• View in gallery

    Diagram illustrating the position of a specimen (humerus from a pigeon [Columbia livia] cadaver; caudal aspect) in a testing apparatus and mediolateral direction of load during 4-point bending. Dimensions of inner (14 mm) and outer (46 mm) spans are indicated.

  • View in gallery

    Photographs of examples of failure during bending of pigeon cadaver humeri for intact specimens (cranial aspect; A), specimens with an ostectomy gap repaired with a locking plate (lateral aspect; B), and specimens with an ostectomy gap repaired with a nonlocking plate (lateral aspect; C). In panel C, the specimen on the right has a comminuted fracture, whereas the specimen on the left has simple loosening of the screws as indicated by the fact that the proximal screws in the distal segment are not flush with the plate. Bar = 1 cm.

  • View in gallery

    Photographs of examples of failure of pigeon cadaver humeri during torsion for intact specimens (craniomedial aspect; A), specimens with an ostectomy gap repaired with a locking plate (cranial aspect; B), and specimens with an ostectomy gap repaired with a nonlocking plate (cranial aspect; C). Bar = 1 cm.

  • 1. Bennett RA, Kuzma AB. Fracture management in birds. J Zoo Wildl Med 1992; 23:538.

  • 2. Gull JM, Saveraid TC, Szabo D, et al. Evaluation of three miniplate systems for fracture stabilization in pigeons (Columba livia). J Avian Med Surg 2012; 26:203212.

    • Search Google Scholar
    • Export Citation
  • 3. Van Wettere AJ, Redig PT, Wallace LJ, et al. Mechanical evaluation of external skeletal fixator-intramedullary pin tie-in configurations applied to cadaveral humeri from red-tailed hawks (Buteo jamaicensis). J Avian Med Surg 2009; 23:277285.

    • Search Google Scholar
    • Export Citation
  • 4. Hedrick TL, Tobalske BW, Biewener AA. How cockatiels (Nymphicus hollandicus) modulate pectoralis power output across flight speeds. J Exp Biol 2003; 206:13631378.

    • Search Google Scholar
    • Export Citation
  • 5. Wang G, Liaw PK. Bending-fatigue behavior of bulk metallic glasses and their composites. JOM 2010; 62:2533.

  • 6. Biewener AA, Dial KP. In vivo strain in the humerus of pigeons (Columba livia) during flight. J Morphol 1995; 225: 6175.

  • 7. Demner D, Garcia TC, Serdy MG, et al. Biomechanical comparison of mono- and bicortical screws in an experimentally induced gap fracture. Vet Comp Orthop Traumatol 2014; 27:422429.

    • Search Google Scholar
    • Export Citation
  • 8. Dumont ER. Bone density and the lightweight skeletons of birds. Proc Biol Sci 2010; 277:21932198.

  • 9. Blake CA, Boudrieau RJ, Torrance BS, et al. Single cycle to failure in bending of three standard and five locking plates and plate constructs. Vet Comp Orthop Traumatol 2011; 24:408417.

    • Search Google Scholar
    • Export Citation
  • 10. Pater TJ, Grindel SI, Schmeling GJ, et al. Stability of unicortical locked fixation versus bicortical non-locked fixation for forearm fractures. Bone Res 2014; 2:14014.

    • Search Google Scholar
    • Export Citation
  • 11. Hamman D, Lindsey D, Dragoo J. Biomechanical analysis of bicortical versus unicortical locked plating of mid-clavicular fractures. Arch Orthop Trauma Surg 2011; 131:773778.

    • Search Google Scholar
    • Export Citation
  • 12. Overturf SJ, Morris RP, Gugala Z, et al. Biomechanical comparison of bicortical locking versus unicortical far-cortex-abutting locking screw-plate fixation for comminuted radial shaft fractures. J Hand Surg Am 2014; 39:19071913.

    • Search Google Scholar
    • Export Citation
  • 13. Rowe-Guthrie KM, Markel MD, Bleedorn JA. Mechanical evaluation of locking, nonlocking, and hybrid plating constructs using a locking compression plate in a canine synthetic bone model. Vet Surg 2015; 44:838842.

    • Search Google Scholar
    • Export Citation
  • 14. Gordon S, Moens NM, Runciman J, et al. The effect of the combination of locking screws and non-locking screws on the torsional properties of a locking-plate construct. Vet Comp Orthop Traumatol 2010; 23:713.

    • Search Google Scholar
    • Export Citation
  • 15. Bennert BM, Kircher PR, Gutbrod A, et al. Evaluation of two miniplate systems and figure-of-eight bandages for stabilization of experimentally induced ulnar and radial fractures in pigeons (Columba livia). J Avian Med Surg 2016; 30:111121.

    • Search Google Scholar
    • Export Citation
  • 16. Christen C, Fischer I, von Rechenberg B, et al. Evaluation of a maxillofacial miniplate compact 1.0 for stabilization of the ulna in experimentally induced ulnar and radial fractures in pigeons (Columba livia). J Avian Med Surg 2005; 19:185190.

    • Search Google Scholar
    • Export Citation
  • 17. Arens S, Eijer H, Schlegel U, et al. Influence of the design for fixation implants on local infection: experimental study of dynamic compression plates versus point contact fixators in rabbits. J Orthop Trauma 1999; 13:470476.

