- Email: [email protected]
Biomarker Changes in Anterior Cruciate Ligament–Deficient Knees Compared With Healthy Controls
Biomarker Changes in Anterior Cruciate LigamenteDeficient Knees Compared With Healthy Controls Daniel J. Kaplan, B.A., Vanessa G. Cuellar, M.D., Ph.D., Laith M. Jazrawi, M.D., and Eric J. Strauss, M.D.
Purpose: To establish how synovial fluid biomarker concentrations change in patients after anterior cruciate ligament (ACL) tears, with and without associated cartilage injury, with comparisons made to healthy controls. Methods: Patients were prospectively enrolled between January 2013 and December 2014. Inclusion criteria included any patient undergoing knee arthroscopy. Patients with a confirmed ACL tear were allocated to either the ACL tear with cartilage injury group or the ACL tear without cartilage injury group based on intraoperative assessment. Patients who underwent an arthroscopic procedure with no injury history or symptoms in their contralateral knee were asked to provide samples to serve as healthy controls. These subjects may or may not have been the same ones with noted ACL pathology. The concentrations of 20 biomarkers were determined using a multiplex magnetic bead immunoassay. Biomarker concentrations were then compared between the 3 study groups (ACL tears with and without cartilage injury, and uninjured contralateral knees) using an analysis of variance test with pairwise comparisons. The minimal clinically important difference was calculated based on the standard error of measurement. Results: The study included synovial fluid samples from 134 knees: 34 ACL tears without cartilage injury (mean age 34.0 years), 28 ACL tears with cartilage injury (mean age 36.3 years), and 72 healthy controls (mean age 41.1 years). Analysis of variance testing showed significant differences among groups for matrix metalloproteinase-3 (F ¼ 81.8; P < .001), tissue inhibitor of metalloproteinase (TIMP)-1 (F ¼ 7.9; P .001), TIMP-2 (F ¼ 4.5; P ¼ .015); fibroblast growth factor-2 (F ¼ 4.9; P ¼ .011), interleukin-6 (F ¼ 8.2; P ¼ .001), and macrophage inflammatory protein-1 beta (F ¼ 7.3; P ¼ .001). Pairwise comparisons showed no significant differences between ACL tears with, and without cartilage injury, but did show that both groups of ACL tears had significantly higher concentrations of (first P value ¼ ACL tears with and then ACL tears without cartilage injury): matrix metalloproteinase-3 (P < .001; P < .001), TIMP-1 (P < .001; P ¼ .011), interleukin-6 (P ¼ .009; P ¼ .038), and macrophage inflammatory protein-1 beta (P ¼ .003; P ¼ .045) compared with contralateral controls. ACL tears without associated cartilage damage had significantly lower concentrations of TIMP-2 (P ¼ .011) and fibroblast growth factor-2 (P ¼ .014) compared with controls. All biomarker concentration differences that reached statistical significance were also larger than calculated minimal clinically important differences. Conclusions: The current study identified 6 pro- and anti-inflammatory synovial fluid biomarkers whose concentrations after ACL injury were significantly different compared with uninjured controls. No significant differences in synovial fluid biomarker concentrations were seen between ACL injured knees with and without associated cartilage damage. Level of Evidence: Level III, retrospective comparative study of prospectively gathered data.
C
From the Division of Sports Medicine, Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, New York, New York, U.S.A. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received March 16, 2016; accepted November 21, 2016. Address correspondence to Eric J. Strauss, M.D., NYU Hospital for Joint Diseases, 333 East 38th Street, 4th Floor, New York, NY 10016, U.S.A. E-mail: [email protected] Ó 2016 by the Arthroscopy Association of North America 0749-8063/16227/$36.00 http://dx.doi.org/10.1016/j.arthro.2016.11.019
linical and functional outcomes after anterior cruciate ligament (ACL) reconstruction are largely positive with reported success rates ranging from 82% to 95%1-3; however, patients’ postoperative recovery is highly variable and they remain at high risk for posttraumatic osteoarthritis.4-6 Although surgical reconstruction of the ACL reliably restores stability to the knee, degenerative changes develop in 50% to 60% of treated patients at long-term follow-up.7,8 Rupture of the ACL leads to anterior translation of the tibia with an associated impaction of the posterolateral tibial plateau against the midportion of the lateral
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2017: pp 1-9
1
2
D. J. KAPLAN ET AL.
femoral condyle. In up to 80% of affected patients, this impaction injury is manifested as bone marrow edema lesions with increased signal intensity present on T2-weighted, fat-saturated magnetic resonance imaging.9-13 On impact, the articular cartilage overlying the bone marrow edema lesion is exposed to significant loads, potentially causing irreversible local injury and the release of proinflammatory, catabolic cytokines, and chemokines into the joint space.6,14-19 It has been theorized that this alteration in the local microenvironment at the time of ACL injury, and its associated chondral impaction, contributes to the development of generalized post-traumatic osteoarthritis.20 Synovial fluid biomarkers have gained significant recent attention because of their ability to provide a window into the molecular milieu associated with various knee joint pathologies. The purpose of the current study was to establish how synovial fluid biomarker concentrations change in patients after ACL tears, with and without associated cartilage injury, with comparisons made to healthy controls. We hypothesized that proinflammatory biomarkers and those associated with cartilage degradation would be significantly increased after ACL injury compared with uninjured controls, with the highest levels seen among patients with ACL tears with associated cartilage injuries.
Methods Population Patients were prospectively enrolled over a 2-year period, from January 2013 to December 2014, forming a database of subjects that had synovial fluid samples collected. Inclusion criteria for the overall database included any patient indicated for knee arthroscopy (e.g., meniscectomy, meniscal repair, ACL or posterior cruciate ligament reconstruction, etc.). Exclusion criteria included age <18 years, systemic inflammatory disease, autoimmune disease, intra-articular injection in the 3 months before surgery, prior knee surgery, immunomodulatory drug use, chemotherapy within the last year, insufficient synovial fluid aspiration, or cartilage/ meniscal transplantation in addition to arthroscopy. Any patient meeting criteria was invited to provide a synovial fluid sample before arthroscopy. For the current analysis, further inclusion criteria included only patients who subsequently were found to have an ACL injury with or without associated cartilage damage. Intraoperative findings were reported by the surgeon and recorded by a trained research assistant during the initial diagnostic arthroscopy, including soft tissue injury and cartilage pathology based on the International Cartilage Repair Society (ICRS) grading scale. All patients undergoing knee arthroscopy who reported no pain and no history of injury to their
contralateral knee were also invited to provide a sample from the contralateral knee at the time of surgery to serve as a healthy control. These subjects may or may not have been the same ones with noted ACL pathology. As this was a pilot study, all subject knees that qualified were included to maximize data gathering. This was a single-center institutional review board approved study. Synovial Fluid Collection and Storage In the operating room, after sterile preparation, but before arthroscopic portal creation, synovial fluid sampling was performed using the superolateral portal location, with an 18-gauge needle and a 10 cc syringe. The maximal amount of synovial fluid that egressed was collected; however, only a few microliters were necessary for molecular analysis. Synovial fluid samples were transferred to sterile tubes containing a protease inhibitor cocktail solution (Halt Protease Inhibitor Cocktail, EDTA-free; Pierce Biotechnology, Rockford, IL). The samples were stored on ice during transport to the laboratory where they were centrifuged at 1,350 g for 10 minutes and supernatant was aliquoted into sterile tubes, before storage at 80 C. Biomarker Analysis The samples were analyzed for the presence of 20 biomarkers using a multiplex bead assay (Milliplex, Millipore, Billerica, MA), including matrix metalloproteinase-3 (MMP-3), MMP-13, tissue inhibitor of metalloproteinase-1 (TIMP-1), TIMP-2, TIMP-3, TIMP-4, fibroblast growth factor (FGF-2), eotaxin, interferon gamma (IFNg), interleukin-10 (IL-10), platelet-derived growth factor-BB, interleukin-1 receptor antagonist (IL-1Ra), interleukin-1 beta (IL-1b), interleukin-6 (IL-6), monocyte chemotactic protein-1 (MCP-1), macrophage inflammatory protein-1 alpha (MIP-1a), macrophage inflammatory protein-1 beta (MIP-1b), regulated on activation, normal T-cell expressed and secreted, tumor necrosis factor alpha (TNFa), and vascular endothelial growth factor. These specific biomarkers were chosen because they are commonly found in joint pain, inflammation, and cartilage degeneration (Table 1). Before the assay of the synovial fluid, samples were treated with hyaluronidase and then assayed according to the manufacturer’s protocol. Statistical Methods All statistical analysis was done using SPSS 23 (IBM, Armonk, NY). Comparisons between patients with ACL injury with cartilage damage, ACL injury without cartilage damage, and healthy controls were made using the Welch analysis of variance test, because the assumption for homogeneity of variances was violated for most biomarkers when tested for
