- Email: [email protected]
Biomechanical comparison of two different periarticular plating systems for stabilization of complex distal humerus fractures
Clinical Biomechanics 21 (2006) 950–955 www.elsevier.com/locate/clinbiomech
Biomechanical comparison of two different periarticular plating systems for stabilization of complex distal humerus fractures q Alexandra Schwartz a, Richard Oka b, Tim Odell b, Andrew Mahar b
a,b,*
a Department of Orthopaedic Surgery, University of California, San Diego, San Diego, CA, United States Orthopaedic Biomechanics Research Center, Department of Orthopaedics, Children’s Hospital, San Diego, CA 92123-4282, United States
Received 16 February 2006; accepted 25 April 2006
Abstract Background. Complex intra-articular distal humerus fractures are relatively uncommon injuries but are fraught with poor outcomes such as malunion, elbow stiffness and deformity. Various types of internal fixation screw-plate constructs have been developed to improve fixation. Specifically, a 90 offset periarticular system lowers the profile on the lateral epicondyle, yet it is unclear how this design compares to other plate constructs. This study compared the mechanical stiffness and plate surface strains between two types of constructs for stabilization of complex distal humerus fractures. Methods. Identical bi-columnar segmental intra-articular fractures were created in ten epoxy composite left humeri. Models were randomly assigned to two groups (n = 5/group) with either parallel plates or perpendicular plates. Rosette strain gages were placed at the most distal possible space on the lateral plate for both constructs. Models were mechanically tested with estimates of physiologic loads in flexion, extension, varus, valgus axial compression and axial torsion. Data for mechanical stiffness, transverse plate strain and longitudinal plate strain were compared with a one-way ANOVA (P < 0.05). Findings. There was no statistical difference in stiffness in any direction. The longitudinal strain for the 90 construct was significantly lower in axial compression. The 180 system demonstrated significantly lower transverse strains during axial torsion. Interpretation. Both systems demonstrated similar mechanical stiffness theoretically providing similar fracture stabilization. Plate strain differences may affect fragment position, but it is unclear how much plate loading occurs in vivo. Surgeon experience and preference may dictate the choice of a plate construct for this fracture configuration. 2006 Elsevier Ltd. All rights reserved. Keywords: Distal humerus fractures; Periarticular plating; Plate surface strains; Biomechanical stability
1. Introduction Intra-articular fractures of the distal humerus typically involve metaphyseal comminution and articular disruption. While relatively uncommon among joint injuries, complications associated with malunion, elbow stiffness,
q All devices used in this study have been approved by the FDA for this indication. * Corresponding author. Address: Orthopaedic Biomechanics Research Center, Department of Orthopaedics, Children’s Hospital, San Diego, CA 92123-4282, United States. E-mail address: [email protected] (A. Mahar).
0268-0033/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2006.04.018
and subsequent decreased function are still relatively common (Gabel et al., 1987; Henley et al., 1987; Holdsworth and Mossad, 1990; Sanders and Sackett, 1990; Waddell et al., 1988). Fixation of these fractures remains a challenge due to the restricted space for instrumentation at the distal segment, the proximity to nerves of the upper extremity, and the need to maintain repair integrity under a large range of motion and low to moderate loading (O’Driscoll, 2005). Open reduction and internal fixation with dual plating systems is the gold standard for fixation of intra-articular distal humerus fractures. This technique affords the most stable fixation and permits early rehabilitation (Huang et al., 2005; Ring et al., 2003; Soon et al., 2004; Wildburger et al., 1991).
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
Typically, these plates involve a medial reconstruction plate with a lateral column plate. Intra-operative contouring of both the medial and posterolateral plates is time consuming and may not result in optimal fit to the complex anatomy of the distal humerus. The recent introduction of 90 offset periarticular plates presents the surgeon with the option of using pre-contoured plates. The theoretical advantages of periarticular plates include reduce operative time associated with contouring the plates and a greater surface area distally to allow for greater fragment fixation. This system has been shown to produce adequate to excellent clinical results (Yang et al., 2003). Historically, two plate constructs at right angles have been considered to provide optimal rigid stabilization of the fracture (Helfet and Hotchkiss, 1990; Jupiter et al., 1985; Kirk et al., 1990) while more recent biomechanical studies have indicated that opposing plates (180 offset) provided greater stability (Jacobson et al., 1997; Schemitsch et al., 1994; Self et al., 1995). In those studies, the 180 orientation generally provided greater stiffness in torsion, axial compression, lateral bending and flexion/extension. However, the physical appearance/location of the 180 plating system in these studies look more like a lateral ‘‘J’’ shaped plate. This ‘‘J’’ shape begins posteriorly in the diaphysis but rotates laterally in the distal humerus giving the appearance of a 180 offset. While the plate system is described as 180 offset, the design is very similar to the periarticular system that is considered a 90 offset. Recently, the 90 offset method was evaluated compared to a dorsal only orientation of plates (Korner et al., 2004). This study found the 90 orientation to have improved biomechanical stabilization compared to the dorsally oriented plates regardless of plate design. Each of these previous studies examined construct stiffness as their primary dependent variable. While this provides meaningful information regarding overall behavior, maintenance of the articular surface with multiple comminuted fragments may also be affected by plate deformations due to physiologic loading. Relatively small amounts of plate deformation may cause the inserted screws to follow the path of plate deformation. Thus, plate deformations may influence the possibility of loosening and subsequent fracture fragment malunion or nonunion (due to screw migration). This phenomenon, of the failure mode of screws backing out of the lateral column, has been documented in previous biomechanical studies (Self et al., 1995). With relatively limited biomechanical data for periarticular plating of the distal humerus, it remains unclear as to whether perpendicular (90) or parallel (180) periarticular plating systems vary in biomechanical performance. The hypothesis to be tested was that the different constructs would indeed demonstrate differences in biomechanical performance. To test this hypothesis, the purpose of this study was to determine the biomechanical stability and plate strains associated with two different types of periarticular plate constructs for stabilization of complex segmental intra-articular fractures of the distal humerus.