    • Search Google Scholar
    • Export Citation
  • 18. Gottsauner-Wolf F, Grabowski JJ, Chao EYS, et al. Effects of freeze/thaw conditioning on the tensile properties and failure mode of bone-muscle-bone units: a biomechanical and histological study in dogs. J Orthop Res 1995; 13:9095.

    • Search Google Scholar
    • Export Citation
  • 19. Smith G. Comments on torsional testing (lett). Vet Comp Orthop Traumatol 2015; 28:459.

  • 20. Arens S, Schlegel U, Printzen G, et al. Influence of materials for fixation implants on local infection. An experimental study of steel versus titanium dcp in rabbits. J Bone Joint Surg Br 1996; 78:647651.

    • Search Google Scholar
    • Export Citation

Advertisement

Ex vivo biomechanical evaluation of pigeon (Columba livia) cadaver intact humeri and ostectomized humeri stabilized with caudally applied titanium locking plate or stainless steel nonlocking plate constructs

View More View Less
  • 1 Department of Small Animal Clinical Sciences, College of Veterinary Medicine, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 2 Department of Small Animal Clinical Sciences, College of Veterinary Medicine, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 3 Department of Small Animal Clinical Sciences, College of Veterinary Medicine, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 4 Department of Small Animal Clinical Sciences, College of Veterinary Medicine, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 5 Department of Materials Science and Engineering, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 6 Department of Materials Science and Engineering, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 7 Department of Materials Science and Engineering, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 8 Department of Civil and Environmental Engineering, College of Engineering, University of Tennessee, Knoxville, TN 37996.
  • | 9 Department of Civil and Environmental Engineering, College of Engineering, University of Tennessee, Knoxville, TN 37996.

Abstract

OBJECTIVE To evaluate mechanical properties of pigeon (Columba livia) cadaver intact humeri versus ostectomized humeri stabilized with a locking or nonlocking plate.

SAMPLE 30 humeri from pigeon cadavers.

PROCEDURES Specimens were allocated into 3 groups and tested in bending and torsion. Results for intact pigeon humeri were compared with results for ostectomized humeri repaired with a titanium 1.6-mm screw locking plate or a stainless steel 1.5-mm dynamic compression plate; the ostectomized humeri mimicked a fracture in a thin cortical bone. Locking plates were secured with locking screws (2 bicortical and 4 monocortical), and nonlocking plates were secured with bicortical nonlocking screws. Constructs were cyclically tested nondestructively in 4-point bending and then tested to failure in bending. A second set of constructs were cyclically tested non-destructively and then to failure in torsion. Stiffness, strength, and strain energy of each construct were compared.

RESULTS Intact specimens were stiffer and stronger than the repair groups for all testing methods, except for nonlocking constructs, which were significantly stiffer than intact specimens under cyclic bending. Intact bones had significantly higher strain energies than locking plates in both bending and torsion. Locking and nonlocking plates were of equal strength and strain energy, but not stiffness, in bending and were of equal strength, stiffness, and strain energy in torsion.

CONCLUSIONS AND CLINICAL RELEVANCE Results for this study suggested that increased torsional strength may be needed before bone plate repair can be considered as the sole fixation method for avian species.

Abstract

OBJECTIVE To evaluate mechanical properties of pigeon (Columba livia) cadaver intact humeri versus ostectomized humeri stabilized with a locking or nonlocking plate.

SAMPLE 30 humeri from pigeon cadavers.

PROCEDURES Specimens were allocated into 3 groups and tested in bending and torsion. Results for intact pigeon humeri were compared with results for ostectomized humeri repaired with a titanium 1.6-mm screw locking plate or a stainless steel 1.5-mm dynamic compression plate; the ostectomized humeri mimicked a fracture in a thin cortical bone. Locking plates were secured with locking screws (2 bicortical and 4 monocortical), and nonlocking plates were secured with bicortical nonlocking screws. Constructs were cyclically tested nondestructively in 4-point bending and then tested to failure in bending. A second set of constructs were cyclically tested non-destructively and then to failure in torsion. Stiffness, strength, and strain energy of each construct were compared.

RESULTS Intact specimens were stiffer and stronger than the repair groups for all testing methods, except for nonlocking constructs, which were significantly stiffer than intact specimens under cyclic bending. Intact bones had significantly higher strain energies than locking plates in both bending and torsion. Locking and nonlocking plates were of equal strength and strain energy, but not stiffness, in bending and were of equal strength, stiffness, and strain energy in torsion.

CONCLUSIONS AND CLINICAL RELEVANCE Results for this study suggested that increased torsional strength may be needed before bone plate repair can be considered as the sole fixation method for avian species.

Traumatic fractures in birds are common and frequently occur in the mid to distal third of the humerus.1 Fracture repair in birds has historically relied on external fixation or external coaptation; although subjective assessments have indicated good outcomes, the devices can limit function and require daily care and a second anesthetic episode for removal. There is a paucity of information available on avian bone plating, and it is suggested the practice is not common because of the unique morphological shape of avian bones, thin cortices, the necessity to allow flight, and the cost and technical difficulties associated with use of plating equipment.1,2 With technological advancements for plating techniques, many of these hurdles may be overcome. Plates currently are available in sizes appropriate for smaller species, can be conformed to bones in multiple dimensions, and can have locking screws, which may be beneficial in bones with thin cortices. Also, plates can allow early return to function and may not require removal after healing.

