Olecranon fractures account for 10% of all upper-extremity fractures in adults1 and are often caused by a fall from standing height in the elderly population.2 Surgical intervention for these injuries requires accurate identification of the fracture pattern, restoration of joint congruency, repair of the elbow extensor mechanism, and stabilization of the elbow for early joint motion.3,4 Two-part transverse fractures can be operatively treated with tension band wiring, intramedullary nailing, or plates, but the optimal operative technique remains a topic of debate.3,5–12
Plate fixation has become an accepted method for stabilization of olecranon fractures,13–17 and several studies have sought to improve clinical results by quantifying the differences caused by changes in implant design or operative technique. It has been suggested that locking plate designs provide no substantial advantage over nonlocking plates in this application.18,19 Other studies have revealed that changes to the plate design, the plate thickness, and the number, diameter, and length of screws used have minimal influence on fixation stability.20,21 Although these results seem to suggest that the general design of olecranon plates is robust, it has also been shown that design changes affecting proximal ulnar fragment fixation have the ability to greatly influence implant performance.22
Complications and failures of olecranon fracture fixation are typically attributed to implant loosening. The presence of osteoporotic bone or comminution can lead to fixation failure when the forearm is subjected to external loads.23–26 It is known that healing in osteoporotic bone can be slow or incomplete27 and that the presence of implants may increase the risk of healing complications.28 Despite these issues, effective postoperative rehabilitation protocols require patients to perform early range of motion exercises to limit stiffness and facilitate their ability to perform their activities of daily living.12,29,30 These protocols routinely increase the ranges of motion and the external loads applied to the affected joint during a prescribed period. These aggressive but necessary rehabilitation protocols may ultimately lead to premature implant failure, especially in osteoporotic bone.
The purpose of this study was to determine if an additional nonlocking screw directed into the proximal ulnar fragment (Figure 1) may provide additional stability that will arrest fracture displacement caused by external loading. Currently, no biomechanical studies have determined the mechanical advantage of this surgical approach. This study compared the performance of olecranon fracture fixation constructs using an additional nonlocking screw (EXPT) with that of plates that were implanted following the manufacturer's guidelines (CTRL). It was hypothesized that the EXPT group would improve fatigue life by effectively reducing displacement between the proximal and the distal ulna segments during cyclic loading.
Fluoroscopic images of reconstructions with a standard repair method (A) and use of an additional retrograde screw (arrow) (B). The additional screw is directed out of axis of the intended vector for a locking screw and instead goes into the proximal fracture segment.
Materials and Methods
Nine matched pairs of fresh-frozen cadaveric arm specimens (6 female, 3 male; average age, 81.2 years; range, 62–92 years) were used for this study. Sample size was based on an a priori power analysis of a previously published olecranon biomechanical study.20 Prior to mechanical testing, all specimens were stored at −20°C and then fully thawed before implantation. Specimens underwent transhumeral and trans-forearm amputations at the midpoints of the bones, and gross dissection was performed. The radioulnar interosseous ligament, elbow capsule, and triceps were kept intact.
Fracture simulation and reconstruction was performed by 2 fellowship-trained orthopedic surgeons (S.M., S.N.). Plates (3.5-mm VA-LCP Olecranon Plates; DePuy Synthes, West Chester, Pennsylvania) were first properly positioned on intact bones and held in place with Kirschner wires. An oscillating saw was used to create a transverse osteotomy at the center of the sigmoid notch of each specimen. A second osteotomy was made 3 mm distal to the first osteotomy, and the bone fragment was removed so that there was no bony contact between the proximal and the distal portions of the ulna. The gap between the bone fragments was intentionally created so that fragment fixation relied completely on screw fixation and did not use bone fragment compression as a means of stabilization. A standard technique, consisting of 2 distal nonlocking bicortical screws (3.5-mm diameter, 20-mm length) and 2 nonlocking proximal screws (2.7-mm diameter, 50- and 60-mm length), was used for the CTRL group. An additional nonlocking screw (2.7-mm diameter, 30-mm length), targeted from distal to proximal through the plate, aimed toward the tip of the olecranon, was employed for the EXPT group (Figure 1). Although more screws could be used for these reconstructions, the authors sought to isolate the impact of the additional screw in the EXPT group; thus, only 4 or 5 screws were used per specimen. Prepared specimens were stored in a 4°C refrigerator for no longer than 48 hours before testing.
