Introduction
The AAJ has a high degree of mobility that allows rotational movements of the head.1, 2 Atlantoaxial joint instability is a neurosurgical condition that affects the cranial part of the cervical vertebral column, can lead to subluxation of the joint, and may result in compressive cervical myelopathy.3 The congenital form of instability is mainly described for young small- or toy-breed dogs such as Yorkshire Terriers, Chihuahuas, Toy Poodles, Pomeranians, and Pekingese.3–7 Among the causes, there are many pathological processes such as aplasia or hypoplasia of the dens of the axis, malformation of the dens, or abnormalities of the transverse ligament of the atlas.3,5–7 Although a congenital malformation is most common, there are also traumatic cases secondary to fracture or separation of the dens or rupture of the transverse ligament of the atlas.8
Treatment options vary and remain controversial. Although conservative treatments have been described,7 surgical management is indicated in most patients to permit permanent reduction of the luxation and stabilization of the joint.3, 4 Many surgical techniques, including dorsal and ventral fixations, are available. Dorsal AAJ fixation can be performed by use of orthopedic wire, pins, suture material, the nuchal ligament, intermuscular sutures, and metal retractors.3, 4 Ventral stabilization procedures include the use of transarticular lag screws, bone plates, K-wires, or threaded pins with or without PMMA.3–5,9–12
Several reports5,9–18 have described various surgical techniques for treatment of AAJI, but studies evaluating the biomechanical properties of canine AAJs following stabilization with different methods are lacking. To the best of our knowledge, no biomechanical studies have been published that compare the stability provided by dorsal and ventral AAJ fixation or assess the biomechanics of AAJI repairs in toy-breed dogs. The purpose of the study reported here was to compare the biomechanical properties of AAJs in vertebral column specimens from toy-breed dogs stabilized with 4 techniques (dorsal wire, dorsal clamp [application of a modified Kishigami AA tension band], ventral transarticular pin, and augmented ventral transarticular pin fixation) after transection of the AAJ ligaments. We hypothesized that, given the transarticular component, the biomechanical stability of AAJs stabilized with ventral fixation techniques might be greater than that of AAJs stabilized with dorsal fixation techniques and that adding a fixation with wire to the ventral transarticular pin fixation would provide increased AAJ rigidity under shear loading conditions.
Materials and Methods
Sample
The cadavers of 13 toy-breed dogs euthanatized for reasons unrelated to the present study were obtained and stored at–25 °C. Computed tomographya and MRIb of each specimen were performed to evaluate the bony and ligamentous structures of the occipitoatlantoaxial junction and allow exclusion of any specimens with anomalous anatomic features. Specimens were excluded from the study if bony abnormalities or signs of instability in the occipitoatlantoaxial region were identified.
Two days before preparation for experiments, the specimens were thawed at room temperature (approx 21 °C). A transverse cut was made in the skull 2 cm cranial to the tympanic bullae, and the vertebral column and spinal cord were transected at the C5-6 intervertebral disk space. The muscles of the occipitoatlantoaxial region were carefully stripped away with care taken to preserve the integrity of the AAJ ligaments and joint capsule. Once prepared, the specimens were placed in plastic bags and stored at–25 °C until testing.
The day before biomechanical testing, the specimens were thawed at 2 °C. The atlantooccipital joints were fixed with 2 transarticular diverging 1-mm, positive-threaded K-wires. Two other 1-mm K-wires were longitudinally inserted from the C5-C2 vertebral bodies to immobilize the C2-C5 motion segments. Once prepared, the specimens were wrapped in towels soaked in saline (0.9% NaCl) solution, placed in plastic bags, and refrigerated at 2 °C until testing on the following day.
AAJ stabilization techniques
Four AAJ stabilization techniques were performed on each specimen after initial biomechanical tests were completed for each specimen (first with intact ligaments and then with the ligaments transected). The order in which the stabilization techniques was performed on each specimen was randomly assigned. Different sequences of fixation techniques were created so that each technique was performed on each specimen in a different order. Each sequence was then assigned to a different specimen by a random draw method.
For the dorsal wire fixation, a loop of 0.5-mm wirec was passed cranially under the arch of C1 through the epidural space. The loop was then folded back and cut. Both strands were passed through 2 predrilled holes in the spinous process of C2 and twisted together3 (Figure 1).
