Compression for preventing recurrence of venous ulcers

Abstract

Background

Up to 1% of adults will have a leg ulcer at some time. Most leg ulcers are venous in origin and are caused by high pressure in the veins due to blockage or damaged valves.

Venous ulcer prevention and treatment typically involves the application of compression bandages/stockings to improve venous return and thus reduce pressure in the legs. Other treatment options involve removing or repairing veins. Most venous ulcers heal with compression therapy, but ulcer recurrence is common. For this reason, clinical guidelines recommend that people continue with compression treatment after their ulcer has healed.

This is an update of a Cochrane review first published in 2000 and last updated in 2014.

Objectives

To assess the effects of compression (socks, stockings, tights, bandages) for preventing recurrence of venous leg ulcers.

Search methods

In August 2023, we searched the Cochrane Wounds Specialised Register, CENTRAL, MEDLINE, Embase, three other databases, and two ongoing trials registries. We also scanned the reference lists of included studies and relevant reviews and health technology reports. There were no restrictions on language, date of publication, or study setting.

Selection criteria

We included randomised controlled trials (RCTs) that evaluated compression bandages or hosiery for preventing the recurrence of venous ulcers.

Data collection and analysis

At least two review authors independently selected studies, assessed risk of bias, and extracted data. Our primary outcome was reulceration (ulcer recurrence anywhere on the treated leg). Our secondary outcomes included duration of reulceration episodes, proportion of follow‐up without ulcers, ulceration on the contralateral leg, noncompliance with compression therapy, comfort, and adverse effects. We assessed the certainty of evidence using GRADE methodology.

Main results

We included eight studies (1995 participants), which were published between 1995 and 2019. The median study sample size was 249 participants. The studies evaluated different classes of compression (UK class 2 or 3 and European (EU) class 1, 2, or 3). Duration of follow‐up ranged from six months to 10 years. We downgraded the certainty of the evidence for risk of bias (lack of blinding), imprecision, and indirectness.

EU class 3 compression stockings may reduce reulceration compared with no compression over six months (risk ratio (RR) 0.46, 95% confidence interval (CI) 0.27 to 0.76; 1 study, 153 participants; low‐certainty evidence).

EU class 1 compression stockings compared with EU class 2 compression stockings may have little or no effect on reulceration over 12 months (RR 1.70, 95% CI 0.67 to 4.32; 1 study, 99 participants; low‐certainty evidence). There may be little or no difference in rates of noncompliance over 12 months between people using EU class 1 stockings and people using EU class 2 stockings (RR 1.22, 95% CI 0.40 to 3.75; 1 study, 99 participants; low‐certainty evidence).

UK class 2 hosiery compared with UK class 3 hosiery may be associated with a higher risk of reulceration over 18 months to 10 years (RR 1.55, 95% CI 1.26 to 1.91; 5 studies, 1314 participants; low‐certainty evidence). People who use UK class 2 hosiery may be more compliant with compression treatment than people who use UK class 3 hosiery over 18 months to 10 years (RR for noncompliance 0.69, 95% CI 0.49 to 0.99; 5 studies, 1372 participants; low‐certainty evidence).

There may be little or no difference between Scholl UK class 2 compression stockings and Medi UK class 2 compression stockings in terms of reulceration (RR 0.77, 95% CI 0.47 to 1.28; 1 study, 166 participants; low‐certainty evidence) and noncompliance (RR 0.97, 95% CI 0.84.1 to 12; 1 study, 166 participants; low‐certainty evidence) over 18 months.

No studies compared different lengths of compression (e.g. below‐knee versus above‐knee), and no studies measured duration of reulceration episodes, ulceration on the contralateral leg, proportion of follow‐up without ulcers, comfort, or adverse effects.

Authors' conclusions

Compression with EU class 3 compression stockings may reduce reulceration compared with no compression over six months. Use of EU class 1 compression stockings compared with EU class 2 compression stockings may result in little or no difference in reulceration and noncompliance over 12 months. UK class 3 compression hosiery may reduce reulceration compared with UK class 2 compression hosiery; however, higher compression may lead to lower compliance. There may be little to no difference between Scholl and Medi UK class 2 compression stockings in terms of reulceration and noncompliance.

