Reconstruction of the lateral ankle A biomechanical
ligaments
analysis*
MARK R.
COLVILLE,†‡ MD, RICHARD
A. MARDER,§ MD, AND BERTRAM ZARINS,∥ MD
From the
†Division of Orthopaedics and Rehabilitation, The Oregon Health Sciences University, Portland, Oregon, the §University of California at Davis, Department of Orthopaedics, Sacramento, California, and the ∥ Massachusetts General Hospital, Orthopaedic Surgery Service, Boston, Massachusetts
ABSTRACT
tunnels and
graft placement configuration is achieved.
The purpose of this study was to perform a biomechanical analysis of several commonly performed operative procedures used to stabilize the lateral ankle. We performed the Evans, Watson-Jones, and Chrisman-Snook procedures on 15 cadaveric ankles and tested the ankles for stability, motion, and isometry of graft placement. The Evans procedure allowed increased anterior displacement, internal rotation, and tilt of the talus when compared to ankles with intact ligaments. Subtalar joint motion was restricted by the Evans procedure. The Watson-Jones procedure controlled internal rotation and anterior displacement of the talus, but was less effective in controlling talar tilt and also restricted subtalar joint motion. The ChrismanSnook procedure allowed increased internal rotation and anterior displacement of the talus when compared to ankles with intact ligaments. The procedure was effective in limiting talar tilt, but restricted subtalar joint motion. Based on the biomechanical data obtained, we devised a lateral ankle reconstruction with bone tunnels that reproduce the anatomic orientation of both the anterior talofibular and calcaneofibular ligaments. This ankle ligament reconstruction resists anterior displacement, internal rotation, and talar tilt without restricting subtalar joint motion. Clinical relevance: We found considerable mechanical differences among the more commonly performed lateral ankle reconstructions. It is possible to locate bone
so
that
a more
anatomic
Sprains of the lateral ankle ligaments can result in recurrent instability in up to 20% of untreated patients.2.4,5,l2.25,28 Many surgical procedures have been devised to treat recurrent lateral ankle instability; the multitude of operations suggest that no ideal method has been found. These operations can be divided into general categories: 1. Procedures that prevent inversion of the foot (see Fig. 2). Nilsonne,22 Evans,1° and Leel9 advocated tenodesis of the base of the fifth metatarsal to the lateral malleolus using ankle
the peroneus brevis tendon. This limits inversion at the ankle and subtalar joints. 2. Procedures that prevent inversion of the foot and reconstruct the anterior talofibular ligament (see Fig. 3). Watson-Jones29 routed the peroneus brevis tendon through bone tunnels in the distal fibula and talus to replace the anterior talofibular ligament. 3. Procedures that reconstruct both the anterior talofibular ligament and calcaneofibular ligament (see Fig. 4). Elmslie9 replaced the anterior talofibular and calcaneofibular ligaments using a free fascia lata graft. Chrisman and Snook6 described a variation of the Elmslie procedure using a
split portion of the peroneus brevis tendon as a graft. 4. Direct late repair of attenuated lateral ligaments with-
out
3
augmentation.3
In 1969, Chrisman and Snook compared the effectiveness of the Evans, Watson-Jones, and Chrisman-Snook procedures in preventing inversion of the foot and ankle .6 Biomechanical studies8, 14,15,2and clinical reports,3however, suggest that symptoms of ankle instability may be due to excessive anterior translocation and internal rotation of the talus as well as excessive talar tilt. Because the locations of
*
Presented in part at the AAOS annual meeting, San Francisco, California, January 25, 1987. t Address correspondence and reprint requests to: Mark R. Colville, MD, Division of Orthopaedics and Rehabilitation, The Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, OP13B, Portland, OR 97201.
594
595
the
anchoring points for the grafts in the most commonly performed procedures are so different, we thought that considerable differences in mechanical behavior should exist between the procedures. The purposes of this study were to compare the effects of commonly performed lateral ankle ligament reconstructions on the biomechanics of the ankle and subtalar joints, and to determine the optimum sites for graft and bone tunnel placement.
MATERIALS AND METHODS We tested 15 ankles we obtained from four cadavers and nine amputated limbs. The specimens were either fresh or had been frozen for less than 1 month. The average age of the patients at the time of amputation or death was 65 years (range, 28 to 81). The skin and subcutaneous tissues were removed from the ankle. The proximal fibula was fixed to the tibia with a screw in those specimens that had been amputated below the knee. The screw maintained the normal tension and proper orientation of the interosseous membrane. The peroneus longus tendon was removed and sutured to the base of the fifth metatarsal at the insertion of the peroneus brevis tendon. Both the peroneus longus and brevis tendons were split longitudinally in such a way that four tendons of equal cross-sectional area were available to use as grafts in performing the various procedures. The forefoot was amputated at the midmetatarsal level.
Figure
1.
Mounting and instrumentation of ankle preparations testing.
for biomechanical
The
specimens were mounted using three smooth Steinpins, 5 mm in diameter (Fig. 1). The pins were drilled through the tibia in a medial to lateral direction 10 to 15 cm above the ankle joint and secured in Hoffman clamps. Another 5-mm diameter Steinmann pin was drilled in an anteroposterior direction into the neck of the talus. Two Schanz screws, 6 mm in diameter, were drilled vertically through the calcaneus into the talus and crossing the subtalar joint. The subtalar joint was held in eversion during insertion of the pins. The preparation was then mounted rigidly onto the front frame of a Genucom Knee Analysis System (FARO, Lake Mary, FL). The Genucom is a computerized analysis system equipped with a six degrees of freedom electrogoniometer and a force plate. Angular changes were measured in degrees and translations measured in millimeters using the Genucom’s electrogoniometer, mann
which was rigidly mounted to the talus through the inferior Schanz screws. The electrogoniometric measurements are accurate to within 1 and 1 mm. Accurate and constant forces could not be applied by hand using the Genucom’s force plate measurement system. Therefore, known forces and angular torques were applied using a weight and pulley system. Anterior translational forces and rotational forces were applied directly to the talus through the anterior pin
(Fig. 1). Tunnels that were 4.5 mm in diameter were made in locations that would allow us to reproduce the configurations shown in Figures 2 through 5. Figure 2 represents the Evans reconstruction. Figure 3
Figure 2. Evans procedure subtalar joints.
