JBJS | Treatment of Ruptures of the Lateral Ankle Ligaments: A Meta-Analysis*

The Journal of Bone & Joint Surgery, Volume 82, Issue 6

Articles | June 01, 2000

Treatment of Ruptures of the Lateral Ankle Ligaments: A Meta-Analysis*

A. C. M. Pijnenburg, M.D.†; C. N. van Dijk, M.D., Ph.D.†; P. M. M. Bossuyt, Ph.D.†; R. K. Marti, M.D., Ph.D.†

Investigation performed at the Academic Medical Centre, Amsterdam, The Netherlands
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
†Departments of Orthopaedic Surgery (A. C. M. P., C. N. V. D., and R. K. M.) and Clinical Epidemiology and Biostatistics (P. M. M. B.), Academic Medical Centre, P.O. Box 22700, 1100 BE Amsterdam, The Netherlands. E-mail address for A. C. M. Pijnenburg: a.c.pijnenburg@ amc.uva.nl.

The Journal of Bone & Joint Surgery. 2000; 82:761-761

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

text A A A

Abstract

Background: Ruptures of the lateral ankle ligaments are very common; however, treatment remains controversial. The aim of the current study was to perform a meta-analysis of randomized, controlled clinical trials of existing treatment strategies for acute ruptures of the lateral ankle ligaments.

Methods: Randomized, controlled trials reported between 1966 and 1998 were included if they involved acute ruptures of the lateral ankle ligaments. Randomized, controlled trials are defined as comparative studies with an intervention group and a control group in which the assignment of participants to a group is determined by the formal procedure of randomization. Summary measures of effectiveness were expressed as relative risks with use of random effects modeling.

Results: When analyzing the trials, we searched for comparable outcome measures in both short and long-term follow-up studies (studies with six months to 3.8 years of follow-up). This resulted in the analyses of three outcome measures: time lost from work, residual pain, and giving-way. This report summarizes the results of twenty-seven trials. With respect to giving-way, a significant difference was noted between operative treatment and functional treatment (relative risk, 0.23; 95 percent confidence interval, 0.17 to 0.31) in favor of operative treatment and a significant difference was also noted between functional treatment and treatment with a cast for six weeks (relative risk, 0.69; 95 percent confidence interval, 0.50 to 0.94) in favor of functional treatment. With respect to residual pain, no significant difference was found between operative and functional treatment and a significant difference was found between functional treatment and treatment with a cast for six weeks (relative risk, 0.67; 95 percent confidence interval, 0.50 to 0.90). We found minimal or no treatment to result in more residual pain (relative risk, 0.53; 95 percent confidence interval, 0.27 to 1.02) and giving-way (relative risk, 0.34; 95 percent confidence interval, 0.17 to 0.71) than did functional treatment.

Conclusions: We concluded that a no-treatment strategy for ruptures of the lateral ankle ligaments leads to more residual symptoms. Operative treatment leads to better results than functional treatment, and functional treatment leads to better results than cast immobilization for six weeks.

Figures in this Article

Introduction

Introduction | Materials and Methods | Results | Discussion | References

Supination injuries of the ankle ligaments are among the most common injuries^5,39,50. They account for about 25 percent of all injuries of the musculoskeletal system³². There are about 5600 ankle injuries each day in the United Kingdom, 23,000 in the United States, and 1600 in The Netherlands^66,70. The most commonly injured part of the lateral ligament complex is the anterior talofibular ligament^6,13,67. Management strategies for acute ruptures of lateral ankle ligaments can be divided into cast immobilization, operative treatment, and functional treatment. Although ruptures of the lateral ankle ligaments are very common, treatment selection remains controversial. In a number of recent reviews on treatment of acute ruptures of the lateral ankle ligaments, early mobilization and functional treatment have been advocated, although some recent publications have shown superior results after operative treatment^30,44,55,62. We conducted a systematic review of the available literature in order to perform a meta-analysis of the effectiveness of existing treatment strategies for acute ruptures of the lateral ankle ligaments. Such a meta-analysis can aid in the development of evidence-based treatment recommendations.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-A:: Figs. 1-A through 4-B: Results of random effects modeling⁹ used to express summary estimates. Studies meeting the quality criteria are depicted as triangles, and others are depicted as circles. Summary estimates are depicted as diamonds.
Figs. 1-A and 1-B: Studies in which patients who received functional treatment were compared with those who received minimal treatment.
Fig. 1-A: Relative risk for pain.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-B:: Relative risk for giving-way.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-A:: Figs. 2-A and 2-B: Studies in which patients who received operative treatment were compared with those who received functional treatment.
Fig. 2-A: Relative risk for pain.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B:: Relative risk for giving-way.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-A:: Figs. 3-A and 3-B: Studies in which patients who received functional treatment were compared with those who were treated with a cast for six weeks.
Fig. 3-A: Relative risk for pain.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-B:: Relative risk for giving-way.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig.4-A:: Figs. 4-A and 4-B: Studies in which patients who received operative treatment followed by six weeks of cast immobilization were compared with those who were treated with a cast for six weeks.
Fig. 4-A: Relative risk for pain.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-B:: Relative risk for giving-way.

View Larger

Table I: Randomized Clinical Trials Included in Meta-Analysis of Treatment of Lateral Ligament Rupture

*Quality criteria: Was the assignment of patients randomized?/Were all patients who entered a trial accounted for and attributed at its conclusion?/Was the measurement of the outcome blinded? 1 = yes, and 0 = no.†Van der Ent made a distinction between single and multiple ligament ruptures in his operative group.‡Zeegers obtained five-year results with use of a questionnaire; these results were not used for analysis.

Author of Study	Yr. of Study	Treatment	No. of Patients	Duration of Follow-up(yrs.)	Arthrography	Stress Radiography	Quality Criteria*
Freeman¹⁹	1965	Op. + cast, 6 wks.	16	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	18
Strapping, 6 wks.	12
Freeman et al.²¹	1965	Op. + cast, 6 wks.	12	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	14
Strapping, 6 wks.	16
Broström⁶	1966	Op. + cast, 3 wks.	96	3.8	+		1/1/0
Cast, 3 wks.	86
Strapping	102
Prins⁴⁹	1978	Op. + cast, 3 wks.	45	6 mos.	+		0/1/0
Cast, 6 wks.	59
Elastic wrap, 3 wks. (no lesion)	101
Elastic wrap, 3 wks. (lesion)	42
Cast, 3 wks.	51
Grönmark et al.²⁴	1980	Op. + cast, 6 wks.	32	3.8	-	Talar tilt L/R diff.	1/1/0
Cast, 6 wks.	33
Strapping, 6 wks.	30
Niedermann et al.⁴³	1981	Op. + cast, 5 wks.	102	1	+		0/1/0
Cast, 5 wks.	107
Strapping, 5 wks. (no lesion)	146
van Moppes and van den Hoogenband⁶⁸	1982	Op. + cast, 5 wks.	50	1	+	Talar tilt L/R diff. > 5.5°	1/1/0
Cast, 6 wks.	50
Strapping, 6 wks.	50
Cetti⁸	1982	Cast, 6 wks.	65	6 mos.	-	Ant. drawer sign > 3 mm, talar tilt > 6°	1/1/0
Brace + pliton shoe, 6 wks.	65
Evans et al.¹⁸	1984	Op. + cast, 3 wks.	50	2	+	Talar tilt L/R diff.	1/1/0
Cast, 3 wks.	50
Van der Ent†⁶⁵	1984	Op. + cast, 3 wks.	332	1	+		1/1/0
Cast, 3 wks.	40
Eggert et al.¹⁴	1986	Op. + cast, 6 wks.	36	1.3	-	Ant. drawer sign > 8 mm, talar tilt > 8°	1/1/0
Op. + shoe, 6 wks.	38
Shoe, 6 wks.	50
Zwipp et al.^71,72,74	1986, 1986, 1991	Op. + cast, 6 wks.	52	2	+		1/0/0
Op. + phys. therapy	50
Cast, 5 wks.	48
Brace + phys. therapy, 5 wks.	50
Korkala et al.³⁷	1987	Op. + cast, 4 wks.	34	2	-	Ant. drawer sign > 6 mm, talar tilt > 15°, L/R diff. > 3 mm or 10°	1/0/0
Cast, 4 wks.	47
Elastic bandage, 2 wks.	36
Klein et al.³³	1988	Op. + cast, 6 wks.	30	6 mos.	-	Ant. drawer sign > 7 mm, talar tilt > 7°	1/1/0
Cast, 6 wks.	30
Möller-Larsen et al.⁴¹	1988	Op. + cast, 6 wks.	55	1	+		1/0/0
Cast, 5 wks.	55
Strapping, 5 wks.	65
Pace et al.⁴⁷	1990	Op. + cast, 5 wks.	12	2	+		0/1/0
Cast, 4 wks.	18
Klein et al.³⁴	1991	Cast, 6 wks.	30	1	-	Ant. drawer sign > 7 mm, talar tilt > 7°, L/R diff. 7 mm and 7°	1/1/0
Air cast + phys. therapy	30
Stadelmayer et al.⁶⁰	1992	Op. + cast, 6 wks.	30	1	-	Ant. drawer sign > 10 mm, talar tilt > 10°	1/1/0
Cast, 6 wks.	30
Sommer et al.^58,59	1989, 1993	Cast + air cast, 9 wks.	3	6 mos.	+	Talar tilt > 10°	1/1/0
Air cast, 6 wks.	37
Strapping, 8 wks.	33
Dettori et al.^11,12	1994	Cast + phys. therapy, 5 wks.	28	1	+		1/1/0
Air cast + phys. therapy, 5 wks.	22
Elastic bandage + phys. therapy, 5 wks.	24
Zeegers‡⁷⁰	1995	Special shoe + elastic sock, 5 wks.	60	1	+		1/1/0
Elastic sock, 5 wks.	62
Strapping, 3 wks. + elastic sock, 2 wks.	59
Air cast + elastic sock, 5 wks.	62
Kaikkonen et al.²⁹	1996	Op. + air cast, 5 wks.	30	9 mos.	-	Ant. drawer sign L/R diff., talar tilt > 15°	1/1/0
Air cast, 5 wks.	51
Povacz et al.⁴⁸	1998	Op. + air cast, 6 wks.	73	2	-	Talar tilt L/R diff. > 5°	1/1/0
Air cast, 6 wks.	73