There was no information on duration of reulceration episodes, ulceration on the contralateral leg, proportion of follow‐up without ulcers, comfort, or adverse effects.

More research is needed to investigate acceptable modes of long‐term compression therapy for people at risk of recurrent venous ulceration. Future trials should consider interventions to improve compliance with compression treatment, as higher compression may result in lower rates of reulceration.

Plain language summary

Compression therapy for preventing venous leg ulcers returning

Key messages

• A high level of compression may be more effective for preventing ulcers returning than a lower level of compression or no compression.
• In the long term, people may be more likely to continue using stockings with lower compression.

What are venous leg ulcers?

Venous leg ulcers are long‐lasting sores that occur in the legs of people with venous disease. They can be caused by a blockage or sluggish blood flow in the vessels that carry blood back up the legs to the heart.

How are venous ulcers treated?

Compression with bandages or hosiery (stockings, socks, tights) can help heal most venous ulcers. However, venous leg ulcers can come back (recur) after they have healed, or new ulcers can develop in a different location. Continued use of compression therapy after healing may reduce the chance of ulcers recurring.

What did we want to find out?

We wanted to know:

• whether compression prevents the recurrence of venous ulcers anywhere on the treated leg;
• which are the best levels, types, or brands of compression for preventing ulcer recurrence;
• whether people really adapt to the use of compression over the long term; and
• if compression therapy has unwanted effects.

What did we do?

We searched for studies that assessed compression for preventing recurrence of venous ulcers. We compared and summarised their results, and we rated our confidence in the evidence based on factors such as study methods and sizes.

What did we find?

We found eight eligible studies, which included 1995 people with healed venous ulcers. Most participants were women, and the average age ranged from 58 years to 78 years. The shortest study lasted six months, and the longest study evaluated the effects of compression over 10 years. The studies used different levels of compression (UK class 2, UK class 3, European class 1, European class 2, European class 3) and no compression. No studies compared different lengths of compression (e.g. below‐knee compared with above‐knee).

Main results

European class 3 compression stockings may reduce the likelihood of ulcer recurrence compared with no compression. For every 1000 people using European class 3 compression stockings, 210 may have ulcer recurrence, compared with 457 of every 1000 people who use no compression.

There may be little or no difference in the risk of ulcer recurrence with European class 1 compression stockings compared with European class 2 compression stockings.

UK class 3 compression hosiery may be better than UK class 2 compression hosiery for preventing ulcer recurrence. For every 1000 people using UK class 3 compression hosiery, 342 may have ulcer recurrence, compared with 530 of every 1000 people using UK class 2 compression hosiery. However, people using the class 3 hosiery may be more likely to stop treatment compared with those using the class 2 hosiery (215 per 1000 of those using class 3 hosiery compared with 148 per 1000 of those using class 2 hosiery).

There may be little or no difference between two different brands of UK class 2 compression stockings (Scholl and Medi) in terms of risk of ulcer recurrence and likelihood of people stopping treatment.

We found no information about how long the episodes of recurrence lasted, the risk of ulcers developing in the other leg, the proportion of time during the study that participants had no ulcers, level of comfort, or unwanted effects of compression therapy.

What are the limitations of the evidence?

We have little confidence in the evidence because the participants in most studies knew which treatment they were receiving, because there was considerable variation in the duration of the studies, and because some studies enroled few people.

How up‐to‐date is this review?

The evidence in this Cochrane review is current to August 2023.

Summary of findings

Summary of findings 1. EU class 3 compression hosiery compared with no compression for preventing recurrence of venous ulcers.

EU class 3 compression hosiery compared with no compression for preventing recurrence of venous ulcers
Patient or population: people with healed venous leg ulcers Setting: community and acute care settings Intervention: compression hosiery Comparison: no compression
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with no compression	Risk with EU class 3 compression
Reulceration Follow‐up: 6 months	Study population		RR 0.46 (0.27 to 0.76)	153 (1 RCT)	⊕⊕⊝⊝ Lowa,b	EU class 3 compression hosiery may reduce ulcer recurrence compared with no compression at 6 months' follow‐up.
	457 per 1000	210 per 1000 (123 to 347)
Duration of reulceration episodes	Vandongen 2000 did not measure this outcome.
Proportion of follow‐up period without ulcers	Vandongen 2000 did not measure this outcome.
Ulceration on the contralateral leg	Vandongen 2000 did not measure this outcome.
Noncompliance	Vandongen 2000 did not analyse this outcome (the no compression group did not have to comply with any treatment), but 16/72 participants in the intervention group refused to continue wearing stockings.
Comfort	Vandongen 2000 did not measure this outcome.
Adverse effects	Vandongen 2000 did not measure this outcome.
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

^a Downgraded one level for imprecision (small sample size) and one level for risk of bias (lack of blinding).

