If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Robert A. Kaufmann, MD, Department of Orthopaedic Surgery, University of Pittsburgh Medical Center, Kaufmann Medical Building, 3471 Fifth Avenue, Suite 1010, Pittsburgh, PA 15213.
The goal of this study was to test the static and dynamic strength and loosening resistance of the posterior flange of a novel total elbow arthroplasty. We also examined the forces experienced by the ulnohumeral joint and the posterior olecranon during expected elbow use.
Methods
Static stress analysis was performed for 3 flange sizes. Failure testing was conducted on 5 flanges (1 medium size and 4 small sizes). Loading occurred to reach 10,000 cycles. If this was accomplished, the cyclic load was increased until failure occurred. If failure occurred before 10,000 cycles, a lower force was employed. The safety factor for each implant size was calculated, and implant failure or loosening was observed.
Results
Static testing revealed a safety factor of 6.6, 5.74, and 4.53 for the small, medium, and large flanges, respectively. The medium-sized flange completed 10,000 cycles with 1,000 N at 1 Hz, and then the force was increased until it failed at 23,000 cycles. Two small-sized flanges failed at 2,345 and 2,453 cycles, respectively, when loaded with 1,000 N. Two more small flanges were loaded with 729 N for 10,000 cycles, and then the cyclic load was continued until they failed at 17,000 and 17,340 cycles, respectively. No screw loosening was noted in any specimens.
Conclusions
This study demonstrates that the posterior flange withstood static and dynamic forces greater than what is expected during in vivo use of a novel total elbow arthroplasty design. Static strength calculation and cyclic loading demonstrate that the medium-sized posterior flange is stronger than the small-sized posterior flange.
Clinical Relevance
Ensuring that the ulnar body component and the posterior flange maintain secure connectivity with the polyethylene wear component may be beneficial to the proper function of a novel nonmechanically linked total elbow arthroplasty.
A novel uncemented total elbow arthroplasty (KTE, Arrch Orthopedics) has been proposed to mitigate complications previously seen in previous total elbow arthroplasty implants, such as aseptic loosening and stress shielding, by closely recreating the normal anatomy of the elbow.
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
These components have a nonmechanically linked articulation that snaps together with 2-finger tightness and exhibits 7° of varus and valgus laxity, allowing physiologic movement.
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Elbow stability is derived from reconstruction of the medial and lateral collateral ligaments and preservation of the radial head. Polyethylene fixation to the ulnar body component is achieved with a posterior flange, which is a collar that locks the polyethylene liner into position.
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Figure 2Multiple posterior flange constructs were considered to secure the polyethylene component. (1) Original single screw design. (2) Tongue and groove design. (3) Key and slot and screw design. (4) Multiple screw design. (5) Design with a groove through the entire polyethylene. (6) Design of the posterior flange and ulnar body component. (7) Schematic of the total construct including the ulnar body component and posterior flange articulating with the humeral component. (8 and 9) After finite element analysis, the tongue and groove was widened and a chamfer was added to the flange.
A posterior flange groove mates with a projection from the ulnar body component to provide rotational control. A custom-designed flange screw secures the flange to the ulnar body component and is inserted into the ulnar body component along the same axis of the intramedullary canal screw (Fig. 3). To prevent this screw from loosening during use, it incorporates a recess that aligns with a hole in the ulnar body component when the screw is fully seated. A cross-locking screw engages the posterior flange screw and the ulnar body component to prevent this screw from backing out.
Figure 3The posterior flange screw is prevented from backing out through a cross-locking screw that mates with a groove in the flange screw. (Top) Schematic of uncemented total elbow arthroplasty (KTE, Arrch Orthopedics) components including ulnar body component, posterior flange, and the posterior flange screw. (Middle) Crosslocking screws prevent the posterior flange screw from loosening. (Bottom) These crosslocking screws align with grooves in the posterior flange screw to prevent it from loosening.
The maximum force is generated at the beginning of flexion from the fully extended elbow and the beginning of extension from the fully bent elbow, and can be approximated by a weight-in-hand multiplier of 16.
Maximum force is required at these ranges given the poor mechanical advantage of the prime movers of the elbow, such as the brachialis, biceps, and brachioradialis.