951
2. Methods Identical bi-columnar segmental intra-articular fractures were created in ten epoxy composite left humeri (Pacific Research Laboratories, Vashon Island, WA, USA) using a custom cutting jig with a transverse cut 30 mm above the articular surface (Fig. 1). Provisional fixation was first obtained using bone clamps followed by multiple 0.063 Kirschner wires. For proximal fixation, six total screws were used for purchase in the humeral diaphysis for both systems. Three screws were delivered in uni-cortical fashion for both medial and lateral plates. To maintain consistent numbers of screws utilized, five screws were used to fix the distal fragments for both systems. Due to plate design differences, screws through the distal aspect of the lateral plate differed accordingly. Based on the fracture pattern, the lateral plate of the 180 plating construct (Acumed, Inc., Hillsboro, OR, USA) allowed three distal lateral to medial screws across the trochlea to the medial column (Fig. 2). The posterolateral plate of the 90 plating construct (Zimmer, Inc, Warsaw, IN, USA) had six potential screw holes, but to achieve similar screw constructs, only three of these holes were filled with lateral to medial screws (Fig. 3). Lateral plates for both orientations were then instrumented with single T-rosette 120ohm strain gages (Part #: FCA-1-17-1LT, Tokyo Sokki Kenkyujo, Tokyo, Japan) with a reported sensitivity of 0.1%. The purpose for using rosette gages was to simultaneously capture plate deformations in two orthogonal clinically relevant directions. Gages were aligned on each plate to allow recording of surface strain in both the longitudinal and transverse directions relative to the humerus. Gages were applied as close to the distal end of the plate as possible (1.5 cm for 90
Fig. 1. The simulated complex intra-articular fracture.
952
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
Fig. 2. Position/orientation of the 180 plate system.
direction of testing. This ‘‘joint’’ did not include joint capsule or ligamentous structures. For flexion and extension testing, specimens were mounted horizontally in an MTS 858 bi-axial testing machine. (Eden Prairie, MN, USA) The specimens were linked to the piston actuator with the custom designed ulno-humeral joint (Fig. 4). The piston applied ±100 N at 0.5 mm/s for 20 cycles. An identical procedure was used for varus/valgus stiffness with the humerus rotated 90. For axial compression and torsion testing, the specimen was mounted parallel to the loading axis of the piston. For torsion, axial moments were applied for 20 cycles between ±2 N m at 0.5 deg/s using the same custom joint. In compression, the piston applied compressive loads from 10 to 100 N at 0.1 mm/s for 20 cycles. Loading across the elbow joint is complex and not well documented. Thus, the loading parameters described above were taken from previous biomechanical studies on distal humerus fixation (An et al., 1989; Korner et al., 2004; Schemitsch et al., 1994). Multiple directions of loading were selected for potential clinical relevance and because testing in multiple planes has been emphasized in the literature for this type of fracture pattern (Schemitsch et al., 1994). Data for displacement (mm), force (N), angle (radians), torque (N m), and longitudinal/transverse plate strain (microstrain) were collected at 10 Hz for the duration of each test (Fig. 5). Construct stiffness (N/mm or N m/deg) was calculated over a the linear portion of each curve (10–100 N for force, 0.2–2 N m for torque). Construct stiffness and strains in both directions for each test were
Fig. 3. Position/orientation of the 90 plate system.
system, 2.5 cm for 180 system) but identical locations were prohibited due to plate design differences and sufficient area (3 mm · 3 mm) to place the gage. Each gage was connected in a custom Wheatstone 1/2 bridge configuration and powered/amplified using standard equipment (Sensotec, Columbus, OH, USA). The cut proximal diaphysis of each humerus was then potted in a cylindrical mold using two part epoxy resin (Bondo Marhyde, Atlanta, GA, USA) for rigid fixation within the testing rig. Prior to testing, a custom ulno-humeral interface was created using epoxy resin by using the distal humerus as a mold. The mold was then refined to simulate the in vivo position of the ulna relative to the distal humerus and this relationship was then fixed for each
Fig. 4. Example figure of setup for flexion/extension testing. The custom ulno-humeral joint, made of two-part epoxy resin, conforms directly to the distal humerus to apply only flexion/extension of the fixed fragments by the changing the elevation of the actuator assembly. A similar mechanical setup was used to test only in varus/valgus, axial compression and axial torsion directions.
