Article, Orthopedics

Compression stockings in ankle sprain: a multicenter randomized study

a b s t r a c t

Objectives: ankle sprain is a frequently encountered traumatic injury in emergency departments and is associated with important health expenses. However, the appropriate care of this traumatic injury remains a matter of debate. We tested the hypothesis that compression stockings speed up recovery from ankle sprain. Methods: Recent (b48 hours) cases of ankle sprain without other traumatic injury in patients aged between 18 and 55 years were included. Patients were randomly allocated to placebo Jersey or class II compression stockings (Venoflex; Thuasne, Levallois-Perret, France). The primary end point was the time to recovery of normal painless walking without requirement for Analgesic drug. Secondary end points were time to return to sport activity, pain, Analgesic consumption, and ankle edema (bimalleolar and midfoot circumferences).

Results: We randomized 126 patients and analyzed 117 patients (60 in the placebo group and 57 in the compression group). The median time to normal painless walking was not significantly decreased (P = .16). No significant differences were observed in pain, analgesic consumption, and bimalleloar and midfoot circumferences. No safety issue was reported. In the subgroup of patients with regular sport activity, the time to return to sport activity was shorter in patients treated with compression stockings (P = .02).

Conclusions: Compression stockings failed to significantly modify the time to return to normal painless walking in ankle sprain. A beneficial effect was observed only in a subgroup of patients, as compression stockings significantly decreased the time to return to sport activity.

Introduction

Acute lateral ankle ligament injury (ie, ankle sprain) is a very frequent traumatic injury, with an incidence rate of 2.15/1000 person-years in the United States [1]. It represents 3% to 5% of emergency department (ED) visits in the United Kingdom [2]. Fifty percent of these injuries occurred during sports, and thus, ankle sprain is mainly observed in Young individuals, leading to a peak incidence rate of 7.2/1000 person-years between 15 and 19 years of age [1]. The mechanism is an inversion of the

? Conflict of interest: The authors declared no conflict of interest.

?? Funding: The study was funded by a research grant (PO880603/N?IDCRCB: 2009- 100328-49) from the Direction Regionale a la Recherche Clinique Ile de France (Paris, France). Compression stockings were provided free of charge by Laboratoires Thuasne (Paris, France).

* Corresponding author. Service d’Accueil des Urgences, Groupe Hospitalier Pitie- Salpetriere, 47-83 boulevard de l’Hopital, 75013 Paris, France. Tel.: +33 1 42 17 72 49;

fax: +33 1 42 17 72 64.

E-mail address: [email protected] (P. Hausfater).

plantar-flexed foot. Up to 30% of chronic sequelae may be noted if not appropriately treated, including permanent laxity leading to recurrence of sprain [3].

Appropriate treatment of ankle sprain remains a matter of debate, mainly because relatively few randomized trials have been conducted in this very frequent traumatic injury. Rest, ice, compression, and elevation (RICE) are usually proposed during the first days, although insufficient evidence is available to determine their effectiveness [4]. Complete immobilization has been discouraged, but a recent randomized study indicates that a short period of immobilization in a below-knee cast provides a better outcome [5]. In this study [5], patients also received tubular compression stockings because inflammation causes edema, which may increase tissue damage, delays healing, or even favors chronic ankle instability [6]. However, there is little evidence to support the use of compression stockings in ankle sprain [7,8].

The goal of our study was thus to assess the efficacy and safety of compression stockings in acute ankle sprain. We tested the hypothesis that compression stockings use is associated with a shorter time to recovery of normal painless walking.

http://dx.doi.org/10.1016/j.ajem.2014.05.054

Methods

This multicenter prospective study was performed from January 2010 to July 2012, after approval by an ethics committee (Comite de Protection des Personnes, CCP de Paris-Ile de France III, Paris, France), and all patients provided informed written consent. The study was conducted according to Good Clinical Practice standards and the Helsinki Declaration, and the protocol was registered at ClinicalTrial. gov (NCT 01259752). Our study followed the CONSORT recommen- dations for parallel-group randomized trials [9].

Patients

The criteria for inclusion in the study were age between 18 (to ensure skeletal maturity) and 55 years (to avoid any interference with ability to return to walk due to aging) [10] and recent (b48 hours) ankle sprain seen in an ED. The investigators verified that patients were able to understand and follow the protocol, including the ability to complete the autoevaluation form, and could be reached by telephone. The exclusion criteria included the following: (1) any fracture detected using ankle radiography, which was performed in all patients whatever were the Ottawa criteria [11]; (2) any other traumatic injury, including ankle wound requiring suture; (3) history of diabetes associated with neuropathy and/or angiopathy, peripheral arterial disease, neurologic disease, or any disease that could interfere with return to walking; and

(4) pregnant women. Three academic centers participated in the study.