Photograph depicting dorsal AAJ fixation with wire in a specimen prepared from the cranial portion of the cervical vertebral column of a cadaveric toy-breed dog for a study to compare the biomechanical properties of 4 AAJ stabilization techniques (dorsal wire, modified dorsal clamp, ventral transarticular pin, and augmented ventral transarticular pin fixation) under shear loading conditions after transection of the AAJ ligaments. The final study sample included 10 specimens, which were reused so that all 4 techniques were performed on each specimen in random order. The specimen in the photograph is shown mounted in the testing device for biomechanical testing.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph depicting dorsal AAJ fixation with wire in a specimen prepared from the cranial portion of the cervical vertebral column of a cadaveric toy-breed dog for a study to compare the biomechanical properties of 4 AAJ stabilization techniques (dorsal wire, modified dorsal clamp, ventral transarticular pin, and augmented ventral transarticular pin fixation) under shear loading conditions after transection of the AAJ ligaments. The final study sample included 10 specimens, which were reused so that all 4 techniques were performed on each specimen in random order. The specimen in the photograph is shown mounted in the testing device for biomechanical testing.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph depicting dorsal AAJ fixation with wire in a specimen prepared from the cranial portion of the cervical vertebral column of a cadaveric toy-breed dog for a study to compare the biomechanical properties of 4 AAJ stabilization techniques (dorsal wire, modified dorsal clamp, ventral transarticular pin, and augmented ventral transarticular pin fixation) under shear loading conditions after transection of the AAJ ligaments. The final study sample included 10 specimens, which were reused so that all 4 techniques were performed on each specimen in random order. The specimen in the photograph is shown mounted in the testing device for biomechanical testing.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
For the dorsal clamp fixation, a modified
Kishigami AA tension bandd was used. The clamp was developed by the manufacturer independent of the study and was not commercially available. An incision was made in the dorsal atlantooccipital membrane, and the hook of the clamp was inserted under the dorsal lamina of the atlas. The caudal part of the clamp, comprising 3 small spikes on both sides, was placed around the spinous process of C2 and secured through compression of the spikes, which then penetrated the spinous process. The caudal part of the implant was an application device that was removed by twisting after placement of the implant was complete13 (Figure 2).
Photographs depicting the dorsal AAJ clamp (a modified Kishigami AA tension bandd) before (A) and after (B) application for AAJ fixation in a specimen. The application device (arrows) was removed from the implant after fixation.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photographs depicting the dorsal AAJ clamp (a modified Kishigami AA tension bandd) before (A) and after (B) application for AAJ fixation in a specimen. The application device (arrows) was removed from the implant after fixation.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photographs depicting the dorsal AAJ clamp (a modified Kishigami AA tension bandd) before (A) and after (B) application for AAJ fixation in a specimen. The application device (arrows) was removed from the implant after fixation.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
For the ventral transarticular pin fixation, 2 diverging, 1.25-mm positive-threaded K-wires were driven bilaterally across the AAJ, as previously described9, 12 (Figure 3). Pin placement was started close to the midline on the cranioventral surface of C2, caudal to the AA articulation. Pins were then directed craniolaterally through the AAJ and placed into the caudal body of the atlas at an approximate 30° angle to the transverse plane. The protruding ends of the pins were then cut off.
Photograph (A) and diagram (B) showing placement of 2 transarticular pins for ventral AAJ fixation in a specimen. Pin placement was started close to the midline on the cranioventral surface of C2, caudal to the AA articulation. Pins were then directed craniolaterally through the AAJ and placed into the caudal body of the atlas at an approximate 30° angle to the transverse plane. The protruding ends of the pins were subsequently cut off (pins were left untrimmed in the photograph to improve their visibility).
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph (A) and diagram (B) showing placement of 2 transarticular pins for ventral AAJ fixation in a specimen. Pin placement was started close to the midline on the cranioventral surface of C2, caudal to the AA articulation. Pins were then directed craniolaterally through the AAJ and placed into the caudal body of the atlas at an approximate 30° angle to the transverse plane. The protruding ends of the pins were subsequently cut off (pins were left untrimmed in the photograph to improve their visibility).
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph (A) and diagram (B) showing placement of 2 transarticular pins for ventral AAJ fixation in a specimen. Pin placement was started close to the midline on the cranioventral surface of C2, caudal to the AA articulation. Pins were then directed craniolaterally through the AAJ and placed into the caudal body of the atlas at an approximate 30° angle to the transverse plane. The protruding ends of the pins were subsequently cut off (pins were left untrimmed in the photograph to improve their visibility).
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
For the augmented ventral transarticular pin fixation, the previously described transarticular pin fixation was combined with a wire fixation. The two 1.25-mm K-wires were placed as previously described. A hole was then drilled through the body of C2 in the transverse plane. Two cortical 2.0-mm screws were placed in the ventral aspect of the medial margin of both alae atlantis. Then a 0.5-mm wire was passed through the drilled hole in C2, brought around both screws, and twisted (Figure 4).
Photograph (A) and diagram (B) depicting the placement of transarticular pins, cortical screws, and wire for the augmented ventral transarticular pin fixation technique in a specimen.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph (A) and diagram (B) depicting the placement of transarticular pins, cortical screws, and wire for the augmented ventral transarticular pin fixation technique in a specimen.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph (A) and diagram (B) depicting the placement of transarticular pins, cortical screws, and wire for the augmented ventral transarticular pin fixation technique in a specimen.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Biomechanical testing
The occipital bone and the caudal aspect of the vertebral column segment from C4 through C5 of each specimen were partially embedded in PMMA that was placed in special molds that were used to attach the specimen to the materials testing machine. A purpose-built shear testing device driven by a uniaxial servohydraulic materials testing machinee was used (Figure 5). The PMMA-embedded portion of the cranial aspect of the specimen was mounted onto a vertical slide attached to the actuator of the testing machine, and the PMMA-embedded portion of the caudal aspect was fixed to a horizontal slide, which was adjustable to avoid compression or extension of the specimen.2, 15 The actuator induced constant shear motion in a dorsoventral direction. A load cell measured the applied force, and the displacement sensor acquired the linear displacement values. This information was then transmitted to a computer for recording.
