PhD Disseration: Biologic Response to Wear Debris in Total Disc Replacement
Total disc replacement (TDR) was clinically introduced as an alternative to spinal fusion to relieve back pain, maintain mobility of the spine and eliminate the adverse side effects of fusion. More recently, gamma-inert-sterilized ultra-high molecular weight polyethylene (UHMWPE) TDR cores were introduced to replace historical gamma-air-sterilized cores in an effort to reduce UHMWPE wear debris and inflammation. In this study, both implant and periprosthetic tissue retrievals from patients with gamma-inert-sterilized TDRs were evaluated for in vivo performance and biological responses, respectively. As pain was the primary revision reason for all patients, the contributions of implant-related damage and tissue responses to the development of pain were also a focus of this investigation.
After analyzing implants and tissues for 11 TDR patients, detectable UHMWPE wear debris was identified with corresponding macrophage infiltration in six patients with associated implant damage. Neither damage nor TDR bearing design, fixed vs mobile, influenced the amount, size and shape characteristics of wear particles. However, comparisons to a retrieval study of historical devices indicated that the number of UHMWPE particles generated from gamma-inert-sterilized devices were decreased by 99% (p=0.003) and were 50% rounder (p=0.003), confirming the improved wear resistance of the newer devices. Accordingly, periprosthetic tissue reactions were also substantially reduced.
Prospective immunohistochemical investigations for these devices showed, for the first time, that UHMWPE wear-debris induced tissue reactions in the human lumbar spine can be linked to inflammation. First, inflammatory factors were elevated in TDR periprosthetic tissues (n=30) when compared to disc degenerative disease (DDD) patient tissues (n=3) from primary surgery and disc tissues (n=4) from normal autopsy patients with no history of lower back pain. The mean percent area of production for vascular endothelial growth factor (VEGF) (p=0.04), interleukin-1beta (IL-1β), (p=0.01) and substance P (p=0.01) were significantly higher in TDR tissues when compared to tissues obtained from DDD patients. Although platelet derived growth factor-bb (PDGFbb) (p=0.14), tumor necrosis factor-alpha (TNFα) (p=0.06) and nerve growth factor (NGF) (p=0.19) were also increased in the TDR patient tissues, these increases were not significant. Compared to normal disc tissues, the mean percent area for all six factors was statistically increased in TDR tissues (at least p<0.05 for all). Interestingly, no statistical differences were observed between DDD and normal disc tissues. Next, our studies showed that TNFα, IL-1ß, VEGF, NGF and substance P strongly correlated with the number of wear particles and also the number of macrophages for the TDR patient group (at least p<0.05 for all). Finally, the pro-inflammatory/pain factors, TNFα and IL-1ß, and the vascularization factors, VEGF and PDGFbb, significantly correlated with the presence of the neural innervation and hypersensitization agents, NGF and substance P (p<0.01 for all). These findings suggest not only the presence of inflammatory reactions, but the presence of factors that can directly and indirectly contribute to the pain sensitivity.
In addition to wear-debris and subsequent inflammation, increased vascularization was another key histomorphological change observed in the TDR tissues that may be involved in the pathogenesis of particle disease. In brief, the ingrowth of blood vessels may be providing a conduit for nociceptive innervation. Studying vascularity in revision tissues showed the total number of blood vessels was significantly associated with TNFα, IL-1ß, VEGF, PDGFbb, NGF and substance P (at least p<0.05 for all), suggesting an interrelation between vascular changes and inflammatory-mediated responses. Furthermore, analysis at the local level revealed the innervation/pain factors, NGF and substance P, were predominantly localized to vascular channels, suggestive of nerve ingrowth and potential neural-maladaptive plasticity at periprosthetic sites. Lastly, comparing blood vessel number with factor expression and macrophage number in individual images obtained from tissue sections with low and high vascularity suggested a temporal link between TNFα, macrophages and angiogenesis. Taken together, elucidating the pathogenesis of inflammatory particle disease will provide information needed to identify potential therapeutic targets and treatment strategies to mitigate pain.