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
953
Table 3 Transverse plate strain during physiologic loading (mean (SD)) 180 Plates 6
Flex/Ext (strain * 10 ) Torsion (strain * 10 6) Axial (strain * 10 6) Varus/Valgus (strain * 10 6)
25.1 17.3 17.6 18.7
(2.4) (4.6) (6.9) (8.9)
90 Plates 23.3 30.3 12.5 21.4
(18.9) (9.4) (11.8) (11.7)
P-Value 0.84 0.04 0.45 0.73
sion (P < 0.05). For transverse strain, there were no differences between systems for flexion/extension, axial compression or varus/valgus. However, the 180 system demonstrated significantly lower transverse strains in torsion compared to the 90 construct (P = 0.04) (Table 3). Fig. 5. Example data set from the axial compression testing. The longitudinal strain appears in phase with the applied load while the transverse strain appears out of phase. The displacement fluctuation for this particular test varied between 0.1 mm and 0.45 mm. Note: the y-axis units were labeled as such to prevent confusion of the three types of data (displacement, force, microstrain) presented in the figure.
evaluated using a one-way ANOVA between 180 and 90 constructs (P < 0.05) (Statsoft, Inc., Tulsa, OK, USA). 3. Results Implants and models were carefully inspected before and after testing and there was no evidence of implant loosening or plastic deformation of the plates. There did not appear to be deformation within the fixation construct as well. There were no statistical differences in physiologic stiffness between plating systems across tests (Table 1). While not significantly different, the 90 system demonstrated greater stiffness magnitudes in each direction of testing except in varus/valgus. There were no differences in longitudinal strains between constructs for torsion (P = 0.9), varus/valgus (P = 0.8) or flexion/extension (P = 0.1) (Table 2). However, the longitudinal strains for the 90 construct were significantly lower in axial compres-
Table 1 Construct stiffness during physiologic loading (mean (SD))
Flex/Ext (N/mm) Torsion (N m/rad) Axial (N/mm) Varus/Valgus (N/mm)
180 Plates
90 Plates
P-Value
37.8 (3.2) 37.9 (13.9) 272.4 (37.0) 68.4 (12.9)
42.1 (4.8) 41.5 (3.4) 413.3 (173.3) 65.2 (13.4)
0.11 0.58 0.11 0.72
Table 2 Longitudinal plate strain during physiologic loading (mean (SD)) 180 Plates 6
Flex/Ext (strain * 10 ) Torsion (strain * 10 6) Axial (strain * 10 6) Varus/Valgus (strain * 10 6)
45.8 33.1 38.6 26.3
(9.5) (7.1) (20.9) (10.5)
90 Plates 32.6 34.5 15.4 28.8
(15.1) (14.1) (6.3) (15.3)
P-Value 0.14 0.86 0.048 0.78
4. Discussion Intra-articular fractures of the distal humerus remain a challenging surgical situation (Ring et al., 2003; Sanders and Sackett, 1990; Soon et al., 2004). While generally positive outcomes have been reported with open reduction and internal fixation, relatively high rates of malunion, joint stiffness and reduced range of motion have been reported (Gabel et al., 1987; Henley et al., 1987; Holdsworth and Mossad, 1990; Letsch et al., 1989). While biomechanical studies have evaluated the efficacy of different types of plating systems and plate orientation (Helfet and Hotchkiss, 1990; Jacobson et al., 1997; Jupiter et al., 1985; Kirk et al., 1990; Korner et al., 2004; Schemitsch et al., 1994; Self et al., 1995), there remains no consensus as to the construct that affords the greatest stability with the potential for early active and passive rehabilitation. In the current study, the 90 plating system had an equivalent stiffness to the 180 plating system across all tests. The 90 plating system had significantly lower longitudinal strains during axial compressive loading while the 180 plating system had significantly lower transverse strains in axial torsion. Due to the anatomy and biomechanics of the elbow joint, axial compressive load may be a more important structural variable than axial torsion. This may be of additional clinical importance in that typical post-operative rehabilitation protocols primarily use flexion/extension motions that would impart joint compressive loads. The stiffness data reported in the current study are lower in magnitude than those reported previously (Schemitsch et al., 1994). However, the differences in stiffness results may be due to differences in loading magnitudes between studies. Also, other studies have utilized a maximum number of screws for each type of construct while the current study utilized an identical number of screws between systems. Another objective of the current study was to evaluate the lateral plate deformations associated with physiologic loading. While stiffness data provides information about the overall construct, an analysis of plate deformation may yield valuable information regarding the potential
954
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
for changes in plate-screw orientation and thus the possibility of fragment migration resulting in malunion. This type of failure mode has been previously described following biomechanical testing (Self et al., 1995). Data for longitudinal strains demonstrated similar trends to the stiffness data. The torsion and varus/valgus strain data were not different between systems. However, flexion/extension strain were much lower in magnitude and the axial compression strain was statistically less for the 90 system. This may indicate better maintenance of fragment position during those modes of loading. For transverse strains, there were no differences in flexion/extension, varus/valgus or axial compression testing. However, transverse strains during torsion were significantly less for the 180 system. This result is not surprising since the 180 system produces moment arms at a greater distance from the long axis of the humerus and should thus have improved resistance to torsional loads. As with any study, there may be some potential deficiencies to the current study. Synthetic models were used to eliminate problems associated with cadaveric bone quality and anatomical variability. An identical number of screws were used for each system and it may be the clinical scenario that all available screw holes would be utilized to maximize stabilization. As the plate systems differ in terms of the number of holes available for lateral fixation of the distal fracture fragments, it was felt that differing the number of screws for fixation would potentially favor one type of instrumentation. The authors felt that a side-by-side comparison eliminating the number of screws as a potential variable would best allow for evaluation of plate design and material. It should also be noted that there were material differences between the systems. The 180 offset system is made of slightly thicker titanium alloy while the 90 offset system is made from a thinner stainless steel. While stainless steel has approximately twice the modulus of elasticity as titanium, these material properties were not part of the analysis as the actual magnitude of plate deformation measured as microstrain was the variable of primary importance. The strain gages were placed as distal as possible for each construct but, due to differing plate designs, experienced a 10 mm offset between constructs and this could certainly affect the measured strain. As there was little opportunity for gage placement elsewhere on the distal aspect of the plate, the authors felt that an evaluation of the measured plate strains at the two different points may still assist with an understanding of the potential for deformation and screw migration. Finally, the sample size (n = 5/group) may potentially reduce the power of the statistical test, although this sample size has previously been used successfully to test fracture fixation in synthetic models (Fricka et al., 2004). A follow up power analysis on the data that approached or attained significant differences demonstrated power values greater than 0.6. Thus, these statistical comparisons were considered valid, although definitive clinical conclusions should be guarded.