to limit inversion of ankle and
represents the Watson-Jones reconstruction, and consists of anterior and posterior segments. The graft in Figure 3 is routed from a posterior to anterior direction through a hole in the fibula made at the level of the ankle joint and perpendicular to the fibular shaft. After exiting from the
596
bone tunnel sites (i.e., the lengths of the various grafts described above) were measured. Changes in length were measured on a suture fixed to one point and passed through the corresponding tunnel using a hand-held ruler with 1mm increments. Data were obtained from 15 ankles in all but the &dquo;anatomic&dquo; procedure, for which data were obtained from the final 6 consecutive ankles tested. Biomechanical
Figure 3. Watson-Jones procedure to limit inversion of ankle and subtalar joints and replace anterior talofibular ligament.
Figure 4. Chrisman-Snook procedure (split peroneus brevis tendon graft) to replace anterior talofibular and calcaneofibular ligaments. fibula, the graft is passed down through a vertical hole in the talar neck and doubled back upon itself. The ChrismanSnook procedure is shown in Figure 4. In this reconstruction the graft is passed posteriorly through the fibular tunnel. The graft is then routed from a posterior to anterior direction through a tunnel made in the calcaneus.
Isometry testing The ankles to 30°
were
moved from
a
plantar flexion. Changes
position of 20° dorsiflexion in distance between various
testing
The following tests were performed on 15 ankles in the order described: 1. An internal rotation moment of 2 N-m was applied and internal rotation was measured at 10° plantar flexion. 2. An anterior force of 30 N was applied and anterior translocation of the talus (anterior drawer) was measured. During the anterior drawer test the talus was allowed to rotate freely so that the intact deltoid ligament would not restrict anterior translation of the lateral talus. 3. An inversion moment of 2 N-m with the ankle in neutral dorsiflexion/plantar flexion was applied and the tilt of the talus was measured. 4. The Schanz screws were withdrawn until they no longer crossed the subtalar joint. The Steinmann pin transfixing the talonavicular joint was removed. An inversion moment of 2 N-m at neutral ankle position was applied. Total inversion was measured. Inversion occurring at the tibiotalar joint (talar tilt) was subtracted from total inversion to yield subtalar joint motion. The following ligaments and capsule were then cut: the anterior ankle capsule, anterior talofibular ligament, and the calcaneofibular ligament. After ligament transection, specimens were rendered grossly unstable to stress testing. No stress testing was performed on ankles with transected ligaments. This testing has been previously performed.23 The three ankle ligament reconstructions (as in Figs. 2 through 4) were then sequentially performed on each ankle using the previously prepared peroneal tendons attached to the base of the fifth metatarsal. Each reconstructed ankle was subjected to the tests described. After preliminary testing it became apparent that there were large differences between the three ligament reconstructions in terms of isometry and control of various ankle motions. Based on this information, we designed a fourth lateral ankle ligament reconstruction. The goals of this procedure were to achieve a more anatomic placement of the graft and to control all pathologic ankle laxities without limiting subtalar joint motion. In this anatomic reconstruction, a split portion of the peroneus brevis tendon was routed, as shown in Figure 5. Bone tunnels were designed so that graft placement duplicated the orientation of the anterior talofibular ligament and calcaneofibular ligament as closely as possible. The graft was passed from an anterior to posterior direction through a calcaneal tunnel that was oriented so that the posterior opening of the tunnel was at the anatomic attachment point of the calcaneofibular ligament to the calcaneus. The graft was then brought from a posterior to anterior direction
597
The length of the graft between the anterior fibular hole and the talus increased 7 mm (range, 4 to 8), or 17%, as the ankle was plantar flexed. Chrisman-Snook procedure (Fig. 4). The length of the graft between the base of the fifth metatarsal and the anterior fibular hole increased 11 mm (range, 8 to 15), or 15%. The length of the graft between the posterior fibular hole and the calcaneus decreased 12 mm (range, 8 to 18), or 24%. Anatomic procedure (Fig. 5). The length of the graft between the anterior fibular hole and the talus increased 3 mm (range, 2 to 4), or 6%, as the ankle was moved from 20° dorsiflexion to 30° plantar flexion. The length of the graft between the posterior fibular hole and the calcaneus decreased 3 mm (range, 2 to 6), or 10%, during the same motion. Biomechanical
Figure 5. Anatomic reconstruction reproduces orientation of anterior talofibular and calcaneofibular ligaments. through an oblique fibular tunnel; the openings of the tunnel are located as close as possible to the fibular attachment points of the calcaneofibular and anterior talofibular ligaments, respectively. The graft was next passed down through vertical tunnel that had been drilled in the talar neck. The graft was then doubled back upon itself. This ligament reconstruction was then tested in the same way as the other three procedures had been tested. All tendon grafts were sutured into place and secured to the bones with sutures that were passed through drill holes made adjacent to the bone tunnels to eliminate sliding of the graft. Low forces were applied (2 N-m) to prevent loss of fixation. We attempted to tension each reconstruction equally by holding the ankle in eversion and neutral dorsiflexion/plantar flexion as the reconstruction was performed. Results of the biomechanical tests were subjected to repeated measures analysis of variance testing followed by Newman-Keuls post-hoc testing.3° These statistical analyses enabled us to group ligament reconstructions that had similar results for each test performed, and to determine significant differences between the ligament reconstructions. a
testing
Internal rotation of the talus (Fig. 6). The two procedures that limited internal rotation of the talus most were the Watson-Jones and anatomic reconstructions. The average amount of internal rotation that occurred with stress testing of each was 4.4°. This was significantly different than the amount of internal rotation possible after the ChrismanSnook (mean talar rotation, 11.1°) and Evans (mean talar
rotation, 9.5°) procedures (P < 0.001). Anterior translocation of the talus (Fig. 7). The procedures that limited anterior translocation of the talus most were the Watson-Jones (3.4 ± 2 mm) and anatomic reconstructions (3.7 ± 1.1 mm). The Chrisman-Snook and Evans procedures allowed significantly greater anterior translocation (5.1 ± 1.4 mm and 4.9 ± 1.3 mm, respectively) (P < 0.001). The anatomic reconstruction was the only procedure that restored anterior translocation to a distance similar to ankles with intact ligaments. Talar tilt (Fig. 8). Talar tilt measured for ankles with intact ligaments was 4.9° ± 2.2°. The anatomic reconstruction allowed 5.2° ± 1.7° and the Chrisman-Snook procedure permitted 5.6° ± 1.8° talar tilt. These values are statistically
RESULTS
Isometry Evans procedure (Fig. 2). As the ankle was moved from 20° dorsiflexion to 30° plantar flexion, the length of the graft decreased 3 mm (range, 1 to 4). This represented a 5% change in length. Watson-Jones procedure (Fig. 3). The length of the portion of the graft between the base of the fifth metatarsal and the posterior fibular tunnel decreased 3 mm (range, 1 to 4), or 5%. There was no difference between the isometry of this graft and that of the Evans procedure.