This model simulated the kinetics and kinematics that are associated with rising from a chair or ambulating with a walker.19,20,31 Similar to previously published protocols,20,31 increasingly loaded elbow extensions were simulated by applying controlled displacements to the triceps tendon with a universal test frame (ElectroForce 3550; TA Instruments, Eden Prairie, Minnesota) (Figure 2). To create this motion, a flexible steel cable was routed through a set of pulleys to connect the actuator, load cell, and custom-built triceps tendon clamp.32 The actuator displaced the triceps tendon 20 mm in a sinusoidal pattern at 0.2 Hz, which resulted in a range of motion between 90° and approximately 55° of flexion.
Photographs of the biomechanical testing setup. Displacement of a steel cable attached to the triceps tendon creates a downward force, indicated by the arrow (A). The application of the force results in elbow extension (B).
Three-dimensional motion capture techniques were used to measure fracture diastasis and elbow flexion angles during the experiment, closely mimicking previously published work.20 A motion capture system (OptiTrack; NaturalPoint, Inc, Corvallis, Oregon) was calibrated such that 0.2-mm accuracy of marker tracking was achieved. Marker clusters were securely attached to the humerus, proximal ulnar fragment, and distal ulnar fragment (Figure 3A), which permitted measurements of fragment diastasis. Locations of the following 3 points were recorded with an instrumented pointer during static trials: (1) medial epicondyle, (2) lateral epicondyle, and (3) the point defined by intersection of the proximal plane of the fracture and the lateral edge of the plate. Two coordinate systems were created such that their z-axes were parallel to the epicondylar axis. “Dummy” y-axes were then created to mimic the global y-axis. The x-axes were determined by taking the cross product of the z- and “dummy” y-axes. To ensure perpendicularity of all axes in the coordinate systems, the “actual” y-axes were calculated by taking the cross-product of the x- and z-axes. Once the rotation matrices of the 2 coordinate systems were established, the origins were translated such that they overlapped exactly at point 3, at the fracture edge (Figure 3B). Each coordinate system was assigned to move in unison with either the proximal or the distal ulnar marker clusters. Fracture diastasis was measured by calculating the Euclidean distance between origins throughout the course of testing (Figure 3C). Elbow flexion angles were calculated by using Euler angle decomposition techniques between the humeral and the proximal fragment clusters about the epicondylar axis.
Schematic showing the steps taken to calculate diastasis. Marker clusters are attached to each body, and coordinate systems are assigned to the clusters (A). Virtual coordinate systems are created, relative to the proximal (Prox.) and distal (Dist.) ulna fragments, and placed directly on top of each other (B). Relative motion between the ulnar fragments moves the virtual coordinate systems, and the distance between them is calculated (C).
Failure in this experiment was defined as 1 of 2 possible events. First, it was determined if permanent relative displacement of the proximal and distal fragments exceeded 3 mm beyond the initial gap of 3 mm. To check this, the relative displacements of the bone fragments were plotted as a function of time for each trial. Failure was identified if permanent creep of the fixation (ie, the valley of the curve) exceeded the 3-mm threshold (Figure 4A). If this did not occur, failure was identified as the point when instantaneous catastrophic failure of the bone occurred (Figure 4B).
Plots of measured diastasis showing steadily increasing diastasis between segments beyond 3 mm, which was a typical failure mechanism of the standard surgical technique group (A), and secure fixation of the ulnar fragments followed by catastrophic failure, which was the typical failure mode for the additional retrograde screw group (B). The arrow indicates the last cycle that was counted before failure.
The dynamic experiment was performed with a series of repeated motions under increasing load. For the first trial, the triceps tendon was tensioned with approximately 100 N to hold the elbow in 90° of flexion while supporting a 0.6-kg empty fixture. The triceps was then cycled 30 times. The test was paused, an additional 0.5-kg mass was placed on the fixture, and tension on the tendon was increased to maintain the load. Once a static flexion angle of 90° was achieved, the second test, using a 1.1-kg load, was run for 30 cycles. The addition of 0.5-kg masses and 30-cycle trials continued up to a final load of 9.6 kg or until catastrophic failure, whichever came first.
