Effect of Varying Screw Configuration and Bone Density on Reverse Shoulder Glenoid Fixation Following Cyclic Loading

*Roche C; **Flurin PH; ***Wright T; ****Crosby L; *Hutchinson D; *****Zuckerman J
*Exactech, Gainesville, FL; **Bordeaux-Merignac Clinic, FR; ***Univ. of Florida Dept. of Ortho.,
Gainesville, FL; ****Wright State, Dayton, OH; *****NYU Hospital for Joint Diseases, NY, NY

Introduction

Recent successes achieved with reverse shoulder arthroplasty have led to an expansion of its indications and an increase in the number of commercially available designs in the global marketplace. Despite these development efforts, little guidance exists regarding reverse shoulder test methods and little published data exists regarding reverse shoulder performance standards. Of the bench studies that have been published, most were performed under idealized conditions (i.e. uniform bone densities of 30 lb/ft³ in which all compression/locking screws achieve excellent/optimal fixation). Since many of the patients who would receive this type of prosthesis have some form of compromised glenoid bone stock due to age, deformity and/or pathology, the authors contend that future work should consider the performance of reverse shoulder designs in these suboptimal conditions. To this end, the purpose of this study is to evaluate the effect of varying compression/locking screw configurations and bone quality on reverse shoulder glenoid fixation following cyclic loading.

Methodology

The reverse shoulder glenoid components (Equinoxe®; Exactech®, Inc.) were assembled to two different densities (15 and 30 lb/ft³, each conforming to ASTM F 1839) of polyurethane bone substitute blocks (Pacific Research Laboratories; Vashon, WA; 76mm x 57mm x 48mm). The assembly of the glenoid components were performed according to the prescribed surgical technique with one exception: the central hole for the bone cage was oversized so that no press-fit of the bone cage was achieved. Doing so ensured glenoid fixation was achieved by only the screws; thereby, isolating the effect of each configuration. The glenoid components were then fixed to the bone blocks with 4.5mm compression screws (via a 3.2mm pilot hole) in five different screw configurations and were secured to the glenoid plate with locking caps; three samples of each configuration were tested in each density bone block (n=30). (Figure 1)

Figure 1. Glenoid Plate Screw Configurations

Configuration #1 represented the ‘manufacturer recommended’ screw pattern in which one screw was placed in the superior hole and three screws in the three most inferior holes. Configuration #2 represented a condition where only three of the four screws achieved fixation. Configuration #3 represented a condition where only the superior and inferior screws achieved fixation. Configuration #4 represented the conversion of a pegged glenoid to a reverse shoulder. Configuration #5 represented the conversion of a keeled glenoid to a reverse shoulder. It should be noted that configurations #4 and #5 were simulated by first implanting a pegged and keeled glenoid (respectively, using the typical surgical technique: ream, drill, broach) in the bone block and then subsequently fixing the reverse shoulder glenoid components per the prescribed technique.

Each humeral liner component was cyclically loaded for 5,000 cycles about a 55 degrees arc along the glenosphere using a rotatory actuator at a rate of 0.5 Hz as a 750 N compression load was applied with a hydraulic testing machine through the center of the glenosphere via the backside of the bone block. (Figures 2 and 3) It should be noted that this cyclic load represents ~12.5 high-load activities a day (such as getting out of a chair) for one year.¹ Before and after cyclic loading, the stability of the glenoid plate/bone block construct was quantified using a dial indicator (having an accuracy of two microns) by measuring the total motion between the glenoid plate and bone block. A 50 N load was applied (before and after cyclic loading) to the top of the glenoid plate as a dial indicator was placed on the bottom of the glenoid plate to measure the amount of glenoid plate displacement in the inferior/superior direction. Next, the glenoid plate was turned 90 degrees and the 50 N load was applied to the side of the glenoid plate as the dial indicator was placed on the opposite side of the glenoid plate to measure displacement in the anterior/posterior direction. (Figure 4)

Figure 2. Orientation and Position of Applied Cyclic Load

Figure 3. Cyclic Loading Apparatus	Figure 4. Set up to Measure Glenoid Plate Micromotion in the Superior/Inferior (left) and Anterior/Posterior Directions (right)

Results

The average glenoid plate motion in the anterior/posterior direction for each screw configuration in each density block is presented in Table 1. The average glenoid plate motion in the inferior/superior direction for each screw configuration in each density block is presented in Table 2.

Table 1. Comparison of Anterior/Posterior Glenoid Plate Motion

Table 2. Comparison of Inferior/Superior Glenoid Plate Motion

Regarding the effect of screw configuration on glenoid fixation, the maximum difference in glenoid motion before and after cyclic loading occurred with screw configuration #3 in the anterior/posterior direction for both density blocks: 140 and 10 microns for the 15 lb/ft³ and 30 lb/ft³ density blocks, respectively. It should be noted that this component in the 15 lb/ft³ density block had a maximum post-cyclic motion of 178 microns. This value is greater than the generally-accepted 150 micron threshold for osseous on-growth;² no other component in any other configuration had a maximum motion above this threshold.

Regarding the effect of bone density on glenoid fixation, for every component tested in each screw configuration, the amount of motion before and after cyclic loading was greater in the 15 lb/ft³ density blocks than in the 30 lb/ft³ blocks. Of note, in nearly every instance, the 2:1 relationship between densities did not hold true regarding fixation: the amount of motion associated with the 15 lb/ft³ density blocks far exceeded the motion associated in the corresponding 30 lb/ft³ blocks.

Discussion

The results of this study demonstrate that the pattern of screw fixation and bone density has a direct effect on reverse shoulder glenoid fixation following cyclic loading. Furthermore, the results demonstrate that certain ‘worst-case’ combinations of these conditions (e.g. only two screws of fixation in poor quality bone) may not provide sufficient stability to achieve the osseous on-growth necessary for the long-term survival of the prosthesis. Because these variables are patient related and are common in the populations who may receive this type of prosthesis (e.g. patients over 70 years of age who have compromised glenoid bone stock), the authors of this study recommend that at least three screws be used to achieve initial glenoid plate fixation. The primary limitation of this study is the small sample size; larger numbers are required to further elucidate the relationships.

References

1. Anglin C, et al. Mechanical testing of shoulder prostheses and recommendations for glenoid design. J Shoulder Elbow Surg. 2000 Jul-Aug;9(4):323-31.

2. Cameron HU, Pilliar RM, MacNab I. The effect of movement on the bonding of porous metal to bone. J Biomed Mater Res. 1973 Jul;7(4):301-11.

Production of this poster sponsored by:
Exactech, Inc., General Research Fund