Photograph of the setup for biomechanical testing with a purpose-built shear testing device driven by a uniaxial servohydraulic materials testing machine. The PMMA-embedded portion of the cranial aspect of the specimen was mounted onto a vertical slide attached to the actuator, and the PMMA-embedded portion of the caudal aspect was fixed to an adjustable horizontal slide. The actuator induced constant shear motion in a dorsoventral direction.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph of the setup for biomechanical testing with a purpose-built shear testing device driven by a uniaxial servohydraulic materials testing machine. The PMMA-embedded portion of the cranial aspect of the specimen was mounted onto a vertical slide attached to the actuator, and the PMMA-embedded portion of the caudal aspect was fixed to an adjustable horizontal slide. The actuator induced constant shear motion in a dorsoventral direction.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Photograph of the setup for biomechanical testing with a purpose-built shear testing device driven by a uniaxial servohydraulic materials testing machine. The PMMA-embedded portion of the cranial aspect of the specimen was mounted onto a vertical slide attached to the actuator, and the PMMA-embedded portion of the caudal aspect was fixed to an adjustable horizontal slide. The actuator induced constant shear motion in a dorsoventral direction.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
The appropriate loading limits were determined in a preliminary test that was performed on an additional specimen from a cadaveric toy-breed dog (a female Yorkshire Terrier). After each of the 4 AAJ fixation techniques, we tested this specimen under different loading limits and looked for obtention of a sigmoid-shaped load-deformation curve2, 15 with the AAJ ligaments intact and the ligaments transected to destabilize the AAJ (Figure 6). The results of this test were not included in the final statistical analysis; a loading limit of 15 N was selected for use with the study sample.
Graphic showing sigmoid-shaped load-deformation curves for the AAJ in a single specimen that was used to determine shear load limits subsequently applied for the study sample. Testing was performed with the setup shown in Figure 5 with the AAJ ligaments intact (intact), after the ligaments were transected (defect), and after fixation with each stabilization technique. Positive displacement represents motion in the dorsal direction. AVTP = Augmented ventral transarticular pin. DC = Dorsal clamp. DW = Dorsal wire. VTP = Ventral transarticular pin.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Graphic showing sigmoid-shaped load-deformation curves for the AAJ in a single specimen that was used to determine shear load limits subsequently applied for the study sample. Testing was performed with the setup shown in Figure 5 with the AAJ ligaments intact (intact), after the ligaments were transected (defect), and after fixation with each stabilization technique. Positive displacement represents motion in the dorsal direction. AVTP = Augmented ventral transarticular pin. DC = Dorsal clamp. DW = Dorsal wire. VTP = Ventral transarticular pin.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Graphic showing sigmoid-shaped load-deformation curves for the AAJ in a single specimen that was used to determine shear load limits subsequently applied for the study sample. Testing was performed with the setup shown in Figure 5 with the AAJ ligaments intact (intact), after the ligaments were transected (defect), and after fixation with each stabilization technique. Positive displacement represents motion in the dorsal direction. AVTP = Augmented ventral transarticular pin. DC = Dorsal clamp. DW = Dorsal wire. VTP = Ventral transarticular pin.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
For each specimen, 3 cycles of motion in a dorsoventral direction were performed with a constant speed of 0.33 mm/s up to a shear load of 15 N as described for the preliminary test specimen, with the fixation techniques performed in random order. The first 2 cycles were used to precondition the sample, and the third cycle was analyzed as described in previous studies.2,15,19 Range of motion was defined as the total displacement between the loading limits of 15 N in the dorsal and ventral directions (recorded as 15 N in the dorsal direction and–15 N in the ventral direction). Specimens were excluded from the analysis if abnormalities or complications occurred during the biomechanical testing.
A load-to-failure test was performed for each specimen with the fixation type that was applied last for the motion testing. This test was also performed with force applied in a dorsoventral direction. The experimental design planned for 3 specimens to be tested to failure with each technique.
Statistical analysis
Statistical analysis was performed with 2 software packages.20,f,g Nonparametric tests were used because of the small sample size. The calculated ROMs were compared among specimens with intact AAJ ligaments, severed AAJ ligaments, and AAJs stabilized with each of the 4 fixation techniques by use of a Friedman test function21,f that allowed for crossover designs and multiple comparisons and by use of a pairwise Wilcoxon signed rank test function (that also corrects for multiple comparisons).22,f Values of P < 0.05 were considered significant for all analyses. Bonferroni multiple comparison corrections were applied in the context of an experimentwise type I error rate. Descriptive results were provided for load-to-failure testing.
Results
Sample
The 13 specimens were obtained from 6 Yorkshire Terriers (4 males and 2 females; median body weight, 1.60 kg [range, 1.30 to 2.13 kg]) and 7 Chihuahuas (5 males and 2 females; median body weight, 1.17 kg [range, 0.73 to 1.69 kg]). On CT and MRI examination, abnormalities were observed in the occipitoatlantoaxial area in 1 dog (a 1.12-kg female Chihuahua) that also had a fracture of the atlas. This specimen was excluded from the analyses.
