Can we predict failure of percutaneous fixation of femoral neck fractures?

• Published in: Injury 2020-02, Vol.51 (2), p.357-360
• Main Authors: Kane, Christina, Jo, Jacob, Siegel, Jodi, Matuszewski, Paul E., Swart, Eric
• Format: Article
• Language: English
• Subjects: Fixation failure; Hip fracture; Osteoporosis
• ISSN: 0020-1383; 1879-0267
• DOI: 10.1016/j.injury.2019.10.068

•Fixation failure rate for CRPP of femoral neck fractures is still relatively high, especially in varus/displaced fractures (nearly 40%).•Inter-user agreement for most common radiographic features characteristics is high, but agreement is poor for rating screw placement.•The presence of a medial transcervical line on AP injury radiographs is associated with a fourfold increased risk of early postoperative failure. This study evaluated a series of geriatric femoral neck fracture treated with closed reduction percutaneous pinning (CRPP) at a single level-1 trauma center to determine if there are any simple, reliable, radiographic characteristics that can be used to predict increased risk of post-operative failure in nondisplaced and valgus impacted fracture patterns. We conducted a retrospective cohort study of all patients with femoral neck fractures (AO/OTA 31B) who underwent CRPP over a 12-year period at a single Level 1 trauma center. Failure was defined as radiographic failure within the first year after the index operation requiring revision surgery. Common patterns identified on initial review were the presence of a visible medial transcervical line (MTL) felt to indicate a tension-sided failure, a straight inferior calcar (SIC) indicating severe valgus impaction, and quality of intra-operative screw positioning. X-rays of patients were then reviewed for these characteristics in a blinded manner by three different trauma-fellowship trained orthopedic surgeons. Inter-rater reliability was calculated using Fleiss’ Kappa Coefficient. Comparisons of failure rates between groups were made using a Fisher's Exact test. 139 patients who underwent CRPP for a femoral neck fracture and follow-up for at least 90 days were identified and reviewed. There were a total of 19 failures (13.6%) within one year. The patients with a varus fracture had a failure rate of 9/24 (37.5%). Of the valgus/nondisplaced fractures, MTL was identified in 42/115 (36%) patients. Inter-rater agreement was high for the presence of an MTL (84%, Kappa 0.69). Patients with an MTL had a fourfold increase in risk of failure (7/42=17% with an MTL vs. 3/73=4% without, p  0.03). The presence of a SIC and quality of screw placement were not predictive of failure. Varus femoral neck fractures fixed with CRPP have a high rate of failure (37.5%). Nondisplaced or valgus impacted fractures with the presence of a visible medial transcervical line on pre-operative radiographic imaging resulted in a fourf