Introduction

Isolated injuries of the short collateral ligaments of the tarsus are relatively rare in dogs, and information on the clinical behavior and treatment of these injuries is sparse in veterinary textbooks. There are three lateral collateral ligaments of the canine tarsocrural joint. The more superficial long lateral collateral ligament courses from the lateral malleolus of the fibula to the base of the fourth tarsal bone and fifth metatarsal bone with an attachment to the distal part of the calcaneus. The two short lateral collateral ligaments (SLCLs) course deep and in near right angles to the long ligament, the fibulocalcaneal ligament, and the fibulotalar ligament.¹ Although the long collateral ligaments stabilize the tarsocrural joint in extension, the short collateral ligaments are described as predominantly providing tarsocrural joint stability in flexion.² The Manual of Small Animal Surgery by Fossum states that with complete subluxations, the paw deviates to the opposite direction of the ligamentous damage, but it does not provide further description as to whether this relates to extension only or also flexion.²

Carmicheal and Marshall describe the fibulocalcaneal ligament as the substantial short lateral part; however, they state this ligament as taut in extension, whereas other authors have stated this ligament as taut in flexion.^3,4 The textbook by Brinker et al. states that there is only a moderate instability of the tarsocrural joint in cases with isolated rupture of the short collaterals; however, further explanation is lacking.⁵ Diagnosis of SLCL injury is described as demonstration of laxity with internal rotation of a flexed hock in the textbook by Slatter.⁶

Aron and Purinton performed a study in 1985 that looked at the anatomy and function of the tarsal ligaments by means of implanted pins and trigonometric measurements of the distances formed between the pins.⁷ In their study, the long collateral ligaments were always transected first and there was no isolated evaluation of the short collateral ligaments with the long ligaments intact. They described a 2-fold increase in external rotation and a 2.2-fold increase in internal rotation after transection of the lateral collateral ligaments in the flexed hock, as well as a 10-fold increase in valgus and 20-fold increase in varus angulation. On the other hand, transection of the medial collateral ligaments led to a 2.7-fold increase in external rotation, a 1.5-fold increase in internal rotation, a 34-fold increase in valgus angulation, and a 13-fold increase in varus angulation. This should imply that rupture of the ligaments leads predominantly to increased deviation of the paw in the direction opposite the ligamentous damage, namely, varus and internal rotation for the lateral collateral ligaments and valgus and external rotation for the medial collateral ligaments, even with the hock in flexion, which goes in accordance with the description in the textbook by Slatter.⁶

We have encountered cases with an instability of the tarsocrural joint only in flexion; however, attempts to repair the ligaments in accordance with functional statements made in the textbook by Slatter and the study by Aron and Purinton have not resulted in satisfactory outcomes in our hands. Therefore, the purpose of this study was to test the function of each of the SLCLs in flexion and with an intact long ligament under applied external and internal rotation as well as valgus and varus stress. A review of the anatomy led to our first hypothesis, that transection of the SLCLs would result in not only increased varus angulation, but also in increased external rotation, which is in contrast to the statements in the textbook by Slatter.⁶ Our second hypothesis was that both ligaments are major stabilizers in flexion, with neither being more important than the other.

Material and Methods

Eight canine cadaveric hind limbs from dogs weighing 10.4–62.8 kg (mean 46.04 kg) were harvested after euthanasia for reasons unrelated to this study. The mean age at euthanasia was 9.5 yr. The limbs were disarticulated at the level of the stifle joint and frozen at –18°C. They were thawed at 10°C for 48 hr before evaluation. Only limbs with no signs of tarsal osteoarthritis, as confirmed by radiographs, were used. Four limbs (two right and two left) were assigned to one of two groups. In group A, the fibulocalcaneal ligament was transected first, followed by the fibulotalar ligament. In group B, the order of transection was reversed.

