JAAOS - 2026-05-27 - Journal Article
Computed Tomography Overestimates Roentgenographic Posterior Tilt in Geriatric Garden I/II Femoral Neck Fractures.
Greenhill DA, Jain N, Weiss S, Lundy D
Topics
Key Takeaway
CT measured posterior tilt averaged 14° versus 9° on plain radiograph (p<0.001), reclassifying an additional 15% of Garden I/II femoral neck fractures as exceeding the 20° arthroplasty-consideration threshold.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study asked whether posterior tilt measured on CT corresponds to plain radiograph measurement in elderly patients with Garden I/II femoral neck fractures, using paired imaging in 100 patients >65 years. CT consistently overestimated posterior tilt relative to radiograph (14° vs. 9°), and identified 17 patients (17%) above the 20° threshold compared to only 2 (2%) on radiograph. The 15 patients reclassified by CT alone had a mean PT-CT minus PT-XR difference of 24°, a clinically significant discordance with direct arthroplasty-versus-fixation implications.
Key Limitation
No clinical outcomes data are included, so it is unknown whether CT-measured posterior tilt ≥20° carries the same prognostic weight for nonunion and fixation failure as the radiograph-derived threshold on which current practice guidelines are based.
Original Abstract
INTRODUCTION
Current best practice is to consider primary arthroplasty for elderly patients with a Garden I/II femoral neck fracture (FNF) when posterior tilt measures greater than 20° on the lateral radiograph. Although this evidence-based threshold stemmed from plain radiographs, clinicians similarly use it using CT when available. The correspondence of posterior tilt measured on radiograph versus CT is unknown.
METHODS
Patients older than 65 years presenting with a Garden I/II FNF during a 6.5-year period (January 2018 to August 2024) were retrospectively reviewed. Patients without concomitant radiograph and CT of the affected hip or with inadequate radiographs were excluded. Posterior tilt of the FNF was measured on lateral radiographs (PT-XR) and compared with CT (PT-CT).
RESULTS
After review of 1,244 records, a total of 100 patients averaging 82 ± 8 years old met inclusion criteria. Twenty-nine patients underwent internal fixation and 71 hemiarthroplasty. On average, PT-CT was larger than PT-XR (14° vs. 9°; P < 0.001). Lateral radiographs identified two patients (2%) with a posterior tilt ≥20°, while CT identified 17 (17%). For the 15 patients in whom increased posterior tilt was identified only by CT, the mean difference between PT-XR and PT-CT was 24 ± 15°.
CONCLUSION
In patients with Garden I/II FNFs, CT often generates larger posterior tilt measurements than plain radiographs. Using CT, an additional 15% of these patients are identified as having posterior tilt ≥20°. Therefore, advanced imaging of Garden I/II FNFs objectively increases the number of fractures considered too unstable for internal fixation. However, its use as an indication for primary arthroplasty is not yet evidence based.
LEVEL OF EVIDENCE
III.