Study design

This was a multicentric, randomized, parallel-group, superiority trial. Patients were included in the study at their First visit in the ED. The following data were recorded: patient characteristics, including age, sex, weight, height, body mass index; personal history including previous ankle sprain, circumstances of the trauma, and time between trauma and ED visit, and clinical signs. Patients were stratified according to their clinical signs in mild or moderate/severe sprain, although a precise Severity assessment is difficult during the acute phase [12]. Moderate/Severe sprain was defined as sprain with, at least, the following: (1) ankle crack or clack perceived by the patient during the Traumatic event, (2) immediate ankle swelling, (3) immediate unability to bear weight, and (4) large ecchymotic area [13]. We also asked patients about regular (weekly) practice of sport and assessed its frequency (estimated hours/week). A pregnancy test was performed in women. No other specific laboratory tests were performed.

Patients were randomly allocated to 1 of the following 2 groups:

(1) placebo using noncompressive stockings (Tubular Jersey; Maill’Tub SA, Port-Brillet, France), and (2) compression stockings (Venoflex; Thuasne France, Levallois-Perret, France). The compres- sion stockings used were class II and developed a pressure between 15 and 20.3 mm Hg, according to the French norm [14]. Stockings were applied from the tibia tuberosity to the base of the toes. Allocation was obtained by means of a telephone call, and an automatic allocation process and randomization was thus not predictable.

Whatever the allocated group, patients received the following Standard care: (1) immediate application of the RICE protocol in the ED [14]; (2) to provide the same immobilization to all patients, we chose the same orthesis (Ligacast; Thuasne; the duration of immobilization varied [expected to be between 4 and 6 weeks] and was decided by the emergency physician, depending on the severity assessment and the clinical evolution); (3) analgesia was provided using oral acetaminophen, a World Health Organization (WHO) class I drug, on a routine basis and at the discretion of the patients. However, when severe pain occurred despite this regimen, the patient was allowed to take tramadol, a WHO class II drug, for more adequate pain relief. All analgesic drugs taken were recorded on the autoevaluation sheet completed by the patients. Pain was assessed using a visual

analog pain scale (VAS, from 0 = no pain to 100 = highest pain). Anticoagulation was prescribed only for patients with a high risk of Thromboembolic disease [10]. A standardized rehabilitation protocol (2-3 sessions per weeks) was prescribed for all patients, as follows:

(1) for mild sprain, rehabilitation began at day 7 and included muscle strengthening and proprioception exercises to reestablish motor coordination, and (2) in moderate/severe sprain, rehabilitation began at days 15 to 21 and included drainage massage and physiotherapy. Then, as soon as possible (painless rehabilitation rule), strengthening exercises of the ankle everters and dorsiflexors followed by inversion and eversion strengthening as tolerated were performed, and proprioception exercises were performed initially seated and then progressively in weight-bearing postures.

Patients were examined again at a second visit that took place 6 to

9 days after trauma. Additional imaging technique (ultrasound, computed tomographic scan) might be ordered at that time if necessary. Patients requiring either complete immobilization (plaster) or surgery were excluded from the study as well as patients with other diagnosis. Thereafter, a telephone call was performed each week, to decide the best timing (recovery or normal painless walking) of the third visit (15 +- 2 or 30 +- 5 days after trauma). Lastly, a visit was performed 90 days after trauma. When necessary, this last visit could be replaced by a telephone interview.

End points

The main end point was the time to recovery of normal painless walking, without analgesic drug requirement. The secondary efficacy end points were as follows: (1) pain assessed at rest and during walking, (2) bimalleolar and midfoot circumferences measured on both sides using a dedicated apparatus (Circumference measurement tape; SECA France, Semur-en-Auxois, France); all investigators were trained before the study to make these measurements appropriately, and the intraobserver and interobserver coefficients of variation of the measurements (n = 20) were 1.7% and 3.7% for the bimalleolar and 2.0% and 4.5% for the midfoot circumferences, respectively; and

(3) return to work or equivalent daily life activity. The safety end points included all adverse events recorded during the follow-up period. In the subgroup of patients regularly practicing sport (other than walking), we assessed the time to return to sport activity. No other subgroup analysis was planned. No interim analysis was planned.

