EVALUATION OF METABOLIC RESPONSE IN BONE MARROW LESIONS AFTER EXERCISE IN RELATION TO CHANGES IN ADJACENT CARTILAGE T2 VALUES

Abstract

Patellofemoral pain syndrome (PFPS) is one of the most common causes of anterior knee pain in active populations. Due to its multifactorial nature, patellofemoral pain is hard to diagnose and treat. Increased stress in the subchondral bone is associated with PFPS and is correlated to elevated bone remodelling. [18F]NaF PET-MR imaging can localize regions of elevated bone remodelling and provide information about the metabolic processes involved (1, 2). Further, [18F]NaF uptake changes after exercise, thus showing that it is sensitive to changes in bone physiology resulting from acute bone loading (3, 4). For cartilage, T2 relaxation times have been used to study cartilage hydration and microstructure changes after exercise. While some studies have observed a decrease in T2 times after exercise, changes are usually transient and have largely been evaluated in the context of healthy subjects (5). An early sign of persistent abnormal mechanical stress is Bone Marrow Lesions (BMLs), which are strongly associated with pain and structural disease progression. We aimed to reproduce prior work which has shown that acute loading results in a large physiological response in bone regions with BMLs. Further, we wanted to explore if there is breakdown of the cartilage microstructure, either at baseline or in response to loading in cartilage regions adjacent to BMLs. The purpose of this study is to detect changes in the functional response of the knee joint to loading using [18F]NaF PET-MR imaging after a stair-climbing exercise in subjects with unilateral patellofemoral pain. The aims of this feasibility study are 1) to detect acute loading-induced changes in [18F]NaF uptake in bone regions with BMLs and 2) compare consequent T2 values in the cartilage adjacent to BMLs at baseline and in response to loading. 4 subjects (3 F, aged 39.5±15.0 years, BMI 26.1±5.9 kg/m2) with unilateral knee pain and BMLs (3 in patella, 1 in femur) underwent two consecutive 30-min [18F]NaF PET/MRI scans of both knees, at baseline and then repeated after a stair-climbing exercise [112 stairs, up and down]. Bone metabolic activity was quantified by calculating maximum standardized uptake values (SUVmax) in areas of BMLs in the painful leg and the corresponding region in the contralateral leg. T2 maps were calculated from quantitative DESS images [TE(1,2)=6.04, 30.44 ms] in the cartilage regions adjacent to the BMLs and in the same regions in the contralateral leg. Segmentations of cartilage were done manually in MATLAB, while SUVmax in BMLs was calculated using ROIs drawn in Horos. Pre- and post-exercise changes in SUVmax and T2 values were compared to account for loading-induced response in both metabolic and structural parameters in painful vs healthy knees. Increased [18F]NaF uptake was observed post-exercise in areas with BMLs (as compared to contralateral leg regions), confirming that PET findings correspond to structural changes observed in MRI (Fig-1). The average SUVmax values increased from 6.35±3.98 to 10.49±5.42 in the painful leg and from 1.71±1.96 to 4.04±3.21 in the contralateral leg after exercise. The average T2 relaxation times increased from 30.61±1.87 ms to 31.61±1.55 ms in the painful leg and from 27.17±2.73 ms to 27.65±2.29 ms in the contralateral leg (Fig-2). The mean change in SUVmax due to exercise was 4.15±2.42 in the BML vs 2.34±2.06 in the contralateral leg. In adjacent cartilage, a mean change in T2 values of 1.01±1.45 ms in the BML region of the painful leg and 0.47±0.48 ms in the contralateral leg was observed. In this feasibility study, [18F]NaF PET-MR showed acute loading-induced changes in BMLs, with a 2-fold increase in SUVmax in the painful leg vs the contralateral leg. While there seemed to be a higher cartilage T2 in regions adjacent to BMLs compared to the contralateral leg, the change induced by exercise was small.