Implant performance was characterized by assessing the overall failure mechanism, measuring the total number of cycles to failure, and measuring maximum load applied to the proximal ulnar segment. Variations in human anatomy resulted in different amounts of angular displacement of the elbow during 20-mm tendon excursion, so a measure that was sensitive to these changes (total work performed against gravity) was also used.
Bone Quality Characterization
Because the mechanical properties of the bone itself play an important role in implant fixation, several assays were performed to fully characterize the bone specimens used in this study. After mechanical testing was completed, the proximal ulnar fragments were isolated and scanned with an optical 3-dimensional scanner (Einscan; Afinia 3D, Chanhassen, Minnesota) to determine total proximal fragment volume. The fragments were then analyzed with micro-computed tomography (MicroCT 35; Scanco, Bruettisellen, Switzerland) to determine tissue mineral density, bone mineral density, and bone volume/total volume ratio. Finally, maximum screw pull-out loads were determined by performing the ASTM International standard assay for screw pull-out testing33 on the radial head of each specimen.
Statistical analysis was conducted using SigmaStat version 4.0 software (Systat Software, Inc, Erkrath, Germany). Shapiro–Wilk tests were performed to test for normal distribution of the data points. Because the distribution of the data set was non-Gaussian, a paired Mann–Whitney rank sum test was performed to determine differences in total cycles sustained, maximum load, and total work performed. The level of significance was set as P<.05.
There were substantial differences in the failure modes between the groups. The EXPT group failed most commonly due to an instantaneous catastrophic failure (7 of 9 trials), whereas the CTRL group most commonly failed due to relative fragment displacement exceeding 3 mm (7 of 9 trials) (Figure 5A). The EXPT group failed by fragment displacement in 2 of 9 trials, while the CTRL group experienced catastrophic failure once. There was 1 instance where a CTRL specimen completed the loading protocol and did not fail. Although the EXPT group exhibited lower amounts of average diastasis as a function of load applied (Figure 5B), there were no significant differences in this regard.
Survival plot showing the ultimate applied loads and failure mechanisms for each specimen (A). The means (solid lines) ± 1 standard deviation (clouds) of the average diastasis are shown as a function of applied load for each group. Standard deviations ceased to exist when only 1 specimen was remaining (B). Abbreviations: CTRL, standard surgical technique group; EXPT, additional retrograde screw group.
There were no significant differences in terms of number of survived cycles, maximum load, or work performed when comparing the CTRL and EXPT groups. The CTRL group sustained an average of 200 (±167) cycles before failure, while the EXPT group sustained an average of 192 (±131) cycles (P=.930) (Figure 6A). Similarly, the CTRL group sustained an average maximum load of 733.11 (±523.67) N before failure, while the EXPT group sustained an average maximum load of 624.89 (±417.49) N (P=.518) (Figure 6B). Finally, the CTRL group experienced an average of 1123.4 (±1746.6) J of work against gravity before failure, while the EXPT group sustained an average of 1063.6 (±1246.4) J (P=1.00) (Figure 6C). Fracture displacement of greater than 3 mm was observed with loads as small as 1.1 kg.
Box and whisker plots showing the interquartile distribution of cycles sustained before failure (A), maximum load sustained by the triceps tendon before failure (B), and work performed against gravity before failure (C). There were no significant differences between groups. Abbreviations: CTRL, standard surgical technique group; EXPT, additional retrograde screw group.
No significant differences were found between groups when measuring bone fragment volume, tissue mineral density, bone mineral density, or pull-out strength. The CTRL group had an average fragment volume of 6101.8 (±1570.6) mm3, while the EXPT group averaged 6476.6 (±1367.2) mm3 (P=.597). The CTRL group had average tissue mineral density and bone mineral density of 733.7 (±34.6) and 193.5 (±70.4) milligrams of hydroxyapatite per cubic centimeter, respectively, while the EXPT group had 738.6 (±70.4) and 196.3 (±122.8) milli-grams of hydroxyapatite per cubic centimeter, respectively (P=.724 and P=.930). Finally, maximum pull-out loads were 461.1 (±160.1) N and 392.9 (±90.9) N for the CTRL and EXPT groups, respectively (P=.288).