Statistical analysis

The following assumptions were made to calculate the number of patients to be included according to a preliminary nonrandomized observational study [15]. We assumed an ? risk of .05, a ? risk of .05, a risk ratio of 1.3, and a standard deviation (SD) of 0.5 between the 2 groups, which corresponded to a reduction of the median time to normal walking of at least 4 days. We calculated that 54 patients would be required. Anticipating a dropout rate of 10% because of the emergency conditions, we decided to include 120 patients (60 per group; nQuery Advisor 4.0; Statistical Solutions Ltd, Cork, Ireland, UK). No interim analysis was performed. Randomization was performed using a random number table equilibrated every 6 patients and stratified by center and by the initial severity assessment of the sprain. Data are expressed as mean +- SD, median, 25th to 75th interquartiles, percentage, and 95% confidence interval. The comparison of 2 means was performed using the Student t test, the comparison of 2 medians using the Mann-Whitney U test, and proportion using the ?² test or the Fisher exact method, when appropriate. Comparison of several means (bimalleolar and midfoot circumferences, VAS) was performed using nonlinear mixed model for repeated measures [16]. The time from inclusion to return to normal walking without analgesic, return to work, and return to sport activity were analyzed using the Kaplan-Meier method and log-rank test. The analysis was performed as

Fig. 1. Study flowchart. Early lost to follow-up means that the patient missed the second visit. *The study was interrupted in one of these patients because of lack of compliance with the protocol.

an intention-to-treat analysis. Statistical analysis was performed using Statistical Analysis System 9.1.3 software (SAS Institute Inc., Chicago, IL).

Table 1

Comparison of the 2 patient groups at baseline

3. Results	Variables	Placebo (n = 60)	Compression stockings (n = 57)	P
A total of 126 patients were included and randomized, 65 in the	Age (y)	30 +- 8	31 +- 9		.40
placebo group and 61 in the compression stocking group. Four patients	Men	32 (53%)	34 (60%)		.49
were subsequently excluded because Associated fractures were diag- nosed using either ultrasound or computed tomographic scan, 3 in the	Women Height (cm) Weight (kg)	28 (47%) 172 +- 8 70 +- 11	23 (40%) 174 +- 10 71 +- 12		.27 .84
placebo group and 1 in the compression stocking group. Five patients	Body mass index (kg/m²)	23.8 +- 3.3	23.3 +- 3.2	.39
(2 and 3, respectively, in the placebo and compression stocking groups)	History of ankle sprain	9 (15%)	9 (16%)	.90
were lost early on follow-up (no second visit). Thus, we analyzed 117	Regular sport activity	35 (58%)	39 (68%)	.26
patients, 60 in the placebo group and 57 in the compression stocking	Duration of sport (h/wk)	2 (1-3)	2 (1-3)		.31
group. Finally, 108 patients, 54 in the placebo group and 54 in the	Daily activity Mainly physical	30 (50%)	39 (68%)	.042
compression stocking group, completed the full study (Fig. 1).	Mainly static	30 (50%)	18 (32%)
There were no significant differences in the baseline character-	Type of trauma

istics of the 2 groups, except for the daily activity (Table 1). Using the Kaplan-Meier method, there was no significant difference in the time to recovery of normal painless walking (Fig. 2A). The median

Sport 25 (42%) 26 (46%) .07

Work 14 (23%) 21 (37%)

Others 21 (35%) 10 (18%)

Injured side

time was 13 (10-18) days in the compression stocking group and 21	Right	32 (53%)	27 (47%)
(16-24) days in the placebo group. In the subgroup of patients with	Left	28 (47%)	30 (53%)	.52

regular sport activity, the time to return to sport activity was significantly shorter in patients treated with compression stockings (Fig. 2B). The median time was 38 (30-60) days in the compression stocking group and 60 (35-81) days in the placebo group.

Time since injury (h) Bimalleolar circumference (mm)	10 (2-18)	3 (1-13)	.14
Traumatized side	269 +- 24	269 +- 21	.78
Nontraumatized side	252 +- 18	255 +- 18	.42
Midfoot circumference (mm) Traumatized side	248 +- 21	251 +- 21	.89
Nontraumatized side	244 +- 19	246 +- 19	.47
Initial severity assessment

Table 2 provides a description of secondary efficacy end points. There were no significant differences between groups. Fig. 3 shows the

evolution of pain intensity at rest and on moving and the bimalleolar	Large ecchymotic area	5 (8%)	5 (9%)	1.00
and midfoot circumferences in the injured foot in the 2 groups. A	Immediate swelling	45 (75%)	42 (74%)	.87
significant decrease was observed over time but without significant	Impossible immediate weight bearing	32 (53%)	26 (46%)	.40

difference between groups. In the noninjured foot, no significant differences were observed over time (data not shown).

