Bisphosphonates reduce bone mineral loss at ligament entheses after joint injury1

Bisphosphonates reduce bone mineral loss at ligament entheses after joint injury1
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Bisphosphonates reduce bone mineral loss at ligament enthesesafter joint injury 1 M. R. Doschak Ph.D. y *, J. M. LaMothe Ph.D. y , D. M. L. Cooper B.Sc. y ,B. Hallgrimsson Ph.D. y , D. A. Hanley M.D. z , R. C. Bray M.D. x  and R. F. Zernicke Ph.D. k y McCaig Centre for Joint Injury & Arthritis Research, University of Calgary, Calgary, AB, Canada  z Department of Medicine, University of Calgary, Calgary, AB, Canada  x Department of Surgery, University of Calgary, Calgary, AB, Canada  k Faculty of Kinesiology, University of Calgary, Calgary, AB, Canada  Summary Objective  : To examine the effects of anterior cruciate ligament (ACL) insufficiency, and subsequent bisphosphonate (BP) antiresorptivetherapy, on the bone mineral interface at the enthesis of remaining ligamentous restraints. Methods  : We measured bone mineral geometry (and subsequent adaptation) at the medial collateral ligament (MCL) srcin, using micro-computed tomography ( m CT). Groups of normal control, 6 and 14 wk anterior cruciate ligament transected (ACLX), and 6 wk ACLX e BP(risedronate) dosed rabbits were evaluated. Samples were then processed histologically, and the results of mineral adaptation andprogression of osteoarthritis (OA) compared to joint laxity values obtained from previous biomechanical testing of the MCL-complex. Results  :  m CT defined the MCL srcin as a symmetrical, metaphyseal depression that contained soft-tissue elements, including fibrocartilageand ligament d as seen in subsequent histological sections. In contrast, the insertions from ACLX animals lost significant bone mineral, with anMCL-insertion volume 1.2 times that of normal controls at 6 wk ACLX, which further increased to 2.3 times that of normal controls at 14 wkACLX. Significant differences were also measured between 6 and 14 wk ACLX and age-matched normal controls in volume of cortical bonecontaining the MCL insertion. However, there were no significant differences in the percentage of cortical bone to underlying trabecular boneat the MCL insertion. When comparing  m CT mineral adaptation at the MCL-enthesis with historical MCL-complex laxity data, the values for laxity after ACLX increased proportionately as bone mineral at the insertion was lost, and subsequent use of the BP risedronate reduced bothmineral loss and MCL-complex laxity. Conclusion  : Compared to the untreated ACLX condition, administering bisphosphonate immediately after loss of the ACL conserved bonemineral at the MCL enthesis, suggesting the potential to therapeutically influence joint-complex laxity and OA progression. ª  2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved. Key words  : Osteoarthritis, Risedronate, Micro-computed tomography, Enthesis, Medial collateral ligament, Anterior cruciate ligamenttransection, Joint laxity. Introduction Following anterior cruciate ligament (ACL) rupture in theknee, a rapid phase of periarticular bone loss often occurs,prior  to the subchondral sclerosis of end-stage osteoarthritis(OA) 1 e 3 . That loss occurs, in part, as both trabecular andcortical bone remodel to adapt to altered loading, but alsodue to the reduced limb usage and related disuseosteopenia. In animal ACL-transection (ACLX) models,significant decreases in femoral and tibial bone miner aldensity (BMD) were measured for both dogs 1 and rabbits 4 ,particularly in periarticular cancellous bone. In humanstudies, the loss of bone mineral in the initial stages after ACL disruption has been well established 5 e 7 , and wasmeasured in the distal femur as a 21% decrease in BMD 5 .As chronic knee instability (and progression in jointpassive laxity) often results in the bone e ligament jointcomplex after loss of the ACL 8,9 , we questioned whether theloss of bone mineral at the ligament insertions (or entheses)was a contributing factor to increased joint laxity. We haverecently shown that bisphosphonate (BP) antiresorptivetherapy can significantly reduce passive laxity in thebone e medial collateral ligament (MCL) e bone complexafter ACLX, suggesting that bone mineral adaptation atthe MCL insertion was influenced with antiresorptivetherapy 10 . We define MCL-complex laxity as the distancebetween tension and compression of the femur  e MCL e tibiacomplex during axial loading.Thus, the objective of this study was to assess in an OAmodel whether antiresorptive therapy altered mineral loss atthe bone e ligament insertion. We examined and quantifiedthe changes in bone mineral at the femoral srcin of theMCL, using micro-computed tomography ( m CT). We also 1 This research was funded by the Canadian Institutes of HealthResearch. Presented in part at the Ostearthritis Research SocietyInternational (OARSI) Congress in Berlin, Germany, October 2003,and printed in Osteoarthritis and Cartilage 2003;11:S85, P236(abstract).