Unicompartmental versus bicompartmental joint space width measures: Which reflect whole joint structural damage better? Data from IMI-APPROACH

Eva A Bax; Roel J H Custers; Nienke van Egmond; Moyo C Kruyt; Hassan Rayegan; Margreet Kloppenburg; Francisco J Blanco; Ida K Haugen; Francis Berenbaum; Frank W Roemer; Harrie Weinans; Simon C Mastbergen; Mylène P Jansen

doi:10.1016/j.joca.2025.11.011

Unicompartmental versus bicompartmental joint space width measures: Which reflect whole joint structural damage better? Data from IMI-APPROACH

Eva A Bax^*
, Roel J H Custers
, Nienke van Egmond
, Moyo C Kruyt
, Hassan Rayegan
, Margreet Kloppenburg
, Francisco J Blanco
, Ida K Haugen
, Francis Berenbaum
, Frank W Roemer
, Harrie Weinans
, Simon C Mastbergen
, Mylène P Jansen^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

OBJECTIVE: To investigate the associations between whole joint cartilage and meniscal morphology on MRI and radiographic joint space width (JSW) measures and in knee osteoarthritis (KOA), to determine whether bicompartmental measures demonstrate stronger associations than unicompartmental ones, and to evaluate their correlations with Kellgren and Lawrence grading.

DESIGN: A cross-sectional analysis of baseline radiographs and MRIs from 262 KOA participants in the prospective, multicenter IMI-APPROACH cohort was conducted. Radiographic measures included minimum joint space width (mJSW), fixed location JSW (JSW(x)), mean JSW, and joint line convergence angle (JLCA), assessed using fully automated software. JSW was evaluated both unicompartmentally and bicompartmentally. Cartilage morphology, full-thickness cartilage loss, meniscal extrusion, tears, and maceration were assessed using the semi-quantitative MRI Osteoarthritis Knee Score to summarize whole-joint cartilage and meniscal morphology. Associations of radiographic measures with MRI outcomes were assessed using multivariable linear regression; Spearman correlations with Kellgren and Lawrence (KL) were also evaluated.

RESULTS: MRI-defined meniscal maceration was associated with unicompartmental and bicompartmental JSW measures. Full-thickness cartilage loss was associated with unicompartmental (95% CI [-0.16;-0.02]) and bicompartmental mJSW (95% CI [-0.14;-0.02]), and JLCA (95% CI [0.04;0.22]). Models explained 32-39% of variance for unicompartmental and 23-45% for bicompartmental measures (R²). Bicompartmental measures showed stronger correlations with KL grading than unicompartmental measures (95% CI: -0.31 to -0.02).

CONCLUSIONS: Associations between whole-joint cartilage and meniscal degeneration are similar for uni- and bicompartmental JSW, with bicompartmental JSW showing stronger correlations with KL grades. These findings support including both compartments in radiographic assessment to improve structural evaluation in KOA.

Original language	English
Pages (from-to)	297-303
Number of pages	7
Journal	Osteoarthritis and Cartilage
Volume	34
Issue number	2
Early online date	28 Nov 2025
DOIs	https://doi.org/10.1016/j.joca.2025.11.011
Publication status	Published - Feb 2026

Keywords

Imaging
Joint space width
Knee osteoarthritis
Radiographs

Access to Document

10.1016/j.joca.2025.11.011Licence: CC BY

PIIS1063458425012397Final published version, 1.69 MBLicence: CC BY

Cite this

Bax, E. A., Custers, R. J. H., van Egmond, N., Kruyt, M. C., Rayegan, H., Kloppenburg, M., Blanco, F. J., Haugen, I. K., Berenbaum, F., Roemer, F. W., Weinans, H., Mastbergen, S. C., & Jansen, M. P. (2026). Unicompartmental versus bicompartmental joint space width measures: Which reflect whole joint structural damage better? Data from IMI-APPROACH. Osteoarthritis and Cartilage, 34(2), 297-303. https://doi.org/10.1016/j.joca.2025.11.011

