ACR Meeting Abstracts

ACR Meeting Abstracts

  • Meeting Abstracts
    • All Meetings
    • ACR Convergence 2026
    • PRSYM 2026
    • ACR Convergence 2025
    • Download Abstract Supplements
  • Keyword Index
  • My Favorites
    • View & print my favorites
  • Search

Home » Meeting Abstracts » ACR Convergence 2024

Abstract Number: 1190

Cross-sectional Associations Among Sonographic and Radiographic Features and Symptoms Scores in a Population-Based Cohort

Katherine Yates1, Carolina Alvarez2, Sahar Sawani3, Liubov Arbeeva4, Todd Schwartz5, Serena Savage-Guin6, Jordan Renner7, Catherine Bakewell8, Minna Kohler9, Janice Lin10, Jonathan Samuels11 and Amanda Nelson3, 1University of North Carolina Medical Center, Chapel Hill, NC, 2University of North Carolina at Chapel Hill, Miami, FL, 3University of North Carolina at Chapel Hill, Chapel Hill, NC, 4University of North Carolina, Chapel Hill, Carrboro, NC, 5University of North Carolina-Chapel Hill, Chapel Hill, NC, 6Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill,, Chapel Hill, NC, 7Thurston Arthritis Research Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, 8Intermountain Healthcare, Salt Lake City, UT, 9Massachusetts General Hospital, Boston, MA, 10Stanford University, Palo Alto, CA, 11NYU Langone, Rye Brook, NY

Share on X (Twitter) Share on Facebook Share on LinkedIn Share on Email
Print

Meeting: ACR Convergence 2024

Keywords: Imaging, Osteoarthritis, Plain Film, radiography, Ultrasound

Session Information

Date: Sunday, November 17, 2024

Title: Osteoarthritis – Clinical Poster I

Session Type: Poster Session B

Session Time: 10:30AM-12:30PM

Background/Purpose: Ultrasound (US) allows for visualization of features of knee osteoarthritis (KOA) including osteophytes, meniscal extrusion, synovitis, and cartilage damage. The current analysis aimed to determine the association of features of KOA on US with patient reported symptoms and radiographic KOA (rKOA).

Methods: Data were from participants in the population-based Johnston County Health Study, who provided demographic information, symptomatic assessments, and imaging. Self-reported pain, aching or stiffness (PAS, 0-3, none to severe) was assessed on most days of any one month in the past 12 months. Posteroanterior semi-flexed radiographs were read for Kellgren Lawrence Grade (KLG) by an expert musculoskeletal radiologist (JBR). Standardized knee US images were obtained by a trained sonographer (SSG). Each set of images was read independently by 2 randomly selected expert readers (CJB, MJK, JL, JS); discrepancies were adjudicated by a 3rd independent reader. Knee outcome definitions included any knee PAS, at least moderate severity PAS, rKOA (KLG≥2), severe rKOA (KLG≥3), and symptomatic KOA (sxKOA, rKOA and any PAS).

We fitted logistic regression models to test the associations (adjusted odds ratios and 95% confidence intervals) between each US feature and KOA outcomes. Models were adjusted for age, sex, race, ethnicity, BMI, education, Charlson comorbidity index, and history of knee injury. Models were defined at the knee level so that generalized estimating equations were used to account for within-person correlation between knees. Effect modification between all covariables and US features was assessed using two-way interaction terms at a 0.05 alpha level and reported in stratified models.

Results: A total of 499 individuals (35% men, 26% Black, 9% Hispanic, mean age 55 [range 35-70] years, mean BMI 33 kg/m2, and 51% with < college degree) with 983 knees, 313 (31%) of which had rKOA, were included. Cartilage damage and the presence of effusion/synovitis were the most frequent features of KOA on US (Table 1). Additionally, 37% had an effusion, 13-18% had meniscal extrusion/damage, and 11% had calcium crystal deposition.

Large medial or lateral osteophytes or popliteal cysts were associated with any and at least moderate PAS (Table 2). Osteophytes, both medial and lateral, were strongly associated with rKOA, severe rKOA, and sxKOA. Smaller magnitude associations were observed with the presence of effusion/synovitis, cartilage damage, and popliteal cysts (Table 3).