Llgamems
Jones
anoorc
Figure 6. Internal rotation of the talus in relation to the tibia with intact ligaments and after the various reconstructions.
598
Figure 7. Anterior translocation of the lateral talus in relation to the tibia. Rotation of the talus was allowed so that the deltoid ligament did not restrict anterior translocation.
Figure joint.
8. Inversion of the talus in relation to the tibiotalar
similar to
one
another. Talar tilt
was
significantly greater
(P < 0.001) after the Watson-Jones (mean, 7.6° ± 4.2°) and Evans (mean, 9.4° ± 3.3°) procedures. Subtalar motion (Fig. 9). The anatomic ligament reconstruction allowed the greatest subtalar joint motion (15.8° ± 7.1°). This value was statistically similar to the amount of subtalar motion in ankles with intact ligaments (18° ± 7.9°). The amount of subtalar motion after the Watson-Jones 0
(7.0° ± 3.4°), Evans (6.2° ± 4.1°), and Chrisman-Snook (5.4° ± 2.9°) procedures was statistically similar. These three procedures significantly restricted subtalar motion com-
pared to ankles with intact ligaments. DISCUSSION The abnormal laxities that must be corrected to restore functional stability to an ankle are poorly understood.&dquo;, 13
Figure 9. Subtalar motion tilt from total inversion.
as
calculated
by subtracting
talar
This lack of knowledge is reflected in the biomechanical differences between the major types of ligament reconstructions. The Evans procedure is designed to limit talar tilt by limiting inversion of the foot.l° The Watson-Jones reconstruction re-creates the anterior talofibular ligament in addition to limiting inversion of the foot.29 Elmslie,9 Chrisman and Snook,’ and others’,&dquo;, 17,20,26 showed that patients who have chronic ankle instability have excessive talar inversion; they recommended reconstruction of both the anterior talofibular and calcaneofibular ligaments. Measurement of talar tilt has been used by many authors&dquo; 11,21 to establish indications for surgery in patients who have chronic instability and to evaluate results after reconstruction. Brostrom3 investigated 60 patients who had symptomatic ankle instability. He detected an increased anterior drawer sign at physical examination. At surgery he found that the anterior talofibular ligament was attenuated in all of these ankles, but the calcaneofibular ligament was less frequently involved. Based on these observations, he advocated direct reconstruction of the anterior talofibular ligament in all unstable ankles, and repair of the calcaneofibular ligament only if necessary. Strain gauge analysis of normal ankle ligaments con-
ducted by the authors’ and biomechanical testing performed by Johnson and Markolfl4 indicate that the anterior talofibular ligament is a primary restraint to anterior translocation, internal rotation, and inversion of the talus at all flexion angles tested. Brostrom3 failed to demonstrate a single patient with isolated rupture of the calcaneofibular ligament in his clinical investigation; Rasmussen23 found that the anterior talofibular ligament always failed before the calcaneofibular ligament during cadaveric stress tests. We believe that symptoms in patients who have recurrent ankle instability are caused by excessive internal rotation, anterior translocation, and tilt of the talus. These excessive motions are usually due to laxity of the anterior talofibular
599
in some ankles, also the calcaneofibular ligament. The ideal ligament reconstruction should recreate the anterior talofibular ligament and, if necessary, the calcaneo-
ligament and,
fibular ligament in an anatomic fashion without restricting subtalar joint motion. We found that proper bone tunnel placement was critical to achieve relative graft isometry. The limb of the graft in the Watson-Jones procedure that re-creates the anterior talofibular ligament can be made more nearly isometric by moving the fibular tunnel distally so that the graft exits the tunnel anteriorly at the attachment of the anterior talofibular ligament. Similarly, the transverse fibular tunnel described in the Chrisman-Snook procedure does not produce anatomic graft placement. An oblique fibular tunnel through the attachment sites of the anterior talofibular and calcaneofibular ligaments will result in nearly isometric graft
placement. We found large differences between the ligament reconstructions in their ability to control excessive ankle laxities. An important limitation to our study was the dependence of the stability of these reconstructions on our technique and
tensioning methods. Although we attempted to standardize the methods for performing the reconstructions, we were unable to measure the uniformity of our fixation and tensioning. We therefore felt that it was appropriate to stress test the reconstructions at relatively low forces. Due to limitations in our testing apparatus, we were unable to apply a simulated weightbearing load across the ankle joint. While applying such a load would more closely duplicate physiologic conditions, we feel the comparative data obtained relating to the effects of the various reconstructions is valid. The Evans procedure was least able to produce normal ankle and subtalar joint mechanics. The Evans procedure restored internal rotation of the talus to a degree similar to specimens with intact ligaments; however, anterior translocation and tilt of the talus were poorly controlled and subtalar joint motion was restricted. The Chrisman-Snook procedure was effective in limiting talar tilt and resulted in motion similar to ankles with intact ligaments. However, subtalar joint motion was restricted. The Watson-Jones procedure was most effective in reducing internal rotation and anterior translocation of the talus. Talar tilt was less well controlled, and was significantly increased (P < 0.001) when compared to laxity of ankles with intact ligaments. Subtalar joint motion was restricted after this procedure. The anatomic reconstruction that we described restores most closely the normal mechanics of the ankle and subtalar joints (Fig. 7). Internal rotation, talar tilt, and anterior translocation were all controlled well, and subtalar joint motion was not restricted. Clinical studies have shown that the Evans, WatsonJones, and Chrisman-Snook procedures are effective in patients with symptomatic lateral ankle instability.6.l0.29 Although we have shown that these procedures have significant mechanical differences, they have withstood the test of time, and each enjoys a high clinical success rate.