No treatment-related adverse event was reported (Table 2). Adverse events reported were related to analgesic administration and falls. The investigators considered that none of these adverse events were related to the treatment tested (placebo vs compression stockings).

Crack or clack perceived at injury Severity class	25 (42%)	24 (42%)	.96
Mild	6 (10%)	3 (5%)	.49
Moderate/Severe	54 (90%)	54 (95%)
Initial pain at rest (/100)	28 +- 21	34 +- 25	.23
Initial pain at walking (/100)	68 +- 19	64 +- 27	.52

Data are mean +- SD, median (25th-75th interquartiles), or number (percentage).

Fig. 2. A, Recovery of normal walking without analgesics (primary end point) in patients treated with placebo (n = 60) and compression stockings (n = 57). B, Subgroup analysis (patients practicing sport): return to sport activity in patients treated with placebo (n = 27) and compression stocking (n = 31). P value refers to the between-group difference (log-rank test).

Discussion

In this multicenter controlled study evaluating the use of compression stockings in ankle sprain, we observed that they did not significantly speed up recovery of normal painless walking (primary end point). Likewise, we did not observe a significant effect on pain or edema, as shown by the ankle and foot diameters. Nevertheless, in the subgroup of patients with regular sport activity, compression stockings significantly reduced the time to return to sport activity.

Appropriate treatment of ankle sprain remains controversial, mainly because few randomized trials have been conducted. In our study, standard treatment comprised the RICE protocol, although its effectiveness remains to be proven [4]; a short period of immobili- zation using the same type of orthesis because this appears to provide a better outcome [17]; and a standard rehabilitation protocol, although its efficacy also remains to be proven [18]. The compression bandage has been reported to provide less improvement at 10 days and 1 month as compared with ankle brace [19], and comparable results were observed in another small trial [20]. However, these 2

Table 2

Comparison of efficacy end points in the 2 patient groups

Variables	Placebo (n = 60)	Compression stockings (n = 57)	P
No. of reEducation sessions	10 (8-12)	10 (8-12)		.20
Use of crutches	21 (35%)	17 (30%)	.55

studies did not permit the assessment of the effect of compression stockings per se [19,20]. Watts and Amstrong [21] did not observe any significant improvement using double Tubigrip (Molnlycke Health Care, Gothenburg, Sweden) as compression bandage, but their study was associated with considerable (50%) dropout because of follow-up lost and this type of compression bandage is thought to provide minimal pressure (ie, b15 mm Hg), which is considered insufficient to reduce edema at the ankle level. In a recent randomized trial, Sultan et al [8] showed a significant improvement in the Foot and Ankle Outcome score (+12%) and Short Form 12 version 2 (SF12v2) quality of life score (+17%) at 8 weeks using compression stockings, but a limited number of patients were included (n = 36). The control group received Tubigrid, and this study did not identify a unique primary end point. It is also worth noting that the conclusion of very small trials should be always be interpreted with caution. Our study suggests that beneficial effects of compression stockings in ankle sprain are limited because they were observed only in the subgroup of patients with sport activity.

In our study, pain was assessed by both pain intensity and intake of concomitant analgesics. We have previously proposed 2 ways in defining Pain severity, the first related to the patient’s perception of pain intensity and the second related to the amount of analgesics required to obtain pain relief [22]. Although our pain-related end points were based on these 2 approaches, no significant differences were noted between groups. Pain at rest may not be sufficient to describe patient’s discomfort, particularly in patients with traumatic injury. However, even when we considered pain upon moving, no significant difference was noted between the groups. Moreover, it is likely that concomitant

No. of days with crutches	18 (13-31)	15 (14-21)	.12	analgesic intake included both resting pain and pain on moving.
Thromboprophylaxis	12 (20%)	12 (21%)	.89	Therefore, we can conclude that compression stockings do not really
Time to return to work (d)	8 (5-23)	7 (1-18)	.20	affect pain process in ankle sprain. This result contrasts with that

Data are mean +- SD or number (percentage). NSAID, nonsteroidal anti-inflammatory drugs.

