SEARCH

SEARCH BY CITATION

Keywords:

  • synovium;
  • ACL;
  • osteoarthritis;
  • remodeling;
  • qPCR

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

While impossible in humans, the mechanisms of early cartilage, bone and meniscal damage can be quantified after anterior cruciate ligament (ACL) injury in animal models. We utilized an ovine model to determine if the mRNA expression of inflammatory and degradative molecules (IL-1β, IL-6, MMP-1, 2, 3, and 13) in the synovium correlated to changes in joint tissues 2 weeks post-ACL surgery, to test the hypothesis that synovial inflammation is a marker of these changes and possibly their originator. Nine “idealized” ACL autografts were performed and compared with three sham and six normal animals. Using validated protocols, early osteophyte formation, articular cartilage, and meniscal damage were quantified. Synovium was harvested and mRNA expression quantified using qPCR. Multiple linear regression analysis (MLRA) was utilized to correlate synovial mRNA expression in treated and contra-lateral limbs, from all treatment groups with corresponding joint scores. Synovial mRNA expression was significantly elevated in all experimental and sham joints. The MLRA model was a significant predictive tool (p = 0.001, R2 = 0.70) of gross tissue scores with significant contributions from IL-1β, IL-6, and MMP-3. Findings suggest that this set of synovial biomarkers is predictive (p < 0.009) of early gross changes of joint tissues after arthrotomy and likely directly involved in the relevant mechanisms, particularly early osteophyte formation, in vivo. © 2011 Orthopaedic Research Society Published by Wiley Periodicals, Inc. J Orthop Res 29: 1185–1192, 2011

Anterior cruciate ligament (ACL) tears are common, with meniscal pathology and osteoarthritic changes to bone and articular cartilage frequently developing in joints with untreated ACL injuries.1 ACL reconstructive surgery can effectively restore stability to the knee, but nonetheless a significant number of reconstructed patients go on to develop osteoarthritis (OA). The biological reasons for the progression of OA in mechanically stable joints are currently not known.

Synovial inflammation is known to play a central role in articular cartilage degradation in inflammatory arthritis,2 but that is not the case in the early stages of OA where synovitis could be either the cause or simply a secondary effect of joint damage.3 However initiated, synovial inflammation could possibly be either a passive marker of disease, or a potential driving force in the onset of early joint changes. Interleukins (IL-1β and IL-6) and matrix metalloproteinases (MMP-1, 2, 3, and 13) in the synovium are potential biomarkers of the onset of OA.4–8 Increased levels of IL-1β within the synovial fluid of ACL deficient patients have also been shown to correlate with increased levels of cartilage degradation.9 Patients diagnosed with early OA have increased expression of the pro-inflammatory mediators IL-1β and IL-610 in their synovium. IL-6 is a cofactor to the catabolic effects of IL-1β while increasing the synthesis of MMP-13 in osteoarthritic joints.11 MMP-1 is produced mainly (but not exclusively) by synovial cells12 and degrades collagen types I and III,13 and type II collagen in cartilage.5 MMP-2 degrades aggrecan and other extracellular matrix proteoglycans, as well as collagen types I, II, and III.14 MMP-3 breaks down extracellular matrix proteoglycans14 and MMP-13 participates in the breakdown of type II collagen.12

To study the potential relevance of these molecules in vivo, we have developed an “idealized” ACL reconstruction model in which the native ovine ACL is cored out at the femoral end and immediately reattached, restoring normal biomechanics as completely as possible while also limiting any biological response to a graft material. In this model, despite relative biomechanical perfection, we have observed early osteophyte formation and some discrete articular cartilage damage that is intriguingly variable between animals. We have also seen some on-going low-grade synovitis in these animals and, based on its apparent relationship to the degrees of tissue changes seen, speculated that it may be actively involved in the remodeling or damage to the tissues of interest in the joint.

