Compression Wrapping for Acute Closed Extremity Joint Injuries: A Systematic Review

This systematic review was carried out according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist.²³  A protocol was submitted at PROSPERO (#CRD42020153123).

The following databases were searched for relevant studies: the Cochrane Library, MEDLINE (using the PubMed.com interface), and Embase (using the Embase.com interface). Search strategies were developed by 2 reviewers (V.B., D.C.B.), and both index terms and text words were used. We searched the databases for all dates through November 3, 2019. Search strategies can be found in the Appendix.

After removing duplicates, 2 reviewers (V.B., D.C.B.) independently screened the titles and abstracts and then evaluated the full texts for relevance. Any discrepancies were discussed, and if no consensus could be reached, a third reviewer (D.Z. or E.S.) was consulted. The inclusion and exclusion criteria are shown in Table 1.

Data concerning study design, study population, intervention, outcome measures, and study quality were extracted by 2 reviewers independently (V.B. and D.C.B.). Data are presented as standardized mean differences (SMDs) with 95% CIs for continuous outcomes and by risk ratios (RRs) with 95% CIs for dichotomous outcomes. If only raw data were available, SMDs and RRs with their 95% CIs were calculated using Review Manager (version 5.3; The Cochrane Collaboration, London, UK)²⁴ . As suggested by Cohen,²⁵  SMDs of around 0.2 were considered a small effect; 0.5, a moderate effect; and 0.8, a large effect. When SMDs were not available, mean differences (MDs) were presented. Significant P values were <.05.

We used the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach²⁶  to determine the certainty of evidence for each outcome. The GRADE approach assesses the limitations in study design, indirectness (research that does not directly compare the interventions of interest delivered to the populations of interest), imprecision (due to low sample size, wide CIs, or lack of data), inconsistency (examination of heterogeneity), and publication bias. The certainty of evidence can be downgraded due to shortcomings in each of these domains. The limitations in study design were independently examined by 2 reviewers (V.B., D.C.B.) using the Cochrane Collaboration Risk of Bias 2 (RoB 2) tool²⁷  for randomized controlled trials (RCTs) and the Risk Of Bias In Non-randomised Studies of Interventions (ROBINS-I) tool for nonrandomized studies.²⁸  In both RoB 2 and ROBINS-I, the initial evaluation of quality for the body of evidence of all included studies was high. The final level of evidence can be graded as high, moderate, low, or very low.

A total of 1193 references were identified with the search strategies. After removal of duplicates, the titles and abstracts of 636 articles were screened for eligibility (see Table 1), followed by full-text screening of 75 articles. At this stage, 35 articles were excluded based on study design (mostly narrative reviews or ongoing trials without published results), 6 on population (not a joint injury, healthy volunteers without actual injury, or not an acute injury), 22 on intervention (eg, compression bandage as preventive measure, compression as only a part of the treatment), 3 on outcome (none of the prioritized outcomes were reported), and the full text of 1 publication could not be obtained. Eight studies were assessed in the systematic review. The Figure gives an overview of the study-selection process.

An overview of study characteristics is provided in Table 2. Six studies were RCTs,^20,29–33  whereas 2 were nonrandomized trials.^34,35  Only 3 studies were published in the last 10 years (2011,²⁰  2014,²⁹  and 2015³⁴ ), 2 studies were published in 2005³¹  and 2006,³⁰  and 3 studies were published before 2000 (1984,³⁵  1991,³³  and 1995³² ). All investigations included patients with ankle sprains and were performed in an in-hospital setting.

The interventions were class 2 compression stockings (pressure of 15–20.3 mm Hg) in 1 study,²⁹  elastic wrap or bandage in 5 studies,^{20,30,31,33,34}  compression bandage in 2 studies,^32,35  and a Tubigrip bandage (Mölnlycke Health Care, Norcross, GA) in 1 study.²⁰  These interventions were compared with a noncompressive stocking, an Aircast Air Stirrup or ankle brace (DJO Global Inc, Lewisville, TX), a splint, no treatment, or elevation of the foot.