Successful plate fixation has been described in a few reports. For example, plates were applied relatively easily and rapidly to the ulna of pigeons (Columba livia).2 In that study,2 a dorsally placed 1.3-mm, stainless steel 11-hole adaption plate designed for human phalanges was placed on the ulnas and resulted in good flight ability at 28 days in all of the birds.

Biomechanical testing is frequently performed on bones of humans and small mammals and provides valuable information about the use of a given construct before placement in vivo. Previous biomechanical testing of avian bones has been limited to 1 study3 in which the investigators focused on optimization of a single external fixation technique in raptors. Avian species may provide a unique opportunity for evaluating the application of new locking plates because of the reduced cortical mass and dimension of their bones. Biomechanical evaluation of avian bones may form the basis for use of these systems in avian species or, potentially, in other small bones with reduced cortical dimensions.

The specific objectives of the study reported here were to evaluate the stiffness, strength, and strain energy of intact avian humeri and humeri with a 1-mm ostectomy gap stabilized by use of a titanium locking plate construct or stainless steel nonlocking plate construct. We hypothesized that the intact humeri would possess the highest strength and stiffness in both bending and torsion and that no significant biomechanical differences would exist between the repaired ostectomized humeri.

Materials and Methods

Samples

Cadavers of 16 healthy adult feral pigeons (body weight, 300 to 400 g) were obtained from a government wildlife service. All pigeons had been euthanized (cervical dislocation) for reasons unrelated to the present study. Approval by an institutional animal care and use committee was not needed because cadavers, rather than live birds, were obtained for the study.

All birds were evaluated and weighed immediately after they were euthanized to ensure they had been healthy and represented a homogenous adult study population. For inclusion, birds had to be free of orthopedic disease (as determined during postmortem examination) as well as free of other visible abnormalities that could have affected health. Immediately after postmortem evaluation, pigeon cadavers were sagittally cut on the midline to separate the right and left sides. Each half was labeled with an identification number and weight, and the halves were stored separately in plastic bags. Samples were frozen and stored at −4°C until biomechanical testing was performed.

Experimental procedures

Specimens were thawed at 4°C over a 12-hour period. Two orthogonal-view radiographs of each humerus were obtained. Specimens were excluded if there was radiographic evidence of osteolysis, fracture, or other boney imperfections.

Humeri were randomly assigned to 1 of 3 groups (10 specimens/group). Group 1 (control group) consisted of intact humeri, and groups 2 and 3 represented fracture repair groups. Group 2 (locking plate group) consisted of humeri with an ostectomy gap and a 6-hole 4.0-mm titanium locking plate with 1.6-mm self-tapping locking screws made of TiAl6V4 titanium alloy.a Group 3 (nonlocking plate group) consisted of humeri with an ostectomy gap and a 6-hole, 316L stainless steel DCP with 1.5-mm self-tapping cortical screws.b Randomization was performed by placing all bird halves in an opaque bag and then blindly selecting a bird half from the bag and assigning it to a group; group assignment was in the order of intact, locking plate, and nonlocking plate. Humeri from the same bird were never assigned to the same group. If a group contained the left and right humerus from the same bird, one half was returned to the bag and another bird half selected.

Liquid plasticc was used as potting material on the proximal and distal ends of each humerus. Potting material was embedded in rectangular high density plastic (bending analysis) or cylindrical black iron steel tubing (torsion analysis). Jigs were custom designed and constructed for the potting process to ensure appropriate alignment of end pieces and to aid in reproducibility.

Plating

For the repair groups (groups 2 and 3), transverse ostectomy was performed at the midpoint of the long axis of the humerus, which was followed by 1 of 2 repair techniques. Procedures were performed by a resident in a veterinary surgery training program (BGD) under the supervision of a board-certified veterinary surgeon (JJB or JPW). For purposes of reproducibility, all ostectomies were performed after embedding in potting material. The liquid plastic was ≥ 1 mm from the fracture repair to ensure that no portion of the construct was implanted into the potting material. All specimens were wrapped in gauze soaked in saline (0.9% NaCl) solution from thawing until testing.

For group 2, 2 monocortical locking screws were placed in each segment with a single bicortical screw placed nearest the fracture site, which centered the plate over the ostectomy site. Calipers were used to verify a 1-mm gap remained after fracture fixation. A plate was applied to the caudal aspect of the humerus. It was contoured to the bone until there was no visual gap between the plate and bone. Screws were tightened manually until the screw head was flush with the plate. For group 3, 3 bicortical nonlocking screws were placed in each segment, and the same procedure was used as for group 2. Despite the fact self-tapping screws were used for group 3, an initial hole for each screw was created by use of a tap.

Biomechanical testing

Humeri of all 3 groups were subjected to biomechanical testing. Before the study was conducted, a power analysis (power = 80% and α = 0.05) was performed by use of estimates and early experimental results. Results of power analysis indicated a minimum of 5 samples/group were required.

Cyclic bending—Five specimens from each group were subjected to 4-point cyclic bending by use of a materials testing systemd with an outer span of 46 mm and inner span of 14 mm (Figure 1). Cyclic bending was performed at 30 N for 1,000 cycles at a frequency of 1 Hz.4,5 Cycling loads were chosen on the basis of forces measured for in-flight cockatiels and pigeons.4,6 Mean slope of the load-deformation curve was calculated and compared with the final slope to determine degree of loosening or stiffening. Specimens were monitored for audible or visible signs of failure. After cycling was completed, samples were examined to detect screw loosening, bone fracture, or other signs of fatigue.