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES
3
Table 1. Description of Biomarkers Analyzed in This Study, Organized Based on the Primary Function Type of Biomarker Proinflammatory
Biomolecule MMP-3 MMP-13 TNFa IL-1b IL-6
MCP-1 MIP-1a MIP-1b RANTES IFNg
Anti-inflammatory
Growth factors
Eotaxin TIMP-1, -2, -3, -4 IL-10
FGF-2 PDGF-BB IL-1Ra VEGF
Basic Function(s) Stromelysin that degrades collagen II, IV, IX, X, XI, proteoglycans, fibronectin, laminin, and elastin; activates other MMPs Collagenase that degrades collagen I, II, III, IV, IX, X, XIV Cytokine produced by macrophages, lymphocytes, natural killer cells, neutrophils, and involved in acute phase of inflammatory response. Can induce cell apoptosis Cytokine produced by activated macrophages, important mediator of the inflammatory response; activates cyclooxygenase-2 Cytokine secreted by T cells and macrophages acting in proinflammatory function, especially during infection or after trauma. Can stimulate IL10 and IL-1Ra production to act in negative feedback loop of inflammatory response Chemokine that recruits monocytes, memory T cells, and dendritic cells to the site of inflammation by tissue injury or infection Chemokine produced by macrophages, activates neutrophils, eosinophils, and basophils in addition to other cytokines such as IL-1, IL-6, TNFa Chemokine recruits leukocytes to site of inflammation, activates natural killer cells Cytokine that is critical for innate and adaptive immunity against viral and bacterial infections; activator of macrophages and MHC-II expression Small chemokine that selectively recruits eosinophils Inhibition of MMPs; chondroprotective role Cytokine with anti-inflammatory effects via downregulation of Th1 cytokine expression, MHC-II antigens, and macrophage activators; enhances B-cell survival, blocks NF-kB activity Promotes angiogenesis, wound healing, and granulation tissue formation Growth factor that plays a role in embryonic development, cell proliferation, cell migration, and angiogenesis Cytokine that inhibits the proinflammatory effects of IL-1b Growth factor that induces angiogenesis
FGF, fibroblast growth factor; IFNg, interferon gamma; IL, interleukin; IL-1Ra, interleukin-1 receptor antagonist; IL-1b, interleukin-1 beta; MCP-1, monocyte chemotactic protein-1; MHC, major histocompatibility complex; MIP-1a, macrophage inflammatory protein-1 alpha; MIP-1b, macrophage inflammatory protein-1 beta; MMP, matrix metalloproteinase; NF-kB, nuclear nactor-kappa-Beta; PDGF-BB, platelet-derived growth factor-BB; RANTES, regulated on activation, normal T-cell expressed and secreted; Th1, T-helper cell type 1; TIMP, tissue inhibitor of metalloproteinase; TNFa, tumor necrosis factor alpha; VEGF, vascular endothelial growth factor.
by Levene’s statistic (P < .05). The assumption for normality was met when tested for by the Shapiro-Wilk test (P < .05). Post hoc pairwise comparisons were made with Games-Howell tests. The minimal clinically important difference (MCID), which can be used to help determine if the difference between 2 values has clinical relevance beyond statistical significance, was determined from the standard error measurements as described and validated in previous studies.21-23 For a biomarker to be of clinical relevance, the absolute difference between the mean concentration of the biomarker in the ACL-injured group with or without cartilage damage and the mean concentration in the control group would need to be larger than the calculated MCID (e.g., [mean of [MMP-3] in ACL injury with cartilage damage mean of [MMP-3] in the control group] > MCID).
Results Patient Demographics The study included synovial fluid samples from 134 knees that included 34 ACL tears without
associated cartilage damage (mean age 34 8.12 years), 28 ACL tears with cartilage associated damage (36.29 9.04 years), and 72 healthy controls (41.06 14.5 years). The average time from knee injury to sample collection was 5.5 weeks (range, 3-12 weeks). Among the patients with ACL tears with associated cartilage lesions found at the time of surgery, there were 16 ICRS grade 2 lesions, 10 ICRS grade 3 lesions, and 2 ICRS grade 4 lesions. Biomarker Comparisons Analysis of variance testing showed significant differences among groups for MMP-3 (F ¼ 81.8; P < .001), IL6 (F ¼ 8.2; P ¼ .001), MIP-1b (F ¼ 7.3; P ¼ .001), TIMP-1 (F ¼ 7.9; P < .001), TIMP-2 (F ¼ 4.5; P ¼ .015), and FGF2 (F ¼ 4.9; P ¼ .011) (Table 2). No significant differences were found between groups for the remainder of synovial fluid analytes assessed. Ninety-five percent confidence intervals (CI) can be seen in Table 2. Pairwise comparisons showed no significant differences between ACL tears with and without cartilage
4
Table 2. Means and Standard Deviations of Biomarker Concentrations With ANOVA F and P Values Means (pg/mL) Marker Class Proinflammatory
95% CI 460,803-585,140 0.0-48.6 0-3.5 158.1-642.8 809.3-1,159.2 12.1-18.1 80-284.8 6.9-23.0 67,177.0-79,891.1 39,852.2-47,577.8 10,830.5-16,161.8 370.0-588.7 1.8-8.9 12,271.6-15,298.7 70.1-118 375.9-442.4
% D From Control MCID 218.3 35,791
1,804.8
118.5
72.3
1.8
31.5 14.0
4,599.5 4,159.5
6.0
1,027.9
Healthy ACL-CI 95% CI % D From Control MCID Control 95% CI 537,985 477,704-598,266 227.5 35,226 164,285 126,234-202,336 28 0.0-67.0 0 0-0 0.89 0.0-1.8 1.19 0.3-2.1 191 58.0-323.4 804.8 66.0 21 1.1-41.3 945 793.2-1,095.8 752 565.1-938.2 13 11.4-16.7 52.9 1.9 9 6.8-10.6 281 71.8-490.9 872 209.4-1,534 17 9.4-23.6 26 18.3-34.3 70,108 63,291-76,924 25.3 4,773.0 55,939 49,149-62,727 42,431 40,139.3-44,724.0 19.2 3,426.4 52,542 49,084.2-58,009.0 13,762 11,064.9-16,459.9 12,798 9,801.4-14,208 426 323.4-528.6 484 337.9-629.7 1 0-2.9 1 0.1-2.6 12,376 11,688.6-13,064.1 15.6 791.7 14,659 13,229.2-16,088.5 169 75.5-263.3 864 69.8-1,658.3 398 358.5-438.4 452 414.1-490.0
NOTE. Any biomarker not listed was not detected on analysis. Bolded: found to be significant on ANOVA testing. D, change; 95% CI, 95% confidence interval; ACL-CD, anterior cruciate ligament with cartilage damage; ACL-CI, anterior cruciate ligament with cartilage intact; ANOVA, analysis of variance; FGF, fibroblast growth factor; IL, interleukin; MCID, minimal clinically important difference; MCP-1, monocyte chemotactic protein-1; MIP-1b, macrophage inflammatory protein-1 beta; MMP, matrix metalloproteinase; PDGF-BB, platelet-derived growth factor-BB; RANTES, regulated on activation, normal T-cell expressed and secreted; TIMP, tissue inhibitor of metalloproteinase; TNFa, tumor necrosis factor alpha; VEGF, vascular endothelial growth factor.
D. J. KAPLAN ET AL.
Biomarker ACL-CD MMP-3 522,972 MMP-13 20 1.42 TNFa IL-6 400 MCP-1 984 MIP-1b 15 RANTES 183 Eotaxin 15 Anti-inflammatory TIMP-1 73,534 TIMP-2 45,215 TIMP-3 13,496 TIMP-4 479 IL-10 5 Growth factors FGF-2 13,785 PDGF-BB 94 VEGF 409
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES
Fig 1. MMP-3 means versus ACL groups. It can be seen that contralateral controls significantly differed from knees with ACL tears. Cartilage injury, however, did not appreciably affect MMP-3 concentration. (ACL, anterior cruciate ligament; MMP, matrix metalloproteinase; SD, standard deviation.)
injury, but did show that both groups of ACL tears had significantly higher concentrations of [ACL tears with cartilage damage P value, % change from control; ACL tears without cartilage damage P value, % change from control]: MMP-3 (P < .001, 218.3%; P < .001, 227.5%), IL-6 (P ¼ .009, 1804%; P ¼ .038, 804.8%), MIP-1b (P ¼ .003, 72.3%; P ¼ .045, 52.9%), and TIMP-1 (P < .001, 31.5%; P ¼ .011, 25.3%) compared with contralateral controls (Figs 1-4; Table 3).
Fig 2. TIMP-1 means versus ACL groups. Again we see contralateral controls significantly differed from knees with ACL tears. Cartilage injury again did not appreciably affect the biomarker concentration. (ACL, anterior cruciate ligament; SD, standard deviation; TIMP, tissue inhibitor of metalloproteinase.)