The synthesized findings are available in Table 3. In 3 RCTs^20,29,31  and 2 nonrandomized trials,^34,35  the effect of a compression or elastic bandage on pain was assessed. Researchers in 2 RCTs^20,31  and 1 nonrandomized trial³⁴  reported the outcome of “reduction of pain,” measured on a visual analog scale. Reduction in pain did not differ when a compression or elastic bandage was compared with a splint, an Aircast brace, or no support (SMD = 0.41, 95% CI = −0.80, 1.61, P = .51³⁴ ; SMD = 0.05, 95% CI = −0.61, 0.71, P = .88²⁰ ; SMD = 0.64, 95% CI = −0.04, 1.32, P = .07,³¹  respectively). Linde et al³⁵  demonstrated no difference in being free from pain while walking after 4 days (RR = 1.28, 95% CI = 0.78, 2.11, P = .33) or 8 days (RR = 1.39, 95% CI = 0.98, 1.95, P = .006). Moreover, Bendahou et al²⁹  showed no difference in time to recovery of normal painless walking (P = .20), pain at rest (SMD = −0.32, 95% CI = −0.68, 0.05, P = .09), or pain with walking (SMD = −0.14, 95% CI = −0.50, 0.22, P = .45) after 6 to 9 days.

Swelling, measured as a change in ankle volume (mL), ankle swelling (cm), or bimalleolar circumference (cm), was studied in 3 randomized trials^29,31,33  and 1 nonrandomized trial.³⁴  No reduction in swelling was evident in 2 RCTs and 1 nonrandomized trial (SMD = −0.14, 95% CI = −0.50, 0.22, P = .45²⁹ ; SMD = 0.55, 95% CI = −0.13, 1.22, P = .11³¹ ; SMD = 0.34, 95% CI = −0.22, 0.89, P = .23,³⁴  respectively) when comparing a compression or an elastic bandage with a splint, no treatment, or an Aircast ankle brace. In their RCT, Rucinski et al³³  found less reduction in swelling when an elastic bandage was applied than with no compression (SMD = 2.02, 95% CI = 0.90, 3.15, P = .0004).

Ankle-joint function, measured using the Karlsson and Peterson score,³⁶  was evaluated in 3 randomized trials.^20,31,32  In 1 study,³¹  researchers identified an increase in ankle-joint function after 10 days (SMD = −0.77, 95% CI = −1.45, −0.08, P = .03) and 1 month (SMD = −0.71, 95% CI = −1.40, −0.03, P = .04), where as another study²⁰  revealed no difference after 10 days (SMD = −0.34, 95% CI = −1.16, 0.49, P = .42) or 1 month (SMD = −0.29, 95% CI = −1.11, 0.53, P = .49) when comparing an elastic bandage with an Aircast ankle brace or no support. Leanderson and Wredmark³²  found no difference in ankle-joint function after 3 to 5 days, 2 weeks, or 4 weeks when they compared a compression bandage with an Air Stirrup ankle brace (P > .05).

The authors of only 1 RCT³²  examined the effect of a compression bandage versus an Air Stirrup ankle brace on ROM. No change occurred in active ROM after 3 to 5 days, 2 weeks, or 4 weeks.

Beynonn et al³⁰  measured the time to return to normal walking, time to return to stair climbing, and time to return to full weight bearing using an elastic wrap or an Air Stirrup ankle brace. None of the outcomes differed between interventions.

The time to return to work was measured in 3 randomized studies.^20,29,32  One RCT³²  showed a benefit for the use of an Air Stirrup ankle brace compared with a compression bandage (median difference = −3.8 days, P < .05). In 2 other studies,^20,29  investigators failed to show a benefit for a compression bandage versus no compression bandage (SMD = −0.50, 95% CI = −1.17, 0.16, P = .14) or noncompressive stockings (median difference = −1 day, P = .20).

Bendahou et al²⁹  also studied the time to return to sport and demonstrated a benefit for the use of a compression bandage versus noncompressive stockings (median difference = −22 days, P < .02).

An overview of the risk of bias of the RCTs, as assessed with the RoB 2 tool, is given in Table 4. Three studies^29–31  had an adequate randomization process; of these, 2 studies^29,30  also had adequate allocation-concealment procedures. In 3 studies,^20,32,33  the researchers mentioned that they randomized the participants without further explanation, and in 4 studies,^20,31,–,33  the researchers did not indicate if allocation to the treatment groups was blinded. All investigations had a low risk of bias due to missing outcome data. In most studies, participants and people applying the interventions were aware of the assigned intervention, although blinding was often not possible due to the nature of the interventions (eg, elastic bandage compared with an ankle brace). The lack of blinding might have been influential in the work of Beynnon et al³⁰  because the patients assessed the outcome; knowledge of the intervention might have affected the assessments. The authors of only 1 study²⁹  published their protocol. No other researchers indicated if a prespecified plan for analysis was available. However, we had no reason to believe the results were selected from multiple outcome measures or multiple analyses of the data. One study²⁹  had an overall low risk of bias; concerns about the risk of bias were present for all other studies.