Biomechanical testing
Two additional specimens were excluded from the final analyses. One specimen (from a 1.30-kg female Yorkshire Terrier) was excluded because the biomechanical test results after AAJ ligament transection were similar to the test results obtained when the ligaments were intact, likely indicating that transection of the ligaments was incomplete or insufficient to render the joint unstable during preparation for the former set of tests. A complication occurred during the ventral transarticular pin fixation for another specimen (from a 1.17-kg male Chihuahua); a fracture of the central surface of C2 occurred during pin placement, leading to increased instability. All of the biomechanical tests were performed for this specimen, but the results were not included in analyses. The final study sample comprised specimens from 10 cadavers (5 Yorkshire Terriers and 5 Chihuahuas).
There were significant differences in ROM among specimens with intact AAJ ligaments, severed AAJ ligaments, and AAJs stabilized with the 4 fixation techniques (Friedman test; P < 0.001) and between AAJs with intact and transected ligaments prior to stabilization procedures (pairwise Wilcoxon signed rank test; P = 0.029). The median ROM for specimens with intact and transected AAJ ligaments was 11.78 mm (range, 6.89 to 20.77 mm) and 13.09 mm (range, 9.18 to 20.91 mm), respectively (Figure 7). All AAJ stabilization techniques except for dorsal clamp fixation were associated with significantly (P = 0.029 for all comparisons by use of pairwise tests) decreased ROM, compared with the results for specimens with intact and transected ligaments. The median ROM after the dorsal wire and dorsal clamp fixation techniques was 9.63 mm (range, 5.65 to 11.76 mm) and 12.73 mm (range, 8.84 to 17.36 mm), respectively (P = 0.029). No significant (P = 0.21) difference in median ROM was found between the 2 ventral fixation techniques, although the median value was slightly lower for the augmented ventral transarticular pin fixation (8.78 mm [range, 3.90 to 11.91 mm]) than for the ventral transarticular pin fixation (9.11 mm [range, 3.68 to 13.06 mm]).
Box-and-whisker plots showing the ROM (total displacement between loading limits of 15 and–15 N in 1 testing cycle) for AAJs of 10 specimens (obtained from 5 Chihuahuas and 5 Yorkshire Terriers) during biomechanical testing with the AAJ ligaments intact, after transection of the ligaments, and after fixation with each stabilization technique. Boxes represent the interquartile (25th to 75th percentile) range, horizontal lines within boxes represent the median value, whiskers indicate the largest and smallest values within 1.5 times the interquartile range, and circles show outliers outside of that range. The brackets indicate significant (P = 0.029) differences between the tested conditions in pairwise comparisons. See Figure 6 for remainder of key.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Box-and-whisker plots showing the ROM (total displacement between loading limits of 15 and–15 N in 1 testing cycle) for AAJs of 10 specimens (obtained from 5 Chihuahuas and 5 Yorkshire Terriers) during biomechanical testing with the AAJ ligaments intact, after transection of the ligaments, and after fixation with each stabilization technique. Boxes represent the interquartile (25th to 75th percentile) range, horizontal lines within boxes represent the median value, whiskers indicate the largest and smallest values within 1.5 times the interquartile range, and circles show outliers outside of that range. The brackets indicate significant (P = 0.029) differences between the tested conditions in pairwise comparisons. See Figure 6 for remainder of key.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Box-and-whisker plots showing the ROM (total displacement between loading limits of 15 and–15 N in 1 testing cycle) for AAJs of 10 specimens (obtained from 5 Chihuahuas and 5 Yorkshire Terriers) during biomechanical testing with the AAJ ligaments intact, after transection of the ligaments, and after fixation with each stabilization technique. Boxes represent the interquartile (25th to 75th percentile) range, horizontal lines within boxes represent the median value, whiskers indicate the largest and smallest values within 1.5 times the interquartile range, and circles show outliers outside of that range. The brackets indicate significant (P = 0.029) differences between the tested conditions in pairwise comparisons. See Figure 6 for remainder of key.
Citation: American Journal of Veterinary Research 82, 10; 10.2460/ajvr.82.10.802
Single load-to-failure testing was performed for the last tested technique in each specimen (n = 10; 2 each for the dorsal clamp and augmented ventral transarticular pin fixations and 3 each for the dorsal wire and ventral transarticular pin fixations). Except for the dorsal clamp results, failures occurred most commonly in regions adjacent to the AAJ.
Failure following dorsal clamp fixation occurred as displacement of the caudal part of the clamp along the spinous process of C2 in one specimen and luxation C3-4 articulation in the other specimen. The median load at failure for the dorsal clamp fixation was 133.42 N (range, 93.61 to 173.25 N). Failure following the dorsal wire fixation occurred as a fracture of C3 in one specimen, a transverse occipital fracture in another specimen, and loosening of 1 transarticular atlantooccipital pin in another specimen. The median load at failure for the dorsal wire fixation was 122.11 N (range, 41.10 to 234.05 N).
Failure following ventral transarticular pin fixation occurred as luxation at the C2-3 articulation in 1 specimen, fracture of C3 in another specimen, and loosening of 1 transarticular pin in another specimen. The median load at failure following the ventral transarticular pin fixation was 141.48 N (range, 93.74 to 148.28 N). Failure following the augmented ventral transarticular pin fixation occurred as loosening of 1 atlantooccipital transarticular pin in one specimen and luxation at the C3-4 articulation in the other specimen. For this technique, the median load at failure was 95.99 N (range, 85.39 to 106.58 N).