*Address correspondence and reprint requests to: Michael R.Doschak, Biomedical Peptide-Conjugation Laboratory, Departmentof Chemical and Materials Engineering, Room 838 Chem Mat Eng,Universityof Alberta,Edmonton,AB,Canada,T6G 2G6. Tel:1-780-492-4645; Fax: 1-780-492-2881; E-mail: mdoschak@ualberta.caReceived 4 November 2004; revision accepted 18 April 2005. OsteoArthritis and Cartilage   (2005)  13,  790 e 797 ª  2005 OsteoArthritis Research Society International. Published by Elsevier Ltd. All rights reserved.doi:10.1016/j.joca.2005.04.015 InternationalCartilageRepairSociety 790  investigated whether short-term antiresorptive therapy (i.e.,with risedronate BP) altered bone mineral changes at theMCL enthesis. Materials and methods SAMPLES A total of 28 samples, harvested from the medial femoralcondyle of age-matched 1 yr old skeletally mature NewZealand White rabbits, were analyzed. All animals weresourced from the same supplier (Riemens Fur Ranches, St.Agathe, ON) and randomly divided into four groups.We surgically transected the ACL in three experimentalgroups of rabbits. The first and second groups remaineduntreated for 6 wk ( n  Z 8) and 14 wk ( n  Z 6), respectively,and served as age-matched ACLX controls. The third groupwas dosed (0.01 mg/kg s.c.) daily with risedronate BP for 6wk ( n  Z 6). The final group of rabbits was evaluated as age-matched, unoperated normal controls ( n  Z 8).Animals were sacrificed by barbiturate overdose, and thefemur  e MCL e tibia complex was dissected free. The MCLwas completely transected in its midsubstance, leavinga portion of the MCL at the surface of the insertion asa landmark. The medial and lateral femoral condyles wereseparated using a bandsaw, and the femoral epiphysis cutin the transverse plane proximal to the patello-femoralgroove (Fig. 1). Samples were placed into a polystyrenevial, capped to reduce dehydration, and held in an uprightposition using a foam jig to standardize the transversescanning plane for subsequent samples. MICRO-COMPUTED TOMOGRAPHY ( l CT) OF CALCIFIEDMINERAL The medial femoral condyle was scanned in its entirety,using an X-ray microtomograph ( m CT, SkyScan 1072,Aartselaar, Belgium). The condyles were scanned at100 kV through 180 (  with a rotation step of 0.9 ( . Allsamples were scanned at 12 ! magnification that producedserial cross-sectional images composed of isotropic19.4 mm 3 voxels. A low-pass Gaussian filter was used tofilter the raw image data to reduce noise, and a fixedthreshold was used on all samples to extract the mineral-ized tissue phase. The filtered binary image data were thenexported to SCION IMAGE (Scion Image Pty Ltd, Frederick,Maryland, USA) for 2-dimensional (2-D) morphometricmeasures or to ANALYZE TM 4.0 (Mayo Clinic, Rochester,MN, USA) for 3-dimensional volumetric rendering. Mea-sured indices included cortical bone volume of the insertion,MCL-insertion volume (i.e., space occupied by soft tissue),and percentage of cortical bone to trabecular bone at theMCL insertion. MORPHOMETRIC MEASUREMENTS (FROM 2-D  m CT IMAGES) MCL-insertion volume  Lines were drawn parallel to the postero-medial andantero-medial surfaces of the medial femoral condyle. Thepoints of intersection between these lines and the out-ermost region of cortical bone containing the MCL insertwere connected. The area defined by that line and thecortical bone containing the MCL insert was defined as theMCL insert area. That process was repeated for everysection containing the MCL insertion, summed, andmultiplied by  m CT section thickness to achieve MCL insertvolume. Peri-insertional accreted bone and osteophyteswere not included in measurements (Fig. 2). Sampling of cortical bone volume and percentage cortical/trabecular bone  Perpendicular to the line connecting the anterior andposterior limits of the MCL