JGJulien Galley

Julien Galley

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Julien Galley has a total of 27 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • What Is the Validity of Traction MR Arthrography for Detecting Lesions of the Ligamentous-Fossa-Foveolar Complex in Patients Undergoing Joint-preserving Surgery Through Surgical Hip Dislocation?

    by Vera M. Stetzelberger, Ester Brosio, Jonathan Laurençon, Darius Marti, Florian Schmaranzer, Julien Galley, Daniel Guntern, Joseph M. Schwab, Moritz Tannast

    Background: Accurate preoperative characterization of intraarticular abnormalities is important when planning hip-preserving surgery in young, active patients. Lesions of the ligamentum teres and the associated acetabular fossa and fovea capitis (forming the ligamentous-fossa-foveolar complex [LFFC]) are frequently observed intraoperatively, although their relationship with hip pain in patients without osteoarthritis remains uncertain. These lesions have received limited attention in preoperative imaging. Questions/purposes: (1) What is the intraobserver and interobserver reliability of traction MR arthrography for assessing lesions of the LFFC? (2) What is the diagnostic performance of traction MR arthrography for detecting these lesions compared with intraoperative findings at surgical hip dislocation? Methods: Between April 2019 and April 2022, we retrospectively identified 214 hips from a consecutive cohort of patients undergoing surgical hip dislocation at a single tertiary hip preservation center. We performed this surgery for a variety of indications, including for symptomatic femoroacetabular impingement, childhood hip diseases, and for surgically treating acetabular fractures. During this period, preoperative MR arthrography was part of the routine preoperative work-up in all patients for assessment of intraarticular abnormalities, planning for intraoperative instrument and graft needs, and prognostic purposes. MR arthrography results were not considered when choosing surgical approach. Patients without traction MR arthrography were excluded. A total of 70 patients were available for final analysis. Traction MR arthrography served as the index test and intraoperative assessment during surgical hip dislocation as the reference gold standard. The decision to perform surgical hip dislocation was based on the clinical and standard radiographic evaluation and generally not influenced by traction MR arthrography findings. Six observers (three board-certified orthopaedic surgeons with expertise in hip-preserving surgery and three board-certified radiologists with specific expertise in musculoskeletal imaging) independently conducted measurements twice using an adapted version of a previously validated grading system (grade 0, normal; grade I, degeneration; grade II, partial defect; grade III, complete defect); for diagnostic performance, findings were dichotomized as lesion absent (grade 0) or present (grade > 0). Intraclass correlation coefficients (ICCs) with 95% confidence interval (CI) were calculated to assess intraobserver reliability and interobserver reproducibility of the adapted grading system. The diagnostic performance of traction MR arthrography for detecting lesions of the LFFC using intraoperative findings as a gold standard was determined by calculating sensitivity, specificity, positive and negative predictive values, accuracy, and positive and negative likelihood ratios. Results: Intraobserver reliability ICCs ranged from 0.54 to 0.69 for ligamentum teres lesions, 0.21 to 0.64 for fossa lesions, and 0.30 to 0.64 for fovea lesions. Interobserver reproducibility ranged from 0.71 to 0.84 for fossa lesions, 0.73 to 0.77 for fovea lesions, and 0.78 to 0.79 for ligamentum teres lesions. Traction MR arthrography demonstrated limited diagnostic performance for lesion detection. The sensitivity was 86% for ligamentum teres lesions, 79% for fossa lesions, and 75% for fovea lesions, whereas specificity was low at 32%, 52%, and 50%, respectively. Overall accuracy ranged from 54% to 67%. Conclusion: In this cohort of patients undergoing surgical hip dislocation, contrast-enhanced traction MR arthrography demonstrated limited ability to detect and grade lesions of the LFFC preoperatively. Future prospective studies using standardized imaging protocols are needed to determine whether diagnostic performance can be improved. Level of Evidence: Level III, diagnostic study.