Figure 1—
Figure 1—

Diagram illustrating the position of a specimen (humerus from a pigeon [Columbia livia] cadaver; caudal aspect) in a testing apparatus and mediolateral direction of load during 4-point bending. Dimensions of inner (14 mm) and outer (46 mm) spans are indicated.

Citation: American Journal of Veterinary Research 78, 5; 10.2460/ajvr.78.5.570

Bending to failure—After cyclic bending was completed, specimens were loaded at a crosshead velocity of 0.05 mm/s until failure, which generated a load-deformation curve. Strain energy, yield strength, and ultimate strength were recorded. The mechanism of failure was recorded as catastrophic or noncatastrophic. Catastrophic failure involved bone breakage and a representative acute decrease in load during displacement. Noncatastrophic failure was represented by a slow gradual decrease in stiffness (ie, screw pullout or implant fatigue). Samples were monitored during testing and by use of video recordings to detect audible or visible signs of fracture.

Cyclic torsion—The remaining 5 specimens from each group were cycled in torsion between −3° to 3° for 1,000 cycles at a frequency of 1 Hz by use of a materials testing system.7,e The corresponding load was measured at each rotational displacement. Cycling torsion was chosen on the basis of forces measured for in-flight cockatiels and pigeons.4,6 Torsion cycling also was based on preliminary data for 2 intact specimens, which indicated that cyclic torsion of −3° to 3° (total movement, 6°) was tolerated well (data not shown). Mean slope of the load-deformation curve was calculated and compared with the final slope to determine the degree of stiffening or loosening. Specimens were monitored for audible or visible signs of failure. High-definition cameras were used to aid in visual examination of potential failures. Non-plated surfaces of the bones were lightly dusted with a high-contrast white-and-black–speckled acrylic-based spray paint.f This technique can be useful for monitoring propagation of cracks, and the minimal amount of spray paint used was not likely to alter mechanical properties. After cyclic torsion was completed, specimens were examined to detect screw loosening, bone fracture, or other signs of fatigue.

Torsion to failure—After cyclic torsion was completed and specimens were visually inspected for failure, constructs were loaded in torsion at a rate of 0.1°/s until failure, which generated a torsional load-deformation curve. Failure load and failure angles were recorded. The mechanism of failure was recorded as catastrophic or noncatastrophic, as described for bending to failure. Samples were monitored during testing and by use of video recordings to detect audible or visible signs of fracture.

Calculation of variables

Stiffness was calculated as the mean of the slopes of the load-deformation curves for each cycle. Mean slope (stiffness) was compared with the slope of the 1,000th cycle as an assessment of potential implant loosening, shifting, fatigue, or microfracture. Ultimate failure, displacement at ultimate failure, ultimate torque, and rotation at ultimate torque were obtained directly from the data for the highest load achieved by a given construct over the course of the loading period.

Yield point is classically defined as the load at which a material begins to deform plastically. In the present study, yield point was defined as the load at which the first visible or audible signs of loosening or breakage of implants or bone fracture were detected or an acute negative deflection in the load-deformation curve was seen. Yield point equaled the ultimate failure value if there was complete failure. The displacement at yield point was obtained directly from the data.

Strain energy is the ability of a material to absorb energy and deform without fracturing. Practically, it is measured as the area under the load-deformation curve. The linear trapezoidal method was used to calculate strain energy from the initiation of the test to the yield point in bending and the ultimate torque in torsion.

Statistical analysis

Data were analyzed by use of a 1-way multivariate ANOVA. Dependent variables included those determined during bending cycles (stiffness), bending failure (strain energy, displacement at yield point, yield point, displacement at ultimate failure, and ultimate failure), torsion cycles (stiffness), and torsion failure (strain energy, displacement at ultimate failure, and ultimate failure); each treatment group served as the independent variable. Significance was set at values of P < 0.05. All analyses were conducted by use of a commercially available statistics program.g

Results

Specimens

During initial examination, there was a simple short spiral fracture of the mid-diaphysis of the right humerus of 1 pigeon; thus, another humerus was used. All remaining humeri appeared visibly normal, and examination of radiographs did not reveal abnormalities. All specimens successfully completed testing, except for 1 construct from group 3, which was replaced because 1 screw was stripped.

Cyclic bending

Humeri repaired with the nonlocking plate were significantly (P = 0.003) stiffer than were humeri repaired with the locking plate. No plastic deformation or screw loosening was detectable for either repair group. No significant difference was detected for final stiffness, compared with mean stiffness. No other significant differences were detected with regard to cyclic bending (Table 1).

Table 1—

Mean ± SD and range values for cyclic bending and bending to failure for humeri specimens from 5 pigeon (Columbia livia) cadavers.

 Cyclic bendingBending to failure
Group*Stiffness (N/mm)Strain energy (N-mm)Displacement at yield point (mm)Yield point (N)Displacement at ultimate failure (mm)Ultimate failure (N)
Intact
 Mean ± SD299.5 ± 109.1a93.5 ± 19.4a0.70 ± 0.13170.0 ± 6.3a0.70 ± 0.13170.0 ± 6.3a
 Range105.6 to 363.069.2 to 114.50.55 to 0.85159.2 to 175.90.55 to 0.85159.2 to 175.9
Locking plate
 Mean ± SD203.2 ± 54.6b42.5 ± 11.7b0.517 ± 0.08190.1 ± 10.7b0.662 ± 0.29393.8 ± 10.7b
 Range135.7 to 278.631.1 to 58.70.440 to 0.60980.2 to 107.30.440 to 1.17080.2 to 107.3
Nonlocking plate
 Mean ± SD387.9 ± 44.8c61.7 ± 29.9b0.792 ± 0.341100.0 ± 27.5b1.620 ± 1.390133.0 ± 66.5b
 Range335.8 to 446.213.5 to 86.50.349 to 1.29062.2 to 130.00.349 to 3.66062.2 to 237.4

Groups comprised specimens of intact humeri, humeri with an ostectomy gap repaired with a locking plate, and humeri with an ostectomy gap repaired with a nonlocking plate.