Discussion
The dorsal wire, ventral transarticular pin, and augmented ventral transarticular pin fixation techniques used in the present study resulted in improvement of AAJI created by transection of the AAJ ligaments in vertebral column specimens from cadaveric toy-breed dogs. Specimens stabilized with these 3 techniques following ligament transection had similar biomechanical properties and greater rigidity, compared with results for the same specimens with intact and severed ligaments, under shear loading conditions.
In 2 specimens, the ROM measured with intact ligaments was substantially greater than those observed for the other specimens. This could have resulted from inadvertent damage to the AAJ capsule and stabilizing structures during specimen preparation, leading to an increased ROM before the remaining tests were performed.
Surgical stabilization of the AAJ is a challenge because of the complicated anatomy, mainly in toy breed dogs that have very fine and often immature bone structures.11,14,18,23 The main objective of surgical treatment for AAJI in dogs is to reduce subluxation and provide a rigid and permanent fixation of the AAJ.6,7,9,12 In human medicine, this is a highly developed field of investigation, and numerous biomechanical studies have been performed. However, in veterinary medicine, biomechanical studies are lacking, especially for toy-breed dogs.
A dorsal surgical approach was initially used to allow access to the joint and avoid potential iatrogenic injury to vital structures.24 However, because the articular surface remains intact with this approach, the long-term stability achieved by dorsal fixation is provided by fibrous scar tissue formation, and AAJ fusion cannot be achieved.6,7,12,18,23,24 Furthermore, dorsal stabilization techniques resist loading only in flexion.15 As there is no completely effective dorsal fixation method that resists rotational and shear forces acting on the AAJ,18 other displacements of the joint could delay the formation of fibrous scar tissue and increase the risk of fixation failure and subluxation.12
Although the dorsal wire fixation significantly reduced ROM during shear loading in the present study, several risks are associated with the technique. Perforation of the spinous process of the axis in toy-breed dogs can weaken the bone and lead to stabilization failure.3,5,18 Moreover, the wire could cut through the lamina of the atlas, especially in immature dogs with softer bones.5,18,24 In addition, in passing the wire, there is an important risk of spinal cord injury, and severe complications such as apnea or cardiac arrest have been described.3–5,14,23,24 Denny et al4 used this method to treat 13 dogs. At the end of surgery, 4 of these dogs experienced respiratory problems and cardiac arrest.4
To decrease the risk of iatrogenic spinal cord injury associated with passing the wire loop under the arch of the atlas, Pujol et al16 used Kishigami AA tension bands for treatment of AAJI in a study of 8 toy-breed dogs. The clinical results suggested that the use of this tension band results in stable and durable fixation for ≥ 1 year.16 Riedinger et al15 performed a biomechanical investigation of various implants, including a dorsal AAJ clamp similar to that used in the present study, in AAJs from Beagle cadavers; the clamp in that study was constructed as an adaptation of the Kishigami AA tension band, and the results indicated that this was the least stable of the evaluated techniques.
In the present study, the clamp used was also a modified Kishigami AA tension band, and it provided less stability of the AAJ than the dorsal wiring method, with no significant difference in ROM between specimens after severing of the AAJ ligaments and those stabilized with the dorsal clamp fixation technique. Whereas the study by Riedinger et al15 used specimens from Beagles, we performed the technique on specimens from toy-breed dogs with extremely fine spinous processes; we consider it possible that this anatomic difference did not allow the correct placement of the attachment spikes of the device across the bone in the present study. These findings were compatible with the suggestion by Riedinger et al15 that stability provided by use of a dorsal clamp may be considerably lessened if smaller implants are used in immature toy-breed dogs. Furthermore, during the load-to-failure tests in the present study, 1 failure of the modified Kishigami AA tension band fixation occurred by displacement of the part of the clamp attached to the spinous process. Therefore, although the modified Kishigami AA tension band prevents the risk of spinal cord injury owing to its placement on the dorsal arch,16 we consider that in the long term, this technique may be more at risk of failing and displacement of the clamp could lead to a lesion of the spinal cord resulting from reluxation, especially in toy-breed dogs that have very thin spinous processes. However, the risk of reluxation may depend on the timing of a potential failure. With this dorsal fixation technique, long-term stability might be provided by formation of fibrous tissue, and if failure takes place soon after implant placement, insufficient fibrous tissue formation could lead to an increased risk of reluxation. Furthermore, we described our dorsal clamp as a modified Kishigami AA tension band, but although the 2 implants are similar, their means of attachment is not the same. We have not made a direct comparison between these 2 implants and cannot draw valid conclusions about their respective stability in toy-breed dogs.