5
Fig 3. IL-6 means versus ACL groups. Contralateral controls significantly differed from knees with ACL injuries for IL-6 concentration. Although there is an appreciable difference between ACL tears with and without associated cartilage injury, it is not significant. (ACL, anterior cruciate ligament; IL, interleukin; SD, standard deviation.)
ACL tears without cartilage damage had significantly lower concentrations of TIMP-2 (P ¼ .011) and FGF-2 (P ¼ .014) compared with controls (Figs 5 and 6; Table 3). All biomarker concentration differences that reached statistical significance were also larger than calculated MCIDs (Table 2).
Fig 4. MIP-1b means versus ACL groups. Although the absolute concentrations are relatively small compared with the other biomarkers, contralateral controls significantly differed from knees with ACL injuries for MIP-1b concentration. Cartilage injury did not significantly affect concentration. (ACL, anterior cruciate ligament; MIP-1b, macrophage inflammatory protein-1 beta; SD, standard deviation.)
6
D. J. KAPLAN ET AL.
Table 3. Games-Howell Post Hoc Comparisons of Biomarkers ANOVA Biomarker MMP-3 IL-6 MIP-1b TIMP-1 TIMP-2 FGF-2
F 81.8 8.2 7.3 7.9 4.5 4.9
Pairwise Comparisons
P <.00001 .001 .001 .001 .015 .011
ACL-CI vs ACL-CD 0.933 0.276 0.734 0.734 0.595 0.205
ACL-CD vs Control <.00001[ 0.009[ 0.003[ 0.0007[ 0.188 0.673
ACL-CI vs Control <.00001[ 0.038[ 0.045[ 0.011[ 0.011Y 0.014Y
NOTE. [: ACL injury has a significantly higher mean concentration than contralateral. Y: ACL injury has a significantly lower mean concentration than contralateral. ACL-CD, anterior cruciate ligament with cartilage damage; ACL-CI, anterior cruciate ligament with cartilage intact; ANOVA, analysis of variance; FGF, fibroblast growth factor; IL, interleukin; MIP-1b, macrophage inflammatory protein-1 beta; MMP, matrix metalloproteinase; TIMP, tissue inhibitor of metalloproteinase.
Discussion The current study identified 6 specific synovial fluid biomarkers that differed significantly between knees with ACL tears and uninjured contralateral controls (and those differences were larger than calculated MCIDs), but found no differences between knees with ACL tears without cartilage injury compared with ACL tears with cartilage injury. The use of biomarkers as part of the diagnostic workup and treatment decision-making process is gaining considerable attention and has become widely acknowledged in modern medicine. Biomarkers may be identified in the blood, urine, or synovial fluid of patients, and represent objective indicators of normal processes, pathology, or responses to therapeutic intervention. Most studies to date have investigated biomarkers in the blood or urine. However, it is becoming increasingly appreciated that intra-articular pathologies may be more accurately described by analysis of jointspecific synovial fluid.24 Of the biomarkers found to be significantly different between groups, 3 of these cytokines are proinflammatory and catabolic (MMP-3, IL-6, and MIP-1b), whereas the other 3 are anti-inflammatory or anabolic (TIMP-1, TIMP-2, and FGF-2). Each of these biomarkers had differences in mean between injury group and uninjured controls larger than the calculated MCID, indicating clinical relevance as well as statistical significance. These findings confirm data from previous smaller-sized studies that show similar cytokine concentration increases after ACL injury,25,26 as well as in the setting of meniscal tears.27 MMP-3 is an important enzyme for cartilage destruction and remodeling.28 Several studies suggest chondrocytes upregulate MMP expression after articular cartilage injury and during cartilage degradation.29-32 It is believed that MMPs increase during abnormal cartilage
Fig 5. TIMP-2 means versus ACL groups. One of 2 biomarkers where contralateral concentration was actually higher than ACL tear knees. Knees with and without cartilage injury did not differ significantly from each other; however, only those knees with ACL tears and no associated cartilage injury were significantly different from contralaterals. (ACL, anterior cruciate ligament; SD, standard deviation; TIMP, tissue inhibitor of metalloproteinase.)
loading, such as after the impaction injury that occurs during an ACL tear; however, they are also induced by proinflammatory cytokines, in particular IL-1 and TNF.33,34 IL-6 is a well-known cytokine associated with inducing production of acute phase reactants,25 and MIP-1b is a chemokine produced by macrophages that
Fig 6. FGF-2 means versus ACL groups. The second biomarkers where contralateral concentration was higher than ACL tear knees. Knees with and without cartilage injury did not differ significantly from each other; however, only those knees with ACL tears and no associated cartilage injury were significantly different from contralaterals. (ACL, anterior cruciate ligament; FGF, fibroblast growth factor; SD, standard deviation.)
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES
induces the expression of other proinflammatory cytokines.35 It is therefore not surprising that these proinflammatory mediators would be elevated after traumatic ACL injury. TIMP-1 and TIMP-2 are inhibitors of the matrix metalloproteinases, and are known to be upregulated after traumatic events. In contrast to previous ACL studies, we did not find elevations in ACL-deficient knees of IL-1Ra,36 MCP-1,26 or IFNg37 compared with healthy controls. It is logical that IL-1Ra, an antagonist of IL-1,25 is not increased, as we did not find IL-1 to be increased in our ACL-injured patients. MCP-1 is another chemokine and a prominent mediator of inflammation.38 Although our data did not find the mean concentrations significantly different between the ACL-deficient groups and the contralateral controls for MCP-1, the ACL groups both had higher mean concentrations. IFNg has been shown to falsely produce a signal with multiplexed immunoassays,39 which could explain its occasional inclusion in the literature. Prior literature has shown that IL-1 and TNFa, 2 of the most important mediators of inflammation, are typically absent during knee injury.20,25,26,36 Our study confirms this point. The lack of these traditional inflammatory mediators, along with IFNg, points to ACL injury inducing a specific injury response, as opposed to a more generalized inflammatory response. Interestingly, we did not find any significant differences in synovial fluid biomarker concentrations between ACL injured knees with and without cartilage damage. One rationale for this is that the cytokine response could be due to the impaction injury that occurs during the pivot-shift mechanism of ACL rupture. This would explain why a similar cytokine response develops with or without gross articular cartilage damage. It is possible that the local alteration in microenvironment that occurs at the time of ACL injury contributes to the development of generalized post-traumatic osteoarthritis, regardless of associated cartilage injury. However, it is possible that ACL tears with associated cartilage damage may have a more pronounced impact at a later point in time, and if samples were obtained further out from the time of injury, a significant difference would be detected. Previous literature has not separated the ACL patients by association with cartilage damage, so we cannot compare these results with other studies. Although no a priori power analysis was performed as this is a pilot study, for the 6 variables found to be significant, 95% CI between ACL tear subjects with and without cartilage damage had considerable overlap, indicating type II (beta) error is unlikely, and these groups were in fact not significantly different. Another interesting finding is the significantly lower concentrations of TIMP-2 and FGF-2 in the ACL tear groups compared with the contralateral controls. We believe that in the setting of an ACL injury, the
7
anti-inflammatory TIMP-2 and the anabolic FGF-2 may be consumed or downregulated in response to the traumatic event and with exposure to the increased levels of the proinflammatory and catabolic MMP-3, IL-6, and MIP-1b. Four biomarkers were not found to be significantly different on analysis of variance testing, but did have considerably different means, and relatively little overlap of 95% CI including regulated on activation, normal T-cell expressed and secreted, eotaxin, IL-10, and platelet-derived growth factor-BB. However, with the exception of IL-10, the 3 biomarkers had similar mean and 95% CI between the ACL tears with and without cartilage injury groups (and higher biomarker concentrations in the control group), indicating that even if a significant relation exists that was not picked up due to beta error, it would be consistent with our other findings (i.e., cartilage damage does not impact biomarker concentrations). Limitations The current study assesses only a single point in timedshortly after injury and just before surgery. Prior studies have shown an immediate biochemical response after acute knee injury, with variable changes in cytokine concentrations during the postinjury time course.40,41 Another limitation of the current study is the lack of stratification of patients based on the energy level of their injury. We would hypothesize that patients who sustain an ACL rupture with a higher energy mechanism would have a faster and larger biochemical response to the injury. In addition, the current investigation did not correlate clinical and functional outcomes with synovial fluid biomarker levels seen at the time of surgery. We also did not investigate the bone marrow edema lesion presence or size correlation with biomarker levels, which limits our proposed pivot-shift mechanism to a theory. It is also important to note that the specimens obtained as healthy controls provided only a clinical control, because arthroscopic evaluation was not possible, and were from patients indicated for knee arthroscopy in the contralateral leg. Finally, our observations and analysis are limited by our relatively small numbers of patients, opening up the possibility of type II (beta) error.
Conclusions The current study identified 6 pro- and antiinflammatory synovial fluid biomarkers whose concentrations after ACL injury were significantly different compared with uninjured controls and those differences were larger than calculated MCIDs. No significant differences in synovial fluid biomarker concentrations were seen between ACL-injured knees with and without cartilage damage.