The goal of the present study was to quantify these early gross joint tissue changes and determine if synovium can, at a minimum, be used as a marker of these effects. Based on their expected involvement in the relevant tissue processes as noted above, the molecular (mRNA) activity of a specific set of selected inflammatory (IL-1 and 6) and degradative (MMP-1, 2, 3, and 13) molecules was compared in a blinded fashion to gross tissue changes in a population of animals at 2 weeks post-surgery. Our hypothesis was that there would be a quantitative relationship between the physical changes in these major tissues in the joint and the message for these synovial inflammatory and degradative molecules; implicating synovium as an active participant and as a potential marker of their subsequent progression.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

ACL Autograft Surgery

All animal surgery and post-operative care procedures followed protocols approved by the University of Calgary Health Sciences Animal Care Committee and complied with the guidelines of the Canadian Council on Animal Care. Eighteen skeletally mature (3- to 4-year-old) female Suffolk-cross sheep were allocated into three groups: ACL autograft surgical group (n = 9), surgical sham control group with non-breached ACL core (n = 3), and non-operated controls (n = 6). The ovine model is a commonly used orthopedic model for the investigation of various joint pathologies15 including ACL injury.16 The ACL core surgeries were accomplished via arthrotomy to the right stifle joint. Briefly, each surgical animal received an injection of Liquamycin (Pfizer Canada, Inc., Kirkland, QC, Canada) 24 h pre- and post-surgery, Atro-Sa (Rafter Products, Calgary, AB, Canada), Acevet (Vetoquinal, Inc., Lavaltrie, QC, Canada), and Temgesic (Schering-Plough, Hertfordshire, UK) were all administered, as a pre-anesthetic cocktail. Once the animal had been placed under general anesthesia, the stifle joint was opened on the lateral side and the patella was dislocated medially to expose the ACL. In nine sheep, the lateral epicondyle of the femur was exposed as the entry point for a guide pin, drilled along the long axis of the femoral insertion of the ACL (Fig. 1(1,2)). A dry pneumatic drill with a hollow bore coring device was fitted over the guide pin and used to core out the uninjured ACL on its femoral insertion (Fig. 1(3)). Before its final detachment, two perpendicular lines were burned on the core using cautery (Fig. 1(4)), to be used as reference points to aid in the replacement of that core. A very thin small osteotome and mallet were then used to breach the final cortex and complete the release of the core. With the bone core and attached ACL insertion now free, using the guide pin the core could be immediately fixed in place using two crossed Kirschner wires (Zimmer, Inc., Warsaw, IN) (Fig. 1(5,6)). The joint was closed using standard closure. Each animal received an injection of Anafen (Merial Canada, Inc., Daie d'Urfe, QC, Canada) upon recovery. A surgical sham control group (n = 3) was performed by doing the same surgical approach, including a similar arthrotomy and temporary patellar dislocation, but coring only half way through the femur. A group of six non-operated normal control animals were age matched and housed for the same duration of time as the experimental subjects.

thumbnail image

Figure 1. Images of the ACL autograft surgery performed on a cadaveric sheep limb. 1. The femoral ACL insertion was marked and a guide pin was inserted. 2. The ACL guide was removed. 3. The corer followed the guide pin to the femoral ACL insertion. 4. The anatomical location of the ACL autograft was marked using cautery. 5. The ACL autograft was fixed using 2 K-wires (K-wires were also used on Sham cores). 6. The K-wires were cut, and the stifle joint reconstructed.

Download figure to PowerPoint

Tissue Harvest and Gross Morphological Grading

All animals were sacrificed 2 weeks post-injury with Euthanyl (Bimeda-MTC, Cambridge, ON, Canada). Each hind limb was disarticulated and the stifle joint opened and flushed with a DMEM-F12 solution with 1% antibiotics. Synovial tissue was harvested from the distal femoral groove posterior to the patella and directly submersed in DMEM-F12 with antibiotics. The synovium was snap frozen in liquid nitrogen, and stored at −70°C. For the right (treated) and left (contra-lateral non-surgical) stifle joints, articular cartilage was graded by the same observer for gross defects using the previously validated modified Drez17 scoring protocol to quantify six tibial plateau, four femoral condyle, two femoral groove, and two patellar locations. Osteophyte formation was quantified using the modified Grood18 protocol on the margins of the same cartilage locations previously described. Meniscus scoring used the modified Hellio Le Graverand protocol adapted from Adams19 to quantify meniscal damage in three medial and three lateral locations. These three previously validated scoring protocols were combined to create a combined morphological scoring system. Protocol scores and descriptions are summarized in Table 1. In order to quantify changes to the tissues of the entire joint, and to create an adequate scale for future studies on manipulations that will cause greater damage, we divided the joint into 14 distinct areas. Although this grading approach allows for a larger total score, it simultaneously increases the sensitivity to detect focal changes, as expected in OA In future studies we will use this system to relate specific areas of change and their severity to altered biomechanics, using the biomechanically normal situation described here as “the baseline.”