Stratified models revealed a strong association between medial osteophytes and sxKOA for both men and women. Associations between lateral osteophytes and sxKOA appeared stronger in younger participants. Associations between effusion and rKOA, meniscal extrusion and PAS, and cartilage damage or popliteal cyst and severe rKOA, appeared stronger in men than in women (data not shown).

Conclusion: The presence of effusion/synovitis and osteophytes on US are commonly seen in KOA and strongly associated with higher KLG and patient reported symptoms. These US features can be assessed at bedside for the evaluation of KOA symptoms. It remains to be determined how these features may progress over time and their longitudinal impact on patient symptoms.

Supporting image 1

Descriptive statistics of features of knee osteoarthritis seen on ultrasound. US=ultrasound; GS=gray scale; CPD=color power Doppler.

Supporting image 2

Logistic regression modeling log odds of each knee outcome and its association with each US feature, separately, using GEE for within-person correlation between knees. Each model adjusts for age, sex, race, ethnicity, BMI, education, CCI, and knee injury. Shown as BOLD when significant at 0.05 level. US=ultrasound; PAS=pain, aching or stiffness; OR=odds ratio; GS=gray scale; CPD=color power Doppler; ne=not estimable for cells with <5 counts.

Supporting image 3

Logistic regression modeling log odds of each knee outcome and its association with each US feature, separately, using GEE for within-person correlation between knees. Each model adjusts for age, sex, race, ethnicity, BMI, education, CCI, and knee injury. Shown BOLD when significant at 0.05 level. US=ultrasound; radiographic knee osteoarthritis (rKOA); sxKOA=symptomatic KOA; KLG= Kellgren Lawrence Grade; PAS=pain, aching or stiffness; OR=odds ratio; GS=gray scale; CPD=color power Doppler; ne=not estimable for cells with <5 counts.


Disclosures: K. Yates: None; C. Alvarez: None; S. Sawani: None; L. Arbeeva: None; T. Schwartz: None; S. Savage-Guin: None; J. Renner: None; C. Bakewell: AbbVie/Abbott, 2, 6, Bristol-Myers Squibb(BMS), 2, Eli Lilly, 2, 6, Janssen, 2, 6, Novartis, 2, 6, Pfizer, 2, 6, UCB, 2, 6; M. Kohler: Janssen, 12, medical advisory board, Novartis, 12, medical advisory board, Setpoint Medical, 5, Springer Publications, 9; J. Lin: None; J. Samuels: None; A. Nelson: MedScape Education, 6, NestleHealth, 6.

To cite this abstract in AMA style:

Yates K, Alvarez C, Sawani S, Arbeeva L, Schwartz T, Savage-Guin S, Renner J, Bakewell C, Kohler M, Lin J, Samuels J, Nelson A. Cross-sectional Associations Among Sonographic and Radiographic Features and Symptoms Scores in a Population-Based Cohort [abstract]. Arthritis Rheumatol. 2024; 76 (suppl 9). https://acrabstracts.org/abstract/cross-sectional-associations-among-sonographic-and-radiographic-features-and-symptoms-scores-in-a-population-based-cohort/. Accessed .

ACR Meeting Abstracts - https://acrabstracts.org/abstract/cross-sectional-associations-among-sonographic-and-radiographic-features-and-symptoms-scores-in-a-population-based-cohort/

My Favorites

Save and print abstracts during your browser session by clicking the “Favorite” button at the bottom of any abstract (must have cookies enabled in browser). See saved favorites.

Abstract Policies

  • ACR Convergence Abstract Embargo Policies
  • ACR Convergence Abstract Permissions & Reprints
  • PRSYM Abstract Policies

ACR Convergence. Where Rheumatology Meets

ACR Convergence 2026

Join us November 6-11 in Orlando, Florida.
See Registration Information

  • Contact ACR
  • Privacy Policy
  • ACR Policies
  • Cookie Preferences

© Copyright 2026 American College of Rheumatology