Within a column, values with different superscript letters differ significantly (P < 0.05).

Bending to failure

Yield points of intact specimens were significantly higher than those for locking plate (P < 0.001) and nonlocking plate (P = 0.005) specimens. There was no significant difference in displacement at yield point among specimens. The ultimate failure value of intact specimens was significantly (P = 0.010) higher than that of locking plate specimens. There were no significant differences for displacement at ultimate failure among specimens. Intact specimens had significantly higher strain energies (P = 0.005) than did locking plate specimens. All intact specimens failed catastrophically in the middle to distal portion of the diaphysis (Figure 2). All locking plate specimens failed catastrophically through the proximal portion of the diaphysis (3/5) or distal portion of the diaphysis (2/5). All nonlocking plate specimens failed through the proximal screw of the distal segment; 3 of 5 failed catastrophically, whereas 2 of 5 failed noncatastrophically. No other significant differences were detected (Table 1).

Figure 2—
Figure 2—

Photographs of examples of failure during bending of pigeon cadaver humeri for intact specimens (cranial aspect; A), specimens with an ostectomy gap repaired with a locking plate (lateral aspect; B), and specimens with an ostectomy gap repaired with a nonlocking plate (lateral aspect; C). In panel C, the specimen on the right has a comminuted fracture, whereas the specimen on the left has simple loosening of the screws as indicated by the fact that the proximal screws in the distal segment are not flush with the plate. Bar = 1 cm.

Citation: American Journal of Veterinary Research 78, 5; 10.2460/ajvr.78.5.570

Cyclic torsion

Intact humeri were significantly stiffer than humeri repaired with the locking plate (P = 0.005) or nonlocking plate (P < 0.001). No plastic deformation or screw loosening was detectable for either repair group. No significant difference was detected in the final stiffness, compared with the mean stiffness. No other significant differences were detected (Table 2).

Table 2—

Mean ± SD and range values for torsion bending and bending to failure for humeri specimens from 5 pigeon cadavers.

 Cyclic torsionTorsion to failure
Group*Stiffness (N-mm/°)Strain energy (N-mm-°)Rotation at utimate torque (°)Ultimate torque (N-mm)
Intact    
Mean ± SD125.0 ± 25.6a10,738 ± 2,520a11.55 ± 1.051,105.0 ± 144.0a
Range117.4 to 151.27,903 to 14,43810.34 to 21.62952.1 to 1,300.0
Locking plate    
Mean ± SD36.2 ± 4.6b4,526 ± 1,171b12.04 ± 3.62371.0 ± 38.5b
Range32.5 to 43.23,430 to 6,3799.54 to 16.20326.7 to 396.2
Nonlocking plate    
Mean ± SD31.4 ± 8.8b7,475 ± 4,664b18.85 ± 7.18350.0 ± 167.0b
Range22.5 to 44.81,255 to 12,76212.16 to 29.30156.6 to 475.4

Within a column, values with different superscript letters differ significantly (P < 0.05).

See Table 1 for remainder of key.

Torsion to failure

Ultimate torque of intact humeri was significantly higher than that of locking plate (P < 0.001) and nonlocking plate (P = 0.005) specimens. There were no significant differences among plated groups for ultimate torque. Intact humeri had significantly higher strain energies (P = 0.033) than did locking plate specimens. Yield points were not reported because all constructs failed catastrophically. All intact humeri failed in a comminuted manner (spiraling proximally) in the middle to distal portion of the humerus (Figure 3). Locking plate specimens failed with fracture lines extending proximally (4/5) or distally (1/5). Nonlocking plate specimens failed with fracture lines extending proximally (3/5) or distally (2/5). No other significant differences were detected (Table 2).

Figure 3—
Figure 3—

Photographs of examples of failure of pigeon cadaver humeri during torsion for intact specimens (craniomedial aspect; A), specimens with an ostectomy gap repaired with a locking plate (cranial aspect; B), and specimens with an ostectomy gap repaired with a nonlocking plate (cranial aspect; C). Bar = 1 cm.

Citation: American Journal of Veterinary Research 78, 5; 10.2460/ajvr.78.5.570

Discussion

Analysis of results of the study reported here revealed that intact humeri of pigeon cadavers were stronger than humeri with a 1-mm mid-diaphyseal ostectomy repaired with a titanium locking plate or stainless steel DCP (ie, nonlocking plate) when loaded to failure in both bending and torsion. Thus, the hypothesis that intact specimens were stronger than specimens of repaired constructs was accepted. Intact humeri were stiffer in bending than humeri repaired with a locking plate and stiffer in torsion than humeri repaired with a locking plate or nonlocking plates. Thus, the hypothesis that intact specimens were stiffer in both bending and torsion than specimens repaired with constructs was rejected. The null hypothesis that humeri with locking plates and nonlocking plates were of equal stiffness and strength in both bending and torsion also was rejected because humeri with nonlocking plates were stiffer in bending.