By modifying the fibular and talar tunnel placement, both the Watson-Jones and Chrisman-Snook procedures can be made more anatomic, thereby allowing more normal ankle mechanics. The Evans, Watson-Jones, and Chrisman-Snook procedures all restrict subtalar joint motion. In cases of combined ankle and subtalar instability, these procedures would be superior to procedures that reconstruct only the ankle ligaments. However, in patients with primary ankle instability, it has been our experience, and that of others, that excessive loss of subtalar motion may lead to chronic pain and functional difficulties on uneven terrain. 27 In most patients with lateral ankle ligament instability, direct reconstruction of the ligaments as described by Brostrom3 is possible, and there is no need for augmentation using tendon grafts. Occasionally, however, augmentation of attenuated tissue is required. From our study we conclude that it is possible to reconstruct the lateral ankle ligaments using tendon grafts in an anatomic fashion while preserving normal joint mechanics. REFERENCES 1. Anderson KJ, Lecocq JF: Operative treatment of injury to the fibular collateral ligament of the ankle. J Bone Joint Surg 36A: 825-832, 1954 2. Brand RL, Collins MDF, Templeton T: Surgical repair of ruptured lateral ankle ligaments. Am J Sports Med 9: 40-44, 1981 3. Broström L: Sprained ankles. VI. Surgical treatment of "chronic" ligament ruptures. Acta Chir Scand 132: 551-565, 1966 4. Broström L: Sprained ankles. V. Treatment and prognosis in recent ligament ruptures. Acta Chir Scand 132: 537-550, 1966 5. Cass JR, Morrey BF: Ankle instability: Current concepts, diagnosis, and treatment. Mayo Clin Proc 59: 165-170, 1984 6. Chrisman OD, Snook GA: Reconstruction of lateral ligament tears of the ankle. An experimental study and clinical evaluation of seven patients treated by a new modification of the Elmslie procedure. J Bone Joint Surg 51A: 904-912, 1969 7. Colville MR, Marder RA, Boyle JJ, et al: Strain measurement in lateral ankle ligaments. Am J Sports Med 18: 196-200, 1990 8. Dias LS: The lateral ankle sprain: An experimental study. J Trauma 19:
266-269, 1979 9. Elmslie RC: Recurrent subluxation of the
ankle-joint. Ann Surg 100: 364-
367, 1934 10. Evans DL: Recurrent instability of the ankle: A method of surgical treatment. Proc R Soc Med 46: 343-344, 1953 11. Freeman MAR: Instability of the foot after injuries to the lateral ligament of the ankle. J Bone Joint Surg 47B: 669-677, 1965 12. Freeman MAR: Treatment of ruptures of the lateral ligament of the ankle. J Bone Joint Surg 47B: 661-668, 1965 13. Freeman MAR, Dean MRE, Hanham IWF: The etiology and prevention of functional instability of the foot. J Bone Joint Surg 47B: 678-685, 1965 14. Johnson EE, Markolf KL: The contribution of the anterior talofibular ligament to ankle laxity. J Bone Joint Surg 65A: 81-88, 1983 15. Kleiger B: The mechanism of ankle injuries. J Bone Joint Surg 38A: 59-70, 1956 16. Laurin CA, Mathieu J: Sagittal mobility of the normal ankle. Clin Orthop
108: 99-104, 1975 17. Laurin CA, Ouellet R, St.-Jacques R: Talar and subtalar tilt: An experimental investigation. Can J Surg 11: 270-279, 1968 18. Leach RE, Namiki O, Paul GR, et al: Secondary reconstruction of the lateral ligaments of the ankle. Clin Orthop 160: 201-211, 1981 19. Lee HG: Surgical repair in recurrent dislocation of the ankle joint. J Bone Joint Surg 39A: 828-834, 1957 20. Leonard MH: Injuries to the lateral ligaments of the ankle. A clinical and experimental study. J Bone Joint Surg 31A: 373-377, 1949 21. McCullough CJ, Burge PD: Rotatory stability of the load-bearing ankle. An experimental study. J Bone Joint Surg 62B: 460-464, 1980 22. Nilsonne H: Making a new ligament in ankle sprain. J Bone Joint Surg 14:
380-381,1932 23. Rasmussen O:
Stability of
the ankle
joint. Analysis
of the function and
600
traumatology of the ankle ligaments. Acta Orthop Scand (Suppl 211) 56: 1-74,1985 24. Rubin G, Witten M: The talar-tilt angle and the fibular collateral ligaments: A method for the determination of talar tilt. J Bone Joint Surg 42A: 311326, 1960 25. Ruth CJ: The surgical treatment of injuries of the fibular collateral ligaments of the ankle. J Bone Joint Surg 43A: 229-239, 1961 26. Shephard E: Tarsal movements. J Bone Joint Surg 33B: 258-263, 1951 27. Snook GA, Chrisman OD, Wilson TC: Long-term results of the Chrisman-
Snook operation for reconstruction of the lateral ligaments of the ankle. J Bone Joint Surg 67A: 1-7, 1985 28. Staples OS: Ruptures of the fibular collateral ligaments of the ankle. Result study of immediate surgical treatment. J Bone Joint Surg 57A: 101-107, 1975 29. Watson-Jones R: Fractures and Other Bone Joint Injuries. Edinburgh, E & S Livingstone, 1940 30. Winer BJ: Statistical Principles in Experimental Design. New York, McGraw Hill, 1971
ligaments
analysis*
MARK R.