The elbow only experiences compressive forces through the coronoid without tensile (distraction) loads occurring even during acts of pulling (such as opening a door) that would intuitively seem to generate joint forces that are tensile in nature.
The force vector acts predominantly in the sagittal plane, and any out-of-plane components that would result in mediolateral joint forces are negligible when compared with axial forces (Fig. 4).
Figure 4The force vector is angled relative to the long axis of the humerus. This angle (ɑ) is directed posterior-to-anterior in elbow extension (elbow distraction) and anterior-to-posterior (elbow compression) in all other positions.
Mechanical factors are substantial contributors to arthroplasty failure, and the posterior flange is tasked with maintaining secure connectivity during expected use of this novel elbow replacement. Therefore, a static strength analysis and cyclic loading test was performed to determine its ability to withstand force application without component loosening.
The goal of failure testing was to load to 10,000 cycles and evaluate for loosening. We hypothesize that the implant would not fail statically or loosen with dynamic loading.
Materials and Methods
Static testing
A stress analysis was performed with an 3559 N force applied to the smallest area of the flange. A safety factor, which is the ratio of the strength to expected strain, was calculated by dividing the yield strength by the stress experienced by the small-, medium-, and large-sized flange. Another stress analysis was performed with an 3559 N force applied to the posterior flange screw. Although the maximum force experienced is 3,000 N, we performed a static analysis using 3,559 N to create a worst-case loading scenario.
The polyethylene was placed into the ulnar body component, and the posterior flange screw was tightened. The humeral body of the implant was embedded in clear resin (Vitacrilic polymethyl methacrylate), and the intramedullary screw for the ulnar side was threaded directly into the test machine actuator. A nut was used to secure washers against the ulnar body to maintain alignment of the assembly during testing. The ulnar body was positioned at 210° relative to the humeral body so that the force would be applied directly through the posterior flange (Fig. 5). Although this force on the posterior flange is expected to be greater than during in vivo use, this loading strategy maximized force application to the posterior flange and prevented subluxation of the humerus. The test assembly was aligned in the test frame (Servohydraulic Test Frame, MTS, Eden Prairie, MN), and a tensile fatigue load algorithm was applied. The elbow was loaded in phosphate-buffered saline at 37 °C to create an environment similar to in vivo conditions.
Figure 5The posterior flange is tested with the ulnar body component embedded in clear resin (Vitacrilic polymethyl methacrylate) and the intramedullary screw for the humeral side of the implant threaded directly into the test machine actuator. The ulnar body was positioned at 210° relative to the humeral body so that the force would be applied directly through the posterior flange. An example of posterior flange fracture after dynamic loading is demonstrated.
Cyclic loading was performed at a frequency of 1 Hz with a target of 10,000 cycles, which was chosen as this is the American Society for Testing and Materials (ASTM) standard for shoulder replacement design evaluation and is similar to this flange component of the total elbow arthroplasty.
We dynamically loaded a force of 1,000 N to achieve 10,000 cycles. We began testing with a medium flange and if the target cycle count was accomplished, the loading continued with the same force. Mechanical failure assessment and fretting were visually characterized once fracture of the flange or dislocation occurred. If the cycle count was not achieved, the force was lowered to achieve 10,000 cycles. If the medium-sized flange achieved the cycle count, we proceeded with smaller flanges as the smaller size made them more vulnerable. A lower force was used if the small-sized flange did not reach 10,000 cycles with 1,000 N. It was hypothesized that a smaller person would use their arm in a manner proportional to their size and experience forces consistent with their weight. With a 10% change in each dimension between the medium and the small, a volumetric difference of 27.1% was calculated. Therefore, a commensurate decrease in the cyclic load to 729 N was used.
Assessment measures
The safety factor during static load testing was calculated. Posterior flange screw loosening and component damage were assessed through visual assessment (ie, back-out) and by comparing posttest tightening torque. The pre- and posttest torques were measured using a torque-measuring wrench, which recorded the torque at final tightening during the pretesting phase and the torque required to begin to loosen the screw in the posttesting phase.
Results
Static strength
Static loading of 3,559 N (Fig. 6) demonstrated a safety factor of 6.6, 5.74, and 4.53 for the large, medium, and small flanges, respectively. The posterior flange screw was also subjected to 3,559 N of static loading (Fig. 6) and demonstrated a safety factor of 5.5.