Summary of findings 2. EU class 1 compression hosiery compared with EU class 2 compression hosiery for preventing recurrence of venous ulcers.

EU class 1 compression hosiery compared with EU class 2 compression hosiery for preventing recurrence of venous ulcers
Patient or population: people with healed venous leg ulcers Setting: community and acute care settings Intervention: EU class 1 compression hosiery Comparison: EUclass 2 compression hosiery
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with EU class 2 compression	Risk with EU class 1 compression
Reulceration Follow‐up: 12 months	Study population		RR 1.70 (0.67 to 4.32)	99 (1 RCT)	⊕⊕⊝⊝ Lowa	EU class 1 compression hosiery compared with EU class 2 compression hosiery may have little or no effect on recurrence at 12 months.
	120 per 1000	204 per 1000 (80 to 518)
Duration of reulceration episodes	Clarke‐Moloney 2014 did not measure this outcome.
Proportion of follow‐up period without ulcers	Clarke‐Moloney 2014 did not measure this outcome.
Ulceration on the contralateral leg	Clarke‐Moloney 2014 did not measure this outcome.
Noncompliance Follow‐up: 12 months	Study population		RR 1.22 (0.40 to 3.75)	99 (1 study)	⊕⊕⊝⊝ Lowa	There may be little or no difference in rates of noncompliance at 12 months between people using EU class 1 compression hosiery and people using EU class 2 compression hosiery.
	100 per 1000	122 per 1000 (40 to 375)
Comfort	Clarke‐Moloney 2014 did not measure this outcome.
Adverse effects	Clarke‐Moloney 2014 did not measure this outcome.
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

^aDowngraded one level for imprecision (small sample size and large CI that incorporates the possibility of benefit and harm) and one level for risk of bias (lack of blinding).

Summary of findings 3. UK class 2 compression hosiery compared with UK class 3 compression hosiery for preventing recurrence of venous ulcers.

UK class 2 compression hosiery compared with UK class 3 compression hosiery for preventing recurrence of venous ulcers
Patient or population: people with healed venous leg ulcers Setting: community and acute care settings Intervention: UKclass 2 compression hosiery Comparison: UK class 3 compression hosiery
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with UK class 3 compression	Risk with UK class 2 compression
Reulceration Follow‐up: 18 months–10 years	Study population		RR 1.55 (1.26 to 1.91)	1314 (5 RCTs)	⊕⊕⊝⊝ Lowa	UK class 2 hosiery may be associated with a higher incidence of recurrence compared with UK class 3 hosiery.
	342 per 1000	530 per 1000 (431 to 653)
Duration of reulceration episodes	No studies measured this outcome.
Proportion of follow‐up period without ulcers	No studies measured this outcome.
Ulceration on the contralateral leg	No studies measured this outcome.
Noncompliance Follow‐up: 18 months–10 years	Study population		RR 0.69 (0.49 to 0.99)	1372 (5 RCTs)	⊕⊕⊝⊝ Lowa	People who use UK class 2 hosiery may be more compliant with compression treatment than people who use UK class 3 hosiery.
	215 per 1000	148 per 1000 (105 to 213)
Comfort	No studies measured this outcome.
Adverse effects	No studies measured this outcome.
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

^aDowngraded one level for indirectness (wide variation in follow‐up between studies) and one level for risk of bias in all studies (high/unclear risk of bias due to lack of blinding/insufficient information on blinding).

Summary of findings 4. Comparison between different brands of compression hosiery (UK class 2) for preventing recurrence of venous ulcers.