5. Conclusions Both systems appear to provide similar stabilization for this complex intra-articular fracture pattern under physiologic loads. From a clinical perspective, a system that provides greater stability during physiologic motion with more distal fixation should decrease the likelihood of fixation failure and thereby decrease the chances for malunion or nonunion. As failure testing was not within the scope of the current study, further efforts to understand failure mechanisms in the in vitro situation are warranted. From these data, the author’s feel that implant selection should still be based first on the complexity of the fracture pattern and then on surgeon preference and/or experience. Acknowledgement This study was supported in part by a research grant from Zimmer, Inc. (Warsaw, IN, USA). References An, K.N., Kaufman, K.R., Chao, E.Y., 1989. Physiological considerations of muscle force through the elbow joint. J. Biomech. 22, 1249– 1256. Fricka, K., Mahar, A.T., Lee, S.S., Newton, P.O., 2004. Biomechanical analysis of antegrade and retrograde flexible intramedullary nail fixation of pediatric femoral fractures using a synthetic bone model. J. Ped. Orthop. 24, 167–171. Gabel, G.T., Hanson, G., Bennett, J.B., Noble, P.C., Tullos, H.S., 1987. Intraarticular fractures of the distal humerus in the adult. Clin. Orthop. Relat. Res. 216, 99–108. Helfet, D.L., Hotchkiss, R.N., 1990. Internal fixation of the distal humerus: a biomechanical comparison of methods. J. Orthop. Trauma 4, 260–264. Henley, M.B., Bone, L.B., Parker, B., 1987. Operative management of intra-articular fractures of the distal humerus. J. Orthop. Trauma 1, 24–35. Holdsworth, B.J., Mossad, M.M., 1990. Fractures of the adult distal humerus. Elbow function after internal fixation. J. Bone Joint Surg. Br. 72, 362–365. Huang, T.L., Chiu, F.Y., Chuang, T.Y., Chen, T.H., 2005. The results of open reduction and internal fixation in elderly patients with severe fractures of the distal humerus: a critical analysis of the results. J. Trauma 58, 62–69. Jacobson, S.R., Glisson, R.R., Urbaniak, J.R., 1997. Comparison of distal humerus fracture fixation: a biomechanical study. J. South. Orthop. Assoc. 6, 241–249. Jupiter, J.B., Neff, U., Holzach, P., Allgower, M., 1985. Intercondylar fractures of the humerus: an operative approach. J. Bone Joint Surg. Am. 67, 226–239. Kirk, P., Goulet, J.A., Freiberg, A., Goldstein, S., 1990. A biomechanical evaluation of fixation methods for fractures of the distal humerus. Orthop. Trans. 14, 674. Korner, J., Diederichs, G., Arzdorf, M., Lill, H., Josten, C., Schneider, E., Linke, B., 2004. A biomechanical evaluation of methods of distal humerus fracture fixation using locking compression plates versus conventional reconstruction plates. J. Orthop. Trauma 18, 286–293. Letsch, R., Schmit-Neuerburg, K.P., Sturmer, K.M., Walz, M., 1989. Intraarticular fractures of the distal humerus: surgical treatment and results. Clin. Orthop. Relat. Res. 241, 238–244. O’Driscoll, S.W., 2005. Optimizing stability in distal humeral fracture fixation. J. Shoulder Elbow Surg. 14, 186S–194S.
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955 Ring, D., Jupiter, J.B., Gulotta, L., 2003. Articular fractures of the distal part of the humerus. J. Bone Joint Surg. Am. A 85, 232–238. Sanders, R.A., Sackett, J.R., 1990. Open reduction and internal fixation of delayed union and nonunion of the distal humerus. J. Orthop. Trauma 4, 254–259. Schemitsch, E.H., Tencer, A.F., Henley, M.B., 1994. Biomechanical evaluation of methods of internal fixation of the distal humerus. J. Orthop. Trauma 8, 468–475. Self, J., Viegas, S.F., Buford Jr., W.L., Patterson, R.M., 1995. A comparison of double-plate fixation methods for complex distal humerus fractures. J. Shoulder Elbow Surg. 4, 10–16.
955
Soon, J.L., Chan, B.K., Low, C.O., 2004. Surgical fixation of intraarticular fractures of the distal humerus in adults. Injury 35, 44–54. Waddell, J.P., Hatch, J., Richards, R., 1988. Supracondylar fractures of the humerus – results of surgical treatment. J. Trauma 28, 1615– 1621. Wildburger, R., Mahring, M., Hofer, H.P., 1991. Supraintercondylar fractures of the distal humerus: results of internal fixation. J. Orthop. Trauma 5, 301–307. Yang, K.H., Park, K.W., Jung, S.H., 2003. Lateral J-plate fixation in comminuted intercondylar fracture of the humerus. Arch. Orthop. Trauma Surg. 123, 234–238.
