Abstract
Introduction
Wear of the UHMWPE tibial component remains a major reason for aseptic loosening and subsequent revision or failure of TKAs [1]. Many retrieval studies measure surface damage patterns as surrogates for the severity of wear, but little is known about how these patterns relate to the volume of material lost. This study (a) examines the wear rate of a cruciate retaining TKA design and (b) relates observed wear patterns to volume loss on the surface. We hypothesize that damage patterns are good predictors for volumetric wear.
Methods
43 revision and 21 postmortem-retrieved MG II (Zimmer Inc.) tibial UHMWPE components were included in this study. Wear scars and damage patterns on the superior articular surfaces were digitized using a video microscope (SmartScope, OGP). Patterns were parsed into four spatially exclusive categories: delamination, polishing, striations and pitting. The surfaces were measured at 100×100µm using a low-incidence laser on a coordinate measuring machine (SmartScope, OGP). Autonomous mathematical reconstruction of the original surface was used [2] to calculate volume changes on the medial and lateral surfaces as an estimate of wear volume [Fig. 1] Total volume loss was calculated within the observed wear scar, and volume loss under each pattern was calculated and normalized to the total volume loss of its insert.
Results
Excluding delaminated components, total wear correlated linearly with time in situ (Pearson's r=0.53) with a volumetric wear rate of 13.0±2.9 mm3/year. Total wear area correlated linearly with total wear volume (Pearson's r=0.44), while delaminated area correlated strongly with total wear volume (Pearson's r=0.80). Excluding delaminated components, striated areas correlated more strongly to total volume loss (Pearson's r=0.54) than did total wear scar area (Pearson's r=0.34), while other patterns showed no correlation [Fig. 2]. When present, delaminated areas contributed most to total volume loss in postmortem- and revision-retrievals (58.3% and 38.7% respectively), striations second most (36.2% and 30.7%) and polished areas third most (24.6% and 27.6%), although significant differences were not observed [Fig. 3]. Pitted areas contributed significantly less (heteroscedastic t-test p=0.010) to total wear on postmortem- (1.3±1.7%) than on revision-retrievals (11.3±19.6%), although they were observed with a slightly higher frequency in the postmortem group (81% vs. 76%).
Discussion
We found that damage patterns were not reliable surrogates for material volume loss. Other than delamination, area of striated patterns best predicted and contributed most to volume loss. Furthermore, our data suggests that polished and striated patterns in the absence of pitting are markers of a well-functioning UHWMPE TKR. While other groups have reported striations on retrievals [3], this damage pattern remains widely unrecognized in retrieval and knee simulator studies, with the mechanism poorly understood. The higher contribution of pitting to total wear volume in revision-retrieved TKR suggests that the fatigue wear mechanism leading to pitting contributes to the need for the early revision of the implant.
