Elsevier

The Knee

Volume 21, Issue 2, March 2014, Pages 549-552
The Knee

Do long leg supine CT scanograms correlate with weight-bearing full-length radiographs to measure lower limb coronal alignment?

https://doi.org/10.1016/j.knee.2013.05.009Get rights and content

Abstract

Introduction

The gold standard for measuring knee alignment is the lower limb mechanical axis (MA) using weight-bearing lower limb full-length x-ray (FLX). However, CT scanograms (CTS) are becoming increasingly popular in view of lower radiation exposure, speed of data acquisition and supine positioning. We compared the correlation and degree of agreement of knee joint coronal alignment using these two imaging modalities.

Method

From our series of complex primary and revision knee arthroplasty patients, we selected those with both FLX and CTS recorded onto digital PACS. The coronal alignments were assessed in 24 knees and the valgus/varus angles relative to the MA were measured. Results were analysed statistically using the paired samples t-test, Pearson's correlation coefficient, intra-class correlation coefficient, Cohen's kappa and Passing and Bablok regression to assess potential equality of methods.

Results

The mean MA was 180.5° (165°–200°) for the CTS and 181° (164°–202°) for the FLX. The CTS MA angle data between the assessors were highly correlated (r = 0.971, p < 0.001) as were FLX MA angle measurements (r = 0.988, p < 0.001). 41.7% of the CTS and 37.5% of the FLX were in varus alignment, while 50% of the CTS and 43.8% of the FLX were in valgus alignment. Malalignment > 5° was revealed by 18.8% of the CTS and 35.4% of the FLX.

Conclusion

Overall, good agreement was observed in MA angle data between the two imaging modalities, but reproducibility may be problematic. In the malaligned limb, weight-bearing FLX still remains a vital imaging modality. CTS should be used with caution in view of the under-detection of malalignment.

Introduction

The population of people over the age of 65 years is estimated to double in the next 30 years [1], and with it, the prevalence of knee osteoarthritis and arthroplasty is likely to increase. The total number of knee arthroplasty procedures entered into the National Joint Registry (NJR) during 2010 was 81,979, an increase of 5.7% compared with 2009 [2].

Knee alignment is a major determinant of load distribution through the knee and is thought to play a role in disease progression of osteoarthritis (OA), as well as the outcome of knee replacement [3], [4], [5]. In knee OA, varus alignment may increase the risk of medial progression, while valgus alignment increases the risk of lateral progression with a resultant decline in function [2], [6]. In knee arthroplasty, varus malalignment of the tibial component increases stress on the polyethylene insert predisposing to wear [7], while valgus malalignment results in asymmetrical forces on the proximal tibia [8]. Malalignment can also cause patella mal-tracking [9]. Malalignment of greater than 5° in either varus or valgus direction has been shown to be associated with greater functional decline when compared with knees with less malalignment [3].

The assessment of mechanical axis (hip–knee–ankle angle) of the lower limb is regarded as the gold standard method of measuring knee alignment [10], [11], [12]. This can be achieved using full-length lower limb x-rays or the CT scanogram. The full-length x-rays (FLX) allows for weight bearing but it involves significant radiation exposure. Another potential problem is in the case where the patient has difficulty in weight bearing or is wheelchair bound. Much less irradiation is generated by the CT scanogram (CTS), and patients are positioned supine; however, knee alignment may not be well represented in this position.

The aim of the present study was to compare the use of these two imaging modalities in assessing lower limb coronal alignment; the extent of agreement between modalities and the level of assessor reliability (inter and intra-observer) were assessed.

To our knowledge there are no published studies comparing full-length lower limb x-rays with CT Scanograms.

Section snippets

Cases

The mechanical axis (MA) angles for a total of 24 cases were assessed (24 FLX with the corresponding 24 CTS) in 15 patients. The mean patient age was 59 years (range 40–90 years). These patients had undergone imaging for alignment assessment in view of complex degenerative deformity or to plan for revision TKA. Knees with FLX images with corresponding CTS images, images with visible landmarks for measurement, well positioned prostheses (in the TKA cases), absence of fixed flexion deformity and

Results

The CTS mechanical axis (MA) angle data for assessor 1 and assessor 2 is highly correlated (r = 0.971, n = 24, p < 0.001), showing good agreement (ICC = .971, p < 0.001) and analysis using the paired samples t-test does not indicate a statistically significant mean difference between the two assessors (t = 1.387, df = 23, p = 0.179).