Unlike dorsal fixation, ventral stabilization techniques have the advantage of allowing access to the articular surface, with removal of the articular cartilage promoting bone fusion and permanent fixation of the AAJ.7,9,14,15,17,23 Various complications have been reported with ventral fixation techniques such as hemorrhage, lesions of the spinal cord, migration or breakage of implants, esophageal stricture, or tracheal injury.5,12–14,17 As an alternative to the traditional ventral midline approach and to minimize the amount of dissection around the trachea and larynx as a means to potentially reduce the risk of iatrogenic injury, a ventral parasagittal approach has been described that allows better surgical exposure and protection of vital structures during fixation.17 In 1981, Sorjonen and Shires9 described the ventral stabilization technique in which K-wires were placed bilaterally through the AAJ in dogs, and alternatives such as the combination of K-wires with PMMA have subsequently been described.6,12,13 Compared with transarticular screw fixation, the advantage of the use of K-wires is that it is possible to place additional K-wires to increase joint stability.13 However, a challenge in toy-breed dogs is the fragility of bone structures, which can lead to intraoperative complications during transarticular pin placement, such as small fractures of the central surface of C2 or pin fracture during transarticular insertion.13 In a retrospective study, Aikawa et al13 found that 31 of 103 positive-profile threaded fixation pins inserted during surgical treatment of AAJI in dogs had broken and that 16 of these were placed transarticularly; the pin failures were identified on follow-up radiographs 1 to 105 months after surgery.
In human medicine, numerous surgical studies25–27 have evaluated and compared 1-, 2- and 3-point posterior AAJ fixation techniques. Investigations have shown that 3-point fixation, combining bilateral transarticular fixation with posterior wire fixation (Gallie technique), provides the most effective resistance to minimize movement at the AAJ25–28 and results in a smaller median ROM value,26 compared with that for posterior wire fixation alone. The same results were found during axial rotation, which could make 3-point fixation the treatment of choice for AAJI.25–27 Regarding the failure rate with 3-point fixation in people, a clinical investigation by Henriques et al25 found that this was almost negligible. Transferring these findings to veterinary medicine suggests that the augmented ventral transarticular pin fixation technique described in the present study could be a useful alternative technique of AAJ fixation in dogs because it would represent a 3-point fixation. In our study, a fracture occurred in 1 specimen at the level of the central surface of C2 during the placement of transarticular K-wires. Although the results were excluded from statistical analyses, the biomechanical tests were still carried out with this specimen to subjectively assess the differences in stability among the tested techniques. After ventral transarticular pin placement, the ROM was reduced from 11.8 mm to 6.2 mm in this specimen. With the augmented pin fixation, the ROM was further reduced to 3.6 mm, suggesting that adding a wire to the transarticular fixation as described in this report might provide additional stability should a complication such as bone fracture or implant loosening occur.
Fixation with multiple implants distributes the forces, provides better stability, and reduces the risk of implant failure and related complications associated with reluxation of the AAJ.23 Platt et al14 retrospectively evaluated outcomes following multiple implant fixation with a ventral approach in 19 dogs, with 17 dogs evaluated 4 to 8 weeks after surgery. Neurologic examination revealed improvement in 16 of 17 dogs.4 Generally, fixation with multiple implants includes the use of PMMA, and the volume of PMMA used can be very large, depending on the technique. Although PMMA increases stabilization, results of some studies6,10,12,14,29 have revealed disadvantages of its use, including thermal damage, increased risk of infection, and necrosis of adjacent structures. In addition, the available bone surface is also less in toy-breed dogs than in larger dogs; thus, adding PMMA could hinder movement cycles owing to the large implant size relative to the dog's anatomy. The augmented fixation technique in the study reported here allows the stability advantages of multiple implants but requires less surface area than those methods and avoids potential complications associated with PMMA.
Many previous studies3–5,10,12 have compared the success of dorsal and ventral surgical AAJ stabilization approaches and evaluated the outcomes of various techniques in dogs. Thomas et al5 evaluated the outcomes of 7 dogs that underwent dorsal wire fixation. Failure in 5 of 7 dogs resulted from wire breakage or fracture of the dorsal arch. These results were consistent with the findings of Denny et al,4 who suggested that dorsal fixation techniques should probably be abandoned because of the high risks of perioperative death and relapse. Ventral stabilization techniques are also associated with complications such as implant migration or breakage. Schulz et al12 evaluated the management of AAJI with ventral transarticular pins and PMMA in 9 dogs and reported acute postoperative complications in 4 of 9 dogs. Another group evaluated the outcomes and complications associated with ventral fixation by use of screws, pins, and PMMA and identified complications in 4 of 12 dogs after surgery.10 A literature review3 of AA subluxation and its treatment in dogs described failure rates of 25% for dorsal fixation procedures and 18% for ventral fixation procedures.
Several authors have compared transarticular fixation with (Gallie technique–based) wire fixation of the AAJ in human patients or human specimens. Transarticular fixation provides substantial benefits, such as greater stability and a higher fusion success rate, compared with wire fixation techniques for the management of AAJI,25–28,30 and numerous clinical investigations and fatigue tests have been performed in human medicine. Investigators of 1 study31 reported that wire implants experienced fatigue and loosening after cyclic loading. On the basis of these results, they recommended adjuvant fixation to better stabilize the AAJ after wire fixation.31
The biomechanical objective of surgical AAJ stabilization is to achieve greater stiffness in the joint. Our data suggested that the dorsal wire fixation technique and both ventral pin fixation techniques resulted in greater biomechanical stability, compared with AAJ specimens with the ligaments intact or severed, and the dorsal wire and augmented ventral pin techniques had greater stabilization than that achieved by use of the dorsal clamp technique. The dorsal fixation techniques have been proposed as being less dangerous and easier than ventral fixation techniques in dogs.6,15,16 However, drawing a parallel with human medical findings suggests that the dorsal wire fixation would be more prone to fatigue over time, because the method frequently only induces pseudoarthrosis and problems with rigidity of the fixation can develop in the long term.