8
D. J. KAPLAN ET AL.
References 1. Roe J, Pinczewski LA, Russell VJ, Salmon LJ, Kawamata T, Chew M. A 7-year follow-up of patellar tendon and hamstring tendon grafts for arthroscopic anterior cruciate ligament reconstruction: Differences and similarities. Am J Sports Med 2005;33:1337-1345. 2. Hui C, Salmon LJ, Kok A, Maeno S, Linklater J, Pinczewski LA. Fifteen-year outcome of endoscopic anterior cruciate ligament reconstruction with patellar tendon autograft for “isolated” anterior cruciate ligament tear. Am J Sports Med 2011;39:89-98. 3. Spindler KP, Wright RW. Prospective study of osseous, articular, and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy. N Engl J Med 2008;359:2135-2142. 4. Feagin JA Jr. The syndrome of the torn anterior cruciate ligament. Orthop Clin North Am 1979;10:81-90. 5. Lohmander LS, Ostenberg A, Englund M, Roos H. High prevalence of knee osteoarthritis, pain, and functional limitations in female soccer players twelve years after anterior cruciate ligament injury. Arthritis Rheum 2004;50: 3145-3152. 6. Theologis AA, Kuo D, Cheng J, et al. Evaluation of bone bruises and associated cartilage in anterior cruciate ligament-injured and reconstructed knees using quantitative t(1r) magnetic resonance imaging: 1-year cohort study. Arthroscopy 2011;27:65-76. 7. Daniel D, Stone M, Dobson B, Fithian D, Rossman D, Kaufman K. Fate of the ACL-injured patient: A prospective outcome study. Am J Sports Med 2014;22:149-152. 8. Ferretti A, Conteduca F, De Carli A, Fontana M, Mariani PP. Osteoarthritis of the knee after ACL reconstruction. Int Orthop 1991;15:367-371. 9. Dimond P, Fadale P, Hulstyn M, Tung G, Greisberg J. A comparison of MRI findings in patients with acute and chronic ACL tears. Am J Knee Surg 1998;11:153-159. 10. Graf BK, Cook DA, De Smet AA, Keene JS. “Bone bruises” on magnetic resonance imaging evaluation of anterior cruciate ligament injuries. Am J Sports Med 1993;21:220-223. 11. Kaplan P, Walker C, Kilcoyne R, Brown D, Dussault R. Occult fracture patterns of the knee associated with anterior cruciate ligament tears: Assessment with MR imaging. Radiology 1992;183:835-838. 12. Kijowski R, Sanogo ML, Lee KS, et al. Short-term clinical importance of osseous injuries diagnosed at MR imaging in patients with anterior cruciate ligament tear. Radiology 2012;264:531-541. 13. Spindler KP, Schils JP, Bergfeld JA, et al. Prospective study of osseous, articular, and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy. Am J Sports Med 1993;21: 551-557. 14. Anderson DD, Chubinskaya S, Guilak F, et al. Posttraumatic osteoarthritis: Improved understanding and opportunities for early intervention. J Orthop Res 2012;29: 802-809. 15. Johnson DL, Urban WP, Caborn DN, Vanarthos WJ, Carlson CS. Articular cartilage changes seen with magnetic resonance imaging-detected bone bruises associated
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
30.
31.
with acute anterior cruciate ligament rupture. Am J Sports Med 1998;26:409-414. Lahm A, Uhl M, Erggelet C, Haberstroh J, Mrosek E. Articular cartilage degeneration after acute subchondral bone damage: An experimental study in dogs with histopathological grading. Acta Orthop Scand 2004;75: 762-767. Lewis JL, Deloria LB, Oyen-Tiesma M, Thompson RC, Ericson M, Oegema TR. Cell death after cartilage impact occurs around matrix cracks. J Orthop Res 2003;21: 881-887. Martin JA, Brown T, Heiner A, Buckwalter JA. Posttraumatic osteoarthritis: The role of accelerated chondrocyte senescence. Biorheology 2004;41:479-491. Martin JA, Buckwalter JA. Post-traumatic osteoarthritis: The role of stress induced chondrocyte damage. Biorheology 2006;43:517-521. Akesen B, Demirag B, Budak F. Evaluation of intraarticular collagenase, TIMP-1, and TNF-alpha levels before and after anterior cruciate ligament reconstruction. Acta Orthop Traumatol Turc 2009;43:214-218. Rejas J, Pardo A, Ruiz MÁ. Standard error of measurement as a valid alternative to minimally important difference for evaluating the magnitude of changes in patient-reported outcomes measures. J Clin Epidemiol 2008;61:350-356. Wyrwich KW, Nienaber N, Tierney WM, Wolinsky FD. Linking clinical relevance and statistical significance in evaluating intra-individual changes in health-related quality of life. Med Care 1999;37:469-478. Wyrwich KW, Tierney WM, Wolinsky FD. Further evidence supporting an SEM-based criterion for identifying meaningful intra-individual changes in health-related quality of life. J Clin Epidemiol 1999;52:861-873. Ruiz-Romero C, Blanco FJ. Proteomics role in the search for improved diagnosis, prognosis and treatment of osteoarthritis. Osteoarthr Cartil 2010;18:500-509. Higuchi H, Shirakura K, Kimura M, et al. Changes in biochemical parameters after anterior cruciate ligament injury. Int Orthop 2006;30:43-47. Cuellar VG, Cuellar JM, Golish SR, Yeomans DC, Scuderi GJ. Cytokine profiling in acute anterior cruciate ligament injury. Arthroscopy 2010;26:1296-1301. Cuellar JM, Scuderi GJ, Cuellar VG, Golish SR, Yeomans DC. Diagnostic utility of cytokine biomarkers in the evaluation of acute knee pain. J Bone Joint Surg Am 2009;91:2313-2320. Emonard H, Grimaud JA. Matrix metalloproteinases. A review. Cell Mol Biol 1990;36:131-153. Billinghurst RC, Dahlberg L, Ionescu M, et al. Enhanced cleavage of type II collagen by collagenases in osteoarthritic articular cartilage. J Clin Invest 1997;99:1534-1545. Mehraban F, Kuo SY, Riera H, Chang C, Moskowitz RW. Prostromelysin and procollagenase genes are differentially up-regulated in chondrocytes from the knees of rabbits with experimental osteoarthritis. Arthritis Rheum 1994;37: 1189-1197. Shlopov BV, Gumanovskaya ML, Hasty KA. Autocrine regulation of collagenase 3 (matrix metalloproteinase 13) during osteoarthritis. Arthritis Rheum 2000;43:195-205.
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES 32. Dahlberg L, Billinghurst RC, Webb G, et al. Selective enhancement of collagenase-mediated cleavage of resident type II collagen in cultured osteoarthritic cartilage and arrest with a synthetic inhibitor that spares collagenase 1 (Matrix metalloproteinase 1). Arthritis Rheum 2000;43:673-682. 33. Harkey MS, Luc BA, Golightly YM, et al. Osteoarthritisrelated biomarkers following anterior cruciate ligament injury and reconstruction: A systematic review. Osteoarthr Cartil 2015;23:1-12. 34. Van de Loo FAJ, Joosten LAB, Van Lent PLEM, Arntz OJ, Van den Berg WB. Role of interleukin-1, tumor necrosis factor alpha, and interleukin-6 in cartilage proteoglycan metabolism and destruction. Effect of in situ blocking in murine antigen- and zymosan-induced arthritis. Arthritis Rheum 1995;38:164-172. 35. Matute Wilander A, Kåredal M, Axmon A, Nordander C. Inflammatory biomarkers in serum in subjects with and without work related neck/shoulder complaints. BMC Musculoskelet Disord 2014;15:103. 36. Cameron ML, Fu FH, Paessler HH, Schneider M, Evans CH. Synovial fluid cytokine concentrations as
37.
38.
39.
40.
41.
9
possible prognostic indicators in the ACL-deficient knee. Knee Surg Sports Traumatol Arthrosc 1994;2:38-44. Ahlen M, Roshani L, Liden M, Struglics A, RostgardChristensen L, Kartus J. Inflammatory cytokines and biomarkers of cartilage metabolism 8 years after anterior cruciate ligament reconstruction: Results from operated and contralateral knees. Am J Sports Med 2015;43:1460-1466. Deshmane SL, Kremlev S, Amini S, Sawaya BE. Monocyte chemoattractant protein-1 (MCP-1): An overview. J Interferon Cytokine Res 2009;29:313-326. Scuderi GJ, Woolf N, Dent K, et al. Identification of a complex between fibronectin and aggrecan G3 domain in synovial fluid of patients with painful meniscal pathology. Clin Biochem 2010;43:808-814. Swärd P, Frobell R, Englund M, Roos H, Struglics A. Cartilage and bone markers and inflammatory cytokines are increased in synovial fluid in the acute phase of knee injury (hemarthrosis)da cross-sectional analysis. Osteoarthr Cartil 2012;20:1302-1308. Catterall JB, Stabler TV, Flannery CR, Kraus VB. Changes in serum and synovial fluid biomarkers after acute injury (NCT00332254). Arthritis Res Ther 2010;12:R229.