The findings from this study suggest that the use of an additional nonlocking screw successfully arrests displacements of the proximal ulna segment during loading. However, this added stability does not result in improved fatigue life and ultimately changes the failure mechanism from slipping to catastrophic failure. This is contrary to the authors' initial hypothesis: the position and vector of an extra screw would provide improved fatigue life to external loads. Based on this result, it is suggested that the clinical application of an additional 3.5-mm retrograde screw into the proximal ulna fragment be avoided.
The results of this study are somewhat similar to those of previously published studies. The plates tested by Edwards et al20 sustained an average of 187 cycles before failure, whereas the implants tested in the CTRL and EXPT groups sustained an average of 200 and 192 cycles, respectively. The failure mechanisms from the CTRL group closely matched those of an experiment performed by Buijze et al19 in which fragment displacement of 2 mm always preceded catastrophic failure. Interestingly, all specimens experienced catastrophic failure prior to 3 mm of fragment diastasis in the study by Edwards et al.20 Specimens tested by Wellman et al21 showed a tendency to exhibit slipping behaviors between 0 and 3 mm, but these tests were not run to failure. It is unclear if a 1-mm difference in the threshold of fragment displacement would make a difference in the results of the previous studies. In this study, reducing the threshold to 2 mm of fragment diastasis would have minimal impact on results. In general, the fragments were securely held in place immediately prior to a slipping event, where diastasis increased from near 0 to a value exceeding 3 mm during the course of 3 to 10 cycles.
Comparisons of maximum load and work against gravity are rational and necessary improvements of the previously published studies that have used this protocol.20,31 This opinion is supported for several reasons. Simply counting the number of cycles to failure does not accurately account for the degree to which the repair construct is burdened in this experimental setup. More specifically, the maximum loads applied to the triceps and total amount of work performed have linear discontinuities that occur every 30 cycles. Reporting cycles to failure can also be misleading because the moment arm of the triceps tendon varies across specimens. The measurement of maximum load applied to the triceps tendon provides a better and more direct insight into causes of failure. Calculation of work against gravity also accounts for these small changes. For example, small but substantial differences in elbow flexion angle were observed between specimens; this was likely due to variations in anatomy. Changes in range of motion within specimens were also observed throughout the course of testing. These changes were caused by stretching of the triceps tendon, which seems unavoidable but should be accounted for. Although comparisons of maximum load and work did not detect any significant differences between groups in this case, it may be worthwhile to perform this measure in other experiments.
This study had several limitations. As with any cadaveric assay, the model represents fixation immediately after surgery. Therefore, the experiment provides insight into fixation strength only immediately after operative intervention. In this specific case, the “day 0” time point may be of the utmost importance, as it also represents scenarios in which union is incomplete but external loads are applied to the joint during early rehabilitation. The protocol relies solely on the triceps tendon to create elbow extension; in reality, co-contraction of muscles occurs to control elbow flexion. There were substantial amounts of variability in the bone measures that were made after testing, but none had a strong correlation with the primary results. As is often the case with cadaveric studies, the variability in specimens may have attributed to underpowered results. Future studies could be better controlled by limiting specimen selection to a certain range of dual-energy x-ray absorptiometry T-scores.
Early motion after operative fixation of olecranon fractures is critical to the clinical success of the procedure. Unfortunately, early motion places additional stress on the fixation construct and may risk early failure, especially in patients with poor bone quality. Therefore, implant designs and surgical techniques continue to be refined in an effort to provide adequate fixation strength immediately after surgical intervention. It has been suggested that passive and assisted range of motion activities can begin anywhere between 1 day and 1 week postoperatively.18,34 In this study, it was shown that small loads applied to the forearm can elicit 3 mm of fracture displacement or instigate catastrophic failure, regardless of the fixation construct. This finding reinforces the concept that rehabilitation protocols should be extremely conservative in the elderly population. Additionally, it indicates the need for improvement in implant design.
When considering the overall results of the current study, the use of an additional nonlocking screw should be avoided in the clinic. Although it seems to do a good job arresting the relative motion of bone fragments, an additional 3.5-mm diameter screw in the already “high traffic” proximal ulnar bone fragment creates an additional stress riser on an already fragile and small bone. This results in an undesirable failure mechanism. These results provide a step forward toward future research questions that may lead to improved implant design. Although these studies are ongoing, olecranon fractures in the elderly should be addressed with fixation that omits the additional nonlocking screw, longer periods of immobilization should be used, and conservative rehabilitation protocols should be employed.