Add to My JBJS

Table I: Randomized Clinical Trials Included in Meta-Analysis of Treatment of Lateral Ligament Rupture

Author of Study	Yr. of Study	Treatment	No. of Patients	Duration of Follow-up(yrs.)	Arthrography	Stress Radiography	Quality Criteria*
Freeman¹⁹	1965	Op. + cast, 6 wks.	16	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	18
Strapping, 6 wks.	12
Freeman et al.²¹	1965	Op. + cast, 6 wks.	12	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	14
Strapping, 6 wks.	16
Broström⁶	1966	Op. + cast, 3 wks.	96	3.8	+		1/1/0
Cast, 3 wks.	86
Strapping	102
Prins⁴⁹	1978	Op. + cast, 3 wks.	45	6 mos.	+		0/1/0
Cast, 6 wks.	59
Elastic wrap, 3 wks. (no lesion)	101
Elastic wrap, 3 wks. (lesion)	42
Cast, 3 wks.	51
Grönmark et al.²⁴	1980	Op. + cast, 6 wks.	32	3.8	-	Talar tilt L/R diff.	1/1/0
Cast, 6 wks.	33
Strapping, 6 wks.	30
Niedermann et al.⁴³	1981	Op. + cast, 5 wks.	102	1	+		0/1/0
Cast, 5 wks.	107
Strapping, 5 wks. (no lesion)	146
van Moppes and van den Hoogenband⁶⁸	1982	Op. + cast, 5 wks.	50	1	+	Talar tilt L/R diff. > 5.5°	1/1/0
Cast, 6 wks.	50
Strapping, 6 wks.	50
Cetti⁸	1982	Cast, 6 wks.	65	6 mos.	-	Ant. drawer sign > 3 mm, talar tilt > 6°	1/1/0
Brace + pliton shoe, 6 wks.	65
Evans et al.¹⁸	1984	Op. + cast, 3 wks.	50	2	+	Talar tilt L/R diff.	1/1/0
Cast, 3 wks.	50
Van der Ent†⁶⁵	1984	Op. + cast, 3 wks.	332	1	+		1/1/0
Cast, 3 wks.	40
Eggert et al.¹⁴	1986	Op. + cast, 6 wks.	36	1.3	-	Ant. drawer sign > 8 mm, talar tilt > 8°	1/1/0
Op. + shoe, 6 wks.	38
Shoe, 6 wks.	50
Zwipp et al.^71,72,74	1986, 1986, 1991	Op. + cast, 6 wks.	52	2	+		1/0/0
Op. + phys. therapy	50
Cast, 5 wks.	48
Brace + phys. therapy, 5 wks.	50
Korkala et al.³⁷	1987	Op. + cast, 4 wks.	34	2	-	Ant. drawer sign > 6 mm, talar tilt > 15°, L/R diff. > 3 mm or 10°	1/0/0
Cast, 4 wks.	47
Elastic bandage, 2 wks.	36
Klein et al.³³	1988	Op. + cast, 6 wks.	30	6 mos.	-	Ant. drawer sign > 7 mm, talar tilt > 7°	1/1/0
Cast, 6 wks.	30
Möller-Larsen et al.⁴¹	1988	Op. + cast, 6 wks.	55	1	+		1/0/0
Cast, 5 wks.	55
Strapping, 5 wks.	65
Pace et al.⁴⁷	1990	Op. + cast, 5 wks.	12	2	+		0/1/0
Cast, 4 wks.	18
Klein et al.³⁴	1991	Cast, 6 wks.	30	1	-	Ant. drawer sign > 7 mm, talar tilt > 7°, L/R diff. 7 mm and 7°	1/1/0
Air cast + phys. therapy	30
Stadelmayer et al.⁶⁰	1992	Op. + cast, 6 wks.	30	1	-	Ant. drawer sign > 10 mm, talar tilt > 10°	1/1/0
Cast, 6 wks.	30
Sommer et al.^58,59	1989, 1993	Cast + air cast, 9 wks.	3	6 mos.	+	Talar tilt > 10°	1/1/0
Air cast, 6 wks.	37
Strapping, 8 wks.	33
Dettori et al.^11,12	1994	Cast + phys. therapy, 5 wks.	28	1	+		1/1/0
Air cast + phys. therapy, 5 wks.	22
Elastic bandage + phys. therapy, 5 wks.	24
Zeegers‡⁷⁰	1995	Special shoe + elastic sock, 5 wks.	60	1	+		1/1/0
Elastic sock, 5 wks.	62
Strapping, 3 wks. + elastic sock, 2 wks.	59
Air cast + elastic sock, 5 wks.	62
Kaikkonen et al.²⁹	1996	Op. + air cast, 5 wks.	30	9 mos.	-	Ant. drawer sign L/R diff., talar tilt > 15°	1/1/0
Air cast, 5 wks.	51
Povacz et al.⁴⁸	1998	Op. + air cast, 6 wks.	73	2	-	Talar tilt L/R diff. > 5°	1/1/0
Air cast, 6 wks.	73

View Larger

Table II: Test for Homogeneity

*High Q values are associated with heterogeneity.

Comparison	Q Values*
Pain	Giving-Way
Minimal treatment compared with functional treatment	Q = 0.28 (p = 0.87)	Q = 1.00 (p = 0.61)
Op. treatment compared with functional treatment
All studies	Q = 42.70 (p < 0.001)	Q = 5.47 (p = 0.49)
High-quality studies	Q = 40.08 (p < 0.001)	Q = 3.87 (p = 0.28)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment only
All studies	Q = 42.48 (p < 0.001)	Q = 2.99 (p = 0.88)
High-quality studies	Q = 10.86 (p = 0.012)	Q = 1.97 (p = 0.80)
Functional treatment compared with 6 wks. of cast treatment
All studies	Q = 2.29 (p = 0.94)	Q = 12.78 (p = 0.08)
High-quality studies	Q = 0.34 (p = 0.99)	Q = 0.45 (p = 0.98)

Add to My JBJS

Table II: Test for Homogeneity

*High Q values are associated with heterogeneity.

Comparison	Q Values*
Pain	Giving-Way
Minimal treatment compared with functional treatment	Q = 0.28 (p = 0.87)	Q = 1.00 (p = 0.61)
Op. treatment compared with functional treatment
All studies	Q = 42.70 (p < 0.001)	Q = 5.47 (p = 0.49)
High-quality studies	Q = 40.08 (p < 0.001)	Q = 3.87 (p = 0.28)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment only
All studies	Q = 42.48 (p < 0.001)	Q = 2.99 (p = 0.88)
High-quality studies	Q = 10.86 (p = 0.012)	Q = 1.97 (p = 0.80)
Functional treatment compared with 6 wks. of cast treatment
All studies	Q = 2.29 (p = 0.94)	Q = 12.78 (p = 0.08)
High-quality studies	Q = 0.34 (p = 0.99)	Q = 0.45 (p = 0.98)

View Larger

Table III: Summary Estimates of Effectiveness

*The data is given as the relative risk, with the 95 percent confidence interval in parentheses.

Comparison	No. of Trials	Pain*	Giving-Way*
Minimal treatment compared with functional treatment	3	0.53 (0.27 to 1.02)	0.34 (0.17 to 0.71)
Op. treatment compared with functional treatment
All studies	7	0.52 (0.18 to 1.53)	0.23 (0.17 to 0.31)
High-quality studies	5	0.51 (0.14 to 1.94)	0.22 (0.15 to 0.32)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment
All studies	9	0.81 (0.40 to 1.67)	0.90 (0.69 to 1.18)
High-quality studies	5	1.08 (0.56 to 2.08)	0.98 (0.71 to 1.36)
Functional treatment compared with 6 wks. of cast treatment
All studies	10	0.67 (0.50 to 0.90)	0.69 (0.50 to 0.94)
High-quality studies	5	0.79 (0.52 to 1.18)	0.58 (0.40 to 0.84)

Add to My JBJS

Table III: Summary Estimates of Effectiveness

*The data is given as the relative risk, with the 95 percent confidence interval in parentheses.