Purpose: To establish how synovial fluid biomarker concentrations change in patients after anterior cruciate ligament (ACL) tears, with and without associated cartilage injury, with comparisons made to healthy controls. Methods: Patients were prospectively enrolled between January 2013 and December 2014. Inclusion criteria included any patient undergoing knee arthroscopy. Patients with a confirmed ACL tear were allocated to either the ACL tear with cartilage injury group or the ACL tear without cartilage injury group based on intraoperative assessment. Patients who underwent an arthroscopic procedure with no injury history or symptoms in their contralateral knee were asked to provide samples to serve as healthy controls. These subjects may or may not have been the same ones with noted ACL pathology. The concentrations of 20 biomarkers were determined using a multiplex magnetic bead immunoassay. Biomarker concentrations were then compared between the 3 study groups (ACL tears with and without cartilage injury, and uninjured contralateral knees) using an analysis of variance test with pairwise comparisons. The minimal clinically important difference was calculated based on the standard error of measurement. Results: The study included synovial fluid samples from 134 knees: 34 ACL tears without cartilage injury (mean age 34.0 years), 28 ACL tears with cartilage injury (mean age 36.3 years), and 72 healthy controls (mean age 41.1 years). Analysis of variance testing showed significant differences among groups for matrix metalloproteinase-3 (F ¼ 81.8; P < .001), tissue inhibitor of metalloproteinase (TIMP)-1 (F ¼ 7.9; P .001), TIMP-2 (F ¼ 4.5; P ¼ .015); fibroblast growth factor-2 (F ¼ 4.9; P ¼ .011), interleukin-6 (F ¼ 8.2; P ¼ .001), and macrophage inflammatory protein-1 beta (F ¼ 7.3; P ¼ .001). Pairwise comparisons showed no significant differences between ACL tears with, and without cartilage injury, but did show that both groups of ACL tears had significantly higher concentrations of (first P value ¼ ACL tears with and then ACL tears without cartilage injury): matrix metalloproteinase-3 (P < .001; P < .001), TIMP-1 (P < .001; P ¼ .011), interleukin-6 (P ¼ .009; P ¼ .038), and macrophage inflammatory protein-1 beta (P ¼ .003; P ¼ .045) compared with contralateral controls. ACL tears without associated cartilage damage had significantly lower concentrations of TIMP-2 (P ¼ .011) and fibroblast growth factor-2 (P ¼ .014) compared with controls. All biomarker concentration differences that reached statistical significance were also larger than calculated minimal clinically important differences. Conclusions: The current study identified 6 pro- and anti-inflammatory synovial fluid biomarkers whose concentrations after ACL injury were significantly different compared with uninjured controls. No significant differences in synovial fluid biomarker concentrations were seen between ACL injured knees with and without associated cartilage damage. Level of Evidence: Level III, retrospective comparative study of prospectively gathered data.
C
From the Division of Sports Medicine, Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, New York, New York, U.S.A. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received March 16, 2016; accepted November 21, 2016. Address correspondence to Eric J. Strauss, M.D., NYU Hospital for Joint Diseases, 333 East 38th Street, 4th Floor, New York, NY 10016, U.S.A. E-mail: [email protected] Ó 2016 by the Arthroscopy Association of North America 0749-8063/16227/$36.00 http://dx.doi.org/10.1016/j.arthro.2016.11.019
linical and functional outcomes after anterior cruciate ligament (ACL) reconstruction are largely positive with reported success rates ranging from 82% to 95%1-3; however, patients’ postoperative recovery is highly variable and they remain at high risk for posttraumatic osteoarthritis.4-6 Although surgical reconstruction of the ACL reliably restores stability to the knee, degenerative changes develop in 50% to 60% of treated patients at long-term follow-up.7,8 Rupture of the ACL leads to anterior translation of the tibia with an associated impaction of the posterolateral tibial plateau against the midportion of the lateral
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2017: pp 1-9
1
2
D. J. KAPLAN ET AL.
femoral condyle. In up to 80% of affected patients, this impaction injury is manifested as bone marrow edema lesions with increased signal intensity present on T2-weighted, fat-saturated magnetic resonance imaging.9-13 On impact, the articular cartilage overlying the bone marrow edema lesion is exposed to significant loads, potentially causing irreversible local injury and the release of proinflammatory, catabolic cytokines, and chemokines into the joint space.6,14-19 It has been theorized that this alteration in the local microenvironment at the time of ACL injury, and its associated chondral impaction, contributes to the development of generalized post-traumatic osteoarthritis.20 Synovial fluid biomarkers have gained significant recent attention because of their ability to provide a window into the molecular milieu associated with various knee joint pathologies. The purpose of the current study was to establish how synovial fluid biomarker concentrations change in patients after ACL tears, with and without associated cartilage injury, with comparisons made to healthy controls. We hypothesized that proinflammatory biomarkers and those associated with cartilage degradation would be significantly increased after ACL injury compared with uninjured controls, with the highest levels seen among patients with ACL tears with associated cartilage injuries.
Methods Population Patients were prospectively enrolled over a 2-year period, from January 2013 to December 2014, forming a database of subjects that had synovial fluid samples collected. Inclusion criteria for the overall database included any patient indicated for knee arthroscopy (e.g., meniscectomy, meniscal repair, ACL or posterior cruciate ligament reconstruction, etc.). Exclusion criteria included age <18 years, systemic inflammatory disease, autoimmune disease, intra-articular injection in the 3 months before surgery, prior knee surgery, immunomodulatory drug use, chemotherapy within the last year, insufficient synovial fluid aspiration, or cartilage/ meniscal transplantation in addition to arthroscopy. Any patient meeting criteria was invited to provide a synovial fluid sample before arthroscopy. For the current analysis, further inclusion criteria included only patients who subsequently were found to have an ACL injury with or without associated cartilage damage. Intraoperative findings were reported by the surgeon and recorded by a trained research assistant during the initial diagnostic arthroscopy, including soft tissue injury and cartilage pathology based on the International Cartilage Repair Society (ICRS) grading scale. All patients undergoing knee arthroscopy who reported no pain and no history of injury to their
contralateral knee were also invited to provide a sample from the contralateral knee at the time of surgery to serve as a healthy control. These subjects may or may not have been the same ones with noted ACL pathology. As this was a pilot study, all subject knees that qualified were included to maximize data gathering. This was a single-center institutional review board approved study. Synovial Fluid Collection and Storage In the operating room, after sterile preparation, but before arthroscopic portal creation, synovial fluid sampling was performed using the superolateral portal location, with an 18-gauge needle and a 10 cc syringe. The maximal amount of synovial fluid that egressed was collected; however, only a few microliters were necessary for molecular analysis. Synovial fluid samples were transferred to sterile tubes containing a protease inhibitor cocktail solution (Halt Protease Inhibitor Cocktail, EDTA-free; Pierce Biotechnology, Rockford, IL). The samples were stored on ice during transport to the laboratory where they were centrifuged at 1,350 g for 10 minutes and supernatant was aliquoted into sterile tubes, before storage at 80 C. Biomarker Analysis The samples were analyzed for the presence of 20 biomarkers using a multiplex bead assay (Milliplex, Millipore, Billerica, MA), including matrix metalloproteinase-3 (MMP-3), MMP-13, tissue inhibitor of metalloproteinase-1 (TIMP-1), TIMP-2, TIMP-3, TIMP-4, fibroblast growth factor (FGF-2), eotaxin, interferon gamma (IFNg), interleukin-10 (IL-10), platelet-derived growth factor-BB, interleukin-1 receptor antagonist (IL-1Ra), interleukin-1 beta (IL-1b), interleukin-6 (IL-6), monocyte chemotactic protein-1 (MCP-1), macrophage inflammatory protein-1 alpha (MIP-1a), macrophage inflammatory protein-1 beta (MIP-1b), regulated on activation, normal T-cell expressed and secreted, tumor necrosis factor alpha (TNFa), and vascular endothelial growth factor. These specific biomarkers were chosen because they are commonly found in joint pain, inflammation, and cartilage degeneration (Table 1). Before the assay of the synovial fluid, samples were treated with hyaluronidase and then assayed according to the manufacturer’s protocol. Statistical Methods All statistical analysis was done using SPSS 23 (IBM, Armonk, NY). Comparisons between patients with ACL injury with cartilage damage, ACL injury without cartilage damage, and healthy controls were made using the Welch analysis of variance test, because the assumption for homogeneity of variances was violated for most biomarkers when tested for
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES
3
Table 1. Description of Biomarkers Analyzed in This Study, Organized Based on the Primary Function Type of Biomarker Proinflammatory
Biomolecule MMP-3 MMP-13 TNFa IL-1b IL-6
MCP-1 MIP-1a MIP-1b RANTES IFNg
Anti-inflammatory
Growth factors
Eotaxin TIMP-1, -2, -3, -4 IL-10
FGF-2 PDGF-BB IL-1Ra VEGF
Basic Function(s) Stromelysin that degrades collagen II, IV, IX, X, XI, proteoglycans, fibronectin, laminin, and elastin; activates other MMPs Collagenase that degrades collagen I, II, III, IV, IX, X, XIV Cytokine produced by macrophages, lymphocytes, natural killer cells, neutrophils, and involved in acute phase of inflammatory response. Can induce cell apoptosis Cytokine produced by activated macrophages, important mediator of the inflammatory response; activates cyclooxygenase-2 Cytokine secreted by T cells and macrophages acting in proinflammatory function, especially during infection or after trauma. Can stimulate IL10 and IL-1Ra production to act in negative feedback loop of inflammatory response Chemokine that recruits monocytes, memory T cells, and dendritic cells to the site of inflammation by tissue injury or infection Chemokine produced by macrophages, activates neutrophils, eosinophils, and basophils in addition to other cytokines such as IL-1, IL-6, TNFa Chemokine recruits leukocytes to site of inflammation, activates natural killer cells Cytokine that is critical for innate and adaptive immunity against viral and bacterial infections; activator of macrophages and MHC-II expression Small chemokine that selectively recruits eosinophils Inhibition of MMPs; chondroprotective role Cytokine with anti-inflammatory effects via downregulation of Th1 cytokine expression, MHC-II antigens, and macrophage activators; enhances B-cell survival, blocks NF-kB activity Promotes angiogenesis, wound healing, and granulation tissue formation Growth factor that plays a role in embryonic development, cell proliferation, cell migration, and angiogenesis Cytokine that inhibits the proinflammatory effects of IL-1b Growth factor that induces angiogenesis
FGF, fibroblast growth factor; IFNg, interferon gamma; IL, interleukin; IL-1Ra, interleukin-1 receptor antagonist; IL-1b, interleukin-1 beta; MCP-1, monocyte chemotactic protein-1; MHC, major histocompatibility complex; MIP-1a, macrophage inflammatory protein-1 alpha; MIP-1b, macrophage inflammatory protein-1 beta; MMP, matrix metalloproteinase; NF-kB, nuclear nactor-kappa-Beta; PDGF-BB, platelet-derived growth factor-BB; RANTES, regulated on activation, normal T-cell expressed and secreted; Th1, T-helper cell type 1; TIMP, tissue inhibitor of metalloproteinase; TNFa, tumor necrosis factor alpha; VEGF, vascular endothelial growth factor.
