Stress shielding (i.e. reduction in bone strains) in the distal ulna is commonly noted following ulnar head replacement arthroplasty. Optimal design parameters for distal ulnar implants, including the length of the stem, are currently unknown. The purpose of this study was to investigate the effect of stem length on bone strains along the length of the ulna.

Strain gauges were applied to each of eight cadaveric ulnae to measure bending loads at six locations along each ulna’s length (approximately 1.5, 2.5, 4.0, 6.0, 8.0, and 13.0cm from the ulnar head). The proximal portion of each bone was secured in a custom-designed jig. A materials testing machine applied loads (5–30N) to the ulnar head while native strains were recorded. The ulnar head was removed and the loading procedure repeated for cemented stainless steel stems 3 and 7cm in length, according to a previously reported technique (Austman et al, CORS 2006). Other stem lengths between 3 and 7cm were tested in 0.5cm intervals with a 20N load applied only. Data were analyzed using a two-way repeated measures ANOVA (á=0.05).

In general, distal bone strains increased as stem length decreased (e.g. average microstrains at the second distal-most gauges: 138±13 (7cm), 147±15 (6cm), 159±21 (5cm), 186±40 (4cm), 235±43 (3cm)). The native strains were different from all stem lengths for the four distal-most gauges (p< 0.05). No differences were found between any stem length and the native bone at the two proximal-most gauges. The 3cm stem replicated the native strains more closely than the 7cm, over all applied loads (e.g. average microstrains at the third gauge level for a 25N load: 357±59 (native), 396±74 (3cm), 257±34 (7cm)).

No stem length tested matched the native strains at all gauge locations. The 3cm stem results were closer to the native strains than the 7cm stem for all loads at gauges overtop of the stem. Overall, the 3cm stem produced the highest strains, and thus would likely result in less distal ulnar bone resorption after implantation. These results suggest that shorter (approximately 3cm) stems should be considered for distal ulnar implants to potentially reduce stress shielding, although this must be balanced by adequate stem length for fixation.

Purpose: To develop an experimental testing method to measure bone strains as a function of multiple implant stem designs in a single specimen, and to show the efficacy of this method with an application in the distal ulna.

Methods: Twenty-four strain gauges were applied to the surface of an isolated cadaveric ulna to measure anterior-posterior (AP) and medial-lateral (ML) bending loads at six locations along its length. The bone was potted in a custom-designed jig and positioned in a materials testing machine. Loads (5-25N) were applied to the ulnar head while strains were recorded. The ulnar head was removed and an 8cm threaded rod (diameter=5.8mm) was cemented into the canal, and subsequently removed after cement curing. This established a threaded cement mantle that would accept various threaded stem designs. To show the efficacy of this technique, testing was repeated with 5 and 7cm stems. The entire canal was then filled with cement and testing repeated to determine the effect of the residual cement void.

Results: All 24 strain gauges provided quality signals throughout the testing period. Strain varied linearly with load (R-squared=0.94–0.99). The initial threaded rod was easily removed, and there was no difficulty in placing subsequent stems within the mantle. Comparing the 5 and 7cm stems, little difference in strains was observed for the most proximal gauges (2%), with higher variations in the stem exit regions (17%). The cement-filled canal exhibited distal strains similar to the intact baselines (average 2% difference at 25N).

Conclusions: A reliable method has been developed that allows multiple stems to be tested in a single bone. Observed strain differences are therefore a function of implant parameters only (such as stem length), and are not influenced by differences in bone properties as occurs when testing multiple specimens. The layer of threaded bone cement did not impact the native bone strains. This experimental method will be useful to compare stem designs in a variety of bones, avoiding the need for large numbers of specimens due to the repeated measure experimental protocol.

Purpose: Anterior flanges have been added to the humeral components of some total elbow arthroplasty systems. Surgeons have the option of placing a wedge of bone or bone cement between the anterior surface of the humerus and the flange in an effort to improve implant stability and load transfer. The purpose of this study was to quantify the cortical strains in the humerus after arthroplasty for different materials placed behind the flange.

Methods: Five fresh-frozen cadaveric distal humeri were thawed and cleaned of all soft tissues. Strain gauges were applied to the anterior and posterior surfaces to record bending and axial strains. The bending gauges were positioned just proximal to the location of the flange tip. Cantilever bending and axial compression were applied using a materials testing machine. Following intact testing, the humeral component of a total elbow was implanted by an experienced surgeon and fixed using bone cement. Testing was repeated three times, each with a different material behind the flange: no graft (simulating a humeral component without an anterior flange), cancellous bone graft, and cement graft. Strains were normalized to the intact state and for the applied moments. Data were analyzed using repeated-measure ANOVAs (p< 0.05).

Results: For bending, the strain values were approximately 80% of the intact values with no graft material, 80% with the bone graft, and 87% with the cement graft. These differences among the graft materials were not significant (p=0.5). Similar results were found for the axial strains (p=0.3).

Conclusions: The intention of the anterior flange is to transfer a portion of the load carried by the implant stem to the distal humerus, thereby reducing stress-shielding and improving strength of the construct. In this investigation that employed bending and axial loads, the presence of an anterior flange had no significant effect on load transfer through the distal humerus regardless of graft material used. This would suggest that for the humeral component employed in this study, the flange might not be fulfilling its intended purpose.