COLVILLE,†‡ MD, RICHARD
A. MARDER,§ MD, AND BERTRAM ZARINS,∥ MD
From the
†Division of Orthopaedics and Rehabilitation, The Oregon Health Sciences University, Portland, Oregon, the §University of California at Davis, Department of Orthopaedics, Sacramento, California, and the ∥ Massachusetts General Hospital, Orthopaedic Surgery Service, Boston, Massachusetts
ABSTRACT
tunnels and
graft placement configuration is achieved.
The purpose of this study was to perform a biomechanical analysis of several commonly performed operative procedures used to stabilize the lateral ankle. We performed the Evans, Watson-Jones, and Chrisman-Snook procedures on 15 cadaveric ankles and tested the ankles for stability, motion, and isometry of graft placement. The Evans procedure allowed increased anterior displacement, internal rotation, and tilt of the talus when compared to ankles with intact ligaments. Subtalar joint motion was restricted by the Evans procedure. The Watson-Jones procedure controlled internal rotation and anterior displacement of the talus, but was less effective in controlling talar tilt and also restricted subtalar joint motion. The ChrismanSnook procedure allowed increased internal rotation and anterior displacement of the talus when compared to ankles with intact ligaments. The procedure was effective in limiting talar tilt, but restricted subtalar joint motion. Based on the biomechanical data obtained, we devised a lateral ankle reconstruction with bone tunnels that reproduce the anatomic orientation of both the anterior talofibular and calcaneofibular ligaments. This ankle ligament reconstruction resists anterior displacement, internal rotation, and talar tilt without restricting subtalar joint motion. Clinical relevance: We found considerable mechanical differences among the more commonly performed lateral ankle reconstructions. It is possible to locate bone
so
that
a more
anatomic
Sprains of the lateral ankle ligaments can result in recurrent instability in up to 20% of untreated patients.2.4,5,l2.25,28 Many surgical procedures have been devised to treat recurrent lateral ankle instability; the multitude of operations suggest that no ideal method has been found. These operations can be divided into general categories: 1. Procedures that prevent inversion of the foot (see Fig. 2). Nilsonne,22 Evans,1° and Leel9 advocated tenodesis of the base of the fifth metatarsal to the lateral malleolus using ankle
the peroneus brevis tendon. This limits inversion at the ankle and subtalar joints. 2. Procedures that prevent inversion of the foot and reconstruct the anterior talofibular ligament (see Fig. 3). Watson-Jones29 routed the peroneus brevis tendon through bone tunnels in the distal fibula and talus to replace the anterior talofibular ligament. 3. Procedures that reconstruct both the anterior talofibular ligament and calcaneofibular ligament (see Fig. 4). Elmslie9 replaced the anterior talofibular and calcaneofibular ligaments using a free fascia lata graft. Chrisman and Snook6 described a variation of the Elmslie procedure using a
split portion of the peroneus brevis tendon as a graft. 4. Direct late repair of attenuated lateral ligaments with-
out
3
augmentation.3
In 1969, Chrisman and Snook compared the effectiveness of the Evans, Watson-Jones, and Chrisman-Snook procedures in preventing inversion of the foot and ankle .6 Biomechanical studies8, 14,15,2and clinical reports,3however, suggest that symptoms of ankle instability may be due to excessive anterior translocation and internal rotation of the talus as well as excessive talar tilt. Because the locations of
*
Presented in part at the AAOS annual meeting, San Francisco, California, January 25, 1987. t Address correspondence and reprint requests to: Mark R. Colville, MD, Division of Orthopaedics and Rehabilitation, The Oregon Health Sciences University, 3181 SW Sam Jackson Park Road, OP13B, Portland, OR 97201.
594
595
the
anchoring points for the grafts in the most commonly performed procedures are so different, we thought that considerable differences in mechanical behavior should exist between the procedures. The purposes of this study were to compare the effects of commonly performed lateral ankle ligament reconstructions on the biomechanics of the ankle and subtalar joints, and to determine the optimum sites for graft and bone tunnel placement.
MATERIALS AND METHODS We tested 15 ankles we obtained from four cadavers and nine amputated limbs. The specimens were either fresh or had been frozen for less than 1 month. The average age of the patients at the time of amputation or death was 65 years (range, 28 to 81). The skin and subcutaneous tissues were removed from the ankle. The proximal fibula was fixed to the tibia with a screw in those specimens that had been amputated below the knee. The screw maintained the normal tension and proper orientation of the interosseous membrane. The peroneus longus tendon was removed and sutured to the base of the fifth metatarsal at the insertion of the peroneus brevis tendon. Both the peroneus longus and brevis tendons were split longitudinally in such a way that four tendons of equal cross-sectional area were available to use as grafts in performing the various procedures. The forefoot was amputated at the midmetatarsal level.
Figure
1.
Mounting and instrumentation of ankle preparations testing.
for biomechanical
The
specimens were mounted using three smooth Steinpins, 5 mm in diameter (Fig. 1). The pins were drilled through the tibia in a medial to lateral direction 10 to 15 cm above the ankle joint and secured in Hoffman clamps. Another 5-mm diameter Steinmann pin was drilled in an anteroposterior direction into the neck of the talus. Two Schanz screws, 6 mm in diameter, were drilled vertically through the calcaneus into the talus and crossing the subtalar joint. The subtalar joint was held in eversion during insertion of the pins. The preparation was then mounted rigidly onto the front frame of a Genucom Knee Analysis System (FARO, Lake Mary, FL). The Genucom is a computerized analysis system equipped with a six degrees of freedom electrogoniometer and a force plate. Angular changes were measured in degrees and translations measured in millimeters using the Genucom’s electrogoniometer, mann
which was rigidly mounted to the talus through the inferior Schanz screws. The electrogoniometric measurements are accurate to within 1 and 1 mm. Accurate and constant forces could not be applied by hand using the Genucom’s force plate measurement system. Therefore, known forces and angular torques were applied using a weight and pulley system. Anterior translational forces and rotational forces were applied directly to the talus through the anterior pin
(Fig. 1). Tunnels that were 4.5 mm in diameter were made in locations that would allow us to reproduce the configurations shown in Figures 2 through 5. Figure 2 represents the Evans reconstruction. Figure 3
Figure 2. Evans procedure subtalar joints.
