Diet-induced Obesity, Systemic Inflammation, and the Development of Hip and Shoulder Osteoarthritis: Insights from a Rat Model

Purpose: Obesity is one of the primary risk factors for the onset and progression of osteoarthritis (OA). Using a rat model of high-fat/high-sucrose (HFS) diet-induced obesity (DIO), we have demonstrated a strong significant relationship between body fat, systemic inflammatory mediators, and knee OA, using a Modified Mankin scoring system. Although several studies demonstrate a relationship between obesity and hand OA, the relationship between body fat, systemic inflammatory mediators, and structural OA-like changes in other synovial joints, like the shoulder and hip, is inconsistent. The purpose of this study was to determine whether HFS-induced obesity would result in OA-like changes in the shoulder and hips of rats. It was hypothesized that HFS-fed animals would exhibit OA-like changes in the hip and shoulder, and these changes would be associated with elevated serum levels of systemic inflammatory mediators. Methods: Sixteen 12-week old male Sprague-Dawley rats were allocated to either an HFS diet (DIO, 40% fat, 45% sucrose, n=9) or chow diet (13% fat, 0% sucrose, n=7) for a 12-week obesity induction period. After the obesity induction period, body mass, and composition (Dual Energy X-ray Absorptiometry) were quantified, and animals were euthanized. Shoulder (humeral and scapular cartilage) and hip (femoral and acetabular cartilage) joints from all animals were harvested, fixed in formalin, and scored using the Mankin criteria. OARSI subscores for bone changes and synovitis were also determined for each joint. Serum inflammatory mediators were profiled using a 27-plex Luminex ® assay. Outcomes were compared by diet using a one-way ANOVA or non-parametric statistics (Mankin and OARSI scores) at α=0.05. Relationships between body fat, body mass, systemic inflammatory mediators, and joint damage were evaluated using Spearman correlations. Results: DIO animals gained significantly more body mass (p=0.002) and body fat (p<0.001) compared to chow. Although a variety of pro-inflammatory mediators were increased in the serum from DIO animals, histological scores were statistically similar between DIO and chow animals (Table 1, p>0.05). No significant relationships were found between body mass or body fat and damage in hip and shoulder. Serum leptin, however, was positively associated with increased hip synovitis scores across all animals (R=0.57, p=0.02) and a number of mediators were positively associated with increased acetabular cartilage damage (IL-4, IL-5, IL-13, IL-17; R≥0.50, p≤0.05). No significant positive relationships were calculated between any measure of OA damage at the shoulder and serum inflammation. Of note, an increased number of DIO animals had OA damage in both the shoulder and hip compared to chow animals (multi-joint: 4/9 DIO animals, 1/7 chow animals). When compared to animals with damage at only one joint site, animals with multi-joint OA had significantly more histological damage, increased shoulder synovitis (p<0.05) a trend toward increased hip synovitis (p=0.13), and increased serum levels for IL-1α and IL-4 (Figure 1, p<0.05). Conclusions: Hip and shoulder joints do not consistently sustain OA damage after 12-weeks of DIO in rats, potentially due to the known genetic heterogeneity of these rats, refuting our primary hypothesis. However, systemic inflammatory alterations resulting from DIO induction in this model system may lead to a higher incidence of multi-joint OA with the HFS diet over the long term. Future work will evaluate whether OA develops more slowly in hip and shoulder joints when compared to previous reports in DIO knee joints, possibly due to the unique environment of the knee (i.e. Hoffa's fat pad, presence of menisci, hinge joint configuration). However, potential links between DIO, systemic inflammation, synovium changes and OA-like alterations in the hips and shoulders of DIO rats were identified, and these links will be further investigated over longer time periods.Tabled 1Shoulder and Hip OA-like Changes for Obese and Chow Animals.GroupShoulder JointHip JointHumerusScapulaSynovitisFemurAcetabulumSynovitisChow2 (2–14)–0 (0–4)3 (2–13)4 (0–6)0 (0–1)DIO2 (2–14)0 (0–14)0 (0–2)2 (0–14)2 (0–14)0 (0–1)Single-Joint OA2 (2–11)–0 (0–1)2 (0–10)2 (0–6)0 (0–1)Multi-Joint OA14 (10–14)*0 (0–14)0 (0–4)*5 (0–14)ˆ7 (5–14)*0 (0–1)ˆData are shown as median (minimum-maximum); *indicates p<0.05 compared to single-joint OA group; ˆindicates p<0.15 compared to single-joint OA group Open table in a new tab Data are shown as median (minimum-maximum); *indicates p<0.05 compared to single-joint OA group; ˆindicates p<0.15 compared to single-joint OA group