Different brands of UK class 2 compression hosiery for preventing recurrence of venous ulcers
Patient or population: people with healed venous leg ulcers Setting: community and acute care settings Intervention: Scholl UK class 2 compression stockings Comparison: Medi UKclass 2 compression stockings
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with Medi	Risk with Scholl
Reulceration Follow‐up: 18 months	Study population		RR 0.77 (0.47 to 1.28)	166 (1 RCT)	⊕⊕⊝⊝ Lowa	UK class 2 compression stockings manufactured by Scholl compared with those manufactured by Medi may have little or no effect on incidence of ulcer recurrence at 18 months.
	315 per 1000	243 per 1000 (148 to 403)
Duration of reulceration episodes	Franks 1995 did not measure this outcome.
Proportion of follow‐up period without ulcers	Franks 1995 did not measure this outcome.
Ulceration on the contralateral leg	Franks 1995 did not measure this outcome.
Noncompliance Follow‐up: 18 months	Study population		RR 0.97 (0.84 to 1.12)	166 (1 RCT)	⊕⊕⊝⊝ Lowa	There may be little or no difference in noncompliance with compression treatment between people who use Scholl UK class 2 compression stockings and people who use Medi UK class 2 compression stockings.
	837 per 1000	812 per 1000 (703 to 937)
Comfort	Franks 1995 did not measure this outcome.
Adverse effects	Franks 1995 did not measure this outcome.
*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

^a Downgraded one level for imprecision (small sample size) and one level for risk of bias (lack of blinding).

Background

See Appendix 1 for a glossary of terms.

Description of the condition

Venous ulceration is a chronic recurring condition. One Scottish study found that 45% of people with open leg ulcers had them for more than 10 years (Callam 1985). The prevalence of chronic venous ulcers increases with age and with the progression of chronic venous disease. At least 10% of the population will develop a chronic wound in their lifetime, and the injury‐related mortality rate is 2.5% (Peréz 2018; Ruckley 2002). There is a considerable cost to the affected person in prescription charges (dressings, drugs, and bandages), increased laundry bills due to discharge from the ulcer, time off work attending nurse/doctor consultations, pain, isolation, and distress (Charles 1995). Two surveys undertaken in the north of England estimated that venous ulceration had a point prevalence of approximately 0.30 cases per 1000 population in the UK (Cullum 2016; Gray 2018), whilst one of these surveys estimated a point prevalence of 0.11 per 1000 for mixed arterial/venous leg ulceration (Cullum 2016). One review of studies of the prevalence of complex wounds suggests that there are limited high‐quality data for estimating the burden of venous leg ulceration in low‐ and middle‐income countries (Cullum 2016).

Treatment of leg ulceration is very expensive for health services, mainly owing to nursing time. The estimated cost of leg ulcer treatment for the UK National Health Service (NHS) was GBP 300 million in 1992 and GBP 941 million in 2022 (Bosanquet 1992; Urwin 2022). The average NHS cost of treating venous ulcers over 12 months in 2018 was estimated at GBP 7600 per ulcer. However, the cost of managing uncured leg ulceration compared with cured leg ulceration was 4.5 times higher: GBP 13,445 versus GBP 2981 (Guest 2018).

The reported prevalence of venous ulcers is 1% to 3% of the adult population worldwide, with some people experiencing a repeated cycle of ulceration, healing, and recurrence (Baker 1991; Bergan 2006). Around three‐quarters of leg ulcers are due to blood flow changes in the leg veins. These changes are caused by blockages (occlusion), weak vales (venous incompetence), or both (Callam 1985). The resulting ulcers are known as venous ulcers, venous stasis ulcers, or varicose ulcers.

Occlusion or incompetence in the veins of the leg leads to increased intravascular pressure. A visible sign of this condition is distended, tortuous (varicose) veins. Increased intravascular pressure may lead to varicose eczema, lower leg oedema, and deposition of scar tissue (fibrin) and iron pigments in the skin. A possible consequence is skin breakdown or delayed healing when the leg is injured (Eberhardt 2014).

Treatment of venous ulcers and prevention of ulcer recurrence aims to reduce the pressure in the veins. This can be accomplished by surgically removing superficial or perforating veins (or both), injecting an irritant solution to block incompetent veins (sclerotherapy), or applying compression. Some people are ineligible for venous surgery, and others prefer not to undergo the procedure. Surgery on the deep veins is experimental, unevaluated, and not widely practised.