A paucity of information has been published on avian biomechanical testing. Standard testing procedures, such as load, rate, and number of cycles, are not known. Testing variables for the present study were based on forces measured for in-flight cockatiels and pigeons.4,6 Pigeon humeri were chosen for use because their size is similar to that of birds commonly kept as pets, they are large enough for application of a small plating system, and they could be easily obtained and were readily available. Additionally, studies4,6 have been conducted on in vivo flight mechanics of pigeons. Bending and torsional testing were selected because avian humeri do not have appreciable axial compression relative to torsional and bending loads.6,8 The purpose of the present study was to maximize the work of the implant; thus, a 1-mm ostectomy gap was chosen because it could be reliably created and avoided load sharing. Mediolateral bending was selected to mimic in vivo flight forces (upstroke and downstroke). The plate was placed on the caudal aspect of the humerus because this would maximize the area moment of inertia of the plate in a live bird, and access to the caudal aspect of the humerus is feasible by means of a dorsal surgical approach.1

In the present study, nonlocking plate constructs were significantly stiffer in bending than were locking plate constructs. This difference may have been related, in part, to the plate material because the locking plate was titanium and the nonlocking plate was stainless steel. On the basis of the modulus of the materials (material property) and moment of inertia of each plate (plate geometry), overall stiffness of the titanium locking plate construct was estimated to be 40% that of the bending stiffness of the titanium locking plate alone (bending modulus E9 = 110 GPa; moment of inertia I4 = 1.54 mm). Similarly, estimated stiffness of the stainless steel nonlocking plate construct was 32% that of the stainless steel plate alone (bending modulus E9= 200 GPa; moment of inertia I4 = 3.81 mm). This 8% difference of normalized values suggested that the locking plate may have transferred plate stiffness to the overall construct better than did a nonlocking plate for these avian humeri; however, studies of identical plate materials and geometries would be necessary before a definitive conclusion could be drawn.

We hypothesized that the remaining variation between the 2 plating systems would be dominated by the locking and nonlocking configurations and the interface between the plate and bone provided by the screws. Investigators of several studies10–12 that involved the use of stainless steel locking plates have found that total monocortical locking constructs are as stiff as bicortical nonlocking constructs. Results of the study reported here are in contrast to results of those studies,10–12 possibly because of differences in plate contouring or the fact that human antebrachial and clavicular bone do not accurately represent the thin cortices of avian bone.

Stiffness is undoubtedly a major component for successful fracture healing. Ideal measures of stiffness to achieve the most successful fracture repair are not known in birds, but a balance of adequate stabilization while minimizing stress protection undoubtedly would be ideal. Interestingly, titanium locking plate constructs failed via bone fracture through multiple screws in all specimens, whereas stainless steel nonlocking plate constructs failed via screw pullout in 2 of 5 specimens. This difference likely was the result of angular stability afforded by locking systems.

In a recent study,13 investigators evaluated effects of the configuration of locking and nonlocking screws in a locking compression plate in a composite bone model. Those investigators found that use of a bicortical locking screw adjacent to the fracture gap resulted in increased strength in both bending and torsion, compared with results for a bicortical nonlocking screw in that position, and that an all-locking configuration was stiffest and strongest. Because torsional stress is clinically important in avian bones,6 bicortical screw application nearest the fracture gap was chosen for the study reported here. However, results of the present study are not consistent with results for that previous study13 in that the humeri with locking plates were not significantly stiffer or stronger than the humeri with nonlocking plates. The difference in results could have been attributable, in part, to the plate materials. However, on the basis of shear modulus and plate geometry, we hypothesized that differences in bending were more likely a result of the overall configuration and less attributable to differences in plate materials. Although all-locked constructs have been found to be superior to all-non-locked constructs in torsion,13,14 monocortical screws are less stiff and strong in torsion than are bicortical locking screws7 and bicortical nonlocking screws.10 Monocortical screw application was chosen for the present study to conform with manufacturers’ clinical recommendations, which attempted to minimize disruption to the medullary canal.

In contrast to nonlocking plates in the present study and locking plates in another study,13 locking plates in the present study were not perfectly contoured to the bone. Bending stiffness is expected to decrease with increasing length between the plate and bone, which could be the reason that no significant differences were found between humeri with locking plates and nonlocking plates for torsional stiffness and strength. Finally, avian bone has relatively thin cortices and brittle features.8 The composite bone models used in previous studies13 may not have appropriately represented the features of avian bone.

Two types of plating materials with different ductile properties were used to measure strain energy in the present study. Ductile properties of these materials could have influenced the overall construct strain energy or energy that the repair could absorb before yielding. Strain energy of the intact specimens was significantly greater in both bending and torsion than was strain energy of the locking plate specimens. It may be necessary that all screws in the locking plate be bicortical screws to mimic the strain energy of intact bone.