@article{16092bc34b04496a8e16e2b7f983d093,

title = "Unicompartmental versus bicompartmental joint space width measures: Which reflect whole joint structural damage better? Data from IMI-APPROACH",

abstract = "OBJECTIVE: To investigate the associations between whole joint cartilage and meniscal morphology on MRI and radiographic joint space width (JSW) measures and in knee osteoarthritis (KOA), to determine whether bicompartmental measures demonstrate stronger associations than unicompartmental ones, and to evaluate their correlations with Kellgren and Lawrence grading.DESIGN: A cross-sectional analysis of baseline radiographs and MRIs from 262 KOA participants in the prospective, multicenter IMI-APPROACH cohort was conducted. Radiographic measures included minimum joint space width (mJSW), fixed location JSW (JSW(x)), mean JSW, and joint line convergence angle (JLCA), assessed using fully automated software. JSW was evaluated both unicompartmentally and bicompartmentally. Cartilage morphology, full-thickness cartilage loss, meniscal extrusion, tears, and maceration were assessed using the semi-quantitative MRI Osteoarthritis Knee Score to summarize whole-joint cartilage and meniscal morphology. Associations of radiographic measures with MRI outcomes were assessed using multivariable linear regression; Spearman correlations with Kellgren and Lawrence (KL) were also evaluated.RESULTS: MRI-defined meniscal maceration was associated with unicompartmental and bicompartmental JSW measures. Full-thickness cartilage loss was associated with unicompartmental (95\% CI [-0.16;-0.02]) and bicompartmental mJSW (95\% CI [-0.14;-0.02]), and JLCA (95\% CI [0.04;0.22]). Models explained 32-39\% of variance for unicompartmental and 23-45\% for bicompartmental measures (R²). Bicompartmental measures showed stronger correlations with KL grading than unicompartmental measures (95\% CI: -0.31 to -0.02).CONCLUSIONS: Associations between whole-joint cartilage and meniscal degeneration are similar for uni- and bicompartmental JSW, with bicompartmental JSW showing stronger correlations with KL grades. These findings support including both compartments in radiographic assessment to improve structural evaluation in KOA.",

keywords = "Imaging, Joint space width, Knee osteoarthritis, Radiographs",

author = "Bax, \{Eva A\} and Custers, \{Roel J H\} and \{van Egmond\}, Nienke and Kruyt, \{Moyo C\} and Hassan Rayegan and Margreet Kloppenburg and Blanco, \{Francisco J\} and Haugen, \{Ida K\} and Francis Berenbaum and Roemer, \{Frank W\} and Harrie Weinans and Mastbergen, \{Simon C\} and Jansen, \{Myl{\`e}ne P\}",

note = "Publisher Copyright: {\textcopyright} 2025 The Authors",

year = "2026",

month = feb,

doi = "10.1016/j.joca.2025.11.011",

language = "English",

volume = "34",

pages = "297--303",

journal = "Osteoarthritis and Cartilage",

issn = "1063-4584",

publisher = "W.B. Saunders Ltd",

number = "2",

}

Bax, EA, Custers, RJH, van Egmond, N, Kruyt, MC, Rayegan, H, Kloppenburg, M, Blanco, FJ, Haugen, IK, Berenbaum, F, Roemer, FW, Weinans, H , Mastbergen, SC & Jansen, MP 2026, 'Unicompartmental versus bicompartmental joint space width measures: Which reflect whole joint structural damage better? Data from IMI-APPROACH', Osteoarthritis and Cartilage, vol. 34, no. 2, pp. 297-303. https://doi.org/10.1016/j.joca.2025.11.011