Comparison	No. of Trials	Pain*	Giving-Way*
Minimal treatment compared with functional treatment	3	0.53 (0.27 to 1.02)	0.34 (0.17 to 0.71)
Op. treatment compared with functional treatment
All studies	7	0.52 (0.18 to 1.53)	0.23 (0.17 to 0.31)
High-quality studies	5	0.51 (0.14 to 1.94)	0.22 (0.15 to 0.32)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment
All studies	9	0.81 (0.40 to 1.67)	0.90 (0.69 to 1.18)
High-quality studies	5	1.08 (0.56 to 2.08)	0.98 (0.71 to 1.36)
Functional treatment compared with 6 wks. of cast treatment
All studies	10	0.67 (0.50 to 0.90)	0.69 (0.50 to 0.94)
High-quality studies	5	0.79 (0.52 to 1.18)	0.58 (0.40 to 0.84)

View Larger

Table IV: Quality Sheet for Inclusion of Studies

Title:
Year:
No. of patients included and excluded:
Percentage followed:
Patient selection:	Stress radiograph (yes/no)	Arthrography (yes/no)
Criteria used for	stress radiograph:	arthrography:
Randomization:	yes/no	Design:
Treatment
Functional:	Duration:
Operative:	After treatment:	Duration:
Outcome variables:
Quality score:
Adequate randomization	yes/no
Included patients accounted for and attributed at conclusion	yes/no
Measurement of outcome blinded	yes/no

Add to My JBJS

Table IV: Quality Sheet for Inclusion of Studies

Title:
Year:
No. of patients included and excluded:
Percentage followed:
Patient selection:	Stress radiograph (yes/no)	Arthrography (yes/no)
Criteria used for	stress radiograph:	arthrography:
Randomization:	yes/no	Design:
Treatment
Functional:	Duration:
Operative:	After treatment:	Duration:
Outcome variables:
Quality score:
Adequate randomization	yes/no
Included patients accounted for and attributed at conclusion	yes/no
Measurement of outcome blinded	yes/no

View Larger

Table V: Operative Complications

*Complications were not mentioned in studies by Eggert et al.¹⁴, Freeman et al.^19,21, Kaikkonen et al.²⁹, Möller-Larsen et al⁴¹., Klein et al.³³, Prins⁴⁹, and Pace et al.⁴⁷.

Author of Study*	Total No. of Patients with Operation	Complications	No. of Patients with Complication
Broström⁷	96	Wound-healing disturbance	1
Sensory deficits	7
Scar neuroma	2
Stadelmayer et al.⁶⁰	30	Wound-healing disturbance	2
Povacz et al.⁴⁸	73	Wound-healing disturbance	2
Paresthesia	8
Zwipp and Krettek⁷¹	102	Wound-healing disturbance	2
Paresthesia	6
van Moppes and van den Hoogenband⁶⁸	50	Hyperesthesia	1
Niedermann et al.⁴³	102	Suppurating wound infection	3
Paresthesia	5
Scar neuroma	7
Evans et al.¹⁸	50	Lung embolus	1
Sensory deficits	6
Hyperesthesia	2
Sudeck atrophy	1
Van der Ent⁶⁵	332	Wound-healing disturbance	12
Sensory deficits	4
Hyperesthesia	13
Korkala et al.³⁷	34	Deep venous thrombosis	3

Add to My JBJS

Table V: Operative Complications

*Complications were not mentioned in studies by Eggert et al.¹⁴, Freeman et al.^19,21, Kaikkonen et al.²⁹, Möller-Larsen et al⁴¹., Klein et al.³³, Prins⁴⁹, and Pace et al.⁴⁷.

Author of Study*	Total No. of Patients with Operation	Complications	No. of Patients with Complication
Broström⁷	96	Wound-healing disturbance	1
Sensory deficits	7
Scar neuroma	2
Stadelmayer et al.⁶⁰	30	Wound-healing disturbance	2
Povacz et al.⁴⁸	73	Wound-healing disturbance	2
Paresthesia	8
Zwipp and Krettek⁷¹	102	Wound-healing disturbance	2
Paresthesia	6
van Moppes and van den Hoogenband⁶⁸	50	Hyperesthesia	1
Niedermann et al.⁴³	102	Suppurating wound infection	3
Paresthesia	5
Scar neuroma	7
Evans et al.¹⁸	50	Lung embolus	1
Sensory deficits	6
Hyperesthesia	2
Sudeck atrophy	1
Van der Ent⁶⁵	332	Wound-healing disturbance	12
Sensory deficits	4
Hyperesthesia	13
Korkala et al.³⁷	34	Deep venous thrombosis	3

Materials and Methods

Introduction | Materials and Methods | Results | Discussion | References

The aim of this study was to perform a meta-analysis of the effectiveness of existing treatment strategies for acute ruptures of the lateral ankle ligaments. We started with two hypotheses: (1) a no-treatment strategy for acute ruptures of the lateral ankle ligaments leads to more residual symptoms than does treatment of these injuries, and (2) operative treatment followed by functional aftertreatment leads to fewer residual symptoms than does functional treatment alone.

Randomized clinical trials of different forms of treatment for acute ruptures of the lateral ankle ligaments, reported between 1966 and 1998, were eligible for inclusion in our systematic review. Studies involving treatment of recurrent ankle injuries or chronic instability were not included. Studies with inadequate follow-up, defined as a follow-up of less than 60 percent of the patients or of an unclear percentage of the patients or a follow-up that was carried out exclusively with questionnaires, were excluded as well.

Trials comparing cast treatment with functional treatment, operative treatment with cast treatment, and operative treatment with functional treatment as well as studies comparing different forms of functional treatment were included. Strapping, bracing, use of an orthosis (such as an air cast), elastic wrapping, and use of special shoes for at least five weeks were considered to be functional treatment. A short period of cast immobilization (up to three weeks) was also considered to be a form of functional treatment as the immobilization was carried out for such a short period of time.

We searched the COCHRANE, MEDLINE, and EMBASE (Excerpta Medica) databases, supplemented by cross-bibliographic checks of the reference lists of the published reviews. The following Medical Subject Heading (MeSH) terms were used: ankle, ankle injuries, ankle joint, lateral ligament, clinical trial, randomized clinical trial, random allocation, double-blind method, single-blind method, and placebo⁶⁹. We made no limitation regarding the language in which the article was written. We contacted several authors personally to ask them whether they were aware of unpublished studies. This did not lead to additional data.

In most trials, an arthrogram or stress radiographs, or both, had been made to analyze the extent of damage to the lateral ligament complex. As the criteria for determining ligament rupture on arthrograms often were not provided, the findings as given by the author were accepted. Findings derived from stress radiographs were accepted if a clear description of talar tilt and the anterior drawer sign was stated.

Inadequacies in trial design may lead to overoptimistic estimates of treatment effectiveness³. Each trial was rated on the basis of three quality characteristics that are known to be most prone to bias²⁵: Was the assignment of patients randomized? Were all patients who entered the trial properly accounted for and attributed at its conclusion? Was the measurement of the outcome blinded? Randomization was defined as any form of random allocation of treatments to patients with concealment of allocation to the patients and clinicians at the point of inclusion in the study⁵³.

Eligible articles, blinded with regard to author and institution, were assessed by three independent investigators. These investigators checked the criteria for inclusion and exclusion, scored the three quality features, and extracted the data (Table IV). Differences between raters were resolved by consensus.

We only used outcome variables that allowed us to compare and summarize trial results. This means that the same outcome variables had to have been used in more than one trial. We used relative risks to express the relative effectiveness of treatment strategies^10,23. If two treatment strategies are compared in terms of residual symptoms, a relative risk lower than unity implies that fewer patients who underwent the former treatment had residual symptoms; if the relative risk exceeds one, the opposite is true.

A test of homogeneity of treatment effectiveness across similar trials was performed with use of the Q statistic (Table II). High Q values are associated with heterogeneity in trial results. Random effects modeling was used to obtain summary estimates of the relative risk.

Two analyses were performed. The first analysis was based on all included studies, and the second was performed on all high-quality studies. A study was considered high-quality when proper randomization had been performed with concealment of allocation and all included patients had been attributed at the conclusion.

Results

Introduction | Materials and Methods | Results | Discussion | References

A total of forty-two randomized, controlled trials were identified. Fifteen of them had to be excluded. Two studies were excluded because only physical examination had been used for diagnosis^27,45. Several studies^{4,5,15,42,45,46,54} excluded patients with a ligament rupture. Two studies were excluded because a large percentage of patients had been lost to follow-up^1,26, and four studies were excluded because follow-up had been conducted with use of a questionnaire only^2,15,26,35. Data from four studies was not analyzed because it was not possible to derive the relevant outcome variables^{5,28,38,57,58}.