by Levene’s statistic (P < .05). The assumption for normality was met when tested for by the Shapiro-Wilk test (P < .05). Post hoc pairwise comparisons were made with Games-Howell tests. The minimal clinically important difference (MCID), which can be used to help determine if the difference between 2 values has clinical relevance beyond statistical significance, was determined from the standard error measurements as described and validated in previous studies.21-23 For a biomarker to be of clinical relevance, the absolute difference between the mean concentration of the biomarker in the ACL-injured group with or without cartilage damage and the mean concentration in the control group would need to be larger than the calculated MCID (e.g., [mean of [MMP-3] in ACL injury with cartilage damage mean of [MMP-3] in the control group] > MCID).
Results Patient Demographics The study included synovial fluid samples from 134 knees that included 34 ACL tears without
associated cartilage damage (mean age 34 8.12 years), 28 ACL tears with cartilage associated damage (36.29 9.04 years), and 72 healthy controls (41.06 14.5 years). The average time from knee injury to sample collection was 5.5 weeks (range, 3-12 weeks). Among the patients with ACL tears with associated cartilage lesions found at the time of surgery, there were 16 ICRS grade 2 lesions, 10 ICRS grade 3 lesions, and 2 ICRS grade 4 lesions. Biomarker Comparisons Analysis of variance testing showed significant differences among groups for MMP-3 (F ¼ 81.8; P < .001), IL6 (F ¼ 8.2; P ¼ .001), MIP-1b (F ¼ 7.3; P ¼ .001), TIMP-1 (F ¼ 7.9; P < .001), TIMP-2 (F ¼ 4.5; P ¼ .015), and FGF2 (F ¼ 4.9; P ¼ .011) (Table 2). No significant differences were found between groups for the remainder of synovial fluid analytes assessed. Ninety-five percent confidence intervals (CI) can be seen in Table 2. Pairwise comparisons showed no significant differences between ACL tears with and without cartilage
4
Table 2. Means and Standard Deviations of Biomarker Concentrations With ANOVA F and P Values Means (pg/mL) Marker Class Proinflammatory
95% CI 460,803-585,140 0.0-48.6 0-3.5 158.1-642.8 809.3-1,159.2 12.1-18.1 80-284.8 6.9-23.0 67,177.0-79,891.1 39,852.2-47,577.8 10,830.5-16,161.8 370.0-588.7 1.8-8.9 12,271.6-15,298.7 70.1-118 375.9-442.4
% D From Control MCID 218.3 35,791
1,804.8
118.5
72.3
1.8
31.5 14.0
4,599.5 4,159.5
6.0
1,027.9
Healthy ACL-CI 95% CI % D From Control MCID Control 95% CI 537,985 477,704-598,266 227.5 35,226 164,285 126,234-202,336 28 0.0-67.0 0 0-0 0.89 0.0-1.8 1.19 0.3-2.1 191 58.0-323.4 804.8 66.0 21 1.1-41.3 945 793.2-1,095.8 752 565.1-938.2 13 11.4-16.7 52.9 1.9 9 6.8-10.6 281 71.8-490.9 872 209.4-1,534 17 9.4-23.6 26 18.3-34.3 70,108 63,291-76,924 25.3 4,773.0 55,939 49,149-62,727 42,431 40,139.3-44,724.0 19.2 3,426.4 52,542 49,084.2-58,009.0 13,762 11,064.9-16,459.9 12,798 9,801.4-14,208 426 323.4-528.6 484 337.9-629.7 1 0-2.9 1 0.1-2.6 12,376 11,688.6-13,064.1 15.6 791.7 14,659 13,229.2-16,088.5 169 75.5-263.3 864 69.8-1,658.3 398 358.5-438.4 452 414.1-490.0
NOTE. Any biomarker not listed was not detected on analysis. Bolded: found to be significant on ANOVA testing. D, change; 95% CI, 95% confidence interval; ACL-CD, anterior cruciate ligament with cartilage damage; ACL-CI, anterior cruciate ligament with cartilage intact; ANOVA, analysis of variance; FGF, fibroblast growth factor; IL, interleukin; MCID, minimal clinically important difference; MCP-1, monocyte chemotactic protein-1; MIP-1b, macrophage inflammatory protein-1 beta; MMP, matrix metalloproteinase; PDGF-BB, platelet-derived growth factor-BB; RANTES, regulated on activation, normal T-cell expressed and secreted; TIMP, tissue inhibitor of metalloproteinase; TNFa, tumor necrosis factor alpha; VEGF, vascular endothelial growth factor.
D. J. KAPLAN ET AL.
Biomarker ACL-CD MMP-3 522,972 MMP-13 20 1.42 TNFa IL-6 400 MCP-1 984 MIP-1b 15 RANTES 183 Eotaxin 15 Anti-inflammatory TIMP-1 73,534 TIMP-2 45,215 TIMP-3 13,496 TIMP-4 479 IL-10 5 Growth factors FGF-2 13,785 PDGF-BB 94 VEGF 409
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES
Fig 1. MMP-3 means versus ACL groups. It can be seen that contralateral controls significantly differed from knees with ACL tears. Cartilage injury, however, did not appreciably affect MMP-3 concentration. (ACL, anterior cruciate ligament; MMP, matrix metalloproteinase; SD, standard deviation.)
injury, but did show that both groups of ACL tears had significantly higher concentrations of [ACL tears with cartilage damage P value, % change from control; ACL tears without cartilage damage P value, % change from control]: MMP-3 (P < .001, 218.3%; P < .001, 227.5%), IL-6 (P ¼ .009, 1804%; P ¼ .038, 804.8%), MIP-1b (P ¼ .003, 72.3%; P ¼ .045, 52.9%), and TIMP-1 (P < .001, 31.5%; P ¼ .011, 25.3%) compared with contralateral controls (Figs 1-4; Table 3).
Fig 2. TIMP-1 means versus ACL groups. Again we see contralateral controls significantly differed from knees with ACL tears. Cartilage injury again did not appreciably affect the biomarker concentration. (ACL, anterior cruciate ligament; SD, standard deviation; TIMP, tissue inhibitor of metalloproteinase.)
5
Fig 3. IL-6 means versus ACL groups. Contralateral controls significantly differed from knees with ACL injuries for IL-6 concentration. Although there is an appreciable difference between ACL tears with and without associated cartilage injury, it is not significant. (ACL, anterior cruciate ligament; IL, interleukin; SD, standard deviation.)