High‐compression bandaging has increased ulcer healing rates, and the use of these bandages is widespread (Cullum 2000). However, people with healed ulcers are still at risk of ulcer recurrence: 12‐month recurrence rates range from 26% to 69% (Monk 1982; Moffatt 1995; Vowden 1997).

Description of the intervention

There are many ways of applying compression, including bandages, stockings, wraps, and combinations of these modalities. Using devices that generate external pressure around the leg can improve venous return and thus reduce venous reflux. There are no internationally agreed standards for classifying compression systems, which means it can be difficult to make valid comparisons.

The different manufacturers of medical and support hosiery follow different standards to describe the compression capability of the various garments; most use the UK standard (Table 5) or the European (EU) standard (Table 6). Manufacturers in the USA do not categorise compression hosieries by compression class but provide descriptions of the levels of pressure generated (Clarke‐Moloney 2014). In the UK, stockings are classified according to the force required to stretch them and hence the compression level they can apply to a limb (Table 5); but different specifications apply to hosiery supplied through hospitals and community pharmacists. Other countries use different classification systems. An international collaboration proposed a set of guidelines for classifying compression devices in clinical studies (Rabe 2008), but there is no universally accepted tool to date.

1. Classification of compression stockings (UK).

Class	Descriptor	Ankle pressure	Indication
1	Light support	14–17 mmHg	Treatment of varicose veins
2	Medium support	18–24 mmHg	Treatment of severe chronic hypertension and severe varicose veins, and prevention of venous leg ulcers
3	Strong support	25–35 mmHg	Treatment of more severe varicosities and prevention of venous leg ulcers

2. Classification of compression stockings (EU).

Class	Descriptor	Ankle pressure	Indication
1	Light support	18.4–21.1 mmHg	Mild varices, venous hypertension in pregnancy, heaviness and fatigue in leg
2	Medium support	25.2–32.3 mmHg	Pronounced varices, moderate oedema, inflammation of superficial veins after resolution of mild ulceration
3	Strong support	36.5–46.6 mmHg	Severe varicose veins, post‐thrombotic syndrome, pronounced oedema, prevention of venous ulcers
4	Heavy support	> 59 mmHg	Lymphoedema, elephantiasis

Compression bandages are further divided according to the force needed to extend them (compression level applicable to a limb). In addition, the performance of a bandage is directly dependent on operator training and application technique (specifically, whether the bandage is applied in a spiral or figure of eight, how many layers are applied, and the amount of extension used; Partsch 2008; Shi 2021).

Side effects of compression therapy include skin irritation and itching, decompensation of heart failure, and exceptional but potentially devastating complications such as neurological damage, venous thromboembolism, arterial thrombosis, and skin necrosis. The most severe complications are infrequent and occur in predisposed people or during inappropriate use of compression devices. Clinicians must screen for conditions that increase the risk of complications before prescribing compression therapy, and must check compression devices for proper fit and use (Rabe 2020).

How the intervention might work

Although compression therapy is the first‐line treatment for venous ulcers, its mechanism of action is not fully understood (Partsch 2015). Some authors propose that the application of external pressure to the calf muscle increases interstitial pressure, resulting in increased venous return and reduced venous hypertension (Nicolaides 1993; Partsch 2006). Compression reduces oedema by decreasing capillary filtration, moving fluid from compressed tissues to non‐compressed tissues, and improving lymphatic drainage (Partsch 2015). Nevertheless, there are limited data concerning how long compression hosiery should be worn after healing and which class of compression hosiery is superior for preventing venous leg ulcer recurrences.

Why it is important to do this review

Venous ulcers affect up to three in 1000 people, with higher incidences in older people. This type of evolution of chronic venous disease has a significant negative impact on quality of life and results in considerable costs for the people affected, systems, and healthcare professionals (Nelson 2016). Compression therapy is currently the first‐line treatment (Bergan 2006).

This is an update of a Cochrane review first published in 2000, and previously updated in 2012 and 2014 (Nelson 2000; Nelson 2012; Nelson 2014b). Since the original Cochrane review of compression for the prevention of recurrent venous ulcers, the number of relevant randomised controlled trials (RCTs) has more than doubled, associated with an increase in the variety of compression modalities and refinement of device classifications.