Figure 6Static loading of the flange employed 3559 N applied through the most vulnerable region of the posterior flange.
The medium flange achieved the target cycle number of 10,000 and was reloaded with the same force until failure at the posterior flange screw head was observed at 23,597 cycles. After failure, observation revealed that fretting occurred at the interface between the posterior flange and the ulnar body.
Small flange 1
The first small flange was cyclically loaded with 1,000 N and failed after 2,338 cycles through a posterior flange fracture.
Small flange 2
The second small flange was cyclically loaded with the same 1,000 N force and failed after 1,933 cycles through a posterior flange fracture.
Small flange 3
After 2 failures, we dynamically loaded the third small flange with 729 N, and 10,000 cycles were completed without failure. After this, the force was increased to 802 N and it was cyclically loaded again, causing failure at 10,737 cycles through a posterior flange fracture.
Small flange 4
The fourth small flange was cyclically loaded with 729 N, and 10,000 cycles were completed without failure. The force was increased to 802 N and cyclically loaded again, after which it failed at 12,813 cycles through a posterior flange fracture.
Loosening
The posterior flange screw torque was assessed for loosening (Table). One screw fractured and the remaining 4 screws were immobile until the cross-locking screws were removed. The posttest torque was measured once the cross-locking screws were removed.
TablePre- and Posttest Torque Measurements for Each Test of the Posterior Flange
The main finding of this study was that the posterior flange withstood static and dynamic forces greater than what is expected during in vivo use of a novel total elbow arthroplasty design. The posterior flange screw could only be loosened once the cross-locking screw was removed. In addition, the static strength calculation and cyclic loading demonstrated that the medium-sized posterior flange has greater mechanical properties than the small-sized posterior flange.
Static strength
Even though elbow forces are compressive against the coronoid, we modeled a substantial tensile load to the posterior flange, which is expected to be greater than the tensile load during in vivo clinical use of this device. The high safety factor observed in the stress analysis makes it unlikely to fail mechanically due to a one-time force application. A high safety factor was also demonstrated in the stress analysis of the posterior flange screw.
Dynamic strength
Our goal was to reach at least 10,000 cycles with cyclic loading to conform to ASTM F2028 testing guidelines. ASTM F2028 is intended to investigate the resistance of a total shoulder glenoid component to loosening, which is the most common clinical complication in total shoulder arthroplasty. The guideline determines how much prosthetic glenoid component rocks or pivots following cyclic displacement of the humeral head. We used this guideline as there is no similar regulation for total elbow arthroplasty.
The medium flange failed at 23,000 cycles, after which we tested 4 small flanges, given that their 10% smaller size would make them more vulnerable to loading failure. After 2 small flanges failed at 2,000 cycles, we loaded the remaining 2 small flanges with a smaller force to achieve 10,000 cycles. Therefore, 729 N was applied, which allowed the remaining 2 flanges to complete 10,000 cycles. We used 729 N for the remaining small flanges on the basis of the hypothesis that a smaller person would use their arm in a manner proportional to their size.
Fretting
As the medium flange failed, fretting, which is wear that occurs as 2 materials are repeatedly moved against one another under a load, was observed visually. Fretting causes surface degradation with the formation of grooves and pits. Although fretting was visualized, it was isolated to the region of the posterior flange resulting in mechanical failure. Galvanic corrosion will likely not occur given that the body and posterior flange components and cross-locking screws are machined from ASTM F136 Titanium, which is frequently used for similar implant applications due to excellent biocompatibility, resistance to corrosion, high strength, and ductility.
It was not possible to remove the posterior flange screw when the cross-locking screw was engaged in the groove within the posterior flange screw. Once the engaged cross-locking screw was removed, it became possible to remove the posterior flange screw, and the posttest torque was less than the applied pretorque. Although quantification of the limit for acceptable posttest torque loosening is application dependent given multiple variables such as stress relaxation, plastic deformation, fretting of bolt threads, and screw thread pitch, it is commonly reported to take approximately 10% less torque to loosen a bolt than is needed to tighten it.
Measurements of repeated tightening and loosening torque of seven different implant/abutment connection designs and their modifications: an in vitro study.
