Abstract
Objectives:
Biochemical joint tissue changes are associated with the development of posttraumatic osteoarthritis (PTOA). Serial assessments of serum biomarkers related to cartilage degradation are predictive of symptomatic progression of idiopathic knee OA but have not been explored in patients with symptomatic PTOA following anterior cruciate ligament (ACL) injury. Both serum and synovial biomarker profiles of inflammation and cartilage degradation post-ACL injury and ACL reconstruction (ACLR) have been associated with deleterious changes of magnetic resonance imaging markers of tibiofemoral cartilage composition (i.e., lower proteoglycan density and poor collagen organization) which precede radiographic PTOA development. However, it is unclear if biochemical changes post-ACL injury and ACLR are also associated with clinically relevant knee OA symptoms. The purpose of this study was to determine if changes in biochemical biomarker profiles preoperatively to 6-months post-ACLR were associated with clinically relevant knee symptoms related to PTOA at 12-months post-ACLR.
Methods:
Participants with a primary history of ACL injury who were planning on undergoing ACLR with a bone-patellar tendon-bone autograft were included in the prospective, longitudinal cohort study. Serum was collected from participants at three separate study visits: i) preoperatively within 15 days of ACL injury; ii) 6 months post-ACLR; and iii) 12 months post-ACLR. Blood samples from the antecubital fossa were collected in 5 mL serum separation tube vacutainers, centrifuged at 3000 rpms and -4°C, and stored in aliquots at -80°C both preoperatively and 6 months post-ACLR. At 12 months post-ACLR, participants completed the Knee Injury and Osteoarthritis Outcome survey. Upon completion of the study, serum samples were batch processed using commercial enzyme-linked immunosorbent assays (ELISA) to determine biomarker concentrations of monocyte chemoattractant protein 1 (sMCP-1), cartilage oligomeric matrix protein (sCOMP), matrix metalloproteinase 3 (sMMP-3), and type-II collagen degradation to synthesis ratio (sC2C:sCPII) which assess inflammation, cartilage breakdown, cartilage degradation, and cartilage turnover, respectively. KOOS subscale scores were calculated on a scale of 0-100. Participants were classified as Symptomatic or Asymptomatic based on Englund et al. cut off scores for each KOOS subscale which were previously validated in patients with radiographic and symptomatic knee OA after a traumatic knee injury. A K-means cluster analysis was used to determine serum biochemical biomarker profiles based on z-score changes in sMCP-1, sCOMP, sMMP-3, sC2C:sCPII preoperatively to 6 months post-ACLR. One-way ANOVAs were used to determine statistically significant differences between biochemical biomarker changes and characterize the profiles of each group. Independent t-tests were used to assess differences in KOOS subscale scores between each group profile and P-values were Bonferroni corrected for multiple comparisons (p<0.0125). Cohen’s d effect sizes were calculated to determine the magnitude of differences between groups. Odds ratios (OR) and 95% confidence intervals (CIs) were also calculated to determine the odds of being classified as Symptomatic and reporting clinically relevant, OA-related symptoms.
Results:
Thirty-six participants completed all study visits and were dichotomized into groups based on changes in biochemical biomarker profiles over 6 months post-ACLR (Table 1). Group 1 (n=15) was characterized by increases in sMCP-1 and sCOMP preoperatively to 6-months post-ACLR (sMCP-1: p<0.001, d=1.75, 95% CI=[0.96-2.52] and sCOMP: p<0.001, d=1.45, 95% CI=[0.70-2.20]) and Group 2 (n=21) was characterized by decreases in sMCP-1 and sCOMP preoperatively to 6-months post-ACLR (Figure 1). No statistically significant differences were found for changes in sMMP-3 (p=0.18, d=0.46, 95% CI=[-0.22-1.13]) or sC2C:sCPII (p=0.10, d=-0.57, 95% CI=[-1.25-0.11]) preoperatively to 6-months post-ACLR between groups. Group 1 characterized by serum biochemical biomarker profiles with increasing sMCP-1 and sCOMP preoperatively to 6-months post-ACLR did not demonstrate statistically significant differences in self-reported knee pain (p=0.72, d=-0.12, 95% CI=[-0.79-0.54]), symptoms (p=0.96, d=0.02, 95% CI=[-0.64-0.68]), activities of daily living (p=0.89, d=-0.05, 95% CI=[-0.71-0.62]), sport (p=0.99, d=0.01, 95% CI=[-0.66-0.67]), or quality of life (p=0.99, d=0.00, 95% CI=[-0.66-0.67) at 12 months post-ACLR (Figure 2) compared to Group 2 characterized by serum biochemical biomarker profiles with decreasing sMCP-1 and sCOMP preoperatively to 6-months post-ACLR. There were no statistically significant associations between serum biochemical biomarker groups and symptomatic OA status (OR=2.13, 95% CI=0.51-9.01).
Conclusions:
Pre-operative and post-ACLR changes in biochemical biomarker sampling revealed a group of ACLR patients with increased biomarker concentrations associated with inflammation (sMCP-1) and cartilage breakdown (sCOMP). Serum biomarker changes from the pre-operative to the 6-month post-ACLR time points had no predictive value with respect to patient reporting of clinically relevant knee OA symptoms 12 months post-ACLR. Therefore, serum biochemical changes post-ACLR may play a role in worsening knee joint health as reported in prior studies, but not poor symptom development related to PTOA. Future research is needed to assess the mechanistic role of biochemical changes in symptomatic PTOA development.