Table 1. Primer Sequences Used for qPCR Analysis of Select Synovial Cytokines
Gene Primer SequenceGene Bank Accession No.
18sF5′-TGG TCG CTC GCT CCT CTC C-3′X 03205
 R5′-CGC CTG CTG CCT TCC TTG G-3′ 
1L-1βF5′-CGA ACA TGT CTT CCG TGA TG-3′NM 001009465
 R5′-TCT CTG TCC TGG AGT TTG CAT-3′ 
IL-6F5′-ACA GCA AGG AGA CAC TGG CA-3′NM 001009392
 R5′-GCC GCA GCT ACT TCA TCC GA-3′ 
MMP-1F5′-GGT ATC GGA GGA GAC GCT CA-3′AF 267156
 R5′-GTG CGC ATG TAG AAC CGG TC-3′ 
MMP-2F5′-CTA CCA CCT CCA ACT ACG AT-3′AF 267159
 R5′-CAG AAT GTG GCT ACT AGC AG-3′ 
MMP-3F5′-TTA GAG AAC ATG GGG ACT TTT TG-3′AF 267
 R5′-CGG GTT CGG GAG GCA CAG-3′ 
MMP-13F5′-GGT GAC AGG CAG ACT TGA TGA TAA C-3′AY 091604
 R5′-ATT TGG TCC AGG AGG GAA AGC G-3′ 

qPCR

The synovium was examined for mRNA expression of IL-1β, IL-6, MMP-1, 2, 3, and 13, using real-time reverse transcriptase polymerase chain reaction (qPCR) run in duplicate. qPCR was performed as previously described.20 Briefly, total RNA was extracted from tissue using Qiagen RNeasy kit (Qiagen Sciences, Germantown, MD), treated with DNAse-I according to the manufacturer's instructions and quantified. Total RNA (1 µg) was reverse transcribed to generate single stranded CDNA using Qiagen Omniscript RT kit (Qiagen Sciences). For use in real-time RT-PCR, primers were created and validated from the target mRNAs previously mentioned (summarized in Table 2). BIO-RAD iQ SYBR Green Supermix (12.5 µl) (BIO-RAD, Hercules, CA), molecular biology water (3.5 µl), as well as forward and reverse primer (0.75 µl each) formed the PCR reaction mixture, using 7.5 µl RT for each reaction. iCycler Thermal Cycler (BIO-RAD) was utilized during amplification and detection, validated through inspection of the melting curve (dF/dT vs. temperature) for non-specific peaks. Level of gene expression was normalized to 18s mRNA. iCycler iQ Optical System Software version 3.0a (BIO-RAD) was used to quantify results.

Table 2. Summary of Stifle Joint Injury Scoring Protocols
ScoreDescription
Cartilage
 Modified Drez
  0Normal
  1Surface irregularities, color changes, surface intact
  2Surface fibrillation/fragmentation with no loss of cartilage
  3Surface fibrillation/fragmentation with loss of cartilage
  4Loss of cartilage exposing bone (<10% of total surface area) or extensive damage to cartilage
  5Large areas of exposed bone (>10%)
Osteophytes
 Modified Grood
  0No osteophytes
  1Height < 2 mm, length < 10 mm
  2Height < 2 mm, length < 30 mm
  32 mm < height < 4 mm, length > 30 mm
  44 mm < height<6 mm, length > 50 mm
  5Gross bone deformation significantly altering geometry
Meniscus
 Modified Hellio Le Graverand
  0Normal
  1Minimal fibrillation, degeneration or scar
  2Moderate surface fibrillation, but no tears
  3Severe fibrillation, incomplete tears
  4Complete tears, bucket-handle tears, multiple incomplete tears
  5Degenerative tear

Statistical Analysis

Facilitated by Stata 9.2 for Macintosh (Stata, College Station, TX) between treatment group comparison of gross joint injury as well as mRNA expression levels were investigated utilizing Wilcoxon ranked tests with Bonferroni post hoc adjustments to limit type 1 error. Paired comparisons of joint injury as well as mRNA expression levels among the surgically treated limbs and respective contra-lateral limbs were conducted utilizing Wilcoxon ranked tests for matched pairs.