- Rommens PM, Küchle R, Schneider RU, Reuter M. Olecranon fractures in adults: factors influencing outcome. Injury. 2004;35(11):1149–1157. doi:10.1016/j.injury.2003.12.002 [CrossRef]
- Powell AJ, Farhan-Alanie OM, Bryceland JK, Nunn T. The treatment of olecranon fractures in adults. Musculoskelet Surg. 2017;101(1):1–9. doi:10.1007/s12306-016-0449-5 [CrossRef]
- Lee DH. Treatment options for complex elbow fracture dislocations. Injury. 2001;32(suppl 4):SD41–SD69. doi:10.1016/S0020-1383(01)00160-7 [CrossRef]
- Wiegand L, Bernstein J, Ahn J. Fractures in brief: olecranon fractures. Clin Orthop Relat Res. 2012;470(12):3637–3641. doi:10.1007/s11999-012-2393-5 [CrossRef]
- Chalidis BE, Sachinis NC, Samoladas EP, Dimitriou CG, Pournaras JD. Is tension band wiring technique the “gold standard” for the treatment of olecranon fractures? A long term functional outcome study. J Orthop Surg Res. 2008;3:9. doi:10.1186/1749-799X-3-9 [CrossRef]
- Helm RH, Hornby R, Miller SW. The complications of surgical treatment of displaced fractures of the olecranon. Injury. 1987;18(1):48–50. doi:10.1016/0020-1383(87)90386-X [CrossRef]
- Macko D, Szabo RM. Complications of tension-band wiring of olecranon fractures. J Bone Joint Surg Am. 1985;67(9):1396–1401. doi:10.2106/00004623-198567090-00015 [CrossRef]
- Schneider MM, Nowak TE, Bastian L, et al. Tension band wiring in olecranon fractures: the myth of technical simplicity and osteosynthetical perfection. Int Orthop. 2014;38(4):847–855. doi:10.1007/s00264-013-2208-7 [CrossRef]
- Willinger L, Lucke M, Crönlein M, Sandmann GH, Biberthaler P, Siebenlist S. Malpositioned olecranon fracture tension-band wiring results in proximal radioulnar synostosis. Eur J Med Res. 2015;20:87. doi:10.1186/s40001-015-0184-7 [CrossRef]
- Ring D, Jupiter JB. Fracture-dislocation of the elbow. J Bone Joint Surg Am. 1998;80(4):566–580. doi:10.2106/00004623-199804000-00014 [CrossRef]
- Sahajpal D, Wright TW. Proximal ulna fractures. J Hand Surg Am. 2009;34(2):357–362. doi:10.1016/j.jhsa.2008.12.022 [CrossRef]
- Buijze G, Kloen P. Clinical evaluation of locking compression plate fixation for comminuted olecranon fractures. J Bone Joint Surg Am. 2009;91(10):2416–2420. doi:10.2106/JBJS.H.01419 [CrossRef]
- Bailey CS, MacDermid J, Patterson SD, King GJ. Outcome of plate fixation of olecranon fractures. J Orthop Trauma. 2001;15(8):542–548. doi:10.1097/00005131-200111000-00002 [CrossRef]
- Anderson ML, Larson AN, Merten SM, Steinmann SP. Congruent elbow plate fixation of olecranon fractures. J Orthop Trauma. 2007;21(6):386–393. doi:10.1097/BOT.0b013e3180ce831e [CrossRef]
- Lindenhovius AL, Brouwer KM, Doornberg JN, Ring DC, Kloen P. Long-term outcome of operatively treated fracture-dislocations of the olecranon. J Orthop Trauma. 2008;22(5):325–331. doi:10.1097/BOT.0b013e31817283f7 [CrossRef]
- Hewins EA, Gofton WT, Dubberly J, MacDermid JC, Faber KJ, King GJ. Plate fixation of olecranon osteotomies. J Orthop Trauma. 2007;21(1):58–62. doi:10.1097/01.bot.0000246467.32574.fe [CrossRef]
- Fyfe IS, Mossad MM, Holdsworth BJ. Methods of fixation of olecranon fractures: an experimental mechanical study. J Bone Joint Surg Br. 1985;67(3):367–372. doi:10.1302/0301-620X.67B3.3997942 [CrossRef]