insert in the transverse plane,a line was extended from the deepest point of the MCLinsert. From this newly defined intersection, an arc wasdrawn 100 (  from the posterior-medial line, with a radius of5.25 mm. The region of interest was defined as the areaenclosed by this arc. The cortical bone within this arc wasseparately defined as a region of interest. The area definedby the cortical bone region of interest multiplied by sectionthickness was defined as the volume fraction of corticalbone containing the MCL insert. The ratio of the volume ofcortical bone to the volume of the arc was defined as thepercentage of cortical bone defining the MCL insert. Thisprocedure was undertaken for the deepest point of theinsert, as well as the twentieth and fortieth sections bothproximally and distally to the deepest point of the insert. Themeasurements from the five sections were summed. HISTOLOGICAL ASSESSMENT OF FOCAL CARTILAGEDEGRADATION Following the 30 min  m CT scans, bone samples werefixed in 10% neutral buffered formalin for 72 h. Followingdecalcification in formic acid (Cal-Ex II, Fisher Scientific,ON) for 1 wk, and dehydration in ETOH/xylene, sampleswere paraffin embedded. Frontal sections (7  m m) were cutusing a rotary microtome, and stained using haematoxylinand eosin and safranin-O/fast green, to examine the MCLinsertion tidemark characteristics and to grade articular cartilage degradation. For  all investigations, a modifiedMankin grading system 11,12 (Table I) was used to character-ize the superficial surface structure, cellularity, Safranin-Ostaining characteristics, and tidemark integrity. Whenexamining the histology, the single observer was blindedas to the treatment. Fig. 1. Diagram illustrating isolation of the medial femoral condyle(red dotted line), and the transverse plane of subsequent  m CTslices (green plane). The MCL insertion appeared as a corticaldepression. 791 Osteoarthritis and Cartilage Vol. 13, No. 9  STATISTICAL ANALYSIS Analysis of variance (ANOVA, SPSS for Windows) wasused to detect significant main effects in the measuredvariables.  Post hoc   comparisons were conducted usingTukey’s test (SPSS for Windows), with a significance levelof  P  ! 0.05 used for all statistical tests. For non-parametricmeasures, the Kruskal e Wallis ANOVA was used.We conducted power calculations for bone-relatedmeasures in our NZW rabbit model for both mechanicaland physiological parameters. Baseline data contrastingnormal control and 6 wk ACLX rabbits included bonemineral density, and MCL-complex laxity. Those datasuggested a minimum population size of six per groupprovided an acceptable margin of error to detect a signifi-cant difference with 95% confidence for the related boneindices. Results We measured significant differences in the fractionvolume of cortical bone containing the MCL insertionbetween age-matched normal controls (0.42 G 0.04 mm 3 )and both 6 wk ACLX (0.38 G 0.03 mm 3 ;  P  Z 0.047) and14 wk ACLX (0.36 G 0.05 mm 3 ;  P  Z 0.02). That resulted insignificantly decreasing cortical bone volume after ACLX.Similarly, compared to normal unoperated controls, MCL-insertional volume was significantly greater in the ACLXcohort at 6 wk ( P  ! 0.04) (Figs. 3 and 4) and increasedfurther at 14 wk after ACLX ( P  ! 0.02). When we dosedACLX animals daily with risedronate for 6 wk, however, theloss of MCL-insertional bone was less and not significantlydifferent from normal control animals. No significant differ-ences were measured in the percentage of cortical bone tounderlying trabecular bone beneath the MCL insertion,between 6 wk and 14 wk ACLX and age-matched controls.Three-dimensional rendering of the 2-D  m CT scansillustrated the degree of bony adaptation (entheseal de-pression and peri-insertional osteophytosis) that occurredafter loss of the ACL (Fig. 5).When examining histological sections of the MCL in-sertion, tissue elements such as lamellar bone, fibrocarti-lage, and ligament substance were readily discernable, aswas the periosteal lining adjacent to the MCL insertion. Ashift in the calcified tidemark was observed in 6 wk ACLXanimals, exposing a greater amount of non-mineralized,eosinophilic-stained tissue (Fig. 6). That contrasted withrisedronate-dosed 6 wk ACLX animals, where the tidemark Fig. 2. Diagram demonstrating the intersects drawn to isolate the MCL insertion, for subsequent areal quantitation and volume determination.For ACLX