ACL tears without cartilage damage had significantly lower concentrations of TIMP-2 (P ¼ .011) and FGF-2 (P ¼ .014) compared with controls (Figs 5 and 6; Table 3). All biomarker concentration differences that reached statistical significance were also larger than calculated MCIDs (Table 2).
Fig 4. MIP-1b means versus ACL groups. Although the absolute concentrations are relatively small compared with the other biomarkers, contralateral controls significantly differed from knees with ACL injuries for MIP-1b concentration. Cartilage injury did not significantly affect concentration. (ACL, anterior cruciate ligament; MIP-1b, macrophage inflammatory protein-1 beta; SD, standard deviation.)
6
D. J. KAPLAN ET AL.
Table 3. Games-Howell Post Hoc Comparisons of Biomarkers ANOVA Biomarker MMP-3 IL-6 MIP-1b TIMP-1 TIMP-2 FGF-2
F 81.8 8.2 7.3 7.9 4.5 4.9
Pairwise Comparisons
P <.00001 .001 .001 .001 .015 .011
ACL-CI vs ACL-CD 0.933 0.276 0.734 0.734 0.595 0.205
ACL-CD vs Control <.00001[ 0.009[ 0.003[ 0.0007[ 0.188 0.673
ACL-CI vs Control <.00001[ 0.038[ 0.045[ 0.011[ 0.011Y 0.014Y
NOTE. [: ACL injury has a significantly higher mean concentration than contralateral. Y: ACL injury has a significantly lower mean concentration than contralateral. ACL-CD, anterior cruciate ligament with cartilage damage; ACL-CI, anterior cruciate ligament with cartilage intact; ANOVA, analysis of variance; FGF, fibroblast growth factor; IL, interleukin; MIP-1b, macrophage inflammatory protein-1 beta; MMP, matrix metalloproteinase; TIMP, tissue inhibitor of metalloproteinase.
Discussion The current study identified 6 specific synovial fluid biomarkers that differed significantly between knees with ACL tears and uninjured contralateral controls (and those differences were larger than calculated MCIDs), but found no differences between knees with ACL tears without cartilage injury compared with ACL tears with cartilage injury. The use of biomarkers as part of the diagnostic workup and treatment decision-making process is gaining considerable attention and has become widely acknowledged in modern medicine. Biomarkers may be identified in the blood, urine, or synovial fluid of patients, and represent objective indicators of normal processes, pathology, or responses to therapeutic intervention. Most studies to date have investigated biomarkers in the blood or urine. However, it is becoming increasingly appreciated that intra-articular pathologies may be more accurately described by analysis of jointspecific synovial fluid.24 Of the biomarkers found to be significantly different between groups, 3 of these cytokines are proinflammatory and catabolic (MMP-3, IL-6, and MIP-1b), whereas the other 3 are anti-inflammatory or anabolic (TIMP-1, TIMP-2, and FGF-2). Each of these biomarkers had differences in mean between injury group and uninjured controls larger than the calculated MCID, indicating clinical relevance as well as statistical significance. These findings confirm data from previous smaller-sized studies that show similar cytokine concentration increases after ACL injury,25,26 as well as in the setting of meniscal tears.27 MMP-3 is an important enzyme for cartilage destruction and remodeling.28 Several studies suggest chondrocytes upregulate MMP expression after articular cartilage injury and during cartilage degradation.29-32 It is believed that MMPs increase during abnormal cartilage
Fig 5. TIMP-2 means versus ACL groups. One of 2 biomarkers where contralateral concentration was actually higher than ACL tear knees. Knees with and without cartilage injury did not differ significantly from each other; however, only those knees with ACL tears and no associated cartilage injury were significantly different from contralaterals. (ACL, anterior cruciate ligament; SD, standard deviation; TIMP, tissue inhibitor of metalloproteinase.)
loading, such as after the impaction injury that occurs during an ACL tear; however, they are also induced by proinflammatory cytokines, in particular IL-1 and TNF.33,34 IL-6 is a well-known cytokine associated with inducing production of acute phase reactants,25 and MIP-1b is a chemokine produced by macrophages that
Fig 6. FGF-2 means versus ACL groups. The second biomarkers where contralateral concentration was higher than ACL tear knees. Knees with and without cartilage injury did not differ significantly from each other; however, only those knees with ACL tears and no associated cartilage injury were significantly different from contralaterals. (ACL, anterior cruciate ligament; FGF, fibroblast growth factor; SD, standard deviation.)
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES
induces the expression of other proinflammatory cytokines.35 It is therefore not surprising that these proinflammatory mediators would be elevated after traumatic ACL injury. TIMP-1 and TIMP-2 are inhibitors of the matrix metalloproteinases, and are known to be upregulated after traumatic events. In contrast to previous ACL studies, we did not find elevations in ACL-deficient knees of IL-1Ra,36 MCP-1,26 or IFNg37 compared with healthy controls. It is logical that IL-1Ra, an antagonist of IL-1,25 is not increased, as we did not find IL-1 to be increased in our ACL-injured patients. MCP-1 is another chemokine and a prominent mediator of inflammation.38 Although our data did not find the mean concentrations significantly different between the ACL-deficient groups and the contralateral controls for MCP-1, the ACL groups both had higher mean concentrations. IFNg has been shown to falsely produce a signal with multiplexed immunoassays,39 which could explain its occasional inclusion in the literature. Prior literature has shown that IL-1 and TNFa, 2 of the most important mediators of inflammation, are typically absent during knee injury.20,25,26,36 Our study confirms this point. The lack of these traditional inflammatory mediators, along with IFNg, points to ACL injury inducing a specific injury response, as opposed to a more generalized inflammatory response. Interestingly, we did not find any significant differences in synovial fluid biomarker concentrations between ACL injured knees with and without cartilage damage. One rationale for this is that the cytokine response could be due to the impaction injury that occurs during the pivot-shift mechanism of ACL rupture. This would explain why a similar cytokine response develops with or without gross articular cartilage damage. It is possible that the local alteration in microenvironment that occurs at the time of ACL injury contributes to the development of generalized post-traumatic osteoarthritis, regardless of associated cartilage injury. However, it is possible that ACL tears with associated cartilage damage may have a more pronounced impact at a later point in time, and if samples were obtained further out from the time of injury, a significant difference would be detected. Previous literature has not separated the ACL patients by association with cartilage damage, so we cannot compare these results with other studies. Although no a priori power analysis was performed as this is a pilot study, for the 6 variables found to be significant, 95% CI between ACL tear subjects with and without cartilage damage had considerable overlap, indicating type II (beta) error is unlikely, and these groups were in fact not significantly different. Another interesting finding is the significantly lower concentrations of TIMP-2 and FGF-2 in the ACL tear groups compared with the contralateral controls. We believe that in the setting of an ACL injury, the
7
anti-inflammatory TIMP-2 and the anabolic FGF-2 may be consumed or downregulated in response to the traumatic event and with exposure to the increased levels of the proinflammatory and catabolic MMP-3, IL-6, and MIP-1b. Four biomarkers were not found to be significantly different on analysis of variance testing, but did have considerably different means, and relatively little overlap of 95% CI including regulated on activation, normal T-cell expressed and secreted, eotaxin, IL-10, and platelet-derived growth factor-BB. However, with the exception of IL-10, the 3 biomarkers had similar mean and 95% CI between the ACL tears with and without cartilage injury groups (and higher biomarker concentrations in the control group), indicating that even if a significant relation exists that was not picked up due to beta error, it would be consistent with our other findings (i.e., cartilage damage does not impact biomarker concentrations). Limitations The current study assesses only a single point in timedshortly after injury and just before surgery. Prior studies have shown an immediate biochemical response after acute knee injury, with variable changes in cytokine concentrations during the postinjury time course.40,41 Another limitation of the current study is the lack of stratification of patients based on the energy level of their injury. We would hypothesize that patients who sustain an ACL rupture with a higher energy mechanism would have a faster and larger biochemical response to the injury. In addition, the current investigation did not correlate clinical and functional outcomes with synovial fluid biomarker levels seen at the time of surgery. We also did not investigate the bone marrow edema lesion presence or size correlation with biomarker levels, which limits our proposed pivot-shift mechanism to a theory. It is also important to note that the specimens obtained as healthy controls provided only a clinical control, because arthroscopic evaluation was not possible, and were from patients indicated for knee arthroscopy in the contralateral leg. Finally, our observations and analysis are limited by our relatively small numbers of patients, opening up the possibility of type II (beta) error.
Conclusions The current study identified 6 pro- and antiinflammatory synovial fluid biomarkers whose concentrations after ACL injury were significantly different compared with uninjured controls and those differences were larger than calculated MCIDs. No significant differences in synovial fluid biomarker concentrations were seen between ACL-injured knees with and without cartilage damage.