  • Open Access

    Validation of CLASS MRI for personalized ACL footprints identification

    by Grégoire Thürig, Marc Barrera Usó, Raúl Panadero‐Morales, Julien Galley, Joseph Schwab, Alexander Heimann, Moritz Tannast, Daniel Petek

    Purpose: In modern anterior cruciate ligament (ACL) surgery, the focus is usually on anatomical reconstruction to restore the natural kinematics of the knee. The individual optimal positioning of the ACL footprints (FPs) in primary surgery is still controversial and, especially in revision surgery, difficult to realize surgically. In this regard, a new MRI‐based sequence, the Compressed Lateral and anteroposterior Anatomic Systematic Sequence (CLASS) with marked femoral and tibial FPs as a template, could help. The purpose of this study was to (1) validate the reliability and reproducibility of the localization of femoral and tibial FPs of ACL in the generation of CLASS and (2) compare the identification of ACL FPs by CLASS with previously described methods.Methods: Magnetic resonance imaging (MRI) of uninjured knees from a predominantly young cohort is used to apply the CLASS algorithm. ACL FPs were subsequently identified by a board‐certified radiologist and an orthopaedic knee surgeon. Intraobserver reliability and interobserver reproducibility were assessed. Measurements of the ACL FPs according to established methods were performed and compared with the results from the literature.Results: Identification of ACL FPs and generation of CLASS images resulted in ‘almost perfect’ reliability and reproducibility. Most measurements also showed ‘almost perfect’ consistency. Statistical analysis showed significant variations between the deep‐shallow and high‐low positions when compared to the published literature.Conclusions: The CLASS MRI sequence is a reliable and reproducible method for identifying ACL FPs. The observed variability in the location of the ACL FP underlines the importance of a patient‐specific surgical approach.Level of Evidence: Level II.


  • Open Access

    The Vulcan salute sign: a non-sensitive but specific sign for Morton’s neuroma on radiographs

    by Julien Galley, Reto Sutter, Christoph Germann, Christian W. A. Pfirrmann

    Objectives: To assess the value of the divergence of toes on conventional radiographs of the foot for diagnosing Morton’s neuroma. Methods: This retrospective case–control study was approved by the local ethics committee. In 100 patients with MRI-proven Morton’s neuroma 2/3 or 3/4 (study group) and 100 patients without (control group), conventional weight-bearing dorso-plantar view radiographs were evaluated for the subjective presence of interphalangeal divergence, called the Vulcan salute sign or V-sign, by two blinded, independent musculoskeletal radiologists. Interphalangeal angles (2/3 and 3/4) and intermetatarsal angle I/V were measured. The t test and chi-squared test were used to compare the groups. Diagnostic performance was calculated. Interobserver reliability was assessed using κ statistics and intraclass correlation coefficient (ICC). Results: The difference between the groups was significant (P < 0.05) regarding the presence of the V-sign, which was found in 30 of 100 patients with Morton neuroma and in 3 of 100 control patients, with a sensitivity of 30% and a specificity of 97%. The differences between interphalangeal angles were significant (P < 0.05) between the groups. The interphalangeal angle 2/3 mean values were 7.9° (± 4.8) for the study group vs 5.4° (± 2.6) for the controls; the 3/4 angle values were 6.5° (± 3.8) and 3.4° (± 2.5), respectively. There was no significant difference between the groups in the intermetatarsal angle I/V. Interobserver agreement was substantial for the V-sign, with a κ value of 0.78. The ICC was excellent concerning angle measurements, with all values ≥ 0.94. Conclusion: The Vulcan salute sign on conventional radiographs is specific for Morton’s neuroma.