Sometimes separate articles that reported on the same study but were published on different occasions were found^{14,43,57-59,63,64,71-74}.

The studies that were included in the meta-analysis are summarized in Table I. Except for two studies that included only soldiers^11,19, the trials (at least those in which the patient populations were defined) involved average patient samples from the general population.

Three outcome variables in the short-term follow-up studies were identified: time lost from work, inability to participate in sports, and pain and swelling. The variables identified in the long-term follow-up studies were giving-way, recurrent instability, pain, swelling, range of motion, objective stability (talar tilt and the anterior drawer sign on stress radiographs), and arthrosis. We were able to perform only a limited number of analyses. The only variable that was found in a sufficient number of short-term follow-up trials for us to perform the analysis was time lost from work. Giving-way and pain (the number of patients with giving-way and pain) were found in a sufficient number of long-term follow-up studies. Giving-way is defined as a feeling of uncertainty due to motor incoordination secondary to a disorder of proprioception or mechanical ankle instability³⁰. It was not possible to include recurrent instability, swelling, range of motion, objective instability, or arthrosis in the analysis as these variables had not been used in a sufficient number of trials or had been defined in several ways.

The tests for homogeneity showed substantial variability in some of the comparisons of treatment for pain, both in the group of all studies and in the subgroup of studies meeting the three prespecified quality criteria (Table II).

Because the results of the subgroup analysis of the high-quality studies only were not fundamentally different from the results of the corresponding analysis of all studies, we will discuss the results of the analysis of all studies (Table III).

We found no trials of a no-treatment policy reflecting natural history. However, we identified three studies that included a group of patients who received minimal treatment (Fig. 1-B and Fig. 1-B). Zeegers⁷⁰ included a group treated with an elastic sock, and Korkala et al.³⁷ and Dettori et al.¹¹ included patients treated with an elastic bandage for two weeks. These patient groups had more residual pain (relative risk, 0.53; 95 percent confidence interval, 0.27 to 1.02) and giving-way (relative risk, 0.34; 95 percent confidence interval, 0.17 to 0.71) than the patient groups that were treated with functional support in the same trials.

Seven trials were available for comparison between operative treatment followed by functional treatment and functional treatment only. All seven trials used a similar surgical protocol: approximation of the free ends of the ruptured ligament with sutures and closure of the capsule. Four of six trials in which pain (the number of patients with pain after a certain treatment) was analyzed showed better results after operative treatment, although the difference did not reach significance (Fig. 2-A). A significant difference between the two types of treatment, in favor of operative treatment, was found for giving-way in the seven trials (relative risk, 0.23; 95 percent confidence interval, 0.17 to 0.31) (Fig. 2-B). The mean time lost from work was comparable between operative treatment (forty-four days; range, thirty-four to fifty-seven days) and functional treatment (thirty-nine days; range, thirty-four to forty-seven days).

Ten trials were available for comparison between treatment with a cast for six weeks and functional treatment. Functional treatment led to a significantly better result with regard to pain (relative risk, 0.67; 95 percent confidence interval, 0.50 to 0.90) (Fig. 3-A) and giving-way (relative risk, 0.69; 95 percent confidence interval, 0.50 to 0.94) (Fig. 3-B). The mean time lost from work was fifteen days (range, twelve to eighteen days) in the group with functional treatment and thirty-eight days (range, twenty-eight to forty-eight days) in the group treated with a cast.

Nine trials were available for comparison between operative treatment followed by six weeks of cast immobilization and six weeks of cast immobilization only. The analysis of pain showed no significant difference between the treatments (relative risk, 0.81; 95 percent confidence interval, 0.40 to 1.67) (Fig. 4-A). The analysis of giving-way also showed no significant difference between treatments (relative risk, 0.90; 95 percent confidence interval, 0.69 to 1.18) (Fig. 4-B).

Two trials were available for comparison between different methods of functional treatment. Sommer et al.^58,59 compared three weeks of treatment with a plaster cast and six weeks of treatment with an air cast with six weeks of treatment with an air cast and eight weeks of treatment with strapping. The results for giving-way and pain were similar between treatment groups (relative risk, 1.12 [95 percent confidence interval, 0.7 to 17.22] and 1.00 [95 percent confidence interval, 0.7 to 15.33], respectively). Zeegers⁷⁰ compared four treatment strategies: five weeks in an air cast and an elastic sock, three weeks of treatment with strapping and two weeks in an elastic sock, five weeks of wearing a special shoe and an elastic sock, and five weeks of treatment with an elastic sock. The results in the first three treatment groups were not significantly different with regard to any of the outcome parameters. Treatment with an elastic sock alone, however, resulted in a higher percentage of both residual pain and residual giving-way (relative risk, 0.50 [95 percent confidence interval, 0.13 to 1.91] and 0.26 [95 percent confidence interval, 0.06 to 1.15], respectively). The mean time lost from work was twenty-five days in the group treated with an elastic sock alone and sixteen days in the group treated with strapping.

Discussion

Introduction | Materials and Methods | Results | Discussion | References

The treatment of ruptures of the lateral ankle ligaments is a subject of controversy. In this systematic review of the available literature, we found that treatment that was too short in duration or that did not include sufficient support of the ankle joint tended to result in more residual symptoms. Operative treatment led to the best long-term results.

When performing this meta-analysis, we encountered some difficulties that have to be taken into account when the results of this study are interpreted. We did not find any trials with blind measurement of the outcomes. This could have resulted in biased estimates of treatment effectiveness, producing optimistic impressions of more aggressive interventions.

The majority of studies were poor in quality. Several studies had a poor concealment of treatment allocation or did not account for all enrolled patients. Such a selective reporting of patient outcomes increases the potential for bias, leading to overoptimistic estimates of treatment effectiveness. To explore the extent of this bias, we performed both an analysis of all studies that could be identified and a second, subgroup analysis of the results of the studies that met the prespecified quality criteria. The results of the two analyses did not lead to fundamentally different conclusions, indicating that the size of the bias was only moderate in this particular application.

We had to limit our analysis to outcome variables that were available in a sufficient number of articles. As a consequence, some studies could not be included because of the scoring system that had been used or because important variables had not been analyzed at all.

In the present analysis, functional treatment was used as a category for different forms of treatment. It was not possible to perform a meta-analysis of different forms of functional treatment because of a lack of data. Nevertheless, differences in effectiveness within this category of functional treatment strategies cannot be excluded.

The substantial variability in trial results, as shown by the high values of the Q statistics, calls for additional caution with regard to generalizing the results of our analysis to all settings. As the variability was more pronounced for pain than for giving-way, it is possible that it is more difficult to obtain objective and reproducible measures of pain.

Although the natural history of ankle sprains is not fully known, it has been suggested that even untreated ligament ruptures have a good prognosis. We did not find any trials in which the natural history had been studied, but some trials involved patients treated with only minimal support or for a short duration. Significantly more residual symptoms were found in patients treated with these strategies compared with those who received functional treatment or surgery. We therefore concluded that more residual symptoms will occur if a patient with a rupture of the lateral ankle ligaments does not receive treatment.

The summary measures of effectiveness of operative treatment were better than those of functional treatment. This finding is in conflict with the results of former literature reviews by Kannus and Renström³⁰, Tiling et al.⁶², Ogilvie-Harris and Gilbart⁴⁴, and Shrier⁵⁵, who all found functional treatment to be the treatment of choice. An important difference between these papers and our study is that the previous reviews were descriptive and no statistical analysis was performed. We performed a formal meta-analysis of the data derived from the available clinical trials. Another difference is that, in the former reviews, operative treatment was studied regardless of the aftertreatment. In our analysis, we found operative treatment followed by functional treatment or by a short (three-week) period of cast immobilization to lead to better results than functional treatment alone. We found functional treatment to be better than a longer period of cast treatment. However, operative treatment followed by six weeks of plaster-cast immobilization produced results that were comparable with those after six weeks of cast treatment only. This finding suggests (although direct comparisons were not performed) that operative treatment followed by six weeks of cast treatment is likely to be inferior to operative treatment followed by functional treatment or by a short (three-week) period of cast immobilization. This observation is in accordance with that of Salter and Field⁵¹, who suggested that functional aftertreatment leads to a faster recovery.

We evaluated several variables in both short and long-term follow-up studies and tried to compare them among all identified studies. Time lost from work was the only outcome variable available for analysis of the short-term follow-up studies, and there was a wide variation in this time among these studies. We think that resumption of work does not necessarily give a good indication of recovery of the injured ankle. A return to work depends on the advice that the patient receives, job satisfaction, Workers' Compensation schedules, and type of job. Some jobs can be easily carried out while wearing a cast or orthosis while others cannot.