No. of days with analgesic Number of days with treatment^a	5 (0-8)	3 (0-5)	.11
Day	8 (7-16)	11 (8-21)	.04
Night	7 (8-9)	5 (7-8)	.48
No. of days with orthesis	18 (2-25)	10 (1-19)	.04
Analgesic drug used None	0 (0%)	0 (0%)	1.00
WHO class I	31 (52%)	32 (56%)	.63
WHO class II	30 (50%)	25 (44%)	.51
NSAID	2 (3%)	1 (2%)	1.00
Complication or adverse events
Thrombophlebitis	0 (0%)	0 (0%)	1.00
Others	1 (2%)	3 (5%)	.62
Related to intervention	0 (0%)	0 (0%)	1.00
Nonrelated	1 (2%)	3 (5%)	0.62

^a Treatment refers to placebo or compression stocking.

observed with double Tubigrip, which was paradoxically responsible for an increase in analgesic consumption [21].

We did not observe a significant decrease in ankle and midfoot circumferences. Several explanations can be proposed. We probably assessed these circumferences too late (ie, 1-2 weeks after trauma) when most of the edema has disappeared, including that in the placebo group. However, we cannot exclude the fact that our measurement was not accurate enough to show a reduction in edema because the differences in circumferences between the injured and not injured limb were only 6% for the bimalleolar circumference and 2% for the midfoot circumference, respectively. Our measurement was precise enough to show the edema reduction over time but may have been insufficient to show more subtle differences between groups.

From a therapeutic perspective, safety must be weighted with efficacy and the Risk-benefit ratio should always be taken into

Fig. 3. Evolution of pain at rest (A) and at walking (B) assessed using VAS, bimalleolar (BMC; C) and midfoot (MFC; D) circumferences in the placebo (n = 60) and compression stockings (n = 57) groups. P value refers to the between-group difference (interaction). In all panels, the effect of time was significant (P b .001).

consideration. No adverse effect was observed in the compression stocking group. Thus, although the effects of the compression stockings appear modest, its use might be considered because it is devoid of any adverse effect.

Limitations

Our study has several limitations. First, we observed a small proportion of patients that must be excluded from the final analysis despite randomization because the Initial diagnosis was erroneous. We also observed that some patients lost to follow-up, but the proportion remains lower than that usually observed under emer- gency conditions, in young patients, admitted for a Minor injury [10]. However, the number of patients to be included was adjusted to achieve the predetermined Power of the study. Second, we did not enroll children or elderly patients, and thus, our results may not apply to these 2 populations. Third, our study cannot be considered as having been blinded, although we used noncompressive stockings in the placebo group to limit bias. Four, we did not use a score such as the Karlsson scoring scale [23] or Foot and Ankle Outcome score [24] because these scores do not seem to provide more relevant information as compared with a strong clinical end point, from a patient’s point of view, as return to normal painless walking, work, and sport activity. Moreover, there is no consensus concerning the best score, and these scores have been validated for postoperative assessment and not ankle sprains [25]. Lastly, we did not study long- term outcomes. Further studies are required to answer this question

because edema may increase tissue damage and favor chronic ankle instability [6].

Conclusion

In this multicenter randomized study, we observed that compres- sion stockings failed to significantly modify the time to return to normal painless walking. A beneficial effect was observed only in a subgroup of patients, as compression stocking significantly decreased the time to return to sport activity.

Acknowledgment

We thank Dr David Baker, DM, FRCA (Department of Anesthesiology and Critical Care, Hopital Necker-Enfants Malades, Paris, France), for reviewing the manuscript.

Appendix

Investigators

Dr Khaled Saidi and Dr Ariella Skurnik-Cornu (CHU Pitie- Salpetriere, Paris, France), Pr Michel Scepi and Dr Aiham Ghazali (Poitiers, France), Dr Cecile Blanchard and Dr Etienne Hinglais (CHU Tenon, Paris, France).

Data Monitoring

Research assistant: Stephanie Lallauret (URC Pitie-Salpetriere, Paris, France).

Statistical analysis

The statistical analysis plan was finalized before freezing of the database. Analysis was performed by Nathalie Cozic (URC Pitie-Salpetriere, Paris, France).

Scientific committee

Prof. Pierre Hausfater, Dr Mouhssine Bendahou, Dr Sylvie Besch, and Prof Bruno Riou, all in Paris, France.

References

AJEM

Compression stockings in ankle sprain: a multicenter randomized study

a b s t r a c t

Introduction

Methods

Patients

Study design

End points

Statistical analysis

Discussion

Limitations

Conclusion

Leave a Reply Cancel reply