  • High-resolution in vivo MR imaging of intraspinal cervical nerve rootlets at 3 and 7 Tesla

    by Julien Galley, Reto Sutter, Christoph Germann, Florian Wanivenhaus, Daniel Nanz

    Objectives No routine imaging technology allows reliable visualization of nerve rootlets inside the spinal canal with positive contrast. The stronger MR signal at 7 T, with optimized protocols, may offer a solution. The purpose was to evaluate the potential of 3D Dual-Echo Steady-State (DESS) MR imaging of the cervical spine at 3 and 7 T in assessing the micro-anatomy of the nerve rootlets. Materials/methods This prospective study was approved by the local ethics committee. Twenty-one patients, clinically referred to cervical-spine MRI, underwent additional MR exams at 3 T and 7 T, each of which consisted of a single 3D-DESS series with equal acquisition times. Artifacts, visualization quality, and number of identified rootlets (C2 to C8) were rated by two musculoskeletal radiologists. Results were compared by Wilcoxon tests. Interobserver reliability was assessed using weighted κ statistics and intraclass correlation coefficient (ICC). Results Intraspinal rootlets could successfully be visualized at both field strengths. Rating differences for artifacts and quality of rootlet depiction were not significant for the two field strengths. The mean number of identified rootlets was larger for 7-T than for 3-T MR for every assessed nerve; however, this difference was not statistically significant using the Bonferroni correction (p values ranging from 0.002 to 0.53). Interobserver agreement was substantial to almost perfect (weighted κ values of 0.69 and 0.82). The ICC for the number of identified rootlets was 0.80. Conclusion Non-invasive 3D-DESS MR-imaging at 3 and 7 T has the potential to provide precise assessments of the micro-anatomy of intraspinal cervical nerve roots. Key Points • Cervical rootlets can be successfully visualized with positive contrast using 3D-DESS MR-imaging. • 3D-DESS MR-imaging at 3 and 7 T provides precise assessments of the micro-anatomy of cervical nerves. • The mean number of identified cervical rootlets using 3D-DESS was larger for 7 T than for 3 T MR; however, this difference was not statistically significant.


  • Diagnosis of Periprosthetic Hip Joint Infection Using MRI with Metal Artifact Reduction at 1.5 T

    by Julien Galley, Reto Sutter, Christoph Stern, Lukas Filli, Stefan Rahm, Christian W. A. Pfirrmann

    MRI with metal artifact reduction at 1.5 T was sensitive and specific for the diagnosis of periprosthetic hip joint infection. Background MRI with metal artifact reduction has gained importance in assessment of pain with total hip arthroplasty (THA). However, its role in diagnosis of periprosthetic joint infection (PJI) ha...


  • Open Access

    Revisiting Radiographic L5-S1 Parallelism Using MRI T1 Mapping

    by Julien Galley, Federico Balagué

    Thirty years ago, we reported that parallel aspect of the L5-S1 disc on a lateral view of the spine might be considered to be an initial stage of disk degeneration. The current study represents an attempt to increase the validity of parallel sign on conventional radiograph using MR real T1 mapping. Forty-four young asymptomatic volunteers (mean age 21.6 ± 2.3) underwent lumbar spine MRI, twice the same day, morning and afternoon. Dedicated sequences using the inversion-recovery technique were used to calculate the T1 relaxation time. A region of interest (ROI) representing the nucleus pulposus was defined in each disk. The volunteers were stratified according to the presence or absence of a parallel morphology of L5-S1. Correlation between endplates angles, sacral slopes and T1 values were then evaluated. L5-S1 space looks parallel for angles <10° (mean value 6.9° ± 1.4°). Sacral slope was lower in parallel disks (31.7 ± 4.9° vs. 40.1 ± 5.6°), showing a significant difference of 8.4° (p < 0.05). The T1 relaxation values show a significant difference between the two groups (p < 0.05) with a difference of 96 ms for the morning (1090.9 ± 33.3 ms for the parallel group and 1186.9 ± 41.2 ms for the non-parallel) and 121.9 ms for the afternoon (respectively 1004.7 ± 22.2 ms and 1126.6 ± 12.9 ms). The difference between the two groups suggests that parallel morphology of the L5-S1 disk is associated with lower water content.