to limit inversion of ankle and
represents the Watson-Jones reconstruction, and consists of anterior and posterior segments. The graft in Figure 3 is routed from a posterior to anterior direction through a hole in the fibula made at the level of the ankle joint and perpendicular to the fibular shaft. After exiting from the
596
bone tunnel sites (i.e., the lengths of the various grafts described above) were measured. Changes in length were measured on a suture fixed to one point and passed through the corresponding tunnel using a hand-held ruler with 1mm increments. Data were obtained from 15 ankles in all but the &dquo;anatomic&dquo; procedure, for which data were obtained from the final 6 consecutive ankles tested. Biomechanical
Figure 3. Watson-Jones procedure to limit inversion of ankle and subtalar joints and replace anterior talofibular ligament.
Figure 4. Chrisman-Snook procedure (split peroneus brevis tendon graft) to replace anterior talofibular and calcaneofibular ligaments. fibula, the graft is passed down through a vertical hole in the talar neck and doubled back upon itself. The ChrismanSnook procedure is shown in Figure 4. In this reconstruction the graft is passed posteriorly through the fibular tunnel. The graft is then routed from a posterior to anterior direction through a tunnel made in the calcaneus.
Isometry testing The ankles to 30°
were
moved from
a
plantar flexion. Changes
position of 20° dorsiflexion in distance between various
testing
The following tests were performed on 15 ankles in the order described: 1. An internal rotation moment of 2 N-m was applied and internal rotation was measured at 10° plantar flexion. 2. An anterior force of 30 N was applied and anterior translocation of the talus (anterior drawer) was measured. During the anterior drawer test the talus was allowed to rotate freely so that the intact deltoid ligament would not restrict anterior translation of the lateral talus. 3. An inversion moment of 2 N-m with the ankle in neutral dorsiflexion/plantar flexion was applied and the tilt of the talus was measured. 4. The Schanz screws were withdrawn until they no longer crossed the subtalar joint. The Steinmann pin transfixing the talonavicular joint was removed. An inversion moment of 2 N-m at neutral ankle position was applied. Total inversion was measured. Inversion occurring at the tibiotalar joint (talar tilt) was subtracted from total inversion to yield subtalar joint motion. The following ligaments and capsule were then cut: the anterior ankle capsule, anterior talofibular ligament, and the calcaneofibular ligament. After ligament transection, specimens were rendered grossly unstable to stress testing. No stress testing was performed on ankles with transected ligaments. This testing has been previously performed.23 The three ankle ligament reconstructions (as in Figs. 2 through 4) were then sequentially performed on each ankle using the previously prepared peroneal tendons attached to the base of the fifth metatarsal. Each reconstructed ankle was subjected to the tests described. After preliminary testing it became apparent that there were large differences between the three ligament reconstructions in terms of isometry and control of various ankle motions. Based on this information, we designed a fourth lateral ankle ligament reconstruction. The goals of this procedure were to achieve a more anatomic placement of the graft and to control all pathologic ankle laxities without limiting subtalar joint motion. In this anatomic reconstruction, a split portion of the peroneus brevis tendon was routed, as shown in Figure 5. Bone tunnels were designed so that graft placement duplicated the orientation of the anterior talofibular ligament and calcaneofibular ligament as closely as possible. The graft was passed from an anterior to posterior direction through a calcaneal tunnel that was oriented so that the posterior opening of the tunnel was at the anatomic attachment point of the calcaneofibular ligament to the calcaneus. The graft was then brought from a posterior to anterior direction
597
The length of the graft between the anterior fibular hole and the talus increased 7 mm (range, 4 to 8), or 17%, as the ankle was plantar flexed. Chrisman-Snook procedure (Fig. 4). The length of the graft between the base of the fifth metatarsal and the anterior fibular hole increased 11 mm (range, 8 to 15), or 15%. The length of the graft between the posterior fibular hole and the calcaneus decreased 12 mm (range, 8 to 18), or 24%. Anatomic procedure (Fig. 5). The length of the graft between the anterior fibular hole and the talus increased 3 mm (range, 2 to 4), or 6%, as the ankle was moved from 20° dorsiflexion to 30° plantar flexion. The length of the graft between the posterior fibular hole and the calcaneus decreased 3 mm (range, 2 to 6), or 10%, during the same motion. Biomechanical
Figure 5. Anatomic reconstruction reproduces orientation of anterior talofibular and calcaneofibular ligaments. through an oblique fibular tunnel; the openings of the tunnel are located as close as possible to the fibular attachment points of the calcaneofibular and anterior talofibular ligaments, respectively. The graft was next passed down through vertical tunnel that had been drilled in the talar neck. The graft was then doubled back upon itself. This ligament reconstruction was then tested in the same way as the other three procedures had been tested. All tendon grafts were sutured into place and secured to the bones with sutures that were passed through drill holes made adjacent to the bone tunnels to eliminate sliding of the graft. Low forces were applied (2 N-m) to prevent loss of fixation. We attempted to tension each reconstruction equally by holding the ankle in eversion and neutral dorsiflexion/plantar flexion as the reconstruction was performed. Results of the biomechanical tests were subjected to repeated measures analysis of variance testing followed by Newman-Keuls post-hoc testing.3° These statistical analyses enabled us to group ligament reconstructions that had similar results for each test performed, and to determine significant differences between the ligament reconstructions. a
testing
Internal rotation of the talus (Fig. 6). The two procedures that limited internal rotation of the talus most were the Watson-Jones and anatomic reconstructions. The average amount of internal rotation that occurred with stress testing of each was 4.4°. This was significantly different than the amount of internal rotation possible after the ChrismanSnook (mean talar rotation, 11.1°) and Evans (mean talar
rotation, 9.5°) procedures (P < 0.001). Anterior translocation of the talus (Fig. 7). The procedures that limited anterior translocation of the talus most were the Watson-Jones (3.4 ± 2 mm) and anatomic reconstructions (3.7 ± 1.1 mm). The Chrisman-Snook and Evans procedures allowed significantly greater anterior translocation (5.1 ± 1.4 mm and 4.9 ± 1.3 mm, respectively) (P < 0.001). The anatomic reconstruction was the only procedure that restored anterior translocation to a distance similar to ankles with intact ligaments. Talar tilt (Fig. 8). Talar tilt measured for ankles with intact ligaments was 4.9° ± 2.2°. The anatomic reconstruction allowed 5.2° ± 1.7° and the Chrisman-Snook procedure permitted 5.6° ± 1.8° talar tilt. These values are statistically
RESULTS
Isometry Evans procedure (Fig. 2). As the ankle was moved from 20° dorsiflexion to 30° plantar flexion, the length of the graft decreased 3 mm (range, 1 to 4). This represented a 5% change in length. Watson-Jones procedure (Fig. 3). The length of the portion of the graft between the base of the fifth metatarsal and the posterior fibular tunnel decreased 3 mm (range, 1 to 4), or 5%. There was no difference between the isometry of this graft and that of the Evans procedure.