The assessors displayed good agreement with respect to identification of malalignment using CTS with agreement on classification occurring on 21 of the 24 instances (Cohen's kappa = 

Discussion

Knee alignment is important in the assessment of the status and progression of knee pathology. It is especially important in preoperative planning, post-operative assessment and functional outcome of knee arthoplasty. The determination of knee alignment is also of vital use in many knee scoring systems.

Our study demonstrates a good correlation between full-length weight-bearing x-rays and CT scanograms. However, the conflicting results from the Passing and Bablok regression analyses also

Conflict of interest

We declare there was no conflict of interest in relation to this project.

Acknowledgements

We thank James Bonner (CT Superintendent) as well as the Research and Innovation Department of the North Bristol NHS Trust for their help and advice on this project.

References (22)

  • T.P. Andriacchi

    Dynamics of knee malalignment

    Orthop Clin North Am

    (1994 Jul)
  • T.M. Van Raaij et al.

    Conventional knee films hamper accurate knee alignment determination in patients with varus osteoarthritis of the knee

    Knee

    (2009)
  • S.L. Weinstein

    "2000–2010: the bone and joint decade"

    J Bone Joint Surg. Am.

    (2000)
  • National Joint Registry

    8th Annual Report 2011

  • L. Sharma et al.

    The role of knee alignment in disease progression and functional decline in knee osteoarthritis

    JAMA

    (2001)
  • H. Bäthis et al.

    Alignment in total knee arthroplasty. A comparison of computer-assisted surgery with the conventional technique

    J Bone Joint Surg Br

    (2004 Jul)
  • G.M. Brouwer et al.

    Association between valgus and varus alignment and the development and progression of radiographic osteoarthritis of the knee

    Arthritis Rheum

    (2007)
  • Liau JJ, Cheng CK, Huang CH, Lo WH

    The effect of malalignment on stresses in polyethylene component of total knee prostheses–a finite element analysis

    (2002 Feb)
  • H.P. Hsu et al.

    Effect of knee component alignment on tibial load distribution with clinical correlation

    Clin Orthop Relat Res

    (1989 Nov)
  • S. Matsuda et al.

    Effect of femoral and tibial component position on patellar tracking following total knee arthroplasty: 10-year follow-up of Miller-Galante I knees

    Am J Knee Surg

    (2001 Summer)
  • J.H. Lonner et al.

    Effect of rotation and knee flexion on radiographic alignment in total knee arthroplasties

    Clin Orthop Relat Res

    (1996)
  • Cited by (39)

    • Knee Malignment: Is There a Role for Correction in Primary ACLR?

      2021, Operative Techniques in Sports Medicine
    • Comparison of weight-bearing full-length radiographs and computed-tomography-scan-based three-dimensional models in the assessment of knee joint coronal alignment

      2020, Knee
      Citation Excerpt :

      Gbejuade et al. [24] compared the coronal alignment using weight-bearing LLRs and CT scanograms in 24 cases (the authors recognize the sample size as a limitation of the study) and concluded that overall, good agreement was observed in HKA angle among the two imaging modalities. However, specifically observing patients with malalignment of more than five degrees, Gbejuade et al. [24] concluded that reproducibility might be problematic and CT scans should be used with caution. Holme et al. [25] found significant differences among LLRs and CT-scan measurements of the mechanical axis in a sample of 23 unicompartmental knee arthroplasties and 17 native knees.

    • Computed Tomography Techniques Help Understand Wear Patterns in Retrieved Total Knee Arthroplasty

      2018, Journal of Arthroplasty
      Citation Excerpt :

      Although it has been demonstrated to be more reliable and accurate than plain radiographs, it was not possible to obtain images of the leg in weight-bearing position. Gbejuade et al [35] demonstrated good agreement between weight-bearing lower limb full-length x-rays and CT scans; however, in case of non–weight-bearing malalignment, it may be underdetected, especially in cases of pronounced laxity and instability. Recently, an innovative imaging technique, the EOS 2D/3D imaging system, was used in TKA for studying implant positioning and was validated for knee axis measurement after TKA [36–38]; it uses ultrasensitive multi-wire proportional chamber detector to detect X-rays, which allows to scan patients in standing position.

    View all citing articles on Scopus
    View full text