The ultimate goal of biomechanical testing is to predict clinical applicability or outcome. Except for bending stiffness, repair constructs were found to be significantly weaker and less stiff than intact specimens. Safety factors, defined as maximum strain of bone at failure divided by expected maximum physiologic strain, have been reported for bones of many species, including the avian tibia and humerus.6,8 In the avian humerus, the safety factor is approximately 1.9 in torsion and 3.5 in bending.6 Although strain may not directly correlate with force, the relative deficiency in stiffness and strength in torsion, as indicated by the results reported here, suggested that the repairs might not be appropriate as a primary method of fixation and that augmentation, cage rest, or exercise restriction (or a combination of these factors) may be warranted. However, studies2,15,16 have been conducted to evaluate the use of titanium and stainless steel nonlocking plates for the repair of iatrogenic radial and ulnar osteotomies of pigeons in vivo. The conclusions drawn from those studies2,15,16 include that stainless steel nonlocking plates may be used successfully to promote healing and return to flight following iatrogenic transection of the radius and ulna. In those studies,2,15,16 construct failure occurred as bending or twisting of the plates or as screw pullout. In the present study, neither plate type failed as a result of bending or twisting, although 2 of 5 nonlocking plates failed noncatastrophically as a result of screw pullout. These results suggested that the locking plate may have sufficient strength to resist plastic deformation, which would reduce the rate of screw pullout.

Features unique to avian bones include thin cortices relative to bone diameter and high bone density, which results in brittle biomechanical properties.1,8 Thin cortices and a brittle character create unique challenges for avian fracture repair.1 For example, implant application is challenging and has been assessed by recording surgical errors and subjectively recording intraoperative difficulty. In the study reported here, 1 screw was stripped during nonlocking plate fixation; subjectively, it was difficult to assess appropriate screw tightness for this group because of the small size of the implants and thin cortices of avian bone. Some preliminary trials with nonlocking plates resulted in screw stripping or bone fracture, which prompted the use of a tap in the present study. Screw stripping or bone fractures were not encountered with the locking plates, and a tap was not used for insertion of screws for that system. Screws were considered properly placed if they became acutely more difficult to advance and were flush with the plate. A torque-limiting device could have helped standardize screw placement among specimens, although such a device is not available for clinical use of the small implant used in the present study.

Torque is less of a concern when locking plates are applied, which may be a benefit for use of locking plates in avian bones. Other reported benefits of limited-contact locking plates include preservation of the periosteal blood supply and preservation of the endosteal blood supply if monocortical screws are used as well as the potential for reduced risk of infection.17

Limitations of the study reported here included the small number of constructs tested, which possibly could have led to a type II error. A power analysis for comparing the intact bone, locking plate construct, and nonlocking plate construct was performed to decrease the likelihood of a type II error.

As previously mentioned, screw torque was not controlled during placement of screws in the locking or nonlocking plates. Additionally, feral birds may have variations in bone quality because of differences in nutrition, illness, and age. Performing a physical examination on birds to ensure adulthood and apparent health was intended to minimize this variation. All birds were weighed and selected to have a body weight between 300 and 400 g, which is an appropriate weight for a healthy pigeon, to ensure a homogenous and healthy population. Freezing osseous specimens of dogs has a minimal effect on mechanical properties,18 but the effects are unknown for avian bones. To minimize potential effects, specimens were subjected to only 1 freeze-thaw cycle. Wide variations in data were not detected, which supported consistency in testing.

One limitation of the 4-point bending setup used in the present study was that 2 points of contact were directly on bone within the span of the plate. Ideally, load would be applied on the potted ends or be applied outside the plate span to minimize local stress and mimic in vivo biomechanics. To achieve insertion of screws in 3 holes proximal and 3 holes distal to the ostectomy, plates spanned nearly the entire humeral diaphysis, which left only the most proximal and distal aspects of the metaphysis and epiphysis to be secured in the potting material. As a result of spatial constraints, this setup was considered acceptable. Because the site of failure originated at the ostectomy gap, it is believed that results of the present study are valid. Torsional testing was performed with the potted ends constrained. Lack of axial translation may imprecisely mimic true torsion and complicate extrapolation to clinical applications.19

Finally, the use of different plate materials and screw configurations prohibited direct comparison of the locking and nonlocking plates. These systems were chosen because they were commercial veterinary plating systems that could be fit to a readily available source of consistent avian (pigeon) bones. The authors are unaware of a similarly sized stainless steel locking plate option. Both systems were of similar dimensions, involved use of similar-diameter screws, and could be applied to a small bird for clinical use.

To the authors’ knowledge, the study reported here was the first in which biomechanical testing of plates and screws with avian bones has been evaluated. We found that intact humeri of pigeon cadavers were stiffer and stronger, except for stiffness in bending, than humeri repaired with titanium locking plates or stainless steel DCPs. Locking and nonlocking constructs were of equal stiffness and strength, except that the stainless steel nonlocking plates were stiffer in bending. Intangible benefits such as ease of application of locking plates and decreased risk of screw stripping should be considered in thin, brittle cortical bone.20 Information on avian bone should be of value for use on birds, but this information may also be beneficial for other species with conditions (eg, osteopenia in mammals) that result in bone with thin cortical dimensions or brittle biomechanical properties. Additional studies are needed to assess the suitability of extrapolating results for avian bones to other species and pathological conditions.

Acknowledgments

Supported by the Companion Animal Fund at the University of Tennessee College of Veterinary Medicine and by Kyon AG.

The authors thank Sun Xiaocun for statistical assistance and Misty Bailey for technical assistance.

ABBREVIATIONS

DCP

Dynamic compression plate

Footnotes

a.

Mini (3.5/4) advanced locking plate system (ALPS), Kyon AG, Zurich, Switzerland.

b.

DCP, Veterinary Orthopedic Implants, St Augustine, Fla.

c.

Smooth-On 300Q, Smooth-On Inc, Macungie, Pa.

d.