We have updated this review to summarise the current knowledge on the use of compression in preventing the recurrence of venous ulcers and to provide up‐to‐date evidence for decision‐makers, with the aim of ensuring safe evidence‐based clinical practice.

Objectives

To assess the effects of compression (socks, stockings, tights, bandages) for preventing recurrence of venous leg ulcers.

Methods

Criteria for considering studies for this review

Types of studies

We included RCTs and quasi‐RCTs, with individual or cluster randomisation, that assessed compression for preventing recurrence of venous ulcers. We considered the following comparisons.

• Compression versus no compression

• Different strengths of compression

• Different lengths of compression hosiery (below‐knee versus above‐knee/thigh‐length)

• Compression bandages versus compression hosiery

• Different types or brands of compression hosiery

• Different compression regimens (e.g. long‐stretch, short‐stretch, single‐layer)

There were no restrictions on publication status, date, or language.

We did not expect to identify any cross‐over RCTs. However, should such studies emerge in future updates, we plan to only consider outcome data from the first intervention phase (i.e. prior to cross‐over) to avoid the risk of carry‐over effects.

Types of participants

We included studies that recruited people of any age and either sex with healed venous leg ulcers. We accepted trialists' inclusion criteria regarding the diagnosis of ulcers as venous in origin. If we found studies with mixed populations, and only a subset of the participants met our inclusion criteria, we attempted to obtain data for the subgroup of interest from the trialists. For studies with mixed populations where we could not obtain data on the subgroup of interest, we included all participants in our analysis if at least 50% met our eligibility criteria. We planned to explore the effect of this decision in sensitivity analyses.

Types of interventions

We included studies that evaluated compression bandages or hosiery (tights, stockings, or socks) for prevention of venous leg ulcer recurrence. Studies of intermittent pneumatic compression devices were ineligible, as another Cochrane review covers this intervention (Nelson 2014a). Neither did we include studies of compression for the treatment of venous ulcers, which is the topic of another Cochrane review (Shi 2021).

Types of outcome measures

Reporting of our prespecified outcomes was not an inclusion criterion for this review. If we identified an eligible study that reported none of our outcomes, we contacted the study authors to verify whether they had assessed but not reported a relevant outcome. We provided a narrative description of eligible trials that provided no usable data. We presented the results for all time points reported. We considered whether the time points were sufficiently similar for statistical pooling.

Primary outcomes

• Reulceration (break in the skin) anywhere on the treated leg, irrespective of cause (i.e. proportion of participants with recurrence of ulceration anywhere on the treated leg during the follow‐up period)

Secondary outcomes

• Duration of reulceration episodes during the follow‐up period

• Proportion of follow‐up period without ulcers (i.e. during which the participant was ulcer‐free)

• Ulceration on the contralateral leg (i.e. proportion of participants with contralateral ulceration at any time during follow‐up)

• Noncompliance (i.e. proportion of participants who stopped using compression during the follow‐up period), as defined by the study authors

• Comfort (i.e. the experience of wearing compression stockings), as defined by the study authors

• Adverse effects (e.g. allergic reaction, damage to the skin tissue)

We presented the outcomes at the following two time points after the start of the intervention, if data were available.

• Early outcomes (up to 18 months after the start of the intervention)

• Long‐term outcomes (more than 18 months after the start of the intervention)

For changes to this section, see Differences between protocol and review.

Search methods for identification of studies

Electronic searches

For this third update, the Cochrane Wounds Information Specialist searched the following electronic databases to identify reports of relevant clinical trials on 17 February 2022.

