Correlation of synovial fluid leptin with body habitus in the healthy and osteoarthritic canine knee

Abstract

Purpose: Naturally occurring cranial cruciate ligament (CrCL) insufficiency and progressive osteoarthritis (OA) are common in the canine population and can serve to provide useful information to understand both canine and human knee joint disease. Leptin, an adipocytokine, has been shown to be proportional to adipose mass. In humans, leptin has been associated with radiographic progression of knee OA, markers of cartilage damage, and pain in osteoarthritic knees. The primary goal of this study is to evaluate the relationship between serum and synovial fluid (SF) leptin and body condition score and body fat percentage in dogs. The secondary goal was to characterize the relationship between SF leptin and markers of inflammation, such as IL-1β Methods: Forty-eight client-owned dogs, 19 healthy controls and 29 with secondary OA, were enrolled. Any dog with evidence of systemic disease, based on physical exam and routine hematology was excluded from the study. Informed owner consent was obtained and all procedures were approved by the Clinical Research Committee. Sample Collection - Dogs in the control group were sedated and a single femorotibial joint was randomly selected for arthrocentesis. Synovial fluid was collected using aseptic technique. For dogs in the OA group, SF was obtained intra-operatively from the affected limb prior to arthrotomy for correction of CrCL insufficiency. Blood was collected via standard jugular venipuncture. Morphometric Measures - Body condition score, BCS, was determined on a standard 9-point scalea by 3 individual observers, who were unaware of the other raters’ score. Dogs were then divided into to 2 groups, optimal weight (BCS 4-5) and overweight (BCS >5). Additionally trunk and limb measurements were obtained to calculate body fat percentage, as previously reported1. Osteoarthritis Severity Scoring- In the OA group, radiographs were scored for severity of OA by a radiologist utilizing a previously described scoring system2. Owners were required to fill out a clinical metrology questionnaire (Liverpool Osteoarthritis in Dogs, LOAD). Sample Analysis - Canine specific leptinb and IL-1βc ELISAs were used to measure SF and serum leptin and IL-1β. Data Analysis - Student t-tests were used to test for differences in SF leptin means due to CrCL status, presence of SF or serum IL-1β, and the potential confounder non-steroidal anti-inflammatory (NSAID) use. Linear regressions were used to test for association of synovial leptin with BCS, body fat percent, serum leptin concentration, LOAD, and radiographic score. Multiple linear regression was also used to adjust the linear model for the association of synovial leptin and OA status for BCS. Results: SF leptin was significantly lower than serum leptin (p<0.01) and there was a 5.8ng/mL change in SF leptin for every 10ng/ml increase in serum leptin. Overweight dogs had significantly higher concentration of SF leptin than those that were optimal body weight (p<0.01). SF leptin concentration was associated with both BCS and body fat % (p<0.01). For every 1unit increase in BCS there was a 1.7ng/mL increase in SF leptin and for every 10% increase in body fat, there was a 2.7ng/mL increase in SF leptin. Synovial fluid leptin was significantly higher in dogs with OA by a mean amount of 2.8 (95% CI 0.6-5.1) but this difference decreased and became no longer significant when adjusted for differences in body condition score between OA groups (p=0.25). There was a weak association between SF leptin and LOAD; for every 1ng/mL increase in leptin a 10unit increase in LOAD was observed (R2=0.100; p=0.03). There was no significant association between SF leptin and radiographic score (p=0.24), SF IL-1β (p=0.64) or NSAID use (p=0.55). Conclusions: This is the first study to document the presence of leptin in canine SF. As expected, SF leptin was positively correlated with serum leptin and measures of body habitus. However, SF leptin levels in dogs with CrCL insufficiency were not significantly different from control dogs. Still, there was weak positive association between SF leptin and LOAD, in dogs with OA. As all these dogs had CrCL insufficiency at the time of data collection, the pain and dysfunction from joint instability may have masked small changes due to chronic OA and contributed to the weak nature of this link. The results of this study did not correlate leptin levels with severity of OA, or local concentrations of IL-1β. In the canine model of OA, obesity related OA severity may be related to other adipokines or may purely be biomechanical in nature. Further studies are needed in both people and dogs to determine the role of leptin in OA.