Instron ElectroPuls E1000, Illinois Tool Works Inc, Norwood, Mass.

e.

MTS 858, MTS Systems Corp, Eden Prairie, Minn.

f.

Valspar Micromist spray (Flat), The Valspar Corp, Wheeling, Ill.

g.

SAS for Windows, version 9.4, SAS Institute Inc, Cary, NC.

References

  • 1. Bennett RA, Kuzma AB. Fracture management in birds. J Zoo Wildl Med 1992; 23:538.

  • 2. Gull JM, Saveraid TC, Szabo D, et al. Evaluation of three miniplate systems for fracture stabilization in pigeons (Columba livia). J Avian Med Surg 2012; 26:203212.

    • Search Google Scholar
    • Export Citation
  • 3. Van Wettere AJ, Redig PT, Wallace LJ, et al. Mechanical evaluation of external skeletal fixator-intramedullary pin tie-in configurations applied to cadaveral humeri from red-tailed hawks (Buteo jamaicensis). J Avian Med Surg 2009; 23:277285.

    • Search Google Scholar
    • Export Citation
  • 4. Hedrick TL, Tobalske BW, Biewener AA. How cockatiels (Nymphicus hollandicus) modulate pectoralis power output across flight speeds. J Exp Biol 2003; 206:13631378.

    • Search Google Scholar
    • Export Citation
  • 5. Wang G, Liaw PK. Bending-fatigue behavior of bulk metallic glasses and their composites. JOM 2010; 62:2533.

  • 6. Biewener AA, Dial KP. In vivo strain in the humerus of pigeons (Columba livia) during flight. J Morphol 1995; 225: 6175.

  • 7. Demner D, Garcia TC, Serdy MG, et al. Biomechanical comparison of mono- and bicortical screws in an experimentally induced gap fracture. Vet Comp Orthop Traumatol 2014; 27:422429.

    • Search Google Scholar
    • Export Citation
  • 8. Dumont ER. Bone density and the lightweight skeletons of birds. Proc Biol Sci 2010; 277:21932198.

  • 9. Blake CA, Boudrieau RJ, Torrance BS, et al. Single cycle to failure in bending of three standard and five locking plates and plate constructs. Vet Comp Orthop Traumatol 2011; 24:408417.

    • Search Google Scholar
    • Export Citation
  • 10. Pater TJ, Grindel SI, Schmeling GJ, et al. Stability of unicortical locked fixation versus bicortical non-locked fixation for forearm fractures. Bone Res 2014; 2:14014.

    • Search Google Scholar
    • Export Citation
  • 11. Hamman D, Lindsey D, Dragoo J. Biomechanical analysis of bicortical versus unicortical locked plating of mid-clavicular fractures. Arch Orthop Trauma Surg 2011; 131:773778.

    • Search Google Scholar
    • Export Citation
  • 12. Overturf SJ, Morris RP, Gugala Z, et al. Biomechanical comparison of bicortical locking versus unicortical far-cortex-abutting locking screw-plate fixation for comminuted radial shaft fractures. J Hand Surg Am 2014; 39:19071913.

    • Search Google Scholar
    • Export Citation
  • 13. Rowe-Guthrie KM, Markel MD, Bleedorn JA. Mechanical evaluation of locking, nonlocking, and hybrid plating constructs using a locking compression plate in a canine synthetic bone model. Vet Surg 2015; 44:838842.

    • Search Google Scholar
    • Export Citation
  • 14. Gordon S, Moens NM, Runciman J, et al. The effect of the combination of locking screws and non-locking screws on the torsional properties of a locking-plate construct. Vet Comp Orthop Traumatol 2010; 23:713.

    • Search Google Scholar
    • Export Citation
  • 15. Bennert BM, Kircher PR, Gutbrod A, et al. Evaluation of two miniplate systems and figure-of-eight bandages for stabilization of experimentally induced ulnar and radial fractures in pigeons (Columba livia). J Avian Med Surg 2016; 30:111121.

    • Search Google Scholar
    • Export Citation
  • 16. Christen C, Fischer I, von Rechenberg B, et al. Evaluation of a maxillofacial miniplate compact 1.0 for stabilization of the ulna in experimentally induced ulnar and radial fractures in pigeons (Columba livia). J Avian Med Surg 2005; 19:185190.

    • Search Google Scholar
    • Export Citation
  • 17. Arens S, Eijer H, Schlegel U, et al. Influence of the design for fixation implants on local infection: experimental study of dynamic compression plates versus point contact fixators in rabbits. J Orthop Trauma 1999; 13:470476.

    • Search Google Scholar
    • Export Citation
  • 18. Gottsauner-Wolf F, Grabowski JJ, Chao EYS, et al. Effects of freeze/thaw conditioning on the tensile properties and failure mode of bone-muscle-bone units: a biomechanical and histological study in dogs. J Orthop Res 1995; 13:9095.

    • Search Google Scholar
    • Export Citation
  • 19. Smith G. Comments on torsional testing (lett). Vet Comp Orthop Traumatol 2015; 28:459.

  • 20. Arens S, Schlegel U, Printzen G, et al. Influence of materials for fixation implants on local infection. An experimental study of steel versus titanium dcp in rabbits. J Bone Joint Surg Br 1996; 78:647651.

    • Search Google Scholar
    • Export Citation

Contributor Notes

Address correspondence to Dr. Darrow (bdarrow1@utk.edu).