Multiple Linear Regression Model

Data from the right (experimental) and left (contra-lateral) limbs of each animal combined for a total of 36 joints to make up the data set for the regression model. The data associated with each joint constitute a record within the regression model's data set, each record being composed of five variables: total gross morphological score, and mRNA expression level values of IL-1β, IL-6, MMP-1, 2, 3, and 13. A multiple linear regression analysis (MLRA) was then performed using Stata 9.2 for Macintosh (Stata) to determine whether these biological markers correlated with total gross morphological score. After the first iteration, the data set was tested for influential outliers based on studentized residuals and leverage. Records with absolute residual values greater than 2, and leverage values greater then (2k + 2)/n (k = number of prediction variables, n = sample size) were examined. Cook's D algorithm was also used to determine the influence of each record with any value retuned that was greater then 4/n (n = sample size) examined. Utilizing the criteria stated above, an outlier was removed if it was found to be overly influential and had a high studentized residual or leverage value.21 The regression analysis was repeated, and retested for outliers each time there was record exclusion from the dataset. With outliers removed, the model was tested for normality through the implementation of Shapiro–Wilk W-test for normality.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Gross Joint Morphology

While total scores were still low, surgically treated limbs had statistically significantly greater damage scores (combined scores for osteophyte formation, cartilage damage, and meniscal damage) compared to those of both control limbs (p < 0.05) and respective contra-lateral limbs (p < 0.05) (Fig. 2). Average gross scores were similar to shams but interestingly, the inter-animal variation was greater for the ACL reconstructed group, suggested a greater divergence in responses to the surgery that involved drilling all the way into the joint space (completing the core). Among the physical areas examined for gross cartilage defects in each joint, the most common score was 0 with almost all damage features assigned a score of 1 when present. As summarized in Table 3, this suggests that damage was mild, as expected at this early time-point. However modest these scores are, they do represent a significant effect of surgery. Osteophytes were only present in surgical and sham animals, with the most significant ones along the trochlear ridges, perhaps relating to the patellar portion of each surgical procedure. With the exception of one animal that was excluded from the regression model as an outlier, the meniscus scoring protocol showed infrequent and modest alterations, with gross scores of 1 in only two subjects.

thumbnail image

Figure 2. Gross stifle joint damage scores: sum of grading protocols for cartilage damage, early osteophyte formation, and meniscal damage. (A) Experimental (right) limbs by treatment group. Cartilage changes were similar, osteophyte formation appears to be the determining factor between surgical and control. §p < 0.05 Experimental group versus Control group (Wilcoxon ranked test). (B) Contra-lateral (left) limbs of surgical animals. There was no osteophyte formation observed, and only a small amount of meniscal damage in one animal of this group. #p < 0.05 Experimental group versus Control group (Wilcoxon ranked test for matched pairs).

Download figure to PowerPoint

Table 3. Summary of Gross Morphological Grades and Sample Locationsa
Grading Protocol  Group
SurgicalContra-LateralShamControl
Score 0Score 1Score 2Score 3Score 0Score 1Score 2Score 0Score 1Score 2Score 0Score 1
  • a

    The number of animals that received each specified grade, at each sample location, for each grading protocol.

Cartilage
Tibial plateauMedialAnterior54  9  21 51
  Middle27  27 12  6
  Posterior81  9  21 6 
 LateralAnterior81  9  3  51
  Middle45  81 3  51
  Posterior9   9  3  6 
Femoral condyleAnteriorMedial81  72 3  6 
  Lateral9   81 12 6 
 PosteriorMedial45  63 21 42
  Lateral81  9  21 51
Femoral Groove Proximal333 35111133
  Distal27  54 12 15
Patella Proximal18  45  3 24
  Distal54  72 12 42
Osteophyte
Tibial plateauMedialAnterior81  9  3  6 
  Middle9   9  3  6 
  Posterior72  9  21 6 
 LateralAnterior81  9  3  6 
  Middle9   9  3  6 
  Posterior9   9  3  6 
Femoral condyleAnteriorMedial72  9  21 6 
  Lateral63  9  3  6 
 PosteriorMedial72  9  21 6 
  Lateral36  9  21 6 
Femoral Groove Medial27  9   3 6 
  Lateral54  9  12 6 
Patella Proximal9   9  3  6 
  Distal81  9  3  6 
Meniscus
 MedialAnterior9   81 3  6 
  Middle9   9  3  6 
  Posterior8  19  3  6 
 LateralAnterior9   9  3  6 
  Middle81  9  3  6 
  Posterior81  9  3  6 

mRNA Expression

All synovial mRNA expression levels were significantly elevated in experimental joints when compared to those in normal animals (with the exception of IL-6) (p < 0.05) (Fig. 3) as well as respective contra-lateral controls (p < 0.05) (Fig. 4). Statistical differences were not found between surgical and sham, sham and normal, or normal and contra-lateral (left limb of surgical group) groups in the investigation mRNA expression levels (Fig. 3).