samples, osteophytes were excluded from the measurements. Osteophyte margins were readily discernible against the compactcortical bone, when viewed on the high resolution CT scans.Table I Modified Mankin grading system used to assess the degree of cartilage degradation between treatment groups  Modified Mankin grading systemStructureNormal 0Irregular 1Pannus 2Superficial cellular absence 3Cellular clusters 4Fissures into calcified layers 5Disorganization 6Cellular abnormalitiesNormal 0Hypercellularity 1Clusters 2Hypocellularity 3Matrix stainingNormal 0Reduced staining in radial layer 1Reduced inter-territorial matrix 2Only in pericellular matrix 3Absent 4Minimum score Z 0 (healthy cartilage). Maximum score Z 13(degraded OA cartilage). 792  M. R. Doschak   et al. : Bone mineral loss at ligament entheses  remained constant and maintained the basophilic-stainingof mineralized tissue.When comparing  m CT mineral adaptation at the MCL-enthesis with MCL-complex laxity values obtained fr omprevious biomechanical testing of the MCL-complex 10 , itwas evident that MCL-complex laxity values after ACLXincreased proportionately as bone mineral at the insertionwas lost (Fig. 7). Similarly, when risedronate BP was usedto reduce bone loss after ACLX, a significant improvementin MCL-complex laxity also resulted.Data for laxity values were used from cohorts of rabbitsthat had been mechanically tested in a previous study, asmechanical testing of the MCL-complex in this investigationmay have compromised the sample for subsequent  m CTevaluation. Because our experimental design examineddifferences in cohorts of age-matched skeletally maturerabbits, however, that were sourced from the same supplier (and not in individual animals), measurements comparedacross cohorts would remain valid.When assessing articular cartilage/meniscal gross mor-phology during dissection, deteriorative OA changes wereestablished in the periarticular tissues from all ACLX joints,including significant joint anterior  e posterior (A e P) instabil-ity, meniscal tearing, capsular thickening, and pronouncedosteophytosis. By 14 wk ACLX, that had progressed to thefull-thickness fissure of cartilage, with the full-thickness,focal loss of cartilage in some samples. In contrast, allnormal control joints were unremarkable, presenting withmacroscopically healthy joint structures. Those grossobservations were confirmed with the histological evalua-tion of the cartilage samples. All 6 wk ACLX joints scoredsignificantly higher on the modified Mankin scale comparedto normal control tissues, with loss of staining and cellularityin the superficial cartilage layer, and cartilage fibrillation inmany samples. The risedronate dosed 6 wk ACLX groupalso scored significantly higher on the modified Mankinscale compared to normal control tissues, and were nodifferent to scores obtained from 6 wk ACLX animals(Table II). Discussion Our investigation showed that early loss of periarticular bone mineral, after ACL insufficiency, included loss at theMCL insertion. Such alterations in ligament e bone mineralinterface could influence MCL mechanical functioning,including bone e ligament e bone complex laxity, as mea-sured in a previous study 10 . This potentially identifiesa novel mechanism for progressive joint laxity with timeafter ACL deficiency, and therefore identifies a newtherapeutic target for antiresorptive drug therapy. Thismechanism differs to the premise of using interventions toreduce subchondral bone changes and bone sclerosis in VOLUME OF THE MCL-INSERTION ACLX14wkACLX6wk ACLX-Rised GROUPS    V   O   L   U   M   E   (  m  m   3   ) * Sig Dif to NCON (p<0.05) ** Fig. 3. Volume of the MCL insertion. Values are means G SD.ACLX, anterior cruciate ligament transected; rised, risedronate-dosed.Fig. 4. Two-dimensional  m CT images of exemplar medial femoral condyles, illustrating the MCL insertion (arrow) in time zero control, andsubsequent bone adaptation in 6 wk ACLX, 14 wk ACLX, and 6 wk ACLX-risedronate-dosed animals. Dashed lines denote sagittal cutsdividing distal medial and lateral condyles (lateral condyle not shown). 