Biomechanical comparison of two different periarticular plating systems for stabilization of complex distal humerus fractures q Alexandra Schwartz a, Richard Oka b, Tim Odell b, Andrew Mahar b
a,b,*
a Department of Orthopaedic Surgery, University of California, San Diego, San Diego, CA, United States Orthopaedic Biomechanics Research Center, Department of Orthopaedics, Children’s Hospital, San Diego, CA 92123-4282, United States
Received 16 February 2006; accepted 25 April 2006
Abstract Background. Complex intra-articular distal humerus fractures are relatively uncommon injuries but are fraught with poor outcomes such as malunion, elbow stiffness and deformity. Various types of internal fixation screw-plate constructs have been developed to improve fixation. Specifically, a 90 offset periarticular system lowers the profile on the lateral epicondyle, yet it is unclear how this design compares to other plate constructs. This study compared the mechanical stiffness and plate surface strains between two types of constructs for stabilization of complex distal humerus fractures. Methods. Identical bi-columnar segmental intra-articular fractures were created in ten epoxy composite left humeri. Models were randomly assigned to two groups (n = 5/group) with either parallel plates or perpendicular plates. Rosette strain gages were placed at the most distal possible space on the lateral plate for both constructs. Models were mechanically tested with estimates of physiologic loads in flexion, extension, varus, valgus axial compression and axial torsion. Data for mechanical stiffness, transverse plate strain and longitudinal plate strain were compared with a one-way ANOVA (P < 0.05). Findings. There was no statistical difference in stiffness in any direction. The longitudinal strain for the 90 construct was significantly lower in axial compression. The 180 system demonstrated significantly lower transverse strains during axial torsion. Interpretation. Both systems demonstrated similar mechanical stiffness theoretically providing similar fracture stabilization. Plate strain differences may affect fragment position, but it is unclear how much plate loading occurs in vivo. Surgeon experience and preference may dictate the choice of a plate construct for this fracture configuration. 2006 Elsevier Ltd. All rights reserved. Keywords: Distal humerus fractures; Periarticular plating; Plate surface strains; Biomechanical stability
1. Introduction Intra-articular fractures of the distal humerus typically involve metaphyseal comminution and articular disruption. While relatively uncommon among joint injuries, complications associated with malunion, elbow stiffness,
q All devices used in this study have been approved by the FDA for this indication. * Corresponding author. Address: Orthopaedic Biomechanics Research Center, Department of Orthopaedics, Children’s Hospital, San Diego, CA 92123-4282, United States. E-mail address: [email protected] (A. Mahar).
0268-0033/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2006.04.018
and subsequent decreased function are still relatively common (Gabel et al., 1987; Henley et al., 1987; Holdsworth and Mossad, 1990; Sanders and Sackett, 1990; Waddell et al., 1988). Fixation of these fractures remains a challenge due to the restricted space for instrumentation at the distal segment, the proximity to nerves of the upper extremity, and the need to maintain repair integrity under a large range of motion and low to moderate loading (O’Driscoll, 2005). Open reduction and internal fixation with dual plating systems is the gold standard for fixation of intra-articular distal humerus fractures. This technique affords the most stable fixation and permits early rehabilitation (Huang et al., 2005; Ring et al., 2003; Soon et al., 2004; Wildburger et al., 1991).
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
Typically, these plates involve a medial reconstruction plate with a lateral column plate. Intra-operative contouring of both the medial and posterolateral plates is time consuming and may not result in optimal fit to the complex anatomy of the distal humerus. The recent introduction of 90 offset periarticular plates presents the surgeon with the option of using pre-contoured plates. The theoretical advantages of periarticular plates include reduce operative time associated with contouring the plates and a greater surface area distally to allow for greater fragment fixation. This system has been shown to produce adequate to excellent clinical results (Yang et al., 2003). Historically, two plate constructs at right angles have been considered to provide optimal rigid stabilization of the fracture (Helfet and Hotchkiss, 1990; Jupiter et al., 1985; Kirk et al., 1990) while more recent biomechanical studies have indicated that opposing plates (180 offset) provided greater stability (Jacobson et al., 1997; Schemitsch et al., 1994; Self et al., 1995). In those studies, the 180 orientation generally provided greater stiffness in torsion, axial compression, lateral bending and flexion/extension. However, the physical appearance/location of the 180 plating system in these studies look more like a lateral ‘‘J’’ shaped plate. This ‘‘J’’ shape begins posteriorly in the diaphysis but rotates laterally in the distal humerus giving the appearance of a 180 offset. While the plate system is described as 180 offset, the design is very similar to the periarticular system that is considered a 90 offset. Recently, the 90 offset method was evaluated compared to a dorsal only orientation of plates (Korner et al., 2004). This study found the 90 orientation to have improved biomechanical stabilization compared to the dorsally oriented plates regardless of plate design. Each of these previous studies examined construct stiffness as their primary dependent variable. While this provides meaningful information regarding overall behavior, maintenance of the articular surface with multiple comminuted fragments may also be affected by plate deformations due to physiologic loading. Relatively small amounts of plate deformation may cause the inserted screws to follow the path of plate deformation. Thus, plate deformations may influence the possibility of loosening and subsequent fracture fragment malunion or nonunion (due to screw migration). This phenomenon, of the failure mode of screws backing out of the lateral column, has been documented in previous biomechanical studies (Self et al., 1995). With relatively limited biomechanical data for periarticular plating of the distal humerus, it remains unclear as to whether perpendicular (90) or parallel (180) periarticular plating systems vary in biomechanical performance. The hypothesis to be tested was that the different constructs would indeed demonstrate differences in biomechanical performance. To test this hypothesis, the purpose of this study was to determine the biomechanical stability and plate strains associated with two different types of periarticular plate constructs for stabilization of complex segmental intra-articular fractures of the distal humerus.