thumbnail image

Figure 3. Synovial biomarker mRNA expression levels of each treatment group displayed after logarithmic transformation. §p < 0.05 Experimental group versus Control group (Wilcoxon ranked test).

Download figure to PowerPoint

thumbnail image

Figure 4. Synovial biomarker mRNA expression levels of treated limbs and contra-lateral limbs after logarithmic transformation. #p < 0.05 Experimental group versus Control group (Wilcoxon ranked test for matched pairs).

Download figure to PowerPoint

Multiple Linear Regression Analysis

MLRA significantly predicted stifle joint total gross morphological score through a linear combination of synovial inflammatory and degradative biomarkers. This model was significant at p < 0.001 with an R2 value of 0.70. IL-1β, IL-6, and MMP-3 were the most predictive, with p values of 0.001, 0.009, and <0.001, respectively (Table 4). The data for two stifle joints were removed from the data set as a result of outlier testing. Each removed joint was a member of the surgical group. Removed joints 1 and 2 possessed studentized residual values of −4.2 and 6.9, respectively, while removed joint 1 also had a significantly elevated leverage value of 0.89. Both joints were flagged by the Cook's D algorithm. Outlier tests revealed that the removed joints possessed data values that were significantly different from the trend created by the data set as a whole, therefore the data of these joints were negatively influencing the predictive capabilities of the model. Data set normality was achieved in the final model as reported by the Shapiro–Wilk W-test (p = 0.0834, failing to reject null hypothesis of data set normality).

Table 4. Summarized Multiple Linear Regression Analysis Output
Predicted VariableModel Summary
R2F-Valuep-Value
Grass morphological score (n = 34)0.7010.6<0.001
Predictor VariablesCoefficients
Coeffstp-Value
  1. Bold was used to highlight statistical significance within the model.

IL1-β−1.91E−03−3.750.001
IL-61.82E−032.790.009
MMP-13−1.10E−04−1.360.184
MMP-33.30E−045.02<0.001
MMP-23.71E−021.590.123
MMP-11.36E−050.460.647
Const.3.93E+009.77<0.001

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We have developed a controlled animal model of idealized ACL autograft surgery in which we can examine the effects of the surgical procedure itself on the joint as a whole and define its effects on each joint tissue over time. In this study, at only 2 weeks post-anatomic ACL autograft surgery, we have already shown subtle but statistically significant gross joint changes in that model and in its nearly equally invasive sham control as quantified by gross morphological score, as well as mRNA expression levels when compared to normal control joints. Specifically, mRNA expression levels of IL-1β, MMP-1, 2, 3, and 13 were significantly increased in the surgically treated joints when compared to contra-lateral and normal control limb values while IL-6 was only significantly elevated when compared to contra-lateral limb values. Both sham invasion and the ACL reconstruction induced some gross tissue changes, with more variable inter-animal responses to the “ACL reconstruction,” which differed from shams only by complete coring through bone into the joint cavity. Interestingly, early osteophyte formation was the most influential factor separating surgical animals from control animals. Such osteophyte formation has been reported as early as 3 days–3 weeks post-ACL transection in dogs.22, 23

Most importantly, gross tissue changes and molecular activity of the selected set of synovial cytokines and matrix metalloproteinases were found to be highly correlated when examining the surgical, sham, and control animals and thus a linear combination of synovial mRNA expressions could be used to fairly accurately predict the current level of tissue changes in each joint, regardless of its mode of injury (sham or ACL reconstruction), with IL-1β, IL-6, and MMP-3 making significant contributions to this prediction. Therefore, not only is the synovium a potential marker of joint tissue changes, it appears to be implicated in facilitating those changes by virtue of what it appears to be secreting into the joint. If this correlation continues over time it has interesting possibilities for distinguishing individual responses with minimally invasive techniques and perhaps in predicting or preventing them.