Llgamems
Jones
anoorc
Figure 6. Internal rotation of the talus in relation to the tibia with intact ligaments and after the various reconstructions.
598
Figure 7. Anterior translocation of the lateral talus in relation to the tibia. Rotation of the talus was allowed so that the deltoid ligament did not restrict anterior translocation.
Figure joint.
8. Inversion of the talus in relation to the tibiotalar
similar to
one
another. Talar tilt
was
significantly greater
(P < 0.001) after the Watson-Jones (mean, 7.6° ± 4.2°) and Evans (mean, 9.4° ± 3.3°) procedures. Subtalar motion (Fig. 9). The anatomic ligament reconstruction allowed the greatest subtalar joint motion (15.8° ± 7.1°). This value was statistically similar to the amount of subtalar motion in ankles with intact ligaments (18° ± 7.9°). The amount of subtalar motion after the Watson-Jones 0
(7.0° ± 3.4°), Evans (6.2° ± 4.1°), and Chrisman-Snook (5.4° ± 2.9°) procedures was statistically similar. These three procedures significantly restricted subtalar motion com-
pared to ankles with intact ligaments. DISCUSSION The abnormal laxities that must be corrected to restore functional stability to an ankle are poorly understood.&dquo;, 13
Figure 9. Subtalar motion tilt from total inversion.
as
calculated
by subtracting
talar
This lack of knowledge is reflected in the biomechanical differences between the major types of ligament reconstructions. The Evans procedure is designed to limit talar tilt by limiting inversion of the foot.l° The Watson-Jones reconstruction re-creates the anterior talofibular ligament in addition to limiting inversion of the foot.29 Elmslie,9 Chrisman and Snook,’ and others’,&dquo;, 17,20,26 showed that patients who have chronic ankle instability have excessive talar inversion; they recommended reconstruction of both the anterior talofibular and calcaneofibular ligaments. Measurement of talar tilt has been used by many authors&dquo; 11,21 to establish indications for surgery in patients who have chronic instability and to evaluate results after reconstruction. Brostrom3 investigated 60 patients who had symptomatic ankle instability. He detected an increased anterior drawer sign at physical examination. At surgery he found that the anterior talofibular ligament was attenuated in all of these ankles, but the calcaneofibular ligament was less frequently involved. Based on these observations, he advocated direct reconstruction of the anterior talofibular ligament in all unstable ankles, and repair of the calcaneofibular ligament only if necessary. Strain gauge analysis of normal ankle ligaments con-
ducted by the authors’ and biomechanical testing performed by Johnson and Markolfl4 indicate that the anterior talofibular ligament is a primary restraint to anterior translocation, internal rotation, and inversion of the talus at all flexion angles tested. Brostrom3 failed to demonstrate a single patient with isolated rupture of the calcaneofibular ligament in his clinical investigation; Rasmussen23 found that the anterior talofibular ligament always failed before the calcaneofibular ligament during cadaveric stress tests. We believe that symptoms in patients who have recurrent ankle instability are caused by excessive internal rotation, anterior translocation, and tilt of the talus. These excessive motions are usually due to laxity of the anterior talofibular
599
in some ankles, also the calcaneofibular ligament. The ideal ligament reconstruction should recreate the anterior talofibular ligament and, if necessary, the calcaneo-
ligament and,
fibular ligament in an anatomic fashion without restricting subtalar joint motion. We found that proper bone tunnel placement was critical to achieve relative graft isometry. The limb of the graft in the Watson-Jones procedure that re-creates the anterior talofibular ligament can be made more nearly isometric by moving the fibular tunnel distally so that the graft exits the tunnel anteriorly at the attachment of the anterior talofibular ligament. Similarly, the transverse fibular tunnel described in the Chrisman-Snook procedure does not produce anatomic graft placement. An oblique fibular tunnel through the attachment sites of the anterior talofibular and calcaneofibular ligaments will result in nearly isometric graft
placement. We found large differences between the ligament reconstructions in their ability to control excessive ankle laxities. An important limitation to our study was the dependence of the stability of these reconstructions on our technique and
tensioning methods. Although we attempted to standardize the methods for performing the reconstructions, we were unable to measure the uniformity of our fixation and tensioning. We therefore felt that it was appropriate to stress test the reconstructions at relatively low forces. Due to limitations in our testing apparatus, we were unable to apply a simulated weightbearing load across the ankle joint. While applying such a load would more closely duplicate physiologic conditions, we feel the comparative data obtained relating to the effects of the various reconstructions is valid. The Evans procedure was least able to produce normal ankle and subtalar joint mechanics. The Evans procedure restored internal rotation of the talus to a degree similar to specimens with intact ligaments; however, anterior translocation and tilt of the talus were poorly controlled and subtalar joint motion was restricted. The Chrisman-Snook procedure was effective in limiting talar tilt and resulted in motion similar to ankles with intact ligaments. However, subtalar joint motion was restricted. The Watson-Jones procedure was most effective in reducing internal rotation and anterior translocation of the talus. Talar tilt was less well controlled, and was significantly increased (P < 0.001) when compared to laxity of ankles with intact ligaments. Subtalar joint motion was restricted after this procedure. The anatomic reconstruction that we described restores most closely the normal mechanics of the ankle and subtalar joints (Fig. 7). Internal rotation, talar tilt, and anterior translocation were all controlled well, and subtalar joint motion was not restricted. Clinical studies have shown that the Evans, WatsonJones, and Chrisman-Snook procedures are effective in patients with symptomatic lateral ankle instability.6.l0.29 Although we have shown that these procedures have significant mechanical differences, they have withstood the test of time, and each enjoys a high clinical success rate.
By modifying the fibular and talar tunnel placement, both the Watson-Jones and Chrisman-Snook procedures can be made more anatomic, thereby allowing more normal ankle mechanics. The Evans, Watson-Jones, and Chrisman-Snook procedures all restrict subtalar joint motion. In cases of combined ankle and subtalar instability, these procedures would be superior to procedures that reconstruct only the ankle ligaments. However, in patients with primary ankle instability, it has been our experience, and that of others, that excessive loss of subtalar motion may lead to chronic pain and functional difficulties on uneven terrain. 27 In most patients with lateral ankle ligament instability, direct reconstruction of the ligaments as described by Brostrom3 is possible, and there is no need for augmentation using tendon grafts. Occasionally, however, augmentation of attenuated tissue is required. From our study we conclude that it is possible to reconstruct the lateral ankle ligaments using tendon grafts in an anatomic fashion while preserving normal joint mechanics. REFERENCES 1. Anderson KJ, Lecocq JF: Operative treatment of injury to the fibular collateral ligament of the ankle. J Bone Joint Surg 36A: 825-832, 1954 2. Brand RL, Collins MDF, Templeton T: Surgical repair of ruptured lateral ankle ligaments. Am J Sports Med 9: 40-44, 1981 3. Broström L: Sprained ankles. VI. Surgical treatment of "chronic" ligament ruptures. Acta Chir Scand 132: 551-565, 1966 4. Broström L: Sprained ankles. V. Treatment and prognosis in recent ligament ruptures. Acta Chir Scand 132: 537-550, 1966 5. Cass JR, Morrey BF: Ankle instability: Current concepts, diagnosis, and treatment. Mayo Clin Proc 59: 165-170, 1984 6. Chrisman OD, Snook GA: Reconstruction of lateral ligament tears of the ankle. An experimental study and clinical evaluation of seven patients treated by a new modification of the Elmslie procedure. J Bone Joint Surg 51A: 904-912, 1969 7. Colville MR, Marder RA, Boyle JJ, et al: Strain measurement in lateral ankle ligaments. Am J Sports Med 18: 196-200, 1990 8. Dias LS: The lateral ankle sprain: An experimental study. J Trauma 19:
266-269, 1979 9. Elmslie RC: Recurrent subluxation of the
ankle-joint. Ann Surg 100: 364-
367, 1934 10. Evans DL: Recurrent instability of the ankle: A method of surgical treatment. Proc R Soc Med 46: 343-344, 1953 11. Freeman MAR: Instability of the foot after injuries to the lateral ligament of the ankle. J Bone Joint Surg 47B: 669-677, 1965 12. Freeman MAR: Treatment of ruptures of the lateral ligament of the ankle. J Bone Joint Surg 47B: 661-668, 1965 13. Freeman MAR, Dean MRE, Hanham IWF: The etiology and prevention of functional instability of the foot. J Bone Joint Surg 47B: 678-685, 1965 14. Johnson EE, Markolf KL: The contribution of the anterior talofibular ligament to ankle laxity. J Bone Joint Surg 65A: 81-88, 1983 15. Kleiger B: The mechanism of ankle injuries. J Bone Joint Surg 38A: 59-70, 1956 16. Laurin CA, Mathieu J: Sagittal mobility of the normal ankle. Clin Orthop
108: 99-104, 1975 17. Laurin CA, Ouellet R, St.-Jacques R: Talar and subtalar tilt: An experimental investigation. Can J Surg 11: 270-279, 1968 18. Leach RE, Namiki O, Paul GR, et al: Secondary reconstruction of the lateral ligaments of the ankle. Clin Orthop 160: 201-211, 1981 19. Lee HG: Surgical repair in recurrent dislocation of the ankle joint. J Bone Joint Surg 39A: 828-834, 1957 20. Leonard MH: Injuries to the lateral ligaments of the ankle. A clinical and experimental study. J Bone Joint Surg 31A: 373-377, 1949 21. McCullough CJ, Burge PD: Rotatory stability of the load-bearing ankle. An experimental study. J Bone Joint Surg 62B: 460-464, 1980 22. Nilsonne H: Making a new ligament in ankle sprain. J Bone Joint Surg 14:
380-381,1932 23. Rasmussen O:
Stability of
the ankle
joint. Analysis
of the function and
600
traumatology of the ankle ligaments. Acta Orthop Scand (Suppl 211) 56: 1-74,1985 24. Rubin G, Witten M: The talar-tilt angle and the fibular collateral ligaments: A method for the determination of talar tilt. J Bone Joint Surg 42A: 311326, 1960 25. Ruth CJ: The surgical treatment of injuries of the fibular collateral ligaments of the ankle. J Bone Joint Surg 43A: 229-239, 1961 26. Shephard E: Tarsal movements. J Bone Joint Surg 33B: 258-263, 1951 27. Snook GA, Chrisman OD, Wilson TC: Long-term results of the Chrisman-
Snook operation for reconstruction of the lateral ligaments of the ankle. J Bone Joint Surg 67A: 1-7, 1985 28. Staples OS: Ruptures of the fibular collateral ligaments of the ankle. Result study of immediate surgical treatment. J Bone Joint Surg 57A: 101-107, 1975 29. Watson-Jones R: Fractures and Other Bone Joint Injuries. Edinburgh, E & S Livingstone, 1940 30. Winer BJ: Statistical Principles in Experimental Design. New York, McGraw Hill, 1971