Although our posttest torque loosening values were more than 10% lower, no torque could loosen the posterior flange screw with the cross-locking screws in place, which is how this novel arthroplasty will be employed clinically.
Comparison with current total elbow arthroplasty designs and advantages of this novel design
Nonconstrained prostheses have little intrinsic stability and depend on soft tissue stabilizers, namely the medial and lateral collateral ligaments, to resist varus and valgus forces.
Semiconstrained options were designed in response to the instability concerns encountered with nonconstrained designs and allow for some varus and valgus motion to occur to reduce stress on the bone-cement interface. However, this limit is often reached when the arm is abducted, and loading forces may contribute to premature aseptic loosening and polyethylene wear.
A finite element analysis has demonstrated that newer linked third-generation implants aim to increase contact area creating better load sharing and lower peak stress on the polyethylene bearing than older semiconstrained second-generation designs.
Despite these improvements, any mechanically linked design will transfer greater stress at the cement-bone interface than a method that is not mechanically linked and, yet, achieves stability through ligamentous restraint.
This novel total elbow arthroplasty seeks to avoid complications that have troubled previous semiconstrained total elbow arthroplasty implants by closely recreating the normal anatomy of the elbow and transmitting forces through ligaments.
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Elbow stability is derived from reconstructing the medial and lateral collateral ligaments, which has been demonstrated to reduce loads on components in unlinked and linked designs.
The distal humerus is replaced by a convex smooth metal articular surface, and the proximal ulna is replaced by a concave polyethylene surface. These components have a nonmechanically linked articulation that snaps together and exhibits 7° of varus and valgus laxity.
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Polyethylene fixation to the ulnar body component is achieved with a posterior flange, which resembles a collar that locks the polyethylene wear component into position while allowing subsequent polyethylene exchange without removing the ulnar body component.
Sources of error
Unidirectional force direction
We chose one force vector that caused the greatest force against the posterior flange. Regular elbow use causes different force angles across the elbow, but these are all compressive between the coronoid and distal humerus. Our force was applied at a single angle, which was chosen to maximize the force experienced by the flange and not cause the distal humerus to dislocate. Despite the force vectors that the elbow experiences when being compressive against the coronoid (Fig. 4), we wanted to evaluate the posterior flange integrity and apply a nonanatomic force.
Small sample size
One limitation of this study is that only 5 flanges were tested, and 2 flanges failed before reaching 10,000 cycles. No loosening was identified regardless of cycle count, making it unlikely that additional flanges would have changed our results.
Target cycles are too low
No loosening was identified regardless of how many cycles were achieved. The target number of 10,000 cycles represents approximately 25 higher-load activities per day for 10 years.
Another limitation is that the small size was used for most tests. The small-sized flange was chosen because the reduced dimensions make it more vulnerable to failure.
Did not apply correct force
Although peak joint reaction forces have been estimated to range from 350 N for light activities to 3,000 N for extreme loading, these forces are compressive against the coronoid (Fig. 4).
Smaller forces are to be expected on the posterior flange region of the olecranon during elbow use. Our applied load of 1,000 N likely overrepresents any force this region may experience.
In conclusion, this study demonstrates that the posterior flange withstood static and dynamic forces greater than what is expected during in vivo use of a novel total elbow arthroplasty design. This supports the use of this design in the clinical setting.
References
Kaufmann R.A.
D’Auria J.L.
Schneppendahl J.
Total elbow arthroplasty: elbow biomechanics and failure.
Combs TN, Nelson BK, Jackucki M, et al. Testing of novel total elbow prostheses using active motion experimental setup. J Hand Surg Am. Published online December 13. https://doi.org/10.1016/j.jhsa.2021.10.02
Measurements of repeated tightening and loosening torque of seven different implant/abutment connection designs and their modifications: an in vitro study.
Declaration of interests: R.A.K. owns Arrch Orthopaedics. No benefits in any form have been received or will be received related directly to this article.
00002-6&id=gr1.jpg){kind=link}
00002-6&id=gr2.jpg){kind=link}
00002-6&id=gr3.jpg){kind=link}
00002-6&id=gr4.jpg){kind=link}
00002-6&id=gr5.jpg){kind=link}
00002-6&id=gr6.jpg){kind=link}