951
2. Methods Identical bi-columnar segmental intra-articular fractures were created in ten epoxy composite left humeri (Pacific Research Laboratories, Vashon Island, WA, USA) using a custom cutting jig with a transverse cut 30 mm above the articular surface (Fig. 1). Provisional fixation was first obtained using bone clamps followed by multiple 0.063 Kirschner wires. For proximal fixation, six total screws were used for purchase in the humeral diaphysis for both systems. Three screws were delivered in uni-cortical fashion for both medial and lateral plates. To maintain consistent numbers of screws utilized, five screws were used to fix the distal fragments for both systems. Due to plate design differences, screws through the distal aspect of the lateral plate differed accordingly. Based on the fracture pattern, the lateral plate of the 180 plating construct (Acumed, Inc., Hillsboro, OR, USA) allowed three distal lateral to medial screws across the trochlea to the medial column (Fig. 2). The posterolateral plate of the 90 plating construct (Zimmer, Inc, Warsaw, IN, USA) had six potential screw holes, but to achieve similar screw constructs, only three of these holes were filled with lateral to medial screws (Fig. 3). Lateral plates for both orientations were then instrumented with single T-rosette 120ohm strain gages (Part #: FCA-1-17-1LT, Tokyo Sokki Kenkyujo, Tokyo, Japan) with a reported sensitivity of 0.1%. The purpose for using rosette gages was to simultaneously capture plate deformations in two orthogonal clinically relevant directions. Gages were aligned on each plate to allow recording of surface strain in both the longitudinal and transverse directions relative to the humerus. Gages were applied as close to the distal end of the plate as possible (1.5 cm for 90
Fig. 1. The simulated complex intra-articular fracture.
952
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
Fig. 2. Position/orientation of the 180 plate system.
direction of testing. This ‘‘joint’’ did not include joint capsule or ligamentous structures. For flexion and extension testing, specimens were mounted horizontally in an MTS 858 bi-axial testing machine. (Eden Prairie, MN, USA) The specimens were linked to the piston actuator with the custom designed ulno-humeral joint (Fig. 4). The piston applied ±100 N at 0.5 mm/s for 20 cycles. An identical procedure was used for varus/valgus stiffness with the humerus rotated 90. For axial compression and torsion testing, the specimen was mounted parallel to the loading axis of the piston. For torsion, axial moments were applied for 20 cycles between ±2 N m at 0.5 deg/s using the same custom joint. In compression, the piston applied compressive loads from 10 to 100 N at 0.1 mm/s for 20 cycles. Loading across the elbow joint is complex and not well documented. Thus, the loading parameters described above were taken from previous biomechanical studies on distal humerus fixation (An et al., 1989; Korner et al., 2004; Schemitsch et al., 1994). Multiple directions of loading were selected for potential clinical relevance and because testing in multiple planes has been emphasized in the literature for this type of fracture pattern (Schemitsch et al., 1994). Data for displacement (mm), force (N), angle (radians), torque (N m), and longitudinal/transverse plate strain (microstrain) were collected at 10 Hz for the duration of each test (Fig. 5). Construct stiffness (N/mm or N m/deg) was calculated over a the linear portion of each curve (10–100 N for force, 0.2–2 N m for torque). Construct stiffness and strains in both directions for each test were
Fig. 3. Position/orientation of the 90 plate system.
system, 2.5 cm for 180 system) but identical locations were prohibited due to plate design differences and sufficient area (3 mm · 3 mm) to place the gage. Each gage was connected in a custom Wheatstone 1/2 bridge configuration and powered/amplified using standard equipment (Sensotec, Columbus, OH, USA). The cut proximal diaphysis of each humerus was then potted in a cylindrical mold using two part epoxy resin (Bondo Marhyde, Atlanta, GA, USA) for rigid fixation within the testing rig. Prior to testing, a custom ulno-humeral interface was created using epoxy resin by using the distal humerus as a mold. The mold was then refined to simulate the in vivo position of the ulna relative to the distal humerus and this relationship was then fixed for each
Fig. 4. Example figure of setup for flexion/extension testing. The custom ulno-humeral joint, made of two-part epoxy resin, conforms directly to the distal humerus to apply only flexion/extension of the fixed fragments by the changing the elevation of the actuator assembly. A similar mechanical setup was used to test only in varus/valgus, axial compression and axial torsion directions.
