Michelle L. Delco DVM1 Lawrence J. Bonassar PhD1 John G. Kennedy MD2 Rocky Tuan PhD3 Peter G. Alexander PhD3 Lisa Fortier DVM PhD1.
1Cornell University Ithaca NY USA 2Hospital for Special Surgery New York NY USA 3University of Pittsburgh Pittsburgh PA USA.
M.L. Delco: None. L.J. Bonassar: None. J.G. Kennedy: None. R. Tuan: None. P.G. Alexander: None. L. Fortier: 3B; Arthrex.
Introduction: Each day approximately one in every 10 000 people will sprain their ankle.1 It is estimated that at least half will damage articular cartilage in the process and over 50% will go on to develop post-traumatic osteoarthritis (PTOA) years or decades later. Evidence suggests that direct impact to the cartilage surface during joint injury plays a pivotal roll in initiating and perpetuating the degenerative process. However the relationships between type and magnitude of mechanical injury and the biology of early PTOA remain poorly understood.2 No in vivo models of ankle PTOA currently exist. The objective of this study was to validate the use of a high-speed spring loaded impacting device using the horse as an animal model. We demonstrate that this device is able to deliver an impact injury to equine talar cartilage that would be expected to initiate PTOA in the human ankle.
Methods: Tissue collection and impact injury
Right and left tali were harvested from normal horses (n = 6 ages 2-11 years) immediately following euthanasia and incubated in phenol-free MEM media buffered with HEPES. Osteochondral (OC) blocks were mounted in an adjustable vice grip which allowed the articular surface to be positioned perpendicular to the direction of impact. The articular surface of the talus was impacted in regions corresponding to the highest incidence of naturally occurring osteochondral lesions in humans3 (Figure 1) using a custom-made spring loaded impacting device4 (Figure 2a). Displacement of the impactor tip was measured using a linear variable differential transformer (LVDT). Impact force was measured by a load cell within the impacting device situated proximal to the tip. Cartilage thickness measurements and impact area were measured to calculate stress-strain data. Three tip designs and 6 spring tensions were compared.
Multiphoton imaging and histology
Impacted OC blocks were incubated in media for approximately 1 hour then full-thickness cartilage sections containing the impact or control site were cut off the bone and placed in 1 μM sodium fluorescein for at least 15 minutes prior to imaging to stain dead cells. Samples were imaged with multiphoton microscopy using a Ti:sapphire laser at 780 nm excitation. Images were acquired at the articular surface in the transverse plane. Dead cells and extracellular matrix (ECM) microcracks were quantified and the difference between control and impacted cartilage was assessed using a two sample T-test. Impacted and control cartilage samples were fixed in 4% paraformaldehyde then sectioned and stained with H&E and safranin O/fast green to assess morphology and relative proteoglycan content.
Results: Mechanical data
By adjusting spring tension and varying tip design (diameter and radius of curvature) maximum stresses ranging from greater than 160 MPa to less than 40 MPa were achieved (Figure 2b). Impact times averaged 1.7 ms (±0.04 ms). The spring-loaded impactor set to a spring spacing of 15 mm delivered a consistent maximum stress averaging 46 MPa (± 1.5 MPa). Strain was calculated to be 0.54 (± 0.1) with a strain rate of 435%/sec (±32%/sec).
Cell death microcracking and histology
Impacted samples contained more dead cells than control samples (Figure 3). The difference in cell death between control and impacted samples reached significance between 30 and 60 microns from the articular surface. Although cell death appeared more wide-spread when impacting with the flatter tip of larger diameter cartilage microcracking was less common than with the small-diameter tip. Histologic examination revealed articular surface fibrillation and matrix microcracking in impacted samples (Figures 3 d f).
Discussion: Evidence suggests strain rate to be more important than the magnitude of applied force during traumatic injury of cartilage.2 Many systems deliver loads to cartilage too slowly to be considered a true “impact” injury5 and drop-tower systems are impractical for translation in vivo. Strain rates of 100%/sec are commonly used in in vitro biomechanical studies of cartilage injury however most of these studies are performed on isolated cartilage samples devoid of underlying bone. It has been suggested that strain rates of approximately 500%/s and load rise-times in the millisecond range are required to exceed physiologic loading in vivo5 however the absolute mechanical factors required to initiate PTOA are not known. Here we demonstrate these thresholds for impact injury are achievable using the described spring-loaded device. Similar to the human ankle the equine talocrural joint rarely suffers osteoarthritis in the absence of injury. Additional benefits of using the horse as a model include large joint size allowing straightforward intraarticular placement and manipulation of the impactor during arthroscopic surgery as well as the ability to exercise horses on a treadmill following cartilage injury. The impactor can deliver a repeatable measurable injury to the articular surface of the equine tali resulting in fissures in the ECM and cell death within one hour of impact injury.
Significance: The current ex vivo results set the stage for in vivo studies. Development of an equine talocrural cartilage impact model which creates cartilage injury similar to that expected to cause PTOA in the human ankle will allow us to investigate early events in PTOA at the cellular and microstructural level over time using longitudinal study designs. Use of this model will allow for preclinical testing of preventative strategies such as exercise modification medications regenerative therapies etc. to minimize the long-term development of PTOA.
Acknowledgments: Funded by the Harry M. Zweig Fund for Equine Research (LAF) and NIH grant number T32RR007059 from the National Center for Research Resources (MLD). The authors gratefully acknowledge Alex VanSlyke for assistance with multiphoton imaging.
References: 1.O'Loughlin PF Heyworth BE Kennedy JG. 2010. Current Concepts in the Diagnosis and Treatment of Osteochondral Lesions of the Ankle.The American Journal of Sports Medicine 38(2):392-404.
2.Buckwalter JA Brown TD. 2004. Joint Injury Repair and Remodeling.Clin Orthop Relat Res 423:7-16.
3.Raikin SM Elias I Zoga AC et al. 2007. Osteochondral Lesions of the Talus: Localization and Morphologic Data from 424 Patients Using a Novel Anatomical Grid Scheme.Foot Ankle Int 28(2):154-161.
4.Alexander PG McCarron JA Levine MJ Melvin GM Murray P Manner PA Tuan RS. An in vivo Lapine Model for Impact-Injury and Ostoearthritic Degeneration of Articular Cartilage 2012. For Peer Review Cartilage.
5.Aspden RM Jeffrey JE Burgin LV. 2002. Impact Loading of Articular Cartilage Letter to the Editor. Osteoarthritis and Cartilage 10(7):588-589.