One of the most important variables, which is commonly accepted as a primary functional outcome of treatment of a lateral ligament rupture, is giving-way. A ruptured ligament that does not heal or that heals with elongation can give rise to secondary instability. Giving-way is a sign of secondary mechanical instability and can also be a disorder of proprioception^19-22. In these cases, giving-way seems to be independent of the grade of the initial injury^52,68. Pain was the other long-term follow-up variable for which summary estimates of treatment effect could be made. It remains unclear why so many patients have pain in the involved ankle at the time of long-term follow-up. The heterogeneity of the results and the fact that differences in pain between treatment groups did not reach significance in the random effects analysis suggest that the cause of pain lies in the initial injury and that patient response may be less influenced by the method of treatment.

Analysis of the pooled results showed operative treatment to be superior to functional treatment, yet there are reasons to question the selection of operative treatment as the treatment of choice. Operative treatment is associated with an increased risk of complications, such as disturbance of wound-healing, infection, dystrophy, iatrogenic nerve damage leading to sensory deficit, paresthesia, and so on (Table V). Operative treatment is also associated with higher costs. Because of the high prevalence of ankle injuries, operative repair may be performed by surgeons in training, which may have affected some of the published results. This can account for differences in results among the various trials. In studies with a favorable outcome of operative treatment, patients were operated on by the senior author^6,49,65. Finally, when conservative treatment fails, secondary operative reconstruction of the ruptured ligaments can be performed, with similar good results, even years after the initial injury^{7,9,16-18,31,40,56,61}.

References

Introduction | Materials and Methods | Results | Discussion | References

Allen, M. J., and McShane, M.: Inversion injuries to the lateral ligament of the ankle joint. A pilot study of treatment. British J. Clin. Pract.,39: 282-286, 1985.39282 1985

Andersen, K.; Albers, C.; Mï¿½-Madsen, B.; Niedermann, B.; Simonsen, O.; and Snorrason, N.: Conservative treatment of severe ankle sprains [abstract]. Acta Orthop. Scandinavica,58: 697-698, 1987.58697 1987

Berlin, J. A., and Rennie, D.: Measuring the quality of trials: the quality of quality scales [editorial]. J. Am. Med. Assn.,282: 1083-1085, 1999.2821083 1999

Brakenbury, P. H., and Kotowski, J.: A comparative study of the management of ankle sprains. British J. Clin. Pract.,37: 181-185, 1983.37181 1983

Brooks, S. C.; Potter, B. T.; and Rainey, J. B.: Inversion injuries of the ankle: clinical assessment and radiographic review. British Med. J. (Clin. Res. Ed.),282: 607-608, 1981.282607 1981

Broström, L.: Sprained ankles. V. Treatment and prognosis in recent ligament ruptures. Acta Chir. Scandinavica,132: 537-550, 1966.132537 1966

Broström, L.: Sprained ankles. VI. Surgical treatment of "chronic" ligament ruptures. Acta Chir. Scandinavica,132: 551-565, 1966.132551 1966

Cetti, R.: Conservative treatment of injury to the fibular ligaments of the ankle. British J. Sports Med.,16: 47-52, 1982.1647 1982

Chrisman, O. D., and Snook, G. A.: 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 and Joint Surg.,51-A: 904-912, July 1969.51-A904 1969

Cooper, H., and Hedges, L.: The Handbook of Research Synthesis. New York, Russell Sage Foundation, 1994.

Dettori, J. R.; Pearson, B. D.; Basmania, C. J.; and Lednar, W. M.: Early ankle mobilization, Part I: The immediate effect on acute, lateral ankle sprains (a randomized clinical trial). Milit. Med.,,159: 15-20, 1994.15915 1994

Dettori, J. R., and Basmania, C. J.: Early ankle mobilization, Part II: A one-year follow-up of acute, lateral ankle sprains (a randomized clinical trial). Milit. Med.,,159: 20-24, 1994.15920 1994

Duquennoy, A., and Lisè©lçª D.: Diagnostic de rupture ligamentaire externe; è©ments cliniques du diagnostic de gravitç?¤e l'entorse. Rev. chir. orthop.,61 (Supplement 2): 132-133, 1975.61 (Supplement 2)132 1975

Eggert, A.; Grüber, J.; and Darda, L.: Zur Therapie von AuÞ¥nknöcheldbandverletzungen. Randomstudie zur postoperativen Therapie und frühfunktionellen Behandlungstaktik. Unfallchirurg,89: 316-320, 1986.89316 1986 [PubMed]

Eiff, M. P.; Smith, A. T.; and Smith, G. E.: Early mobilization versus immobilization in the treatment of lateral ankle sprains. Am. J. Sports Med.,22: 83-88, 1994.2283 1994 [PubMed]

Elmslie, R. C.: Recurrent subluxation of the ankle-joint. Ann. Surg.,100: 364-367, 1934.100364 1934 [PubMed]

Evans, D. L.: Recurrent instability of the ankle - a method of surgical treatment. Proc. Roy. Soc. Med.,46: 343-344, 1953.46343 1953

Evans, G. A.; Hardcastle, P.; and Frenyo, A. D.: Acute rupture of the lateral ligament of the ankle. To suture or not to suture?. J. Bone and Joint Surg.,66-B(2): 209-212, 1984.66-B(2)209 1984

Freeman, M. A. R.: Treatment of ruptures of the lateral ligament of the ankle. J. Bone and Joint Surg.,47-B(4): 661-668, 1965.47-B(4)661 1965

Freeman, M. A. R.: Instability of the foot after injuries to the lateral ligament of the ankle. J. Bone and Joint Surg.,47-B(4): 669-677, 1965.47-B(4)669 1965

Freeman, M. A. R.; Dean, M. R. E.; and Hanham, I. W. F.: The etiology and prevention of functional instability of the foot. J. Bone and Joint Surg.,47-B(4): 678-685, 1965..47-B(4)678 1965.

Freeman, M. A., and Wyke, B.: The innervation of the ankle joint. An anatomical and histological study in the cat. Acta Anat.,68: 321-333, 1967.68321 1967 [PubMed]

Friedman, L. M.: Fundamentals of Clinical Trials. Ed. 3. St. Louis, Mosby-Year Book, 1996.

Grönmark, T.; Johnsen, O.; and Kogstad, O.: Rupture of the lateral ligaments of the ankle: a controlled clinical trial. Injury,,11: 215-218, 1980.11215 1980

Guyatt, G. H.; Sackett, D. L.; and Cook, D. J.: Users' guides to the medical literature. II. How to use an article about therapy or prevention. A. Are the results of the study valid? Evidence-Based Medicine Working Group. J. Am. Med. Assn.,,270: 2598-2601, 1993.2702598 1993

Hedges, J. R., and Anwar, R. A.: Management of ankle sprains. Ann. Emerg. Med.,9: 298-302, 1980.9298 1980 [PubMed]

Johannes, E. J.; Sukul, D. M.; Spruit, P. J.; and Putters, J. L.: Controlled trial of a semi-rigid bandage (`Scotchrap') in patients with ankle ligament lesions. Curr. Med. Res. and Opin.,,13: 154-162, 1993.13154 1993

Jongen, S. J. M.; Pot, J. H.; and Dunki-Jacobs, P. B.: De behandeling van de verzwikte enkel. Geneesk. Sport,,25: 98-101, 1992.2598 1992

Kaikkonen, A.; Kannus, P.; and Jarvinen, M.: Surgery versus functional treatment in ankle ligament tears. A prospective study. Clin. Orthop.,326: 194-202, 1996.326194 1996 [PubMed]

Kannus, P., and Renström, P: Current concepts review. Treatment for acute tears of the lateral ligaments of the ankle. Operation, cast, or early controlled mobilization. J. Bone and Joint Surg.,73-A: 305-312, Feb 1991.73-A305 1991

Karlsson, J.; Bergsten, T.; Lansinger, O.; and Peterson, L.: Reconstruction of the lateral ligaments of the ankle for chronic lateral instability. J. Bone and Joint Surg.,,70-A: 581-588, April 1988.70-A581 1988

Keeman, J. N.: Commentaar enkelspecial. Reuma Trauma,1: 34-35, 1990.134 1990

Klein, J.; Schreckenberger, C.; Röddecker, K.; and Tiling, T.: Operative oder Konservative Behandlung der frischen Aussenbandruptur des oberen Sprunggelenks. Eine randomisierte klinische Studie. Unfallchirurg,,91: 154-160, 1988.91154 1988

Klein, J.; Rixen, D.; Albring, T.; and Tiling, T.: Funktionelle versus Gipsbehandlung bei der frischen AuBenbandruptur des oberen Sprunggelenks. Eine randomisierte klinische Studie. Unfallchirurg,94: 99-104, 1991.9499 1991 [PubMed]

Konradsen, L.; Holmer, P.; and Sondergaard, L.: Early mobilizing treatment for grade III ankle ligament injuries. Foot and Ankle,12: 69-73, 1991.1269 1991 [PubMed]

Korkala, O.; Rusanen, M.; Kytömaa, J.; Jokipii, P.; and Avikainen, V.: Treatment of lateral ligament injuries of the ankle. A prospective clinical study [abstract]. Acta Orthop. Scandinavica,,57: 579, 1986.57579 1986

Korkala, O.; Rusanen, M.; Jokipii, P.; Kytomaa, J.; and Avikainen, V.: A prospective study of the treatment of severe tears of the lateral ligament of the ankle. Internat. Orthop.,11: 13-17, 1987.1113 1987

Leanderson, J., and Wredmark, T.: Treatment of acute ankle sprain. Comparison of a semi-rigid ankle brace and compression bandage in 73 patients. Acta Orthop. Scandinavica,,66: 529-531, 1995.66529 1995

McCulloch, P. G.; Holden, P.; Robson, D. J.; Rowley, D. I.; and Norris, S. H.: The value of mobilisation and non-steroidal anti-inflammatory analgesia in the management of inversion injuries of the ankle. British J. Clin. Pract.,39: 69-72, 1985.3969 1985

Marti, R. K.: Bandapparat des oberen sprunggelenkes. 50e Jahrestagung der Deutschen Gesellschaft fur Unfallheilunde. Berlin, Springer, 1986.