8
D. J. KAPLAN ET AL.
References 1. Roe J, Pinczewski LA, Russell VJ, Salmon LJ, Kawamata T, Chew M. A 7-year follow-up of patellar tendon and hamstring tendon grafts for arthroscopic anterior cruciate ligament reconstruction: Differences and similarities. Am J Sports Med 2005;33:1337-1345. 2. Hui C, Salmon LJ, Kok A, Maeno S, Linklater J, Pinczewski LA. Fifteen-year outcome of endoscopic anterior cruciate ligament reconstruction with patellar tendon autograft for “isolated” anterior cruciate ligament tear. Am J Sports Med 2011;39:89-98. 3. Spindler KP, Wright RW. Prospective study of osseous, articular, and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy. N Engl J Med 2008;359:2135-2142. 4. Feagin JA Jr. The syndrome of the torn anterior cruciate ligament. Orthop Clin North Am 1979;10:81-90. 5. Lohmander LS, Ostenberg A, Englund M, Roos H. High prevalence of knee osteoarthritis, pain, and functional limitations in female soccer players twelve years after anterior cruciate ligament injury. Arthritis Rheum 2004;50: 3145-3152. 6. Theologis AA, Kuo D, Cheng J, et al. Evaluation of bone bruises and associated cartilage in anterior cruciate ligament-injured and reconstructed knees using quantitative t(1r) magnetic resonance imaging: 1-year cohort study. Arthroscopy 2011;27:65-76. 7. Daniel D, Stone M, Dobson B, Fithian D, Rossman D, Kaufman K. Fate of the ACL-injured patient: A prospective outcome study. Am J Sports Med 2014;22:149-152. 8. Ferretti A, Conteduca F, De Carli A, Fontana M, Mariani PP. Osteoarthritis of the knee after ACL reconstruction. Int Orthop 1991;15:367-371. 9. Dimond P, Fadale P, Hulstyn M, Tung G, Greisberg J. A comparison of MRI findings in patients with acute and chronic ACL tears. Am J Knee Surg 1998;11:153-159. 10. Graf BK, Cook DA, De Smet AA, Keene JS. “Bone bruises” on magnetic resonance imaging evaluation of anterior cruciate ligament injuries. Am J Sports Med 1993;21:220-223. 11. Kaplan P, Walker C, Kilcoyne R, Brown D, Dussault R. Occult fracture patterns of the knee associated with anterior cruciate ligament tears: Assessment with MR imaging. Radiology 1992;183:835-838. 12. Kijowski R, Sanogo ML, Lee KS, et al. Short-term clinical importance of osseous injuries diagnosed at MR imaging in patients with anterior cruciate ligament tear. Radiology 2012;264:531-541. 13. Spindler KP, Schils JP, Bergfeld JA, et al. Prospective study of osseous, articular, and meniscal lesions in recent anterior cruciate ligament tears by magnetic resonance imaging and arthroscopy. Am J Sports Med 1993;21: 551-557. 14. Anderson DD, Chubinskaya S, Guilak F, et al. Posttraumatic osteoarthritis: Improved understanding and opportunities for early intervention. J Orthop Res 2012;29: 802-809. 15. Johnson DL, Urban WP, Caborn DN, Vanarthos WJ, Carlson CS. Articular cartilage changes seen with magnetic resonance imaging-detected bone bruises associated
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
30.
31.
with acute anterior cruciate ligament rupture. Am J Sports Med 1998;26:409-414. Lahm A, Uhl M, Erggelet C, Haberstroh J, Mrosek E. Articular cartilage degeneration after acute subchondral bone damage: An experimental study in dogs with histopathological grading. Acta Orthop Scand 2004;75: 762-767. Lewis JL, Deloria LB, Oyen-Tiesma M, Thompson RC, Ericson M, Oegema TR. Cell death after cartilage impact occurs around matrix cracks. J Orthop Res 2003;21: 881-887. Martin JA, Brown T, Heiner A, Buckwalter JA. Posttraumatic osteoarthritis: The role of accelerated chondrocyte senescence. Biorheology 2004;41:479-491. Martin JA, Buckwalter JA. Post-traumatic osteoarthritis: The role of stress induced chondrocyte damage. Biorheology 2006;43:517-521. Akesen B, Demirag B, Budak F. Evaluation of intraarticular collagenase, TIMP-1, and TNF-alpha levels before and after anterior cruciate ligament reconstruction. Acta Orthop Traumatol Turc 2009;43:214-218. Rejas J, Pardo A, Ruiz MÁ. Standard error of measurement as a valid alternative to minimally important difference for evaluating the magnitude of changes in patient-reported outcomes measures. J Clin Epidemiol 2008;61:350-356. Wyrwich KW, Nienaber N, Tierney WM, Wolinsky FD. Linking clinical relevance and statistical significance in evaluating intra-individual changes in health-related quality of life. Med Care 1999;37:469-478. Wyrwich KW, Tierney WM, Wolinsky FD. Further evidence supporting an SEM-based criterion for identifying meaningful intra-individual changes in health-related quality of life. J Clin Epidemiol 1999;52:861-873. Ruiz-Romero C, Blanco FJ. Proteomics role in the search for improved diagnosis, prognosis and treatment of osteoarthritis. Osteoarthr Cartil 2010;18:500-509. Higuchi H, Shirakura K, Kimura M, et al. Changes in biochemical parameters after anterior cruciate ligament injury. Int Orthop 2006;30:43-47. Cuellar VG, Cuellar JM, Golish SR, Yeomans DC, Scuderi GJ. Cytokine profiling in acute anterior cruciate ligament injury. Arthroscopy 2010;26:1296-1301. Cuellar JM, Scuderi GJ, Cuellar VG, Golish SR, Yeomans DC. Diagnostic utility of cytokine biomarkers in the evaluation of acute knee pain. J Bone Joint Surg Am 2009;91:2313-2320. Emonard H, Grimaud JA. Matrix metalloproteinases. A review. Cell Mol Biol 1990;36:131-153. Billinghurst RC, Dahlberg L, Ionescu M, et al. Enhanced cleavage of type II collagen by collagenases in osteoarthritic articular cartilage. J Clin Invest 1997;99:1534-1545. Mehraban F, Kuo SY, Riera H, Chang C, Moskowitz RW. Prostromelysin and procollagenase genes are differentially up-regulated in chondrocytes from the knees of rabbits with experimental osteoarthritis. Arthritis Rheum 1994;37: 1189-1197. Shlopov BV, Gumanovskaya ML, Hasty KA. Autocrine regulation of collagenase 3 (matrix metalloproteinase 13) during osteoarthritis. Arthritis Rheum 2000;43:195-205.
BIOMARKER CHANGES IN ACL-DEFICIENT KNEES 32. Dahlberg L, Billinghurst RC, Webb G, et al. Selective enhancement of collagenase-mediated cleavage of resident type II collagen in cultured osteoarthritic cartilage and arrest with a synthetic inhibitor that spares collagenase 1 (Matrix metalloproteinase 1). Arthritis Rheum 2000;43:673-682. 33. Harkey MS, Luc BA, Golightly YM, et al. Osteoarthritisrelated biomarkers following anterior cruciate ligament injury and reconstruction: A systematic review. Osteoarthr Cartil 2015;23:1-12. 34. Van de Loo FAJ, Joosten LAB, Van Lent PLEM, Arntz OJ, Van den Berg WB. Role of interleukin-1, tumor necrosis factor alpha, and interleukin-6 in cartilage proteoglycan metabolism and destruction. Effect of in situ blocking in murine antigen- and zymosan-induced arthritis. Arthritis Rheum 1995;38:164-172. 35. Matute Wilander A, Kåredal M, Axmon A, Nordander C. Inflammatory biomarkers in serum in subjects with and without work related neck/shoulder complaints. BMC Musculoskelet Disord 2014;15:103. 36. Cameron ML, Fu FH, Paessler HH, Schneider M, Evans CH. Synovial fluid cytokine concentrations as
37.
38.
39.
40.
41.
9
possible prognostic indicators in the ACL-deficient knee. Knee Surg Sports Traumatol Arthrosc 1994;2:38-44. Ahlen M, Roshani L, Liden M, Struglics A, RostgardChristensen L, Kartus J. Inflammatory cytokines and biomarkers of cartilage metabolism 8 years after anterior cruciate ligament reconstruction: Results from operated and contralateral knees. Am J Sports Med 2015;43:1460-1466. Deshmane SL, Kremlev S, Amini S, Sawaya BE. Monocyte chemoattractant protein-1 (MCP-1): An overview. J Interferon Cytokine Res 2009;29:313-326. Scuderi GJ, Woolf N, Dent K, et al. Identification of a complex between fibronectin and aggrecan G3 domain in synovial fluid of patients with painful meniscal pathology. Clin Biochem 2010;43:808-814. Swärd P, Frobell R, Englund M, Roos H, Struglics A. Cartilage and bone markers and inflammatory cytokines are increased in synovial fluid in the acute phase of knee injury (hemarthrosis)da cross-sectional analysis. Osteoarthr Cartil 2012;20:1302-1308. Catterall JB, Stabler TV, Flannery CR, Kraus VB. Changes in serum and synovial fluid biomarkers after acute injury (NCT00332254). Arthritis Res Ther 2010;12:R229.