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
953
Table 3 Transverse plate strain during physiologic loading (mean (SD)) 180 Plates 6
Flex/Ext (strain * 10 ) Torsion (strain * 10 6) Axial (strain * 10 6) Varus/Valgus (strain * 10 6)
25.1 17.3 17.6 18.7
(2.4) (4.6) (6.9) (8.9)
90 Plates 23.3 30.3 12.5 21.4
(18.9) (9.4) (11.8) (11.7)
P-Value 0.84 0.04 0.45 0.73
sion (P < 0.05). For transverse strain, there were no differences between systems for flexion/extension, axial compression or varus/valgus. However, the 180 system demonstrated significantly lower transverse strains in torsion compared to the 90 construct (P = 0.04) (Table 3). Fig. 5. Example data set from the axial compression testing. The longitudinal strain appears in phase with the applied load while the transverse strain appears out of phase. The displacement fluctuation for this particular test varied between 0.1 mm and 0.45 mm. Note: the y-axis units were labeled as such to prevent confusion of the three types of data (displacement, force, microstrain) presented in the figure.
evaluated using a one-way ANOVA between 180 and 90 constructs (P < 0.05) (Statsoft, Inc., Tulsa, OK, USA). 3. Results Implants and models were carefully inspected before and after testing and there was no evidence of implant loosening or plastic deformation of the plates. There did not appear to be deformation within the fixation construct as well. There were no statistical differences in physiologic stiffness between plating systems across tests (Table 1). While not significantly different, the 90 system demonstrated greater stiffness magnitudes in each direction of testing except in varus/valgus. There were no differences in longitudinal strains between constructs for torsion (P = 0.9), varus/valgus (P = 0.8) or flexion/extension (P = 0.1) (Table 2). However, the longitudinal strains for the 90 construct were significantly lower in axial compres-
Table 1 Construct stiffness during physiologic loading (mean (SD))
Flex/Ext (N/mm) Torsion (N m/rad) Axial (N/mm) Varus/Valgus (N/mm)
180 Plates
90 Plates
P-Value
37.8 (3.2) 37.9 (13.9) 272.4 (37.0) 68.4 (12.9)
42.1 (4.8) 41.5 (3.4) 413.3 (173.3) 65.2 (13.4)
0.11 0.58 0.11 0.72
Table 2 Longitudinal plate strain during physiologic loading (mean (SD)) 180 Plates 6
Flex/Ext (strain * 10 ) Torsion (strain * 10 6) Axial (strain * 10 6) Varus/Valgus (strain * 10 6)
45.8 33.1 38.6 26.3
(9.5) (7.1) (20.9) (10.5)
90 Plates 32.6 34.5 15.4 28.8
(15.1) (14.1) (6.3) (15.3)
P-Value 0.14 0.86 0.048 0.78
4. Discussion Intra-articular fractures of the distal humerus remain a challenging surgical situation (Ring et al., 2003; Sanders and Sackett, 1990; Soon et al., 2004). While generally positive outcomes have been reported with open reduction and internal fixation, relatively high rates of malunion, joint stiffness and reduced range of motion have been reported (Gabel et al., 1987; Henley et al., 1987; Holdsworth and Mossad, 1990; Letsch et al., 1989). While biomechanical studies have evaluated the efficacy of different types of plating systems and plate orientation (Helfet and Hotchkiss, 1990; Jacobson et al., 1997; Jupiter et al., 1985; Kirk et al., 1990; Korner et al., 2004; Schemitsch et al., 1994; Self et al., 1995), there remains no consensus as to the construct that affords the greatest stability with the potential for early active and passive rehabilitation. In the current study, the 90 plating system had an equivalent stiffness to the 180 plating system across all tests. The 90 plating system had significantly lower longitudinal strains during axial compressive loading while the 180 plating system had significantly lower transverse strains in axial torsion. Due to the anatomy and biomechanics of the elbow joint, axial compressive load may be a more important structural variable than axial torsion. This may be of additional clinical importance in that typical post-operative rehabilitation protocols primarily use flexion/extension motions that would impart joint compressive loads. The stiffness data reported in the current study are lower in magnitude than those reported previously (Schemitsch et al., 1994). However, the differences in stiffness results may be due to differences in loading magnitudes between studies. Also, other studies have utilized a maximum number of screws for each type of construct while the current study utilized an identical number of screws between systems. Another objective of the current study was to evaluate the lateral plate deformations associated with physiologic loading. While stiffness data provides information about the overall construct, an analysis of plate deformation may yield valuable information regarding the potential
954
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955
for changes in plate-screw orientation and thus the possibility of fragment migration resulting in malunion. This type of failure mode has been previously described following biomechanical testing (Self et al., 1995). Data for longitudinal strains demonstrated similar trends to the stiffness data. The torsion and varus/valgus strain data were not different between systems. However, flexion/extension strain were much lower in magnitude and the axial compression strain was statistically less for the 90 system. This may indicate better maintenance of fragment position during those modes of loading. For transverse strains, there were no differences in flexion/extension, varus/valgus or axial compression testing. However, transverse strains during torsion were significantly less for the 180 system. This result is not surprising since the 180 system produces moment arms at a greater distance from the long axis of the humerus and should thus have improved resistance to torsional loads. As with any study, there may be some potential deficiencies to the current study. Synthetic models were used to eliminate problems associated with cadaveric bone quality and anatomical variability. An identical number of screws were used for each system and it may be the clinical scenario that all available screw holes would be utilized to maximize stabilization. As the plate systems differ in terms of the number of holes available for lateral fixation of the distal fracture fragments, it was felt that differing the number of screws for fixation would potentially favor one type of instrumentation. The authors felt that a side-by-side comparison eliminating the number of screws as a potential variable would best allow for evaluation of plate design and material. It should also be noted that there were material differences between the systems. The 180 offset system is made of slightly thicker titanium alloy while the 90 offset system is made from a thinner stainless steel. While stainless steel has approximately twice the modulus of elasticity as titanium, these material properties were not part of the analysis as the actual magnitude of plate deformation measured as microstrain was the variable of primary importance. The strain gages were placed as distal as possible for each construct but, due to differing plate designs, experienced a 10 mm offset between constructs and this could certainly affect the measured strain. As there was little opportunity for gage placement elsewhere on the distal aspect of the plate, the authors felt that an evaluation of the measured plate strains at the two different points may still assist with an understanding of the potential for deformation and screw migration. Finally, the sample size (n = 5/group) may potentially reduce the power of the statistical test, although this sample size has previously been used successfully to test fracture fixation in synthetic models (Fricka et al., 2004). A follow up power analysis on the data that approached or attained significant differences demonstrated power values greater than 0.6. Thus, these statistical comparisons were considered valid, although definitive clinical conclusions should be guarded.