Möller-Larsen, F.; Wethelund, J. O.; Jurik, A. G.; de Carvalho, A.; and Lucht, U.: Comparison of three different treatments for ruptured lateral ankle ligaments. Acta Orthop. Scandinavica,,59: 564-566, 1988.59564 1988

Muwanga, C. L.; Quinton, D. N.; Sloan, J. P.; Gillies, P.; and Dove, A. F.: A new treatment of stable lateral ligament injuries of the ankle joint. Injury,,17: 380-382, 1986.17380 1986

Niedermann, B.; Andersen, A.; Andersen, S. B.; Funder, V.; Jorgensen, J. P.; Lindholmer, F.; and Vuust, M.: Rupture of the lateral ligaments of the ankle: operation or plaster cast? A propective study. Acta Orthop. Scandinavica,,52: 579-587, 1981.52579 1981

Ogilvie-Harris, D. J., and Gilbart, M.: Treatment modalities for soft tissue injuries of the ankle: a critical review. Clin. J. Sport Med.,,5: 175-186, 1995.5175 1995

O'Hara, J.; Valle-Jones, J. C.; Walsh, H.; O'Hara, H.; Davey, N. B.; Hopkin-Richards, H.; and Butcher, R. M.: Controlled trial of an ankle support (Malleotrain) in acute ankle injuries. British J. Sports Med.,26: 139-142, 1992.26139 1992

Oostendorp, R. A. B.: Functionele instabiliteit na het inversietrauma van de enkel en voet: een effectonderzoek van pleisterbandage versus pleisterbandage met fysiotherapie. Geneesk. Sport,,20: 45-55, 1987.2045 1987

Pace, N.; Lo Iacono, E.; Chiusaroli, M.; Serafini, P.; Basconi, G.; and Zanoli, S.: A comparison between conservative treatment and operative treatment (capsulosyndesmorrhaphy) in severe acute capsuloligamentous lateral lesions of the ankle joint. Italian J. Orthop. and Traumatol.,,16: 19-24, 1990.1619 1990

Povacz, P.; Unger, S. F.; Miller, W. K.; Tockner, R.; and Resch, H.: A randomized, prospective study of operative and non-operative treatment of injuries of the fibular collateral ligaments of the ankle. J. Bone and Joint Surg.,80-A: 345-351, March 1998.80-A345 1998

Prins, J. G.: Diagnosis and treatment of injury to the lateral ligament of the ankle. A comparative clinical study. Acta Chir. Scandinavica,Supplementum: 486, 1978.Supplementum486 1978

Ruth, C. J.: The surgical treatment of injuries of the fibular collateral ligaments of the ankle. J. Bone and Joint Surg.,43-A: 229-239, March 1961.43-A229 1961

Salter, R. B., and Field, P.: The effects of continuous compression on living articular cartilage. An experimental investigation. J. Bone and Joint Surg.,42-A: 31-49, Jan 1960.42-A31 1960

Schaap, G. R.; de Keizer, G.; and Marti, K.: Inversion trauma of the ankle. Arch. Orthop. and Trauma Surg.,,108: 273-275, 1989.108273 1989

Schulz, K. F.; Chalmers, I.; Grimes, D. A.; and Altman, D. G.: Assessing the quality of randomization from reports of controlled trials published in obstetrics and gynecology journals. J. Am. Med. Assn.,272: 125-128, 1994.272125 1994

Scotece, G. G., and Guthrie, M. R.: Comparison of three treatment approaches for grade I and II ankle sprains in active duty soldiers. J. Orthop. and Sports Phys. Ther.,,15: 19-23, 1992.1519 1992

Shrier, I.: Treatment of lateral collateral ligament sprains of the ankle: a critical appraisal of the literature. Clin. J. Sport Med.,5: 187-195, 1995.5187 1995 [PubMed]

Snook, G. A.; Chrisman, O. D.; and Wilson, T. C.: Long-term results of the Chrisman-Snook operation for reconstruction of the lateral ligaments of the ankle. J. Bone and Joint Surg.,67-A: 1-7, Jan 1985.67-A1 1985

Sommer, H. M.,, and Arza, D.: Die konservative funktionelle Behandlung der fibularen Kapselbandruptur auch beim Leistungssportler?. Sportverletz.-Sportschaden,1: 25-28, 1987.125 1987 [PubMed]

Sommer, H. M., and Arza, D.: Functional treatment of recent ruptures of the fibular ligament of the ankle. Internat. Orthop.,13: 157-160, 1989.13157 1989

Sommer, H. M., and Schreiber, H.: Die fruh-funktionelle konservative Therapie der frischen fibularen Kapsel-Band-Ruptur aus sozial-okonomischer Sicht. Sportverletz.-Sportschaden,7: 40-46, 1993.740 1993 [PubMed]

Stadelmayer, B.; Dauber, A.; and Pelzl, H.: Operative oder konservative Therapie der Aussenbandruptur am oberen Sprunggelenk?. Unfallchirurgie,18: 37-43, 1992.1837 1992 [PubMed]

St. Pierre, R.; Allman, F., Jr.; Bassett, F. H., III; Goldner, J. L.; and Fleming, L. L.: A review of lateral ankle ligamentous reconstructions. Foot and Ankle,3: 114-123, 1982.3114 1982 [PubMed]

Tiling, T.; Bonk, A.; Hoher, J.; and Klein, J.: Die akute Aussenbandverletzung des Sprunggelenks beim Sportler. Chirurg,65: 920-933, 1994.65920 1994 [PubMed]

van den Hoogenband, C. R.; van Moppes, F. I.; and Coumans, P. F.: Study on clinical diagnosis and treatment of lateral ligament lesion of the ankle joint. Internat. J. Sports Med.,5 (Supplement): 159-161, 1984.5 (Supplement)159 1984

van den Hoogenband, C. R.,, and van Moppes, F. I.:: Die behandlung der lateralen Ligamentrupturen des oberen Sprunggelenkes mit der Coumans-Bandage und direkter Mobilisation. Eine prospective Vergleichsstudie. Hefte Unfallheilk.,189: 1030-1034, 1987.1891030 1987

Van der Ent, F.: Lateral ankle ligament injury. Thesis, Erasmus Universiteit, Rotterdam, 1984.

van Dijk, C. N.: On diagnostic strategies in patients with severe ankle sprain. Thesis, Universiteit van Amsterdam, Amsterdam, 1994.

van Dijk, C. N.; Bossuyt, P. M. M.; and Marti, R. K.: Medial ankle pain after lateral ligament rupture. J. Bone and Joint Surg.,78-B(4): 562-567, 1996.78-B(4)562 1996

van Moppes, F. I., and van den Hoogenband, C. R.: Diagnostic and therapeutic aspects of inversion trauma of the ankle joint. Thesis, Universiteit van Limburg, Maastricht, 1982.

Walker-Dilks, C. J.; McKibbon, K. A.; and Haynes, R. B.: Beyond ACP Journal Club: how to harness Medline for etiology problems [editorial]. ACP J. Club,,121: 10-A11, 1994.12110 1994

Zeegers, A. V. C. M.: Het supinatie van de enkel. Thesis, Universiteit van Utrecht, Utrecht, 1995.

Zwipp, H., and Krettek, C.: Diagnostik und Therapie der akuten und chronischen Bandinstabilitat des unteren Sprunggelenkes. Orthopüde,15: 472-478, 1986.15472 1986

Zwipp, H.; Tscherne, H.; Hoffmann, R.; and Wippermann, B. W.: Therapie der frischen fibularen Bandruptur. Orthopüde,15: 446-453, 1986.15446 1986

Zwipp, H.: Stellungsnahme zur operative Therapie des kapselbandruptur. Sportverletz.-Sportschaden,3: 10-13, 1989.310 1989

Zwipp, H.; Hoffmann, R.; Thermann, H.; and Wippermann, B. W.: Rupture of the ankle ligaments. Internat. Orthop.,,15: 245-249, 1991.15245 1991

Topics

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1-B:: Relative risk for giving-way.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-A:: Figs. 2-A and 2-B: Studies in which patients who received operative treatment were compared with those who received functional treatment.
Fig. 2-A: Relative risk for pain.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 2-B:: Relative risk for giving-way.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-A:: Figs. 3-A and 3-B: Studies in which patients who received functional treatment were compared with those who were treated with a cast for six weeks.
Fig. 3-A: Relative risk for pain.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 3-B:: Relative risk for giving-way.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 4-B:: Relative risk for giving-way.