An overview of the risk of bias of the non-RCTs, as assessed with ROBINS-I, is shown in Table 5. Both nonrandomized trials had a risk of bias due to confounding. In Bilgic et al,³⁴  the choice of treatment was at the discretion of the on-shift physician or resident doctor, which could have introduced serious bias. Both investigations had a low risk of bias due to the selection of the participants, classification of the interventions, and deviations from the intended interventions. Linde et al³⁵  displayed a fairly high loss to follow up (13% at first follow up and 26% at second follow up), leading to a moderate risk of bias due to missing outcomes data. In both cases, the outcome assessors were not blinded to the interventions, which may have led to a serious risk of bias in the measurement of subjective outcomes (eg, pain). Finally, no indication of selective reporting was evident.

The assessed certainty of the body of evidence is detailed in Table 3. The certainty of evidence was downgraded for most outcomes due to limitations in study designs. Furthermore, the certainty of the evidence was downgraded due to indirectness because all studies took place in a hospital setting. The overall certainty was further downgraded for imprecision due to limited sample sizes or large variability of results. Therefore, the certainty of evidence was low to very low for all outcomes.

Compression is a common prehospital intervention for the management of ACEJIs (eg, sprains and strains). However, the effectiveness of compression in the first aid treatment of these injuries has been questioned.

Compression is advocated for treating ACEJIs such as ankle sprains to limit swelling^7,37  and to improve quality of life.³⁷  In theory, compression, both circumferential and sequential, is applied to limit the amount of edema caused by the exudation of fluid from the damaged capillaries^38–40  and prevent possible hypoxic damage to surrounding tissues.²⁰ 

With this review, we sought to identify the best available evidence to inform organizations on the use of compression in a prehospital (first aid) setting for the initial management of a patient with an ACEJI. Of 1193 references identified initially, researchers in 8 studies compared compression of acute ankle injuries (sprains) with the use of a noncompressive stocking, splint, ankle brace, elevation of the injured ankle, or no treatment. We found no investigations of compression for other closed extremity injuries to joints such as the wrist or knee, limiting the results of this review to ankle-joint injuries. All included studies were of low to very low certainty according to the GRADE system, reducing our confidence in the estimate of effects. Our findings must be considered with this restriction in mind.

Previous researchers focused on compression applied with cryotherapy,^41,42  RICE,^14,43  or compression alone, but with applications beyond those available to first aid providers in the prehospital setting.^12,13,37 

van den Bekerom at al¹⁴  performed a systematic review examining the effectiveness of applying RICE therapy within 72 hours after an ankle sprain. They included a single study of compression therapy, which demonstrated that the combination of elastic bandaging and intermittent pneumatic compression was better than elastic bandaging alone in decreasing edema and reducing pain.⁴⁴  However, importantly, this finding is not applicable in the prehospital setting due to the unavailability of intermittent pneumatic compression units.

Fousekis et al¹³  conducted a systematic review to evaluate the effectiveness of elastic bandaging in orthopaedic and sports injury prevention and rehabilitation. Moderate-certainty evidence suggested improved ankle proprioception (ie, enhanced kinesthesia and neuromuscular control) in participants who used elastic bandages, but the evidence was insufficient to support the use of elastic bandages to improve other outcomes, such as joint ROM and stability and functional outcome after injury. For the important outcome of ROM, these results are consistent with ours, failing to show a benefit from a compression bandage versus a noncompressive ankle brace.³² 

The systematic review of Seah and Mani-Babu¹²  addressed the effectiveness of managing ankle sprains (acute and chronic) in the community. Functional treatment options for patients with mild to moderate ankle sprains (including elastic bandaging, soft casting, taping, or orthoses with associated coordination training) produced statistically better outcomes than immobilization for multiple measures.

