Date of Award


Document Type


Degree Name

Master of Science (MS)


Biomedical Engineering


Biomedical Engineering and Imaging

Research Advisor

William M. Mihalko, M.D. Ph.D.


Denis J. DiAngelo, Ph.D. John L. Williams, Ph.D.


Total hip arthroplasty, THA, surgical implants, orthopedic surgery


Introduction. Total hip arthroplasty (THA) surgery is one of the most commonly performed and successful orthopedic procedures in the United States. More than 300,000 primary THAs and 40,000 revision THAs performed in the United States every year. While the need for revision surgeries can stem from a variety of causes, there have been, to the author’s knowledge, no studies attempting to correlate the concentrations of certain inflammatory cytokines to metal ion concentrations found in the tissue surrounding the implant, amount of polyethylene wear, or strength of the interface of the modular taper. The purpose of this study was to begin to look at those factors to see if any were indicative of implant survivorship, as well as to see if metal ion content contributes to implant longevity. The testing for this group of well-functioning implants will be useful as a baseline when comparing the same types of testing for failed implants.

Methods. A total of nineteen cadaveric total hip implants were obtained from two sources, the Medical Education and Research Institute (Memphis, TN) and RestoreLifeUSA (Elizabethton, TN). The bearings for these implants were either metal on polyethylene or ceramic on polyethylene. Synovial fluid and tissue samples were taken from the joint for testing. Head dissociation was performed, in which an Instron 4505 was used in accordance with ASTM Standard F2009-00 to remove the head from the stem of the implant, recording force. Corrosion scoring was performed on taper surfaces by three scorers. The polyethylene acetabular liner was measured on the superior side with a micrometer to determine how much material loss was evident compared to the inferior side. These three values were then correlated to the testing performed using the synovial fluid and tissue. The synovial fluid was analyzed for inflammatory cytokines IL-6, MCP-1, IL-1β, MIP-3α, M-CSF, IL-8, IL-2, and TNF-α using a premixed Luminex screening assay. These results were given in picograms per milliliter. An anterior and posterior synovial tissue sample was analyzed for the presence of metal ions cobalt, chromium, and titanium using Inductively Coupled Plasma Mass Spectrometry (ICP-MS). All of these results were compiled and analyzed together to search for potential correlations.

Results. There were no significant differences in dissociation forces between the groups of implants with head corrosion score 1 and head corrosion score 2. The comparison of MCP-1 to the dissociation force produced a correlation coefficient of 0.64 (p-value 0.05) and the comparison of MIP-3α to the dissociation force produced a correlation coefficient of 0.67 (p-value 0.03). However, when the graphs of these correlations were observed, it seemed likely that this correlation was due to one sample pulling the graph in a positive direction which is demonstrated by the 95% confidence interval (CI) of the correlation coefficient (0.011 to 0.90 for MCP-1, and 0.069 to 0.91 for MIP-3α). When comparing polyethylene wear to the inflammatory cytokine concentrations, no significant correlations were seen. There was a positive correlation between cobalt and chromium levels and dissociation force (r=0.56 for cobalt, r=0.66 for chromium), and a negative correlation between titanium levels and dissociation force (r=-0.30). The positive relationship was opposite of what was expected, as more metal debris should mean the implant surfaces are losing material, which should therefore decrease the strength of the taper connection. The 95% confidence interval for the correlation coefficients included zero for cobalt and titanium, and was fairly wide for chromium (0.11 to 0.90). When observing cytokines and metal ion presence, most relationships were very scattered with low correlation coefficients. However, for cobalt, strong positive relationships were seen for IL-6 (r=0.67, CI: 0.19 to 0.89), MCP-1 (r=0.76, CI: 0.33 to 0.93), and MIP-3α (r=0.60, CI: 0.066 to 0.86). When looking at confidence intervals, there seemed to be a mild correlation between cobalt and IL-6 and a moderate correlation between cobalt and MCP-1. No meaningful relationships were seen for any cytokines with chromium or titanium, so it may be useful to select cytokines known to be responsive to those two metals in particular for future studies. When comparing metal levels between the two corrosion levels seen in the heads, there were no statistically significant differences in any of the metals between implants with a corrosion score of one and those with a corrosion score of two.

Discussion. This study was limited by the fact that the sample size for this study was very low. With only nineteen total implants, it was difficult to draw meaningful conclusions. Additional implants are being recruited in order to increase this sample size for future studies. Additionally, it was difficult for meaningful correlations to be seen when comparing any factor to the inflammatory cytokine concentrations, as these values were clustered around the lower limit of detection. However, this was expected with well-functioning implants. While it is difficult to draw meaningful conclusions when used as a correlation, this data will be useful when comparing cytokine concentrations of a group of failed implants. This group is able to serve as a baseline value for each type of testing performed, and will help to make sense of the same testing of failed implants in the future.