View Larger

Table I: Randomized Clinical Trials Included in Meta-Analysis of Treatment of Lateral Ligament Rupture

Author of Study	Yr. of Study	Treatment	No. of Patients	Duration of Follow-up(yrs.)	Arthrography	Stress Radiography	Quality Criteria*
Freeman¹⁹	1965	Op. + cast, 6 wks.	16	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	18
Strapping, 6 wks.	12
Freeman et al.²¹	1965	Op. + cast, 6 wks.	12	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	14
Strapping, 6 wks.	16
Broström⁶	1966	Op. + cast, 3 wks.	96	3.8	+		1/1/0
Cast, 3 wks.	86
Strapping	102
Prins⁴⁹	1978	Op. + cast, 3 wks.	45	6 mos.	+		0/1/0
Cast, 6 wks.	59
Elastic wrap, 3 wks. (no lesion)	101
Elastic wrap, 3 wks. (lesion)	42
Cast, 3 wks.	51
Grönmark et al.²⁴	1980	Op. + cast, 6 wks.	32	3.8	-	Talar tilt L/R diff.	1/1/0
Cast, 6 wks.	33
Strapping, 6 wks.	30
Niedermann et al.⁴³	1981	Op. + cast, 5 wks.	102	1	+		0/1/0
Cast, 5 wks.	107
Strapping, 5 wks. (no lesion)	146
van Moppes and van den Hoogenband⁶⁸	1982	Op. + cast, 5 wks.	50	1	+	Talar tilt L/R diff. > 5.5°	1/1/0
Cast, 6 wks.	50
Strapping, 6 wks.	50
Cetti⁸	1982	Cast, 6 wks.	65	6 mos.	-	Ant. drawer sign > 3 mm, talar tilt > 6°	1/1/0
Brace + pliton shoe, 6 wks.	65
Evans et al.¹⁸	1984	Op. + cast, 3 wks.	50	2	+	Talar tilt L/R diff.	1/1/0
Cast, 3 wks.	50
Van der Ent†⁶⁵	1984	Op. + cast, 3 wks.	332	1	+		1/1/0
Cast, 3 wks.	40
Eggert et al.¹⁴	1986	Op. + cast, 6 wks.	36	1.3	-	Ant. drawer sign > 8 mm, talar tilt > 8°	1/1/0
Op. + shoe, 6 wks.	38
Shoe, 6 wks.	50
Zwipp et al.^71,72,74	1986, 1986, 1991	Op. + cast, 6 wks.	52	2	+		1/0/0
Op. + phys. therapy	50
Cast, 5 wks.	48
Brace + phys. therapy, 5 wks.	50
Korkala et al.³⁷	1987	Op. + cast, 4 wks.	34	2	-	Ant. drawer sign > 6 mm, talar tilt > 15°, L/R diff. > 3 mm or 10°	1/0/0
Cast, 4 wks.	47
Elastic bandage, 2 wks.	36
Klein et al.³³	1988	Op. + cast, 6 wks.	30	6 mos.	-	Ant. drawer sign > 7 mm, talar tilt > 7°	1/1/0
Cast, 6 wks.	30
Möller-Larsen et al.⁴¹	1988	Op. + cast, 6 wks.	55	1	+		1/0/0
Cast, 5 wks.	55
Strapping, 5 wks.	65
Pace et al.⁴⁷	1990	Op. + cast, 5 wks.	12	2	+		0/1/0
Cast, 4 wks.	18
Klein et al.³⁴	1991	Cast, 6 wks.	30	1	-	Ant. drawer sign > 7 mm, talar tilt > 7°, L/R diff. 7 mm and 7°	1/1/0
Air cast + phys. therapy	30
Stadelmayer et al.⁶⁰	1992	Op. + cast, 6 wks.	30	1	-	Ant. drawer sign > 10 mm, talar tilt > 10°	1/1/0
Cast, 6 wks.	30
Sommer et al.^58,59	1989, 1993	Cast + air cast, 9 wks.	3	6 mos.	+	Talar tilt > 10°	1/1/0
Air cast, 6 wks.	37
Strapping, 8 wks.	33
Dettori et al.^11,12	1994	Cast + phys. therapy, 5 wks.	28	1	+		1/1/0
Air cast + phys. therapy, 5 wks.	22
Elastic bandage + phys. therapy, 5 wks.	24
Zeegers‡⁷⁰	1995	Special shoe + elastic sock, 5 wks.	60	1	+		1/1/0
Elastic sock, 5 wks.	62
Strapping, 3 wks. + elastic sock, 2 wks.	59
Air cast + elastic sock, 5 wks.	62
Kaikkonen et al.²⁹	1996	Op. + air cast, 5 wks.	30	9 mos.	-	Ant. drawer sign L/R diff., talar tilt > 15°	1/1/0
Air cast, 5 wks.	51
Povacz et al.⁴⁸	1998	Op. + air cast, 6 wks.	73	2	-	Talar tilt L/R diff. > 5°	1/1/0
Air cast, 6 wks.	73

Add to My JBJS

Table I: Randomized Clinical Trials Included in Meta-Analysis of Treatment of Lateral Ligament Rupture

Author of Study	Yr. of Study	Treatment	No. of Patients	Duration of Follow-up(yrs.)	Arthrography	Stress Radiography	Quality Criteria*
Freeman¹⁹	1965	Op. + cast, 6 wks.	16	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	18
Strapping, 6 wks.	12
Freeman et al.²¹	1965	Op. + cast, 6 wks.	12	1	-	Talar tilt L/R diff. > 6°	0/1/0
Cast, 6 wks.	14
Strapping, 6 wks.	16
Broström⁶	1966	Op. + cast, 3 wks.	96	3.8	+		1/1/0
Cast, 3 wks.	86
Strapping	102
Prins⁴⁹	1978	Op. + cast, 3 wks.	45	6 mos.	+		0/1/0
Cast, 6 wks.	59
Elastic wrap, 3 wks. (no lesion)	101
Elastic wrap, 3 wks. (lesion)	42
Cast, 3 wks.	51
Grönmark et al.²⁴	1980	Op. + cast, 6 wks.	32	3.8	-	Talar tilt L/R diff.	1/1/0
Cast, 6 wks.	33
Strapping, 6 wks.	30
Niedermann et al.⁴³	1981	Op. + cast, 5 wks.	102	1	+		0/1/0
Cast, 5 wks.	107
Strapping, 5 wks. (no lesion)	146
van Moppes and van den Hoogenband⁶⁸	1982	Op. + cast, 5 wks.	50	1	+	Talar tilt L/R diff. > 5.5°	1/1/0
Cast, 6 wks.	50
Strapping, 6 wks.	50
Cetti⁸	1982	Cast, 6 wks.	65	6 mos.	-	Ant. drawer sign > 3 mm, talar tilt > 6°	1/1/0
Brace + pliton shoe, 6 wks.	65
Evans et al.¹⁸	1984	Op. + cast, 3 wks.	50	2	+	Talar tilt L/R diff.	1/1/0
Cast, 3 wks.	50
Van der Ent†⁶⁵	1984	Op. + cast, 3 wks.	332	1	+		1/1/0
Cast, 3 wks.	40
Eggert et al.¹⁴	1986	Op. + cast, 6 wks.	36	1.3	-	Ant. drawer sign > 8 mm, talar tilt > 8°	1/1/0
Op. + shoe, 6 wks.	38
Shoe, 6 wks.	50
Zwipp et al.^71,72,74	1986, 1986, 1991	Op. + cast, 6 wks.	52	2	+		1/0/0
Op. + phys. therapy	50
Cast, 5 wks.	48
Brace + phys. therapy, 5 wks.	50
Korkala et al.³⁷	1987	Op. + cast, 4 wks.	34	2	-	Ant. drawer sign > 6 mm, talar tilt > 15°, L/R diff. > 3 mm or 10°	1/0/0
Cast, 4 wks.	47
Elastic bandage, 2 wks.	36
Klein et al.³³	1988	Op. + cast, 6 wks.	30	6 mos.	-	Ant. drawer sign > 7 mm, talar tilt > 7°	1/1/0
Cast, 6 wks.	30
Möller-Larsen et al.⁴¹	1988	Op. + cast, 6 wks.	55	1	+		1/0/0
Cast, 5 wks.	55
Strapping, 5 wks.	65
Pace et al.⁴⁷	1990	Op. + cast, 5 wks.	12	2	+		0/1/0
Cast, 4 wks.	18
Klein et al.³⁴	1991	Cast, 6 wks.	30	1	-	Ant. drawer sign > 7 mm, talar tilt > 7°, L/R diff. 7 mm and 7°	1/1/0
Air cast + phys. therapy	30
Stadelmayer et al.⁶⁰	1992	Op. + cast, 6 wks.	30	1	-	Ant. drawer sign > 10 mm, talar tilt > 10°	1/1/0
Cast, 6 wks.	30
Sommer et al.^58,59	1989, 1993	Cast + air cast, 9 wks.	3	6 mos.	+	Talar tilt > 10°	1/1/0
Air cast, 6 wks.	37
Strapping, 8 wks.	33
Dettori et al.^11,12	1994	Cast + phys. therapy, 5 wks.	28	1	+		1/1/0
Air cast + phys. therapy, 5 wks.	22
Elastic bandage + phys. therapy, 5 wks.	24
Zeegers‡⁷⁰	1995	Special shoe + elastic sock, 5 wks.	60	1	+		1/1/0
Elastic sock, 5 wks.	62
Strapping, 3 wks. + elastic sock, 2 wks.	59
Air cast + elastic sock, 5 wks.	62
Kaikkonen et al.²⁹	1996	Op. + air cast, 5 wks.	30	9 mos.	-	Ant. drawer sign L/R diff., talar tilt > 15°	1/1/0
Air cast, 5 wks.	51
Povacz et al.⁴⁸	1998	Op. + air cast, 6 wks.	73	2	-	Talar tilt L/R diff. > 5°	1/1/0
Air cast, 6 wks.	73