In the current review, we focused on applications available to first aid providers in the prehospital setting. We could not demonstrate a benefit for the critical outcome of reduced pain during walking or at rest when comparing a compression bandage with no compression or with noncompressive stockings, splints, or braces (Air Stirrup ankle brace).^{20,29,31,34,35}  For the critical outcome of decreased swelling or edema, the authors of 3 studies^29,31,34  showed no benefit when comparing a compression bandage with noncompressive stockings, splints, or braces. However, in 1 study,³³  researchers found less reduction in swelling with an elastic bandage than with no treatment. Yet this was a small investigation with only 10 participants in each group, which restricts the interpretation of the results. Also, in both treatment arms, the ankle was kept elevated at an angle of 45°, which might have influenced swelling.⁴⁵ 

One RCT³²  yielded less benefit for the important outcome of time to return to work when the use of a compression bandage was compared with an Air Stirrup ankle brace. The authors of 2 other randomized studies^20,29  demonstrated no difference in the time to return to work when comparing the use of a compression bandage with no treatment or noncompressive stockings. For the outcome of time to return to sport, 1 randomized trial²⁹  showed a benefit with the use of a compression bandage versus noncompressive stockings. Nonetheless, the findings from these individual studies are not considered adequate to support a recommendation for the use of compression bandages for patients with ankle-joint injuries.

The strength of this review was the rigorous and transparent use of the PRISMA and GRADE methods to identify the best available evidence for the use of a compression bandage by a first aid provider for an ACEJI in the prehospital setting. To our knowledge, this was the first systematic review to evaluate the use of compression as a standalone intervention for patients with ACEJIs. Although first aid training programs teach RICE and others advocate the use of compression to minimize swelling in ACEJIs,⁷  the evidence on the use of compression wrap alone as a treatment for closed extremity joint injuries is limited.

As pointed out by van den Bekerom et al,¹⁴  no information can be provided about the best way, amount, and duration of compression or the position in which the compression treatment should occur. This is especially true in the prehospital setting because of the level of training and expertise of the providers available during the emergent situation. First aid, or the aid offered before advanced medical care or equipment arrives, varies nationally and internationally.

To help clinicians and guideline developers make informed decisions based on evidence, the GRADE Evidence-to-Decision (EtD) framework is a useful tool. It helps in the development of recommendations that consider the evidence for an intervention in light of desirable and undesirable effects, the certainty of the evidence, values, resource requirements, cost, equity, and acceptability.⁴⁶  We used the GRADE EtD framework to discuss the evidence identified through this systematic review with the ILCOR First Aid Task Force. Concern was expressed regarding the potential for improper application of a compression bandage by unskilled lay providers for closed injuries, such as to the ankle joint, in the prehospital setting. The evidence evaluated in this review does not support the use of compression wraps for acute ankle sprains by lay providers in a first aid setting; however, there was also no evidence of harm from compression bandaging.

The aim of our review was to identify the evidence for the use of a compression bandage by first aid providers in a prehospital setting, but all research considered in this review was performed in a hospital setting. Therefore, the certainty of evidence was downgraded for indirectness. Furthermore, all types of noncompressive techniques that can be used in a first aid setting were included as comparisons. This may have led to confounding of the results. Also, we only identified studies of patients with ankle sprains. Whether the results would be applicable to other joints, such as the wrist, knee, or elbow, is unknown.

In most studies, the authors provided no explanation of how much pressure was applied using the compression or elastic bandages, how the wrap was applied (proximal to distal or distal to proximal), whether the pressure was circumferential or sequential, and how long the compression bandage was worn. All of these can be confounding factors. Additionally, a potential limitation is that a first aid provider, or even a health care professional, may not be able to correctly apply a compression bandage.

Some deviations from the review protocol occurred. First, we decided to include adults and exclude children because of heightened concern for fracture in children with immature bones. Second, the protocol excluded the use of tubular bandages (eg, Tubigrip) and compression stockings. The rationale was that these interventions would be too painful to apply immediately after an acute sprain or strain. However, after discussion with the ILCOR First Aid Task Force members, we determined that these interventions would be feasible, and they were therefore included, as long as compression was applied.

Future research is needed to investigate the use of compression bandages for closed extremity joint injuries other than ankle sprains (ie, those affecting the wrist or knee). More work is required to assess if the application of a compression bandage compared with doing nothing results in greater satisfaction among patients and care providers. Also, how much pressure is needed to produce physiological changes in the body is unclear. No information is available on the economic effects (direct and indirect medical costs, lost wages due to inability to work) of using compression bandages for closed extremity joint injuries. Finally, what benefit compression bandages may have when used in combination with other adjunct therapies in the prehospital setting is unknown.

