Radiostereometric analysis of the initial stability of internally fixed femoral neck fractures under differential loading

Abstract

We examined the feasibility of radiostereometric analysis (RSA) in the assessment of the initial stability of internally fixed femoral neck fractures. The study included 16 patients (mean age 73 years). During surgery, multiple RSA-beads were inserted on both sides of the fracture. Radiographs for RSA were taken in the supine position within the first 3 days and 6, 12, 24, and 52 weeks after surgery. To detect any inducible motion at the fracture-site, radiographs for RSA were taken with the patient resting or applying a load through the fracture. Fracture loading was achieved by the patient pressing the ipsilateral foot as much as tolerated on a force plate while providing a counterforce through both hands. Micromotion exceeding the precision values of RSA (≥0.3 mm for the translation vector and/or ≥1.2 degrees for the rotation vector) was considered significant. Permanent three-dimensional fracture-site displacement was also recorded. Voluntary loading induced fracture-site micromotion, which exhibited a dichotomous distribution. In patients with uncomplicated fracture union, inducible micromotion was detectable only at baseline-if at all. Conversely, fractures that developed a nonunion were characterized by the continuation of inducible micromotion beyond baseline. Permanent fracture-site displacement was, on average, nearly an order of magnitude greater than the inducible micromotion. Fracture unions were characterized by the cessation of permanent fracture-site displacement by 12 weeks. Nonunions presented as outliers in permanent fracture-site displacement. Large-scale studies are warranted to evaluate whether the detection of inducible micromotion beyond baseline could serve as an indicator of insufficient fixation stability. © 2018 The Authors. Journal of Orthopaedic Research® Published by Wiley Periodicals, Inc. on behalf of the Orthopaedic Research Society.

Keywords: femoral neck fracture; fracture healing; internal fixation; radiostereometric analysis.

Figure 1

A diagram of patient recruitment and flow through the study.

Figure 2

Paired RSA images (stereoradiographs) taken from a femoral neck fracture fixed with three cannulated compression screws. Yellow circles denote the locations of the inserted RSA tantalum markers in the femoral head and in the trochanteric region. Calibration cage markers are indicated with white circles. The three orthogonal axes (x, y, z) constituting the coordinate system for measuring translation and rotation are given in the black box.

Figure 3

Setup for differentially loaded RSA with the uniplanar technique. Two X‐ray tubes were positioned at a 40° angle to each other in such a way that the X‐ray beams crossed at the site of the femoral neck fracture. A calibration cage, which was placed under the examination table at a fixed height, contained tantalum markers at defined positions to create a 3‐D coordinate system for calculating fracture marker displacement. The X‐ray tubes were operated simultaneously in order to obtain paired images. During differentially loaded RSA, the patient pressed a force plate as much as tolerated with the foot of the operated limb while providing a counterforce with both hands.

Figure 4

The distribution of measured inducible translation and rotation (vectors) during the first 12 weeks of healing. The number of patients with inducible micromotion gradually decreased during healing. Bars represent the number of patients. Dotted lines (red) represent the precision limits of RSA beyond which inducible micromotion was considered significant.

Figure 5

The amount of inducible fracture‐site translation and rotation was only a fraction of the magnitude of permanent fracture‐site displacement. The dots represent the translation and rotation measured in individual patients (n = 13). Bars represent the mean and 95% confidence interval.

Figure 6

The translation and rotation of individual femoral neck fractures (n = 13) from the time of surgery. Outliers are marked in red. Outlier #1 had malpositioned cannulated compression screws, and outliers #13–#16 developed a nonunion or osteonecrosis.