There were several limitations to the present study. As dorsoventral motion is assumed to be the greatest clinical importance in dogs with AAJI,2 the specimens in this study were tested only under shear loading conditions. No conclusions could be made about the biomechanical properties of these treatments under other movement conditions such as flexion, extension, lateral bending, or axial rotation. In addition, to increase statistical power, all 4 fixation techniques were performed and tested on each specimen. The bone quality may have been decreased by successive implant placement. However, the dorsal AAJ clamp and the dorsal wire fixation techniques did not interfere with the 2 ventral fixation techniques, and the 2 ventral fixation techniques did not interfere with each other. To reduce the potential bias resulting from reuse of each specimen, the order in which the techniques were performed and tested was randomly assigned. Another limitation was the ex vivo study design. During specimen preparation and biomechanical testing, it was possible that structural damage could have occurred, which could have influenced the results. Regarding the congelation effect, a study by Gleizes et al32 found no significant difference between frozen and fresh vertebral column segments from sheep in terms of amplitude and rigidity in biomechanical evaluations. The temperature at which testing was performed could also have played a role, as we tested each fixation technique at room temperature rather than body temperature, and this might have an influence on the performance of the metallic implants. Furthermore, the study was designed to test acute stability; long-term stability of the AAJ is normally achieved through bone fusion, which is facilitated by rigid internal fixation. Further studies that include cyclic testing would be needed to provide information that could represent long-term stability of the AAJ with each fixation technique. Finally, the small number of specimens may have also limited the results of the statistical analysis. However, the authors believe that these limitations did not negate the results found.
The dorsal wire fixation, ventral transarticular pin fixation, and augmented ventral transarticular pin fixation described in the present study had similar biomechanical properties and provided greater rigidity, compared with the dorsal AAJ clamp under shear loading after transection of the AAJ ligaments in toy-breed dogs. Extrapolation from other reports suggests that ventral stabilization would be less prone to fatigue over time and therefore would represent a more stable technique, compared with the dorsal wire fixation. Finally, no significant difference in acute biomechanical stability was found between the 2 ventral transarticular pin fixation techniques; however, the authors considered that the numerically lower median and interquartile (25th to 75th percentile) range data for ROM following the augmented technique suggested this could result in a more restricted ROM than the standard ventral transarticular pin fixation technique in some cases. We also believe that augmented fixation could represent a more stable technique in the long term, even though this was not found in the acute testing, because with the augmented technique, 2 different implants independently stabilize the AAJ. The transverse positioning of the wire within the cranial part of the axis might provide a better bony purchase and potentially reduce the risk of failure over time. Further biomechanical tests with load cycling and clinical studies are needed to assess the long-term stability of the augmented ventral transarticular pin fixation before recommendations can be made for the use of this procedure in dogs.
Acknowledgments
The biomechanical tests were performed at ARTORG Center for Biomedical Engineering Research, University of Bern.
The dorsal AAJ clamps were donated by Eickemeyer.
The study was supported by a grant of the Albert-Heim Stiftung.
The authors declare that there were no conflicts of interest.
The authors thank Angela Beugger for technical assistance.
Abbreviations
| AA | Atlantoaxial |
| AAJ | Atlantoaxial joint |
| AAJI | Atlantoaxial joint instability |
| K-wire | Kirschner wire |
| PMMA | Polymethylmethacrylate |
| ROM | Range of motion |
Footnotes
Philips Brilliance CT 16-slice scanner, Philips AG Healthcare, Zürich, Switzerland.
Panorama HFO 1.0T, Diamond Select MR System, Philips AG Healthcare, Zürich, Switzerland.
Cerclage Draht, Provet AG, Lyssach, Switzerland.
Donated by Eickemeyer, Tuttlingen, Germany.
858 Mini Bionix, MTS Systems Corp, Eden Prairie, Minn.
NCSS Statistical Software, NCSS LLC, Kaysville, Utah.
R Core Team (2014). R: A language and environment for statistical computing, version 4.0.2, R Foundation for Statistical Computing, Vienna, Austria. Available at: r-project.org. Accessed Mar 31, 2021.
References
- 1. ↑
Planchamp B, Bluteau J, Stoffel MH, et al. Morphometric and functional study of the canine atlantoaxial joint. Res Vet Sci 2020;128:76–85.
- 2. ↑
Reber K, Bürki A, Vizcaino Reves N, et al. Biomechanical evaluation of the stabilizing function of the atlantoaxial ligaments under shear loading: a canine cadaveric study. Vet Surg 2013;42:918–923.
- 3. ↑
McCarthy RJ, Lewis DD, Hodgood G. Atlantoaxial subluxation in dogs. Compend Contin Educ Pract Vet 1995;17:215–226.
- 4. ↑
Denny HG, Gibbs C, Waterman A. Atlanto-axial subluxation in the dog: a review of 30 cases and an evaluation of treatment by lag screw fixation. J Small Anim Pract 1988;29:37–47.
- 5. ↑
Thomas WB, Sorjonen DC, Simpson ST. Surgical management of atlantoaxial subluxation in 23 dogs. Vet Surg 1991;20:409–412.