- Newman SD, Mauffrey C, Krikler S. Olecranon fractures. Injury. 2009;40(6):575–581. doi:10.1016/j.injury.2008.12.013 [CrossRef]
- Buijze GA, Blankevoort L, Tuijthof GJ, Sierevelt IN, Kloen P. Biomechanical evaluation of fixation of comminuted olecranon fractures: one-third tubular versus locking compression plating. Arch Orthop Trauma Surg. 2010;130(4):459–464. doi:10.1007/s00402-009-0980-z [CrossRef]
- Edwards SG, Martin BD, Fu RH, et al. Comparison of olecranon plate fixation in osteoporotic bone: do current technologies and designs make a difference?J Orthop Trauma.2011;25(5):306–311. doi:10.1097/BOT.0b013e3181f22465 [CrossRef]
- Wellman DS, Tucker SM, Baxter JR, et al. Comminuted olecranon fractures: biomechanical testing of locked versus minifragment non-locked plate fixation. Arch Orthop Trauma Surg. 2017;137(9):1173–1179. doi:10.1007/s00402-017-2735-6 [CrossRef]
- Wagner FC, Konstantinidis L, Hohloch N, Hohloch L, Suedkamp NP, Reising K. Biomechanical evaluation of two innovative locking implants for comminuted olecranon fractures under high-cycle loading conditions. Injury. 2015;46(6):985–989. doi:10.1016/j.injury.2015.02.010 [CrossRef]
- Chao EY, Inoue N, Koo TK, Kim YH. Biomechanical considerations of fracture treatment and bone quality maintenance in elderly patients and patients with osteoporosis. Clin Orthop Relat Res. 2004;425:12–25. doi:10.1097/01.blo.0000132263.14046.0c [CrossRef]
- Veillette CJH, Steinmann SP. Olecranon fractures. Orthop Clin North Am. 2008;39(2):229–236. doi:10.1016/j.ocl.2008.01.002 [CrossRef]
- Strømsøe K. Fracture fixation problems in osteoporosis. Injury. 2004;35(2):107–113. doi:10.1016/j.injury.2003.08.019 [CrossRef]
- Moroni A, Hoang-Kim A, Lio V, Giannini S. Current augmentation fixation techniques for the osteoporotic patient. Scand J Surg. 2006;95(2):103–109. doi:10.1177/145749690609500205 [CrossRef]
- Giannoudis P, Tzioupis C, Almalki T, Buckley R. Fracture healing in osteoporotic fractures: is it really different? A basic science perspective. Injury. 2007;38(suppl 1):S90–S99. doi:10.1016/j.injury.2007.02.014 [CrossRef]
- Tarantino U, Cerocchi I, Scialdoni A, et al. Bone healing and osteoporosis. Aging Clin Exp Res. 2011;23(2)(suppl):62–64.
- Scharplatz D, Allgöwer M. Fracture-dislocations of the elbow. Injury. 1975;7(2):143–159. doi:10.1016/0020-1383(75)90012-1 [CrossRef]
- Cabanela ME, Morrey BF. Fractures of the proximal ulna and olecranon. In: Morrey BF, ed. The Elbow and Its Disorders. 2nd ed. Philadelphia, PA: WB Saunders; 1993:405–428.
- Hutchinson DT, Horwitz DS, Ha G, Thomas CW, Bachus KN. Cyclic loading of olecranon fracture fixation constructs. J Bone Joint Surg Am. 2003;85(5):831–837. doi:10.2106/00004623-200305000-00010 [CrossRef]
- Shi D, Wang D, Wang C, Liu A. A novel, inexpensive and easy to use tendon clamp for in vitro biomechanical testing. Med Eng Phys. 2012;34(4):516–520. doi:10.1016/j.medengphy.2011.11.019 [CrossRef]
- ASTM International. ASTM F543-13e1: Standard Specification and Test Methods for Metallic Medical Bone Screws. West Conshohocken, PA: ASTM International; 2013.
- Baecher N, Edwards S. Olecranon fractures. J Hand Surg Am. 2013;38(3):593–604. doi:10.1016/j.jhsa.2012.12.036 [CrossRef]