The skin was removed from all specimens from the level of the stifle to just proximal to the main foot pad. One 3.0 mm k-wire was inserted in the distal tibia above the malleoli and a second parallel to the first in the proximal tarsal bones (central and quartal tarsal bones). Correct placement was confirmed by radiographs. The tibia was fixed to a wooden post in a vice, parallel to the table, with two proximally applied screws (size depending on size of tibia). To exclude any influence by the long collateral ligament, the tarsocrural joint was held at a 90° angle (confirmed by a goniometer) by a suture (3/0 Nylon) applied to the skin of the third and fourth toe and suspended to a steady hook 80 cm above the tarsal joint to enable free movement of the foot. Tension of the suture was such that the tarsal joint was held at an exact 90° angle without lifting the paw and was dependent on dog size. All limbs were evaluated with intact ligaments, after transection of the first ligament, and after transection of both ligaments. Valgus and varus angulation were evaluated in the frontal plane and external and internal rotation in the axial plane. For combined evaluation of valgus + external rotation, varus + internal rotation, valgus + internal rotation, and varus + external rotation, measurements were taken in both planes. A camera^a was mounted directly above the tarsus of the specimen at a distance of 80 cm for evaluation in the frontal plane and horizontally, at the level of the tibia, 80 cm from the calcaneus for evaluation in the axial plane. The distal pin was stressed with manual pressure until maximum range of motion was reached and the applied forces were measured with a handheld Newton meter that was attached to the pin between two nuts. Each directional force was repeated three times, and photographs were taken each time. Figures 1 and 2 provide examples of specimens tested in frontal and axial planes. The angles formed between the two pins were measured on the photographs using image evaluation software^b and recorded as the “angular value.” These were compared with the angles formed in the intact specimen. Measurements were performed on each of the three photographs, and the mean values were used for statistical analysis. Mean values were compared with statistical standard software^c after normality of the distributions was confirmed with Kolmogorov-Smirnov tests. Paired t tests were used to compare the intact with the transected specimen values and unpaired t tests to compare groups A and B. The significance level was set at P ≤ 0.05.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Results

The most significant finding was that external rotation increased significantly when both short collateral ligaments were transected (P < .0005) as well as when only one of the ligaments was transected (P = .009). There was a significant difference between group A and B in regard to which ligament was cut first, with higher angular values in group A (P < .0005). The values for the specimen with both ligaments transected were generally higher in group A, although not statistically significant (P = .06).

Varus angulation was also significantly increased with both (P = .001) and only one of the ligaments transected (P = .021), but there was no difference between groups A and B (P = .38). No significant changes could be detected when internal rotational and valgus forces were applied. Tables 1 and 2 show the exact values.

TABLE 1 Angular Values in Degrees and Related P Values

View Fullscreen

TABLE 2 Angular Values in Degrees for Groups A and B and Related P Values Comparing the Groups

View Fullscreen

For the combined axial and frontal plane force evaluations, the biggest change was detected when varus and external rotation force were combined, as would be expected based on the results of the singular forces. When both ligaments were cut, external rotational force alone caused a lateral subluxation of the foot beyond the lateral malleolus. This subluxation could be increased with a combined external rotation and varus force; however, varus force alone did not lead to subluxation. External rotation combined with a valgus force led to reduction of the subluxated foot.

The fibulotalar ligament was found to be wider and thicker than the fibulocalcaneal (Figure 3). It was necessary to incise the joint capsule in order to cut the fibulotalar ligament completely because of its intra-articular location.

View Figure

• Download as Powerpoint
• Download Figure

The force that was necessary to achieve the end of range of motion was 1.25–1.5 N in all samples with one exception: the temporary external rotational force required to create subluxation after transection of both ligaments ranged up to 5 N. This appeared to be necessary in order to lift the trochlea of the talus over the lateral malleolus. After subluxation was achieved, the force could be reduced to 1.5 N again without reduction of the foot.

The overall standard deviation of all measurements of the repeated trials was 0.39°.

Discussion

This is the first study to evaluate the function of the SLCLs of the canine tarsus in isolation from the long collateral ligaments. The results show that an isolated rupture of the SLCLs leads to subluxation of the foot toward lateral, which is in contrast to what has been described in veterinary textbooks.⁶ To avoid incorrect treatment decisions, it is important for the clinician to recognize that this is the opposite of the situation in cases of rupture of the long collateral ligament. To the best of our knowledge, no studies have further analyzed this important clinical difference so far.

Our first hypothesis was confirmed. Whereas an increase in varus angulation after transection of the SLCLs goes in accordance with the published literature, we could also show an increase in external rotation, which is the opposite of what has been previously described in textbooks and other studies. It is also the opposite of the clinical behavior in cases of a rupture of a long lateral collateral ligament, in which the deviation is always medial, that is, to the direction opposite the ligamentous damage.^2,6,7 When both ligaments were transected, we could detect both an increase in external rotation and a subluxation toward lateral, that is, the side of ligamentous damage. If one considers the direction of fibers of the short collateral ligaments, they run orthogonal to the fibers of the long collateral ligament from dorsal to plantar.¹ This explains the short collateral ligaments’ stabilizing function in flexion rather than extension. Taking a closer look at the anatomy and their origin and insertion points, one can realize that external rotation actually increases the distance between these points, whereas internal rotation would bring them closer together. The same applies to varus angulation, in which the distance between origin and insertion points of the ligaments is increased. Although this explains our findings, it makes the controversy regarding the direction of subluxation in the existing literature even more surprising.