The evidence for the use of compression wrap as a treatment for closed extremity joint injuries was limited and of low to very low certainty due to limitations in study design, indirectness, and imprecision. The only evidence we identified was related to ankle sprains. We were unable to demonstrate a beneficial or harmful effect from the application of a compression or elastic bandage compared with no compression or a noncompressive stocking, splint, or brace as a first aid treatment in the prehospital environment. Further out-of-hospital and first aid studies are required to investigate the usefulness of compression bandages for patients with closed extremity joint injuries.

Drs Borra and De Buck are employees of the Belgian Red Cross and receive no other funding. One activity of the Belgian Red Cross is to provide first aid training to laypeople. This work was made possible through funding from the Foundation for Scientific Research of the Belgian Red Cross (Mechelen, Belgium). The American Heart Association and ILCOR supplied the resources to assemble the Task Force and manage records and data.

In addition to Drs Borra, Berry, Zideman, and Singletary, members of the ILCOR First Aid Task Force included Jason C. Bendall, Jestin N. Carlson, Pascal Cassan, Wei-Tien Chang, Nathan P. Charlton, Therese Djarv, Matthew Douma, Jonathan L. Epstein, David S. Markenson, Daniel Meyran, Peter Morley, Aaron Orkin, Tetsuya Sakamoto, Janel Swain, and Jeffrey A. Woodin.

1. “Sprains and strains”[Mesh] OR “Soft Tissue Injuries"[Mesh] OR “athletic injuries”[Mesh] OR strain*[TIAB] OR sprain*[TIAB] OR distortion*[TIAB] OR rupture*[TIAB] OR “ankle injuries”[Mesh] OR “knee injuries”[Mesh] OR “wrist injuries”[Mesh] OR “tendon injuries”[Mesh:NoExp] OR overexertion[TIAB] OR ((ankle[TIAB] OR knee[TIAB] OR wrist[TIAB] OR elbow[TIAB]) AND (injur*[TIAB]))

2. “Compression Bandages"[Mesh] OR ((compression[TIAB] OR elastic[TIAB]) AND (bandag*[TIAB] OR wrap*[TIAB] OR dressing*[TIAB] OR stocking*[TIAB] OR sleeve*[TIAB]))

3. 1 AND 2

1. ‘sprain'/exp OR ‘joint injury'/de OR ‘ankle injury'/exp OR ‘knee injury'/exp OR ‘wrist injury'/exp OR ‘elbow injury'/exp OR ‘ligament and tendon injury'/exp OR ‘muscle injury'/exp OR ‘overexertion'/exp OR ‘Soft Tissue Injury'/exp OR ‘sport injury'/exp OR strain*:ab,ti OR sprain*:ab,ti OR distortion*:ab,ti OR rupture:ab,ti OR overexertion:ab,ti OR ((ankle:ab,ti OR knee:ab,ti OR wrist:ab,ti OR elbow:ab,ti) AND (injur*:ab,ti))

2. ‘Compression Bandage'/exp OR ‘compression stocking'/exp OR ‘compression sleeve'/de OR ((compression:ab,ti OR elastic:ab,ti) AND (bandag*:ab,ti OR wrap*:ab,ti OR dressing*:ab,ti OR stocking:ab,ti OR sleeve:ab,ti))

3. 1 AND 2

1. [mh “Sprains and strains”] OR [mh “Soft Tissue Injuries”] OR [mh “athletic injuries”] OR strain*:ti,ab,kw OR sprain*:ti,ab,kw OR distortion*:ti,ab,kw OR rupture*:ti,ab,kw OR [mh “ankle injuries”] OR [mh “knee injuries”] OR [mh “wrist injuries”] OR [mh ^“tendon injuries”] OR overexertion:ti,ab,kw OR ((ankle:ti,ab,kw OR knee:ti,ab,kw OR wrist:ti,ab,kw OR elbow:ti,ab,kw) AND (injur*:ti,ab,kw))

2. [mh “Compression Bandages”] OR ((compression:ti,ab,kw OR elastic:ti,ab,kw) AND (bandag*:ti,ab,kw OR wrap*:ti,ab,kw OR dressing*:ti,ab,kw OR stocking*:ti,ab,kw OR sleeve*:ti,ab,kw))

3. 1 AND 2