Investigating the association of serum lipid biomarkers with biochemical cartilage composition using quantitative MRI – a pilot study

Abstract

Purpose: There has been considerable interest in the association between osteoarthritis (OA) and metabolic syndrome (MS), whose components include abdominal obesity, diabetes, hypertension and dyslipidemia. OA is increasingly understood as a metabolic disease and recent epidemiologic studies suggest that OA pathogenesis could be linked to altered lipid metabolism. While there is evidence suggesting that the MS-associated OA phenotype is biochemically and structurally distinct from standard OA, the few studies that explored the relationship between serum lipid profile and OA utilized radiographic measurements which are insensitive for early and mild OA. Magnetic resonance imaging (MRI)-based T2 and T1ρ mapping are techniques that measure early degenerative changes of the whole cartilage layer while ultrashort (UTE) T2* mapping focuses on early degenerative changes of the deep calcified cartilage layer. Therefore, the goal of this study was to investigate the association of serum lipid levels with knee cartilage composition, assessed using T2, T1ρ and UTE T2* mapping. Methods: Fourteen study participants were recruited using the following inclusion criteria: 40-70 years of age, Kellgren Lawrence grade (KLG) 0-2 in at least one knee, no previous diagnosis of an inflammatory arthropathy, no knee pain most days of the month, no history of knee injury or knee surgery and no metal implants. Serum lipid biomarkers included total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), non-HDL cholesterol (sum of vLDL and LDL) and triglycerides. All subjects underwent bilateral standing anterior-posterior knee radiographs and 3 Tesla MRI imaging of the knee with the lower KL score using 3D multiecho T2, T1ρ and UTE sequences. Cartilage was segmented semi-automatically into five regions (patella, medial/lateral femur, medial/lateral tibia) for T2*UTE maps and six regions (patella, medial/lateral femur, medial/lateral tibia, trochlea) for T2 and T1ρ maps. T2, T1ρ and T2*UTE maps were computed after segmentation. Global values were calculated as the mean of all knee regions. Partial correlations and linear regression analyses were performed using standardized serum lipid biomarkers as predictors and global T2, T1ρ and UTE as outcomes adjusting for BMI. Results: The mean age of all subjects was 54.4 years (SD ± 4.0) with a higher proportion of males (71.4%), mean body mass index (BMI) was 29.4 (SD ± 2.8) and mean total WOMAC was 1.4 (SD ± 2.5). For serum lipid levels, the mean total cholesterol was 159.9 (SD ± 43.0) mg/dl, mean HDL was 44.8 (SD ± 11.0) mg/dl, mean LDL was 94.2 (SD ± 35.8) mg/dl, mean non-HDL cholesterol was 115.1 (SD ± 40.6) mg/dl and mean triglycerides was 104.7 (SD ± 69.0) mg/dl. As shown in Table 1, higher total cholesterol was significantly associated with higher global higher global UTE T2* (0.82 ms [0.07,1.57], r = 0.59, P = 0.035), higher LDL was significantly associated with higher global UTE T2* (0.80 ms [0.06,1.54], r = 0.58, P = 0.037) and higher non-HDL cholesterol was significantly associated with higher global UTE T2* (0.95 ms [0.27,1.63], r = 0.68, P = 0.011). No significant associations were found between serum lipid levels and global T2 and T1ρ. Conclusions: This cross-sectional study demonstrates that higher total cholesterol, LDL and non-HDL cholesterol were associated with higher UTE T2*, suggesting increased degeneration of the deep calcified cartilage layer. This layer may, therefore, have an important role in dyslipidemia-associated early cartilage degeneration. Potential pathophysiological mechanisms underlying these relationships include oxidized LDL’s role in cartilage destruction and ectopic bone formation, subchondral ischemia secondary to lipid deposits in cartilage and disrupted lipid metabolism leading to upregulation of pro-inflammatory signaling within cartilage and chondrocytes. Mechanisms through which dyslipidemia could contribute to OA development must still be fully elucidated. This study was supported by the National Center for Advancing Translational Sciences, NIH, through UCSF-CTSI Grant Number TL1 TR001871. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.Tabled 1Association of serum lipid biomarkers with cartilage UTE T2*. P-values < 0.05 are in bold.Coefficient[95% Conf. Interval]CorrelationP-valueGlobal UTE T2*Total cholesterol0.82[0.07,1.57]0.590.035HDL-0.29[-1.17,0.60]-0.210.491LDL0.80[0.06,1.54]0.580.037Non-HDL cholesterol0.95[0.27,1.63]0.680.011Triglycerides0.68[-0.15,1.51]0.480.101 Open table in a new tab