These findings are not totally surprising in some ways as traumatic knee injury has been shown previously to produce OA-like inflammatory patterns of IL-1β, MMP-1, and 3 expression in human knee synovium, suggesting that these markers are potential markers of disease progression.24 Synovial inflammation has also been reported and studied in various ovine models of OA.25 Our ovine model focused on an early time-point in an ovine model after ACL autograft surgery. Examining a group of animals at the same time point with varying levels of joint damage, surgical animals hypothetically receiving a more significant injury than sham animals, and respective synovial activation measured by inflammatory and degradative cytokine/metalloproteinase mRNA expression levels was key in establishing a linear equation of the synovium's potential involvement at an early time point of joint injury. Of the potential biomarkers that made significant contributions to the MLRA model, IL-1β appears to have been the most influential. Interestingly IL-1β mRNA expression levels are elevated in surgical animals compared to those of respective contra-lateral limb values. However, within the surgical animals IL-1β mRNA expression levels appear to decline as joint injury scores increase in our model as reflected by the negative MLRA coefficient. At later stages of OA progression IL-1β protein levels in synovial fluid have been reported to correlate negatively with the severity of knee OA and patient pain outcome scores at time points well past the initial inflammatory period.26 However it is unclear of why this negative correlation exists at such an early time point with subtle joint changes. The possibility exists that the negative correlation within elevated mRNA expression levels could be an artifact resulting from small sample size. It has been shown that highly expressed IL-6 possesses the ability to exhibit anti-inflammatory properties by down-regulating catabolic factors such as IL-1β.27 It may be possible that the up-regulation in IL-6 recorded in this study at the early time-point of 2 weeks post-ACL autograft surgery is having a similar effect on the mRNA expression of IL-1β. The inclusion of IL-6 mRNA expression levels provided a significant contribution to this regression model, correlating positively with increasing joint injury scores. Interestingly as well as maintaining its well-documented interactions with articular cartilage, IL-6 has been implicated in the enhancement of osteophyte formation.28 Osteophyte formation contributed significantly to the overall joint injury score of surgical animals. As previously described, osteophyte formation was most commonly found in the margins of the patellar femoral groove, in close proximity to the harvest site of synovium. MMP-3 mRNA expression levels maintained a consistent increase with increasing joint injury scores.

Although the ovine model is a widely accepted orthopedic model utilized in the study of various joint related ailments, careful consideration of experimental findings must be made when relating to human models. For instance, in the absence of initial traumatic injury to the joint, it is not possible to claim that changes within the joint due to ACL autograft surgery would be detectable in a clinically applicable model, as otherwise healthy joints were utilized in this study. As we have not yet defined the natural history of the effects seen here, we cannot state that ACL surgery alone will cause any permanent joint damage. In fact, ACL reconstruction is a highly successful surgery from a clinical perspective. However when ACL autograft surgery is itself is the sole traumatic injury to the joint, it does have an early quantifiable effect on the tissues within the joint, at least in this animal model. While sham surgery was statistically indistinguishable from surgical and normal joints, variance in the sham group suggests that this lack of difference was likely a Beta error of low statistical power. We could not justify increasing that sham sample size for ethical reasons, as it would not alter our fundamental conclusions. Results from the sham treatment group suggest that invading the joint by arthrotomy and coring into the bone is having a quantifiable effect. However, the average effects are slightly greater but certainly more variable between animals when the bone core is extended into the joint space (coring out and immediately reattaching the femoral insertion of the ACL).

To our knowledge this is the first study that has attempted to quantify the potential relationship between the molecular activity of select inflammatory and degradative molecules within synovial tissue and physical changes that have occurred within the joint in response to an ACL graft procedure and to a sham surgery of similar surgical magnitude without the final step of bone invasion into the joint. We have shown that ACL autograft surgery and sham surgery causes subtle yet significant changes to the joint tissues at only 2 weeks post-surgery. These changes correlate strongly with the synovial molecular activation of inflammatory cytokines IL-1β and IL-6, and the degradative metalloproteinase MMP-3. The significance of these effects over time is currently under investigation. Furthermore, we hope to exploit the potential of this animal model in the investigation of various controlled ACL graft tensioning and fixation options and their effects on the joint tissues. We plan to use the data collected in this study to form a baseline for future experiments of longer duration.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The authors would like to acknowledge the contributions of Craig Sutherland and Leslie Jacques in animal surgery and post-operative care. Funding for this study was provided by the Canadian Institutes of Health Research.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES