International

International Orthopaedics (SICOT) (1991) 15:245-249

Orthopaedics © Springer Verlag 1991

Rupture of the ankle ligaments H. Zwipp, R. Hoffmann, H. Thermann, and B. W. Wippermann Trauma Department, Medizinische Hochschule, Hannover, Federal Republic of Germany

Summary. A prospective random&ed trial o f treatment of ruptured ankle ligaments was carried out at our institute. Two hundred patients were randomly assigned to 4 treatment groups. The results at I and 2years after injury showed that functional treatment with a newly designed brace appeared to be the method of choice. This gives good mechanical stability, a shorter time o f f work and the ability to resume full sporting activity within 3 months o f injury. Operation is only indicated for dislocations o f the ankle and foot, ankle ligament rupture with additional intraarticular damage and second stage injuries or re-ruptures. R6sum6. Un essai thbrapeutique prospectif et randomisd concernant les ruptures des ligaments de la cheville a btb rbalisb dans notre service. Deux cents patients ont ainsi btd rdpartis entre quatre diffbrents groupes de traitement. Les rbsultats, 1 ~ 2 ans aprds l'accident, montrent que le traitement fonctionnel, l'aide d'un appareillage d'un nouveau modOle, est la mbthode de choix. II obtient une bonne stabilitd mbcanique, une courte invaliditb et permet la reprise d'une activitb sportive compldte clans les trois mois qui suivent l'accident. La chirurgie n'est indiqube que dans les Iuxations de la cheville ou du pied, dans les ruptures ligamentaires avec des lbsions intra-articulaires associOes et dans les traumatismes rbpbtds ou les ruptures itbratives. Introduction T h e r e has been c o n t r o v e r s y in E u r o p e a b o u t the t r e a t m e n t o f the r u p t u r e d lateral ligament o f the

Reprint request to: H. Zwipp, Trauma Department, Medizinische Hochschule, Konstanty-Gutschow-Strasse 8, W-3000 Hannover, FRG

ankle during the past decade. Most o f these injuries occur during recreational and school sport, a n d in patients aged between 15 and 25 years o f age. The injury may, therefore, involve medical expenses as well as time o f f w o r k a n d disability p a y m e n t s . A study o f the w o r k m a n ' s c o m p e n s a tion system in G e r m a n y showed that during 1985, 13 554 patients were in hospital for an average o f 13 days as a result o f isolated ruptures o f the ankle ligaments [16]. Suggestions for t r e a t m e n t s h o u l d be based on prospective r a n d o m i s e d trials r a t h e r than personal j u d g e m e n t or retrospective studies. A n u m b e r o f clinical trials have shown that immobilisation and o p e r a t i o n p r o d u c e a similar o u t c o m e [1, 2, 6, 12, 15, 24]. Others have published excellent results with strapping, orthoses a n d special shoes [l, 8, 9, 11, 12, 13, 14, 18, 19, 22, 24]. W e r e p o r t a prospective r a n d o m i s e d trial to d e t e r m i n e the best t r e a t m e n t for r u p t u r e d ankle ligaments, carried out in our institute during 1985 a n d 1986.

Material and methods We evaluated 4 groups of 50 randomly selected patients using a _+100 point system based on clinical, radiographic and dynamometric results 3, 12 and 24months after injury. The groups were classified as follows: Group A Group B Group C Group D Group E:

Operation and cast immobilisation Operation and early functional treatment Conservative: cast immobilisation Conservative: early functional treatment Excluded from the study

Randomisation and informed consent The reason and purpose of the study and the need for randomisation were explained to the patients. Assignment to the

246

H. Zwipp et al.: Rupture of the ankle ligaments

Fig. 1. Stress tenography showing: a No contrast medium in the joint indicating a single tear; b Contrast medium is present in the joint indicating a double tear

different groups was based on the group marked on the study charts which were drawn at random. Those who rejected randomisation were classified as group E: only their clinical and radiographic findings were analysed. All patients with a history of a previous supination injury, or in whom there were operative findings of a previous ligament injury or cartilage damage were also excluded.

Study design When the diagnosis of acute rupture of the lateral ligament was made, the clinical instability for talar tilt and the anterior drawer test were graded from 1 + to 3 + . The diagnosis was reviewed when stress tenography [5] was carried out 2 4 - 4 8 h after injury. The radiologist had to decide in the 189 patients examined initially at our institution whether there was a single or double tear present, based on the presence of contrast medium in the joint (Fig. 1). In the 38 patients who had their stress radiography before admission, the diagnosis was based on talar tilt > 7 °, or a difference of > 5 ° compared with the other side and anterior drawer > 5 ram, or a difference of > 3 mm compared with the other side. The surgeon recorded the extent of the injury found at operation in the patients who received surgical treatment. Data were collected anonymously and the patients identified by number only. Follow up was at 3, 12, and 24 months after injury by 3 independent examiners who did not know the extent of the injury. Anteroposterior and lateral standard radiographs were taken, together with anteroposterior and lateral stress views of both ankles. Three radiographic and 20 clinical observations with dynamometric measurements of static force, coordination and proprioception were used to analyse the results. Dynamometric measurement was not done at the 24 month follow up. A _+ 100 point system was used for grading the results.

Statistics The incidence of re-rupture or clinical instability was compared in the different groups by the chi-square test, the level of significance being < 0.05 for all tests. To compare parametric and nonparametric results at different times, analysis of variance and Kruskal Wallis tests were performed on a personal computer with the K W I K S T A T statistical package.

Diagnostic procedures Since operative findings were only available in 102 patients, stress enography was done in 189 in order to quantify better

the extent of the lesion. This was important because otherwise we could not be certain that the severity of injury was comparable in each of the treatment groups. The results of stress tenography were confirmed by the operative findings in 92% of cases. There were 8% false positive and no negative results. The distinction between single and double ligament tears was correct in 78%. An even distribution, based on the radiographic findings, was found in the different groups (chi-square analysis).

Patients We saw 227 patients with ruptured ligaments of the ankle between 15 April 1985 and 31 July 1986. Treatment was randomised for 200 who form the basis of this study. In 27, only the diagnosis and treatment were guided by the protocol, but these patients were treated according to their own wishes (group E) and their results were excluded. There were 136 males and 91 females with an average of 23.6 years (range 9 to 51 years). The left and right sides were almost evenly affected; 51% of injuries occurred during sports and 31.1% at home. In the whole group, 12.9% took part in sport occasionally, 75.2% exercised regularly, and 11.9% were top athletes.

Treatment G r o u p A (operation and cast immobilisation). Operation was carried out under regional or general anaesthesia within 60 h of injury. A lateral epimalleolar incision was used [24]. A plaster cast was applied and split; the leg was elevated. Active movement between 10 ° of extension and 20 ° of flexion was begun on the first day. The patients were in hospital for 1 or 2 days; nonsteroidal anti-inflammatory drugs and D H E heparin were given during this period. A walking cast was applied in the neutral position between the 5th and 8th day after operation and removed after 5 weeks. G r o u p B (operation a n d early functional treatment). Operation and postoperative management were the same as in group A, but after 8 to 10 days a newly designed brace, which provides protection against supination, and incorporates a pronation wedge, was fitted (Fig. 2). It can be worn in a sports shoe and the patients were instructed to keep it on day and night. They could drive a car as soon as full weightbearing was possible without pain. Light work was allowed after 3 weeks. G r o u p C (cast immobilisation). The initial cast, which had been split, was removed after about 3 - 5 days when the swelling had subsided. A new cast was moulded in a pronation and

H. Zwipp et aI.: Rupture of the ankle ligaments

247

Group

Treatment

n excellent g o o d

saris.

A

~..~,,,,~

] +

~,~

52

74%

23%

3%

B

~_,~

3 +

~

50

86%

14%

0%

48

78%

]9%

3%

50

80%

20%

0%

....

C

~[

/

D

Fig. 3. Diagram showing the results in the 4 treatment groups. Average results of 3, 12, 24 month following

100%

A

Fig. 2. Photographs showing the newly designed functional brace worn in groups B and D

eversion (J. Schatzker, personal communication) and kept on for 5 weeks after injury. Group D (early functional treatment). After 3 - 5 days in a split lower leg cast, the special brace was fitted and worn for 5 weeks. Driving a car and return to light work was the same as in group B. In each group, 6 sessions of physical therapy, with special emphasis on proprioception and pronator muscle training, were prescribed after the cast or brace was removed.

Results

In groups A and B there were no wound infections; minor wound breakdown occurred in 2 patients and eventually healed. Dysaesthesia around the wound was present in 6 cases. The extent of the injury (single ligament 34% and double 66%) was evenly distributed in the treatment groups and there were approximately 50 patients in each group (Fig. 3). Age, sex and side of injury were also evenly distributed. At 3 months follow up we were able to assess 185 out of the 200 randomised patients (93%), and found that there was a statistically better range of active movement in group D. Subjective stability was better in those operated on (A and B) with less good results in group C. There were no significant differences in other subjective criteria such as giving way, gait stability and limitation in sports or work. The results of radiographic testing did not confirm the clinical impression and

excellent

87

good

I~

B

satisfactory satisfactory

i

good

100% [] 3 M[] 12M[] 24 M

excellent

89

Fig. 4. Histogram showing that the evaluation score at 3 months was somewhat lower in group A, but this was not statistically significant

showed no statistical differences when all the patients in all the groups were compared. We found a higher percentage of absolutely stable ankle joints in the operated groups (A and B), but this did not reach statistical significance. No patient showed radiographic signs of degenerative joint disease. The evaluation score was somewhat lower in group A (Fig. 4) when compared with the other groups, but analysis of variance showed no statistical difference. At 12 months, 168 patients were reviewed (84%) and there were no significant differences between the 4 groups with regard to clinical and radiographic stability, range of movement, subjective limitation or performance score. After 24 months, 159 patients were available (80%) and after 1 and 2 years follow up there was a more even distribution of the joint stability pattern in the different groups (Fig. 5). The type of after-treatment did not seem to influence stability. The subjective results, however, did not reflect these findings. Neither the rate of recurrent giving

248

H. Zwipp et al.: Rupture of the ankle ligaments

~z~ ....

uam

.....

~

. . . . .

Fig. 5. Histogram showing that there was a more even distribution of joint stability patterns in the 4 groups after 24 months

way nor subjective grading of joint stability was significantly different within the groups. No radiographic signs of progressive degenerative disease were seen. The functional score did not reveal a difference between groups as determined by analysis of variance (Fig. 4). When the results of stress views were analysed separately for single or double ligament tears, we could not demonstrate different results for each period of follow up for clinical or radiographic instability. Discussion

A number of prospective randomised trials have confirmed the benefit of conservative treatment compared to operation [1, 6, 8, 9, 12, 13, 15, 17]. The advantages of functional treatment are early mobilisation with less muscle atrophy, an earlier return of full ankle movement, earlier return to activity and more comfort for the patient. Experimental work has indicated that this type of treatment promotes ligament healing [3, 7, 10, 20, 23]. Our trial did not demonstrate a significant difference between the 4 groups even after 2 years, which is in keeping with other published studies. Only one report has shown better results after operation [4]; in this clinical study 345 patients were treated by operative repair and immobilisation, and compared with 40 patients who were immobilised for 3 weeks. The paper does not indicate whether the cast was applied in pronation or supination, and 29 to 37% of all patients had reported previous supination injuries, so some of the randomised patients may have had chronic instability. The importance of a longer follow up is confirmed by our study. Radiographic stability ( < 5 ° talar tilt and < 5 mm anterior drawer) was found in 89% in groups A and B compared with 78% and

74% for groups C and D respectively after 3 months. At 12 and 24 months the results were similar in all groups (mean 67% + 4%). Although about two-thirds of our patients had absolutely stable ankles at 2 years, we intend to reassess them after 5 years. A significant increase in laxity 4 years after functional treatment in an Aircast splint has been reported [23]; 10 out of 18 patients had a talar tilt of between 8 ° and 14 °, and one had as much as 18 ° in spite of minimal subjective discomfort. Other authors have not reported similar findings with the Aircast splint. Only two papers report a 5 year follow up after functional treatment [8, 9], and they did not demonstrate statistically significant difference between operative and conservative treatment when early and late results were compared. There is no evidence in the literature to support the need for operation. One study [21] has reported that operation is superior to functional treatment, but only 28 out of their 78 patients (37%) were able to return to unrestricted sports activity within 3 months of injury: 15% took more than 12 months for full recovery. Our results in patients treated in a cast alone (group C) were not worse than those treated in a brace (group D), but we believe the shorter period off work in the functional group to be very important. We therefore recommend this type of primary treatment as the treatment of choice for ruptured ankle ligaments; it combines the greatest health and socioeconomic benefit with the least risk when compared to other methods. References 1. Brooks SC, Potter BT, Rainey JB (1981) Treatment of the partial tears of the lateral ligament of the ankle: a prospective trial. Br Med J 606 2. Brostr6m L (1966) Sprained Ankles. V. Treatment and prognosis in recent ligament ruptures. Acta Chir Scand 132:537 3. Clayton ML, Miles JS, Abdulla M (1968) Experimental investigation of ligament healing. Clin Orthop 61 : 146 4. Ent FWC vd (1984) Lateral ankle ligament injury. An experimental and clinical study. Proefschrift, Universiteit Rotterdam, Elinkerijk BV, Utrecht 5. Evans GA, Frenyo SD (1979) The stress-tenogram in the diagnosis of ruptures of the lateral ligament of the ankle. J Bone Joint Surg [Br] 61 : 347 6. Evans GA, Hardcastle, Frenyo AD (1984) Acute rupture of the lateral ligament of the ankle, to suture or not to suture? J Bone Joint Surg [Br] 66:209 7. Gamble JG, Edwards CHC, Max SR (1984) Enzymatic adaption in ligaments during immobilization. Am J Sports Med 12:221 8. Hoogenband CR vd, Moppes FI v, Stapert JWlL, Coumans PF, Greep JM (1982) Konservative Behandlung der fibular-talaren und fibulor-calcanearen Bandverletzung mit Coumans-Bandage, eine prospektive Vergleichsstudie.

H. Zwipp et al.: Rupture of the ankle ligaments

9.

10.

11. 12.

13. 14. 15.

Abstraktband zur 130. Tagung der Vereinigung Nordwestdeutscher Chirurgen, 2.-4. Dezember 1982, Hamburg Hoogenband CR vd, Moppes FI van (1987) Die Behandlung der lateralen Ligamentrupturen des oberen Sprunggelenkes mit der Coumans-Bandage und direkte Mobilisation (eine prospektive Vergleichsstudie), Teil 2. Hefte Unfallheilkd 189:1030 Jnoue M, Gomez MA, Hollis JM, Roux RD, Lee EB, Burleson EM, Woo SL-Y (1986) Medial collateral ligament healing: repair vs non-repair, vol 11. Transactions of the 32nd Annual ORS, New Orleans Lousiana, p 78 Jakob RP, Raemy H, Steffen R, Wetz B (1986) Zur funktionellen Behandlung des frischen Aul3enb~inderrisses mit der Aircast-Schiene, Orthop~ide 15: 434-440 Klein J, Schreckenberger C, R6ddecker K, Tiling TH (1988) Operative und konservative Behandlung der frischen Aul]enbandruptur am oberen Sprunggelenk. Randomisierte klinische Studie. Unfallchirurg 91:154 Korkola O, Rusanen M, Jokipii P, Kyt6maa J, Avikainen V (1987) A prospective study of the treatment of severe tears of the lateral ligament of the ankle. Int Orthop 11:13 Neumann K (1987) Ist die konservativ-funktionelle Behandlung frischer Aul~enbandrupturen am OSG gerechtfertigt? Tell 2. Hefte Unfallheilkd 189:1018 Niedermann B, Anderson A, Byrde-Anderson S, Funder V, J6rgensen JB, Lindholmer E, Fuust M (1981) Rupture

249

16. 17.

18. 19. 20. 21. 22. 23. 24.

of the lateral ligaments of the ankle: Operation or plastercast. Acta Orthop Scand 52:579 REHA 85 Rehabilitation und Rehabilitationsstatistik der gesetzlichen Unfallversicherung (1985) Hauptverband der gewerblichen Berufsgenossenschaften e.V. (ed) Sommer HM, Arza D, Ahrendt J (1987) Behandlungsergebnisse von operativ und konservativ versorgten flbularen Kapselbandrupturen, Teil 1. Hefte Unfallheilkd 189:1012 Stover CN (1980) Air stirrup management of ankle injuries in the athlete. Am J Sports Med 8:360 Stover CN (1986) Functional sprain management of the ankle. Ambulatory Care 11 : 25 Tipton CM, James SL, Mergner WTcheng T (1970) Influence of exercise in strength in medial collateral knee ligaments of dogs. Am J Physiol 218:894 Weise K, Rupf G, Weinelt J (1988) Die laterale Bandverletzung des OSG beim Sport. Aktuel Traumatol 18:54 Wetz B, Steffen R, Raemy H, Jakob RP (1987) Sp~itergebnisse nach konservativer Therapie fibulotalarer Bandl~isionen mit der Aircast-Schiene. Schweiz Sportmed 3:115 Zwipp H (1986) Die antero-laterale Rotationsinstabilit~it des oberen Sprunggelenkes. Hefte Unfallheilkd 177 Zwipp H, Tscherne H, Hoffmann R, Wippermann B (1986) Therapie der frischen Bandruptur, Orthop~ide 15: 446-453