Subluxation toward lateral was reported in a clinical case series by Sjöström et al. in 1994.⁸ However, this fact has never been further evaluated by functional studies, nor has this important difference and its clinical implications been pointed out by textbooks.

Studies on tarsal instabilities have often focused on instabilities in extension but not in flexion. Major veterinary orthopedic textbooks contain contradictory information regarding the importance and function of each of the ligaments in flexion. The textbook by Tobias and Johnston describes the fibulocalcaneal ligament as the substantial portion, being taut in extension.³ This contradicts Aron and Purinton, who describe this ligament as taut in flexion, showing that the true function of the short collateral ligaments in stabilizing the canine tarsus has not been fully understood until now.⁷ This study helps to elucidate their function.

Our second hypothesis, that both the fibulotalar and the fibulocalcaneal collateral ligaments are major stabilizers of the tarsocrural joint in flexion, preventing mainly external rotation and lateral subluxation, was confirmed. Subluxation was only possible when both ligaments were transected, proving neither of the ligaments is more important than the other, in contrast to what has been published in the textbook by Tobias et al.³ Looking at the anatomy, the fibulotalar ligament, which lies intra-articular and originates from the caudomedial side of the lateral malleolus, provides a tight connection between the fibula and the talus. Loss of this tight connection enables separation of these bones and is necessary for subluxation to occur. However, subluxation is still prevented by the fibulocalcaneal ligament as long as it is intact. For clinical cases with subluxation, it can therefore be expected that there is always damage to both the fibulotalar and the fibulocalcaneal ligament.

When external rotational force was applied, the angular values were higher in group A after transection of only the fibulocalcaneal ligament compared with the values in group B after transection of the fibulotalar ligament. However, the values appeared to be generally higher in group A after transection of both ligaments, although this did not reach statistical significance (P = 0.06). Potentially, this result could have been influenced by the variability in cadaver sizes, as small dogs appeared to have higher laxity in their joints. Therefore, this finding should be interpreted with caution, and further studies should clarify whether the fibulocalcaneal ligament contributes slightly more to the prevention of external rotation than the fibulotalar.

Incision of the joint capsule was necessary in order to transect the fibulotalar ligament completely. This may have contributed to further instability; however, the incision was always kept to a minimum, and the area of incision was only directly over the ligament itself.

We found the fibulotalar ligament to be thicker and wider than the fibulocalcaneal ligament, contradicting the description by Aron and Purinton.⁷ Potentially, this ligament might not have been transected completely in their study because its intra-articular location underneath the lateral malleolus makes access and transection difficult. This is one potential contributing factor to explain why the descriptions by Aron and Purinton, on which many textbook descriptions are based, differ from our findings. The main difference, however, is that in all of their specimens, the long collateral ligament had been transected before the short collaterals were approached.

A limitation of this study may be that the applied forces were standardized to reaching maximum range of motion rather than a set force. Manual stress appeared to offer a better control for the endpoint and prevented the force from being exaggerated, especially with the different cadaver sizes. In their study on elbow luxation, Farrell et al. also showed that applying manual force in order to reach the end of range of motion was most appropriate and highly repeatable.⁹ The overall standard deviation of the measurements of the repeated trials was only 0.39° and showed a very high repeatability and low variability between trials. We tried to further overcome this limitation by measuring the force that was necessary to reach the endpoint of range of motion, which showed that it was better not to standardize the applied force. If the force had been standardized to, for example, 2 N, the subluxation that was detected upon external rotation after transection of both ligaments would have been missed because a temporary force of up to 5 N was necessary to achieve this subluxation in order to lift the trochlea of the talus over the lateral malleolus. The forces that were necessary to cause subluxation were very small, showing that the instability that is caused by rupture of the short collateral ligaments is severe.

A further limitation is that a broad range of dog sizes and breeds was used based on the availability of cadavers. Different breeds of dogs may have different laxity in their joints and therefore different angular values of valgus/varus and external/internal rotation. Thus, the absolute values detected in our study should not be used as reference points; instead, further studies should evaluate these parameters for different breeds and sizes of dogs.

Conclusion

The results of this study show that an isolated rupture of the SLCLs of the canine tarsus leads to subluxation of the foot toward lateral, namely, external rotational and varus instability. This instability is minor with rupture of only one of the ligaments but major with rupture of both, as this causes a severe lateral subluxation, especially when external rotation and varus forces are combined. This important difference to the situation following rupture of the long lateral collateral ligament, in which subluxation is toward medial, must be kept in mind when treating clinical patients in order to avoid treatment errors.