these (87%) had a transverse component indicating some degree of lumbopelvic instability. The presence of bilateral sacral fractures was 100% sensitive and 99% specific for lumbopelvic instability. Among fractures with lumbopelvic instability, 27 (66%) were Roy-Camille type 1, 11 (27%) were type 2, and 3 (7%) were type 3.Bilateral sacral ala fractures are strongly associated with lumbopelvic instability and can be used as a very sensitive and specific screening tool. All patients with bilateral sacral fractures on axial computed tomography or magnetic resonance imaging should have close assessment of the sagittal plane images to evaluate for this pathology.Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.

View details for DOI 10.1097/BOT.0000000000000972

View details for PubMedID 29189523