Abstract
Introduction:
Subchondral bone density (SBD) distribution is an important parameter regarding that may be important when considering implant stability. This parameter is a reflection of the loading experienced by the joint throughout the lifetime and may be useful in pre-surgical planning and implant design. Clinically, the question of the glenoid surface preparation for TSA/RSA remains controversial, despite numerous published studies on glenoid bone morphology. To address this question, there exists a need to develop a 3D quantitative method capable of analyzing the complex glenoid bone morphology at different depths from the surface. Computed tomographic osteoabsoptiomery (CT-OAM) evaluates SBD based on the Housfield Unit (HU) value of each pixel. In this pilot study, we aimed to analyze SBD distribution of the glenoid at different depths by means of CT-OAM in male TSA subjects.
Materials and Methods:
A study group of twenty male TSA patients (61–69y.o) were included in this study. Each subject obtained a pre-operative CT scan following a standardized protocol on the same CT scanner (1.25 mm slice thickness). Resultant DICOM 2D images were processed in custom-written program (VC++) and the surface of every glenoid was manually traced from the axial slices. Care was taken during the manual tracing process to exclude osteophytes and cyst formations from the resultant surface. Values of HU at every selected pixel on the surface of the glenoid were recorded. Subsequently, the layer of pixels at a 0.5 mm distance from the previous surface was virtually scraped and the HU values of new layer of pixels were recorded. This routine was repeated up to a depth of 5 mm from the glenoid surface, taking measurements on 11 virtual 3D surfaces with a thickness of 0.5 mm. Mean SBD distribution was reported for each layer and differences were compared using ANOVA and Fisher's post-hoc test.
Results:
Apparent differences in mean SBD distribution were identified at every measured depth from the glenoid surface (Fig. 1). Significant differences (Tab.1) were identified between the middle range of studied surfaces (2.5–4.5 mm) when compared to the superficial (0–1.5 mm, p < 0.0001) and deep layers (5 mm, p < 0.0001). The maximum mean value of HU (1635.9 ± 35.5) was measured at 3.5 mm depth and the minimum value of HU was measured on the surface of the glenoid (1445.8 ± 31.3).
Discussion:
The stability of the glenoid component in TSA prostheses is highly dependent on the SBD distribution. Controversy among orthopaedic surgeons exists regarding the depth of reaming required to prepare an arthritic glenoid. Extensive reaming may lead to the violation of the support provided by the denser subchondral bone; however, optimal match between the bone and glenoid component undersurface is highly desirable. This study demonstrates that the density of the bone is sustained up to a depth of approximately 4.5 mm from the glenoid surface, suggesting that an increased reaming may be favorable without compromising bony support.
