Bone Marrow Lesions in Knees with Osteoarthritis: Can Parameters From Dynamic Contrast Enhancement Predict Change in Bone Marrow Lesion Volume or Knee Pain Change?

Andrew D. Gait¹, Timothy F. Cootes¹, Elizabeth J. Marjanovic¹, Matthew J. Parkes², Charles E. Hutchinson³ and David T. Felson⁴, ¹Imaging Sciences, School of Cancer & Enabling Sciences, University of Manchester, Manchester, United Kingdom, ²Arthritis Research UK Centre for Epidemiology, Institute of Inflammation and Repair, University of Manchester, Manchester, United Kingdom, ³Warwick Medical School, University of Warwick, Coventry, United Kingdom, ⁴ARC Epidemiology Unit, University of Manchester, Manchester, United Kingdom

Meeting: 2012 ACR/ARHP Annual Meeting

Keywords: Bone marrow, imaging techniques and osteoarthritis, Knee

Session Information

Title: Osteoarthritis - Clinical Aspects

Session Type: Abstract Submissions (ACR)

Background/Purpose:

Dynamic contrast enhancement is a powerful tool for highlighting features of a medical image which may not otherwise be seen on “static” scans. While used extensively in other settings, it has not been used until now to study bone marrow lesions (BMLs) in osteoarthritis (OA), lesions which vary over time in volume and change with knee pain.

Methods:

We studied 23 patients who had patellofemoral (PF) OA. All met ACR criteria for knee OA and were participating in a trial of PF knee braces. All patients had no treatment at baseline, were on brace at 12 weeks and acquired MRI’s with a gadolinium(Gd)-enhanced dynamic image sequence at both times.

Within these knees, 40 PF or femoral BMLs were manually segmented at baseline and 12 weeks on sagittal Gd-enhanced images (TR 500ms, TE 17ms, FoV 16cm, 384×384). Knee osteoarthritis outcome score (KOOS) pain data was collected at each time.

We assessed dynamic parameters within BMLs in two steps: (1) transforming the segmented BML from the sagittal sequence to the axial dynamic sequence (TR 5.4ms, TE 1.9ms, FoV 14cm, 256×256); (2) using the extended Kety model (Tofts 1997) to calculate the parameters v_e (fractional extracellular volume), v_p (fractional blood plasma volume) and K^trans (transfer coefficient from v_pto v_e:a direct measure of perfusion) at each voxel. The mean of each parameter was calculated for each BML (one to three per patient), taking care to ensure that this was done consistently between the baseline and 12 week images. The distribution of K^trans within the BMLs was heterogeneous and limited to small volumes, so to avoid averaging of diverse values, we examined peak K^trans value in the BML also. To compare with KOOS, mean parameters and peak K^trans value were examined across all BMLs within a knee.

To evaluate the relation of the parameters to BML volumes and KOOS, we calculated Spearman rank correlation coefficients.

Results:

After 12 weeks, change in mean values of the parameters K^trans, v_e and v_p did not correlate either with change in volume of the BML (K^trans: r=0.02; v_e: r=-0.07; v_p: r=0.02) or change in KOOS (K^trans: r=0.18; v_e: r=0.03; v_p: r=0.09). Change in the peak value of K^trans correlated better with change in volume of the BML (r=0.43: p=0.0056, see figure) but not with KOOS (r=-0.11).

Conclusion:

There are no clear correlations between the means of the parameters K^trans, v_e and v_p and change in BML volume or KOOS. The peak K^trans value of a BML may be related to its likelihood of undergoing change.

Disclosure:

A. D. Gait,
None;

T. F. Cootes,
None;

E. J. Marjanovic,
None;

M. J. Parkes,
None;

C. E. Hutchinson,
None;

D. T. Felson,
None.

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/bone-marrow-lesions-in-knees-with-osteoarthritis-can-parameters-from-dynamic-contrast-enhancement-predict-change-in-bone-marrow-lesion-volume-or-knee-pain-change/