The most important findings of this study were that ADAMTS-4, ADAMTS-5, MMP-2 and MMP-9 were all elevated in postoperative samples compared to control samples in the acute phase after ACLR. ADAMTS-4 and − 5, as important mediators of aggrecan cleavage in OA knee cartilage, remained high during the study. These changes in the acute phase after ACLR seem likely to contribute to cartilage degeneration.
Regarding the relationship between intra-articular hemarthrosis and cartilage degeneration, Madhok et al. examined whether hemarthrosis or intra-articular bleeding led to various long-term problems and disorders, such as joint deformity and dysfunction [23]. Moreover, they showed that synovial tissues play critical roles in blood-induced joint disorders [23, 24]. A previous report indicated that exposure of cartilage to blood for 2 days in vitro led to prolonged impairment of joint cartilage [2, 3]. Some animal studies have shown that blood exerts deleterious effects on both cartilage and synovial tissues within short time spans [25, 26]. Therefore, one cause of posttraumatic OA after ACL injury may well be intra-articular hemarthrosis after surgery. A study by Tajima et al. reported that expression levels of gelatinases MMP-2 and − 9 and urokinase-type plasminogen activator (uPA) were elevated by human synovial cells at 24 h after exposure to Hb [2]. Another study found that Hb stimulated expression of ADAMTS-5 and − 9 by synovial cells within 24 h [4]. From our result and such previous studies, aggrecan and collagen cleavage may occur within a day after ACLR. During the study period of the present investigation, expression levels of MMP-2 were decreased on POD7 and levels of MMP-9 were decreased on POD4. However, levels of ADAMTS-4 and − 5 remained higher on POD7 than in control samples. Although no results were obtained beyond 7 days in this study, continuation of this response may result in aggrecan cleavage and subsequent cartilage degeneration. The mechanism of this cartilage degeneration is thought to involve Hb from intra-articular hemarthrosis after ACLR stimulating expression of ADAMTS-4 and − 5, MMP-2 and − 9, and uPA from synovial cells in the short term.
Some studies have reported cytokine concentrations in SF over the short term between pre- and post-ACLR. Zysk et al. reported significantly higher SF concentrations of interleukin (IL)-6 and bone morphogenetic proteins (BMP)-2 at 7 days after ACLR compared to preoperatively [27]. Akesen et al. found that tissue inhibitor of metalloproteinase-1 concentration in SF at 18 h after ACLR was significantly increased compared to preoperatively [28]. Hayward showed that IL-1 and IL-6 concentrations at 1 and 6 h after ACLR were increased compared to preoperatively [29]. These studies showed that inflammatory cytokines were increased in SF shortly after ACLR, and thus may contribute to cartilage degeneration. On the other hand, Roosendaal et al. investigated the effect of blood components on cartilage damage. They found that whole blood or a combination of mononuclear cells and erythrocytes inhibit proteoglycan synthesis and diminish the proteoglycan content in cartilage matrix. Such effects appear independent of IL-1 or tumor necrosis factor (TNF)-alpha production. Erythrocytes reportedly play an important role in blood-induced cartilage degeneration1,6. ADAMTS-4 and − 5 have been shown to cleave the aggrecan core protein at the aggrecanase-specific Glu373- Ala374 bond in the interglobular domain [30]. Those studies also showed that ADAMTS-4 and − 5 are important mediators of aggrecan cleavage in OA knee cartilage. Both proteinases are among the strongest aggrecanases, but ADAMTS-4 in particular is considered to be the main proteinase affecting cartilage in OA [31–33]. Moreover, expression levels of ADAMTS-4 in SF have been found to be particularly high in the early stages of knee OA compared to ADAMTS-5 [34]. According to Rogerson et al., ADAMTS-9 plays an important role in extracellular matrix remodeling during growth and development, and aggrecan in articular cartilage is also a substrate of ADAMTS-9. Accordingly, ADAMTS-9 might have more important roles to play in normal skeletal development compared with ADAMTS-4 and ADAMTS-5 [35]. While previous studies have used cultured articular cartilage to confirm ADAMTS-9 activity, postoperative hemarthrosis in this study was expected to result in low ADAMTS-9 activity from normal cartilage [31]. Animal studies have shown ADAMTS-4 is upregulated in the acute phase after ACL injury [14, 36].
Surgical ACLR reportedly constitutes a second trauma to the injured joint that leads to continuously elevated levels of inflammatory cytokines in SF [37]. Recipients of ACLR show higher cytokine concentrations in SF from the index knee than those treated with rehabilitation alone. However, ACLR is an absolutely necessary surgery to allow individuals to resume sporting activities and live daily life without feelings of knee instability. If surgical ACLR itself contributes to cartilage degeneration, further studies are needed to determine changes in the injured knee and after surgery.
Regarding cases of articular fracture, Justin et al. have reported concentrations of MMPs (MMP-1, -2, -3, -7, -9, -10, -12, and − 13) and aggrecan degradation in SF after acute intra-articular fracture compared to the uninjured joint in the same patients. They examined samples at two timepoints, with the first samples obtained in the first 24 h after injury and the second samples obtained at a mean of 9.5 days (range, 3–21 days) after the initial injury. The results showed that MMP-1, -3, -9, -10 and − 12 levels were elevated in acute injury knees. In addition, a weak association was apparent between elevated MMP concentration and fracture energy [38]. Finally, these MMPs and aggrecan degradation remained elevated at the second timepoint. These results and findings from our own study showed that several proteinases were elevated in the acute period after injury or after surgery. Moreover, ADAMTS-4, ADAMTS-5, MMP-2, and MMP-9 remained elevated at the second timepoint or POD7 after ACLR. Such evidence suggests that these MMPs and ADAMTS may play roles in chronic inflammation and the development of OA. The study by Justin et al. did not refer to causes of elevated proteinase levels [34]. If concentrations of MMPs and ADAMTS remain elevated due to intra-articular hemarthrosis, the effects on cartilage degeneration may be substantial.
For these reasons, removing and washing out articular hemarthrosis after ACLR by aspiration or irrigation may help prevent degeneration of the articular cartilage. Moreover, development of human neutralizing antibodies to ADAMTS-4 and − 5 has been reported and may lead to antibody treatments for OA [39]. The current study should draw attention to the postoperative presence of ADAMTS-4 and − 5 potentially playing a role in progression to OA in the short term after ACLR.
Various limitations must be acknowledged in this study. First, the sample size was small. Many potential cases had declined to undergo puncture out of concerns regarding needle pain, making samples from all points impossible to collect. Also, there were a few cases in which the volume of hemathrosis after ACLR was so small that analysis using ELISA was not possible. As statistical analysis, we did not perform any sample size calculation, because no similar studies have been undertaken previously. We therefore performed post hoc analysis. Second, we did not undertake a long-term analysis of SF after ACLR. If proteinases continue to be present in surgical knees, cartilage degeneration and posttraumatic OA may result. Finally, we did not examine serum results, as previous studies failed to show good correlations between serum and SF concentrations [40, 41].