Fig 1.
Flow diagram of serum (S) and synovial fluid (SF) sampling from control and osteoarthritis (OA) groups.
Samples from the OA group were obtained at three time points and only once from the control group. Samples obtained from stifles in control dogs were labeled as controls while samples from right and left stifles of OA group were labeled as index stifle or contralateral stifle depending on whether the cranial cruciat ligament was intact or not. Only a subset of the dogs in the OA group were recruited for sampling at T2 (4-week post op) and T3 (12-week post op) time points.
Table 1.
Frequency of dog breeds.
Table 2.
Discriminative performance of serum MMP-2 and MMP-3.
Fig 2.
Comparison of synovial fluid biomarkers (IL-8, KC and MCP-1) between control and OA group samples.
Comparison of the distribution of IL-8 (A), KC (B), and MCP-1 (C) synovial fluid concentrations (pg/ml) at the initial visit (T1) for stifles with OA (OA), the stable stifles in the OA dogs (contralateral) and control dogs (control). The horizontal line inside each box is the median and the upper and lower edges of box present the inter-quartile range (IQR). The whiskers are either 1.5 × IQR or the range, whichever is smaller. Dots outside the fence are outliers. *Statistically significant if P < 0.05, ** or if P < 0.01.
Table 3.
Synovial fluid biomarker differences between among groups and over time.
Fig 3.
Comparison of temporal changes of concentrations of synovial fluid IL-8, KC and MCP-1 biomarkers for the OA group dogs.
The first row of graphs are for synovial fluid samples from index stifles (OA) and the second row are for synovial fluid samples from the stable stifles of dogs in the OA group (contralateral). The time points evaluated are initial visit (T1), 4-week recheck (T2) and 12-week recheck (T3). The horizontal line inside each box is the median and the upper and lower edges of box present the inter-quartile range (IQR). The whiskers are either 1.5 × IQR or the range, whichever is smaller. Dots outside the fence are outliers. * if P < 0.05; ** if P < 0.01.
Fig 4.
Discriminative performance of synovial fluid IL-8, KC and MCP-1 biomarkers’ ROC model performance at T1 time point for the OA group compared to control.
The forest plot depicts the area under the curve (AUC) and 95% confidence interval (CI) to quantify the overall discriminative performance. Comparisons are between synovial fluid samples from the index (OA) and contralateral stifles from the OA group and control synovial fluid samples. An AUC = 1 indicates a perfect discriminative ability, whereas an AUC = 0.5 indicates no discriminative ability. It desired to have an AUC ≥ 0.9 for a clinically acceptable performance.
Table 4.
Optimal cut-off points, based on the Younden’s index, for summed scores in each category of histopathological grading of synovitis between OA and control dogs.
Table 5.
Correlation between synovial fluid biomarkers (i.e., IL-8, KC, MCP-1) and histopathological grade of biopsied synovial membrane.