793 Osteoarthritis and Cartilage Vol. 13, No. 9  the initiation and progression of cartilage damage andOA 13,14 .Ligament midsubstance changes also contributed toMCL-complex laxity in the ACL-deficient joint. After ACLinsufficiency, the remaining ligaments rapidly demonstratedthe loss of creep recovery, as a result of increased fiber recruitment to axial loading, with the accompanying loss ofmicroscopically visible crimp 15,16 . That resulted in the initialincrease in MCL-complex laxity that was measured in ACL-deficient joints at 6 wk. However, those ligament midsub-stance changes also occurred in our risedronate-dosedACLX joints. Thus, any net difference in MCL-complex laxitybetween ACLX and risedronate-dosed ACLX joints wouldhave been as a result of other contributing factors, such asMCL-insertional changes in bone mineral.With the loss of ACL-mediated joint restraint, the increasein A e P joint motion, and increased axial loading ofremaining ligamentous restraints (medial and lateral collat-eral ligaments, and the posterior cruciate ligament) isa stimulus for the remodeling activation of adjacent bone.Disuse osteopenia also contributes to the activation of boneremodeling, providing a general periarticular stimulus.Evidence of limb disuse is readily seen when comparinglimb muscle masses between the experimental andcontralateral limbs, with muscle mass in the injured limbmeasuring an average of 87% that of the contralateralcontrol 10 .As a result, bone mineral is resorbed periarticularly, butalso at the bone e ligament entheses. The focal loss ofmineral at ligament insertions would increase the proportionof soft tissue to mineralized tissue in the MCL-complex, andinfluence deformation after load transmission and withsubsequent time-dependant viscoelastic recovery of thesoft tissues across this interface. Thus, the net result ofloading this structure with an increased soft-tissue compo-nent would be an increase in measurable complex de-formation after loading.Loss of peri-entheseal bone mineral could also compro-mise the mechanical strength of the underlying bone duringMCL tensile loading, which may lead to microdamage andmicrofracture of bone surrounding the enthesis. Indeed, wehave previously noted that femoral fracture avulsion (in-stead of ligament midsubstance) was the most commonfailure mode of the rabbit MCL-complex when mechanicallytested to failure 6 wk after ACLX, implying that bone wasthe weakest material in the MCL-complex 9,15 . Another possibility is that meniscal insertions into bone also losebone mineral at their enthesis, and such events wouldpredispose the menisci (and the entire joint) to increaseddisplacement, with the risk of damage through mechanicalmanipulation or even traumatic injury.An interesting finding in this investigation was that bonemineral loss may be occurring from mineralized fibrocarti-laginous tissues, as well as from periarticular bone. Thedistal femoral MCL insertion (i.e., srcin) is fibrocartilagi-nous, with a discontinuous cement line separating fibrocar-tilage from mineralized fibrocartilage and mineralizedlamellar bone 17 . The insertion is therefore comprised oftwo mineralized layers, and both would have been detected‘‘as one’’ in BMD evaluations using imaging modalities. Thetwo layers were indistinguishable when viewed by  m CT incross-section, and appeared as the ‘‘cortical rim’’ of theenthesis. The significant increase in volume of the MCLinsertion suggested that the majority of insertion mineralloss occurred from the mineralized fibrocartilage. As such,this measure indicates that bone mineral may be lostthrough mechanisms other than osteoclast-mediated re-sorption, but likely involves a combination.Calcification is a process that would occur  spontaneouslyin tissues if it was not actively inhibited 18 . Indeed, manyproteins (such as matrix GLA protein (MGP)) interactspecifically with mineral  in vivo   to prevent such spontane-ous calcification, of arteries in particular  19 . At the MCLinsertion, if the layer of fibrocartilage secretes mediatorsand inhibitors of calcification, the loss of the mineral mayrepresent a ‘‘mobilization’’ of loosely aggregated mineralcrystals, due to the changing balance of mineralizationinhibitors in response to injury. Alternatively, as the zone ofmineralized fibrocartilage was thin (ranging from 100 to300  m m), it may be possible that the acidic environment of Fig. 5. Three-dimensional renderings of two-dimensional  m CT images, from exemplar medial femoral condyles, illustrating the MCL-insertionbony depression and bone adaptation in ACLX limbs. 794  M. R. Doschak   et al. : Bone mineral loss at ligament entheses
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks