A combination of laboratory experiment and computational simulation was performed to assess the role of interface porosity on stem migration. The early motion of in vitro prepared cemented femoral components was measured during application of cyclic stair climbing loads. Following testing, transverse sections were obtained and the distribution of pores at the stem-cement interface was determined. Finite element models of cemented stem constructs were developed and a scheme was implemented to randomly assign pores to the stem-cement interface. For a series of 14 in vitro prepared components, pore fractions at the stem-cement interface ranged from 23% to 67%. The majority of pores at the stem-cement interface were less than 1 mm in length with a mean length of 1.27 +/- 2.7 mm and thickness of 0.12 +/- 0.11 mm. For stems with large pore fractions, pores tended to coalesce in longer extended gaps over the stem surface. Finite element and experimental models both revealed strong positive correlations (r(2) = 0.55-0.72; p < 0.0001) between stem-cement pore fraction and stem internal rotation, suggesting that the presence and extent of pores could explain the early motion of the stems. There was an increased volume of cement at risk of fatigue failure with increasing stem migration. Pore fractions greater than 30% resulted in large increases in stem internal rotation, suggesting that attempts to maintain surface porosity at or below this level may be desirable to minimize the risk of clinical loosening.
(c) 2006 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.