Abstract
Introduction
Increasing numbers and incidence rates of noisy (squeaking, scratching or clicking) ceramic-on-ceramic (CoC) total hip arthroplasties (THA) are being reported. The etiology seems to always involve stripe wear producing a stick-slip effect in the bearing which excites vibrations. As stripe wear is also found in silent CoC bearings, a theory has been developed that the vibrations become audible only via amplification through the vibrating stem. This was supported by showing that the excitation frequency and the resonance frequency of the plain stem are similar [1]. However, stem resonance in-vivo would be influenced by the periprosthetic bone damping and transmitting stem vibrations. Thus, if stem resonance is conditional for noisy COC hips, these should show periprosthetic bone different to silent hips.
This study compares stem fit&fill and periprosthetic bone between noisy and silent CoC hips.
Methods
In a consecutive series of 186 primary CoC hips with identical stems, cups (Stryker ABG-II) and femoral heads (Alumina V40, 28mm) a dedicated patient questionnaire [2] identified 38 noisy hips (incidence rate: 20.4%, squeakers: n=23). Stem fit&fill and cortical wall thickness (CWT, medial and lateral) were measured on post-op AP x-rays according to an established method [3, Fig 1]. Measurements were repeated by a single blinded observer in a control group of silent hips matched for gender, age, stem size and follow-up time (4.6yrs). Fit&fill and CWT were compared between the noisy and silent group at proximal, mid-stem and distal level and on the medial and lateral side.
Results
The endosteal canal width was equal in noisy (N) and silent hips (S) at all levels (e.g. proximal: N=39.7±5.5mm, S=41.3±5.7mm, Fig 2). On the lateral side also cortical wall thickness (CWT) was the same at all levels (e.g. proximal: N=2.0±0.8mm, S=1.9±0.9mm). However, on the medial side, noisy hips had higher CWT at proximal (N=4.9±2.8mm, S=3.0±2.1mm, p<0.01) and mid-stem level (N=6.2±2.1mm, N=4.6±1.7mm, p<0.001). Also Fit&fill was slightly higher (proximal: N=66%, S=62%; mid-stem: N=63%, S=59%, p<0.05). Differences and significance levels increased when in the noise group only squeakers were considered.
The mid-stem line, assumed to sit distal to LT, actually cut through LT significantly more often in the noisy (18/38) than in the silent group (9/38, p=0.02) indicating a tendency of noisy stems to sit more proximally.
Discussion
Despite equal endosteal canal widths and lateral cortical wall thickness for noisy and silent hips, noisy hips had significantly thicker medial walls at proximal (+63%) and mid-stem level (+35%) where also fit&fill was higher. Noisy stems also sat more proximal (Fig. 3).
This gives evidence that periprosthetic bone (PPB) may play a role in the development of audible noise in CoC hips by providing particular conditions of support, damping and transmission for an oscillating stem which influences noise frequency and intensity. Comparing PPB at different time points indicated that the differences are less due to post-op remodeling but more to pre-op conditions o