- 6. ↑
Beaver DP, Ellison GW, Lewis DD, et al. Risk factors affecting the outcome of surgery for atlantoaxial subluxation in dogs: 46 cases (1978–1998). J Am Vet Med Assoc 2000;216:1104–1109.
- 7. ↑
Havig ME, Cornell KK, Hawthorne JC, et al. Evaluation of nonsurgical treatment of atlantoaxial subluxation in dogs: 19 cases (1992–2001). J Am Vet Med Assoc 2005;227:257–262.
- 8. ↑
Hansen SC, Bacek LM, Kuo KW, et al. Traumatic atlantoaxial subluxation in dogs: 8 cases (2009–2016). J Vet Emerg Crit Care (San Antonio) 2019;29:301–308.
- 9. ↑
Sorjonen DC, Shires PK. Atlantoaxial instability: a ventral surgical technique for decompression, fixation and fusion. Vet Surg 1981;10:22–29.
- 10. ↑
Sanders SG, Bagley RS, Silver GM, et al. Outcomes and complications associated with ventral screws, pins, and polymethylmethacrylate for atlantoaxial instability in 12 dogs. J Am Anim Hosp Assoc 2004;40:204–210.
- 11. ↑
Dickomeit M, Alves L, Pekarkova M, et al. Use of a 1.5 mm butterfly locking plate for stabilization of atlantoaxial pathology in three toy breed dogs. Vet Comp Orthop Traumatol 2011;24:246–251.
- 12. ↑
Schulz KS, Waldron DR, Fahie M. Application of ventral pins and polymethylmethacrylate for the management of atlantoaxial instability: results in nine dogs. Vet Surg 1997;26:317–325.
- 13. ↑
Aikawa T, Shibata M, Fujita H. Modified ventral stabilization using positively threaded profile pins and polymethylmethacrylate for atlantoaxial instability in 49 dogs. Vet Surg 2013;42:683–692.
- 14. ↑
Platt SR, Chambers JN, Cross A. A modified ventral fixation for surgical management of atlantoaxial subluxation in 19 dogs. Vet Surg 2004;33:349–354.
- 15. ↑
Riedinger B, Bürki A, Stahl C. Biomechanical evaluation of the stabilizing function of three atlantoaxial implants under shear loading: a canine cadaveric study. Vet Surg 2015;44:957–963.
- 16. ↑
Pujol E, Bouvy B, Omaña M, et al. Use of the Kishigami atlantoaxial tension band in eight toy breed dogs with atlantoaxial subluxation. Vet Surg 2010;39:35–42.
- 17. ↑
Shores A, Tepper LC. A modified ventral approach to the atlantoaxial junction in the dog. Vet Surg 2007;36:765–770.
- 18. ↑
Sánchez-Masian D, Lujan-Feliu-Pascual A, Font C, et al. Dorsal stabilization of atlantoaxial subluxation using non-absorbable sutures in toy breed dogs. Vet Comp Orthop Traumatol 2014;27:62–67.
- 19. ↑
Forterre F, Precht C, Riedinger B, et al. Biomechanical properties of the atlantoaxial joint with naturally occurring instability in a toy breed dog. Vet Comp Orthop Traumatol 2015;28:355–358.
- 21. ↑
RDocumentation. Friedman test. Available at: rdocu mentation.org/packages/stats/versions/3.6.2/topics/friedman.test. Accessed Mar 31, 2021.
- 22. ↑
RDocumentation. Wilcox test. Available at: rdocumenta tion.org/packages/stats/versions/3.6.2/topics/pairwisewilcox.test. Accessed Mar 31, 2021.
- 24. ↑
Stalin C, Gutierrez-Quintana R, Faller K, et al. A review of canine atlantoaxial joint subluxation. Vet Comp Orthop Traumatol 2015;28:1–8.
- 25. ↑
Henriques T, Cunningham BW, Olerud C, et al. Biomechanical comparison of five different atlantoaxial posterior fixation techniques. Spine 2000;25:2877–2883.
- 26. ↑
Richter M, Schmidt R, Claes L, et al. Posterior atlantoaxial fixation: biomechanical in vitro comparison of six different techniques. Spine 2002;27:1724–1732.
- 27. ↑
Sim HB, Lee JW, Park JT, et al. Biomechanical evaluations of various C1–C2 posterior fixation techniques. Spine (Phila) 2011;36:E401–E407.
- 28. ↑
Reilly TM, Sasso RC, Hall PV. Atlantoaxial stabilization: clinical comparison of posterior cervical wiring technique with transarticular screw fixation. J Spinal Disord Tech 2003;16:248–253.
- 29. ↑
Martinez SA, Arnoczky SP, Flo GL, et al. Dissipation of heat during polymerization of acrylics used for external skeletal fixator connecting bars. Vet Surg 1997;26:290–294.
- 30. ↑
Oda I, Abumi K, Sell LC, et al. Biomechanical evaluation of five different occipito-atlanto-axial fixation techniques. Spine 1999;24:2377–2382.
- 31. ↑
Dickman CA, Crawford NR, Paramore CG. Biomechanical characteristics of C1–2 cable fixations. J Neurosurg 1996;85:316–322.
- 32. ↑
Gleizes V, Viguier E, Feron JM, et al. Effects of freezing on the biomechanics of the intervertebral disc. Surg Radiol Anat 1998;20:403–407.