TY - JOUR

T1 - Unicompartmental versus bicompartmental joint space width measures

T2 - Which reflect whole joint structural damage better? Data from IMI-APPROACH

AU - Bax, Eva A

AU - Custers, Roel J H

AU - van Egmond, Nienke

AU - Kruyt, Moyo C

AU - Rayegan, Hassan

AU - Kloppenburg, Margreet

AU - Blanco, Francisco J

AU - Haugen, Ida K

AU - Berenbaum, Francis

AU - Roemer, Frank W

AU - Weinans, Harrie

AU - Mastbergen, Simon C

AU - Jansen, Mylène P

PY - 2026/2

Y1 - 2026/2

N2 - OBJECTIVE: To investigate the associations between whole joint cartilage and meniscal morphology on MRI and radiographic joint space width (JSW) measures and in knee osteoarthritis (KOA), to determine whether bicompartmental measures demonstrate stronger associations than unicompartmental ones, and to evaluate their correlations with Kellgren and Lawrence grading.DESIGN: A cross-sectional analysis of baseline radiographs and MRIs from 262 KOA participants in the prospective, multicenter IMI-APPROACH cohort was conducted. Radiographic measures included minimum joint space width (mJSW), fixed location JSW (JSW(x)), mean JSW, and joint line convergence angle (JLCA), assessed using fully automated software. JSW was evaluated both unicompartmentally and bicompartmentally. Cartilage morphology, full-thickness cartilage loss, meniscal extrusion, tears, and maceration were assessed using the semi-quantitative MRI Osteoarthritis Knee Score to summarize whole-joint cartilage and meniscal morphology. Associations of radiographic measures with MRI outcomes were assessed using multivariable linear regression; Spearman correlations with Kellgren and Lawrence (KL) were also evaluated.RESULTS: MRI-defined meniscal maceration was associated with unicompartmental and bicompartmental JSW measures. Full-thickness cartilage loss was associated with unicompartmental (95% CI [-0.16;-0.02]) and bicompartmental mJSW (95% CI [-0.14;-0.02]), and JLCA (95% CI [0.04;0.22]). Models explained 32-39% of variance for unicompartmental and 23-45% for bicompartmental measures (R²). Bicompartmental measures showed stronger correlations with KL grading than unicompartmental measures (95% CI: -0.31 to -0.02).CONCLUSIONS: Associations between whole-joint cartilage and meniscal degeneration are similar for uni- and bicompartmental JSW, with bicompartmental JSW showing stronger correlations with KL grades. These findings support including both compartments in radiographic assessment to improve structural evaluation in KOA.

AB - OBJECTIVE: To investigate the associations between whole joint cartilage and meniscal morphology on MRI and radiographic joint space width (JSW) measures and in knee osteoarthritis (KOA), to determine whether bicompartmental measures demonstrate stronger associations than unicompartmental ones, and to evaluate their correlations with Kellgren and Lawrence grading.DESIGN: A cross-sectional analysis of baseline radiographs and MRIs from 262 KOA participants in the prospective, multicenter IMI-APPROACH cohort was conducted. Radiographic measures included minimum joint space width (mJSW), fixed location JSW (JSW(x)), mean JSW, and joint line convergence angle (JLCA), assessed using fully automated software. JSW was evaluated both unicompartmentally and bicompartmentally. Cartilage morphology, full-thickness cartilage loss, meniscal extrusion, tears, and maceration were assessed using the semi-quantitative MRI Osteoarthritis Knee Score to summarize whole-joint cartilage and meniscal morphology. Associations of radiographic measures with MRI outcomes were assessed using multivariable linear regression; Spearman correlations with Kellgren and Lawrence (KL) were also evaluated.RESULTS: MRI-defined meniscal maceration was associated with unicompartmental and bicompartmental JSW measures. Full-thickness cartilage loss was associated with unicompartmental (95% CI [-0.16;-0.02]) and bicompartmental mJSW (95% CI [-0.14;-0.02]), and JLCA (95% CI [0.04;0.22]). Models explained 32-39% of variance for unicompartmental and 23-45% for bicompartmental measures (R²). Bicompartmental measures showed stronger correlations with KL grading than unicompartmental measures (95% CI: -0.31 to -0.02).CONCLUSIONS: Associations between whole-joint cartilage and meniscal degeneration are similar for uni- and bicompartmental JSW, with bicompartmental JSW showing stronger correlations with KL grades. These findings support including both compartments in radiographic assessment to improve structural evaluation in KOA.

KW - Imaging

KW - Joint space width

KW - Knee osteoarthritis

KW - Radiographs

UR - https://www.scopus.com/pages/publications/105023705715

U2 - 10.1016/j.joca.2025.11.011

DO - 10.1016/j.joca.2025.11.011

M3 - Article

C2 - 41320151

SN - 1063-4584

VL - 34

SP - 297

EP - 303

JO - Osteoarthritis and Cartilage

JF - Osteoarthritis and Cartilage

IS - 2

ER -

Unicompartmental versus bicompartmental joint space width measures: Which reflect whole joint structural damage better? Data from IMI-APPROACH

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this