Interobserver reliability of biplanar radiography is unaffected by clinical factors relevant to individuals at risk of pathological lower limb torsion

Andrew J. Ries, Elizabeth A. Duffy, Michael H. Schwartz, Tom F. Novacheck, Michael M. Chau

Research output: Contribution to journalArticlepeer-review


Background: Assessments of lower limb torsion are ubiquitous in clinical gait analysis practice as pathologic lower limb rotational deformity may contribute to gait abnormalities, anterior knee pain, as well as other debilitating conditions. Understandably, the overall utility of any torsional assessment is dependent on the measurement method's intrinsic accuracy, precision, and robustness to clinical interference factors. Recently, biplanar radiography (BPR) measurements of torsion have been shown to be both accurate and precise, but the robustness of BPR to potential interference factors is unknown. Research question: How robust are BPR lower limb torsional assessments to six potential interference factors: amount of torsion, skeletal maturity, radiograph quality, prior osteotomy, presence of implants, and observer training background and experience? Methods: In this retrospective cohort study, four observers of diverse backgrounds and experience generated digital 3D reconstructions of 44 lower limbs using BPR images obtained during standard of care visits (age range 7–35 years). From each reconstruction, four lower limb torsional parameters were computed: femoral torsion, femorotibial rotation, tibial torsion, and transmalleolar axis equivalent. The mean absolute deviation (MAD) of each torsional parameter – calculated across the four observers – was used as the measure of reliability and tested against all interference factors. Results: Results demonstrated that the average MAD was 2.1 degrees for femoral torsion, 3.0 degrees for transmalleolar axis equivalent, 3.8 degrees for femorotibial rotation, and 4.7 degrees for tibial torsion. None of the six potential interference factors were found to systematically influence BPR reliability across all four torsional parameters. Of the factors found to statistically influence one or more torsional parameter, none affected MAD values to a clinically meaningful extent. Significance: In addition to being accurate and precise, BPR appears to be robust to several clinical factors relevant to children and young adults with or at risk for pathological lower limb torsion.

Original languageEnglish (US)
Pages (from-to)126-131
Number of pages6
JournalGait and Posture
StatePublished - Feb 2023

Bibliographical note

Funding Information:
This study was supported by the generous donors of Gillette Children’s Specialty Healthcare. Additionally, the Gait and Motion Outcomes Fund of Gillette Children’s Foundation supported EAD’s position. We would like to thank EOS® imaging for performing 3D reconstructions used in this study.

Publisher Copyright:
© 2022 Elsevier B.V.


  • Biplanar radiography
  • Femoral torsion
  • Reliability
  • Tibial torsion
  • Transverse plane rotation

PubMed: MeSH publication types

  • Journal Article
  • Research Support, Non-U.S. Gov't


Dive into the research topics of 'Interobserver reliability of biplanar radiography is unaffected by clinical factors relevant to individuals at risk of pathological lower limb torsion'. Together they form a unique fingerprint.

Cite this