5. Conclusions Both systems appear to provide similar stabilization for this complex intra-articular fracture pattern under physiologic loads. From a clinical perspective, a system that provides greater stability during physiologic motion with more distal fixation should decrease the likelihood of fixation failure and thereby decrease the chances for malunion or nonunion. As failure testing was not within the scope of the current study, further efforts to understand failure mechanisms in the in vitro situation are warranted. From these data, the author’s feel that implant selection should still be based first on the complexity of the fracture pattern and then on surgeon preference and/or experience. Acknowledgement This study was supported in part by a research grant from Zimmer, Inc. (Warsaw, IN, USA). References An, K.N., Kaufman, K.R., Chao, E.Y., 1989. Physiological considerations of muscle force through the elbow joint. J. Biomech. 22, 1249– 1256. Fricka, K., Mahar, A.T., Lee, S.S., Newton, P.O., 2004. Biomechanical analysis of antegrade and retrograde flexible intramedullary nail fixation of pediatric femoral fractures using a synthetic bone model. J. Ped. Orthop. 24, 167–171. Gabel, G.T., Hanson, G., Bennett, J.B., Noble, P.C., Tullos, H.S., 1987. Intraarticular fractures of the distal humerus in the adult. Clin. Orthop. Relat. Res. 216, 99–108. Helfet, D.L., Hotchkiss, R.N., 1990. Internal fixation of the distal humerus: a biomechanical comparison of methods. J. Orthop. Trauma 4, 260–264. Henley, M.B., Bone, L.B., Parker, B., 1987. Operative management of intra-articular fractures of the distal humerus. J. Orthop. Trauma 1, 24–35. Holdsworth, B.J., Mossad, M.M., 1990. Fractures of the adult distal humerus. Elbow function after internal fixation. J. Bone Joint Surg. Br. 72, 362–365. Huang, T.L., Chiu, F.Y., Chuang, T.Y., Chen, T.H., 2005. The results of open reduction and internal fixation in elderly patients with severe fractures of the distal humerus: a critical analysis of the results. J. Trauma 58, 62–69. Jacobson, S.R., Glisson, R.R., Urbaniak, J.R., 1997. Comparison of distal humerus fracture fixation: a biomechanical study. J. South. Orthop. Assoc. 6, 241–249. Jupiter, J.B., Neff, U., Holzach, P., Allgower, M., 1985. Intercondylar fractures of the humerus: an operative approach. J. Bone Joint Surg. Am. 67, 226–239. Kirk, P., Goulet, J.A., Freiberg, A., Goldstein, S., 1990. A biomechanical evaluation of fixation methods for fractures of the distal humerus. Orthop. Trans. 14, 674. Korner, J., Diederichs, G., Arzdorf, M., Lill, H., Josten, C., Schneider, E., Linke, B., 2004. A biomechanical evaluation of methods of distal humerus fracture fixation using locking compression plates versus conventional reconstruction plates. J. Orthop. Trauma 18, 286–293. Letsch, R., Schmit-Neuerburg, K.P., Sturmer, K.M., Walz, M., 1989. Intraarticular fractures of the distal humerus: surgical treatment and results. Clin. Orthop. Relat. Res. 241, 238–244. O’Driscoll, S.W., 2005. Optimizing stability in distal humeral fracture fixation. J. Shoulder Elbow Surg. 14, 186S–194S.
A. Schwartz et al. / Clinical Biomechanics 21 (2006) 950–955 Ring, D., Jupiter, J.B., Gulotta, L., 2003. Articular fractures of the distal part of the humerus. J. Bone Joint Surg. Am. A 85, 232–238. Sanders, R.A., Sackett, J.R., 1990. Open reduction and internal fixation of delayed union and nonunion of the distal humerus. J. Orthop. Trauma 4, 254–259. Schemitsch, E.H., Tencer, A.F., Henley, M.B., 1994. Biomechanical evaluation of methods of internal fixation of the distal humerus. J. Orthop. Trauma 8, 468–475. Self, J., Viegas, S.F., Buford Jr., W.L., Patterson, R.M., 1995. A comparison of double-plate fixation methods for complex distal humerus fractures. J. Shoulder Elbow Surg. 4, 10–16.
955
Soon, J.L., Chan, B.K., Low, C.O., 2004. Surgical fixation of intraarticular fractures of the distal humerus in adults. Injury 35, 44–54. Waddell, J.P., Hatch, J., Richards, R., 1988. Supracondylar fractures of the humerus – results of surgical treatment. J. Trauma 28, 1615– 1621. Wildburger, R., Mahring, M., Hofer, H.P., 1991. Supraintercondylar fractures of the distal humerus: results of internal fixation. J. Orthop. Trauma 5, 301–307. Yang, K.H., Park, K.W., Jung, S.H., 2003. Lateral J-plate fixation in comminuted intercondylar fracture of the humerus. Arch. Orthop. Trauma Surg. 123, 234–238.