Session Type: Poster Session (Monday)
Session Time: 9:00AM-11:00AM
Background/Purpose: Composite imaging scores that reflect cartilage damage, bone marrow lesions (BMLs), and effusion-synovitis may enable us to conceptualize knee osteoarthritis (KOA) disease progression as two constructs: 1) disease activity = whole-joint dynamic disease processes that fluctuate over time and relates to knee pain, 2) cumulative damage = joint damage that has accrued over time and relates to radiographic disease severity. We aimed to determine if recently validated composite magnetic resonance (MR) metrics of KOA disease activity and cumulative damage can predict the development of accelerated KOA (AKOA).
Methods: We identified adults from the Osteoarthritis Initiative without radiographic KOA (i.e., Kellgren-Lawrence grade [KL] 0/1) at baseline. Three groups were identified based on annual radiographic disease progression over the first 48-months: 1) AKOA: progressed to KL 3/4; 2) typical KOA: any other increase in KL; 3) no KOA: no change in KL. The visit an adult met the AKOA or typical KOA criteria was defined as the index visit. We assessed cartilage damage in four tibiofemoral regions, BML volume in four tibiofemoral regions, and a whole knee effusion-synovitis volume on 3T MR images with semi-automated programs. We used annual images from two years before to two years after the index visit. Each MR metric was normalized to participant size and standardized from the earliest visit. We calculated disease activity as the sum of the standardized score from all four BML regions and effusion-synovitis. We calculated cumulative damage as the sum of the the standardized score from all four cartilage regions. To determine group differences across time in disease activity and cumulative damage, we used generalized linear mixed models with group (3 levels) and time (up to 5 levels) as independent variables. Additionally, change in the composite metrics from two to one year prior to the index visit was converted to a dichotomous variable to compare the worst tertile (greatest progression) to the other two tertiles. To establish the prognostic capability of disease activity and cumulative damage change, we used separate logistic regression models to determine if change in each composite metric prior to disease development was associated with future AKOA status (AKOA vs no AKOA).
Results: Table 1 provides the baseline characteristics for each group. Starting at one year prior to the index visit, the AKOA group had greater disease activity compared to the typical or no KOA groups (Figure 1A). Starting at the index visit, the AKOA group had worse cumulative damage compared to the typical or no KOA groups (Figure 1B). There were no differences at any time for either composite metrics between the typical and no KOA groups (Figure 1A and 1B). However, adults with the greatest increase in disease activity or cumulative damage from two to one year prior to the index visit were 3.7 and 4.0 times more likely to develop AKOA, respectively (Figure 1A and 1B).
Conclusion: Prior to AKOA development, disease activity and cumulative damage change are prognostic of AKOA development and progression.
To cite this abstract in AMA style:Harkey M, Davis J, Lu B, Price L, Ward R, MacKay J, Eaton C, Lo G, Barbe M, Zhang M, Pang J, Stout A, Michael L, McAlindon T, Driban J. Changes in Novel Composite Scores of Disease Activity and Cumulative Damage Are Prognostic of Accelerated Knee Osteoarthritis: Data from the Osteoarthritis Initiative [abstract]. Arthritis Rheumatol. 2019; 71 (suppl 10). https://acrabstracts.org/abstract/changes-in-novel-composite-scores-of-disease-activity-and-cumulative-damage-are-prognostic-of-accelerated-knee-osteoarthritis-data-from-the-osteoarthritis-initiative/. Accessed October 22, 2020.
« Back to 2019 ACR/ARP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/changes-in-novel-composite-scores-of-disease-activity-and-cumulative-damage-are-prognostic-of-accelerated-knee-osteoarthritis-data-from-the-osteoarthritis-initiative/