View Larger

Table II: Test for Homogeneity

*High Q values are associated with heterogeneity.

Comparison	Q Values*
Pain	Giving-Way
Minimal treatment compared with functional treatment	Q = 0.28 (p = 0.87)	Q = 1.00 (p = 0.61)
Op. treatment compared with functional treatment
All studies	Q = 42.70 (p < 0.001)	Q = 5.47 (p = 0.49)
High-quality studies	Q = 40.08 (p < 0.001)	Q = 3.87 (p = 0.28)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment only
All studies	Q = 42.48 (p < 0.001)	Q = 2.99 (p = 0.88)
High-quality studies	Q = 10.86 (p = 0.012)	Q = 1.97 (p = 0.80)
Functional treatment compared with 6 wks. of cast treatment
All studies	Q = 2.29 (p = 0.94)	Q = 12.78 (p = 0.08)
High-quality studies	Q = 0.34 (p = 0.99)	Q = 0.45 (p = 0.98)

Add to My JBJS

Table II: Test for Homogeneity

*High Q values are associated with heterogeneity.

Comparison	Q Values*
Pain	Giving-Way
Minimal treatment compared with functional treatment	Q = 0.28 (p = 0.87)	Q = 1.00 (p = 0.61)
Op. treatment compared with functional treatment
All studies	Q = 42.70 (p < 0.001)	Q = 5.47 (p = 0.49)
High-quality studies	Q = 40.08 (p < 0.001)	Q = 3.87 (p = 0.28)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment only
All studies	Q = 42.48 (p < 0.001)	Q = 2.99 (p = 0.88)
High-quality studies	Q = 10.86 (p = 0.012)	Q = 1.97 (p = 0.80)
Functional treatment compared with 6 wks. of cast treatment
All studies	Q = 2.29 (p = 0.94)	Q = 12.78 (p = 0.08)
High-quality studies	Q = 0.34 (p = 0.99)	Q = 0.45 (p = 0.98)

View Larger

Table III: Summary Estimates of Effectiveness

*The data is given as the relative risk, with the 95 percent confidence interval in parentheses.

Comparison	No. of Trials	Pain*	Giving-Way*
Minimal treatment compared with functional treatment	3	0.53 (0.27 to 1.02)	0.34 (0.17 to 0.71)
Op. treatment compared with functional treatment
All studies	7	0.52 (0.18 to 1.53)	0.23 (0.17 to 0.31)
High-quality studies	5	0.51 (0.14 to 1.94)	0.22 (0.15 to 0.32)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment
All studies	9	0.81 (0.40 to 1.67)	0.90 (0.69 to 1.18)
High-quality studies	5	1.08 (0.56 to 2.08)	0.98 (0.71 to 1.36)
Functional treatment compared with 6 wks. of cast treatment
All studies	10	0.67 (0.50 to 0.90)	0.69 (0.50 to 0.94)
High-quality studies	5	0.79 (0.52 to 1.18)	0.58 (0.40 to 0.84)

Add to My JBJS

Table III: Summary Estimates of Effectiveness

*The data is given as the relative risk, with the 95 percent confidence interval in parentheses.

Comparison	No. of Trials	Pain*	Giving-Way*
Minimal treatment compared with functional treatment	3	0.53 (0.27 to 1.02)	0.34 (0.17 to 0.71)
Op. treatment compared with functional treatment
All studies	7	0.52 (0.18 to 1.53)	0.23 (0.17 to 0.31)
High-quality studies	5	0.51 (0.14 to 1.94)	0.22 (0.15 to 0.32)
Op. treatment followed by 6 wks. of cast treatment compared with 6 wks. of cast treatment
All studies	9	0.81 (0.40 to 1.67)	0.90 (0.69 to 1.18)
High-quality studies	5	1.08 (0.56 to 2.08)	0.98 (0.71 to 1.36)
Functional treatment compared with 6 wks. of cast treatment
All studies	10	0.67 (0.50 to 0.90)	0.69 (0.50 to 0.94)
High-quality studies	5	0.79 (0.52 to 1.18)	0.58 (0.40 to 0.84)

View Larger

Table IV: Quality Sheet for Inclusion of Studies

Title:
Year:
No. of patients included and excluded:
Percentage followed:
Patient selection:	Stress radiograph (yes/no)	Arthrography (yes/no)
Criteria used for	stress radiograph:	arthrography:
Randomization:	yes/no	Design:
Treatment
Functional:	Duration:
Operative:	After treatment:	Duration:
Outcome variables:
Quality score:
Adequate randomization	yes/no
Included patients accounted for and attributed at conclusion	yes/no
Measurement of outcome blinded	yes/no

Add to My JBJS

Table IV: Quality Sheet for Inclusion of Studies

Title:
Year:
No. of patients included and excluded:
Percentage followed:
Patient selection:	Stress radiograph (yes/no)	Arthrography (yes/no)
Criteria used for	stress radiograph:	arthrography:
Randomization:	yes/no	Design:
Treatment
Functional:	Duration:
Operative:	After treatment:	Duration:
Outcome variables:
Quality score:
Adequate randomization	yes/no
Included patients accounted for and attributed at conclusion	yes/no
Measurement of outcome blinded	yes/no

View Larger

Table V: Operative Complications

*Complications were not mentioned in studies by Eggert et al.¹⁴, Freeman et al.^19,21, Kaikkonen et al.²⁹, Möller-Larsen et al⁴¹., Klein et al.³³, Prins⁴⁹, and Pace et al.⁴⁷.

Author of Study*	Total No. of Patients with Operation	Complications	No. of Patients with Complication
Broström⁷	96	Wound-healing disturbance	1
Sensory deficits	7
Scar neuroma	2
Stadelmayer et al.⁶⁰	30	Wound-healing disturbance	2
Povacz et al.⁴⁸	73	Wound-healing disturbance	2
Paresthesia	8
Zwipp and Krettek⁷¹	102	Wound-healing disturbance	2
Paresthesia	6
van Moppes and van den Hoogenband⁶⁸	50	Hyperesthesia	1
Niedermann et al.⁴³	102	Suppurating wound infection	3
Paresthesia	5
Scar neuroma	7
Evans et al.¹⁸	50	Lung embolus	1
Sensory deficits	6
Hyperesthesia	2
Sudeck atrophy	1
Van der Ent⁶⁵	332	Wound-healing disturbance	12
Sensory deficits	4
Hyperesthesia	13
Korkala et al.³⁷	34	Deep venous thrombosis	3

Add to My JBJS

Table V: Operative Complications

*Complications were not mentioned in studies by Eggert et al.¹⁴, Freeman et al.^19,21, Kaikkonen et al.²⁹, Möller-Larsen et al⁴¹., Klein et al.³³, Prins⁴⁹, and Pace et al.⁴⁷.

Author of Study*	Total No. of Patients with Operation	Complications	No. of Patients with Complication
Broström⁷	96	Wound-healing disturbance	1
Sensory deficits	7
Scar neuroma	2
Stadelmayer et al.⁶⁰	30	Wound-healing disturbance	2
Povacz et al.⁴⁸	73	Wound-healing disturbance	2
Paresthesia	8
Zwipp and Krettek⁷¹	102	Wound-healing disturbance	2
Paresthesia	6
van Moppes and van den Hoogenband⁶⁸	50	Hyperesthesia	1
Niedermann et al.⁴³	102	Suppurating wound infection	3
Paresthesia	5
Scar neuroma	7
Evans et al.¹⁸	50	Lung embolus	1
Sensory deficits	6
Hyperesthesia	2
Sudeck atrophy	1
Van der Ent⁶⁵	332	Wound-healing disturbance	12
Sensory deficits	4
Hyperesthesia	13
Korkala et al.³⁷	34	Deep venous thrombosis	3