• Cochrane Wounds Specialised Register (searched 17 February 2022)

• Cochrane Central Register of Controlled Trials (CENTRAL; 2022, Issue 1) in the Cochrane Library (searched 17 February 2022)

• MEDLINE Ovid including In‐Process & Other Non‐Indexed Citations (1946 to 17 February 2022)

• Embase Ovid (1974 to 17 February 2022)

• CINAHL Plus EBSCO (Cumulative Index to Nursing and Allied Health Literature; 1937 to 17 February 2022)

For MEDLINE Ovid, the Cochrane Wounds Information Specialist combined the subject‐specific strategy with the sensitivity‐ and precision‐maximising version of the Cochrane Highly Sensitive Search Strategy for identifying randomised trials (2008 revision; Lefebvre 2022). The Cochrane Wounds Information Specialist combined the Embase Ovid search with the Ovid Embase filter developed by Cochrane UK (Lefebvre 2022), and combined the CINAHL Plus EBSCO search with the trial filter developed by Glanville 2019. There were no restrictions with respect to language, date of publication, or study setting.

We searched LILACS (Latin American and Caribbean Health Science Information database) and IBECS (Indice Bibliográfico Español de Ciencias de la Salud; ibecs.isciii.es) in collaboration with the Cochrane Brazil Information Specialist on 17 March 2022 (Appendix 2), and we performed top‐up searches on 5 August 2023. We used no filters, but selected the RCTs manually from the LILACS and IBECS databases.

We also searched the following clinical trials registries.

• US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (www.clinicaltrials.gov; searched 23 February 2022)

• World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP; www.who.int/clinical-trials-registry-platform; searched 23 February 2022)

We subsequently conducted systematic top‐up searches of the following databases for RCTs and controlled clinical trials without language, publication year, or publication status restrictions.

• CENTRAL (Cochrane Register of Studies Online (CRSO); 2023, Issue 8)

• MEDLINE (PubMed MEDLINE Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, PubMed MEDLINE Daily, and PubMed MEDLINE; 24 February 2022 to 6 August 2023)

• Embase Wiley (24 February 2022 to 6 August 2023)

• CINAHL EBSCO (24 February 2021 to 6 August 2023)

• AMED Ovid (24 February 2022 to 6 August 2023)

We also performed top‐up searches of WHO ICTRP and ClinicalTrials.gov on 6 August 2023.

Appendix 2 shows the search strategies for the Cochrane Wounds Specialised Register, CENTRAL, MEDLINE Ovid, Embase Ovid, CINAHL Plus EBSCO, and the clinical trials registries.

Details of the search strategies used for the previous versions of the review can be found in Nelson 2014b.

Searching other resources

We aimed to identify other potentially eligible trials or ancillary publications by searching the reference lists of included trials, as well as relevant systematic reviews and health technology assessment reports.

For the original version of this review, we contacted experts in wound care and companies that produce compression stockings/bandages to enquire about unpublished, ongoing, and recently published trials. We scrutinised citations within obtained reviews and papers to identify additional studies. When necessary, we contacted authors of key papers and abstracts to request further information about their trials.

Data collection and analysis

We conducted data collection and analysis following methods described in the Cochrane Handbook for Systematic Reviews of Interventions (McKenzie 2022). For trials that considered multiple interventions in the same group, we only analysed the data of interest. Where we were unable to obtain usable data from a trial, we presented only descriptive data in the review. The Differences between protocol and review section documents changes from previously published versions of the review.

Selection of studies

Two review authors (MAMS and FMJ) assessed the titles and abstracts of all studies identified by the search (minus duplicates) using Covidence, excluding all records that were clearly irrelevant. Two review authors (MAMS and SGJS) independently retrieved and screened the full‐text articles of the remaining references to identify studies for inclusion in the review, recording the reasons for exclusion of ineligible studies. We obtained all available publications for each study. We resolved any disagreements by discussion or by consulting a third review author if necessary. Where the eligibility of a study was unclear, we contacted the study authors. We documented the study selection process in sufficient detail to create a PRISMA flowchart (Liberati 2009).

Data extraction and management

Two review authors (MAMS and SGJS) independently extracted data from the included studies using a predesigned data extraction sheet. We contacted study authors to minimise missing data. We resolved disagreements by discussion with a third review author (SEMBS or AN).

We collected the following information.

• Methods: details of the publication (e.g. year, country, authors, journal), study design, setting, population data (e.g. age), inclusion/exclusion criteria, sample size, number of participants allocated to each treatment group, duration of follow‐up

• Participants: number randomised, number lost to follow‐up/withdrawn, number analysed, mean age and age range, sex

• Interventions: type of intervention and any concomitant treatments

• Outcomes: reported and collected primary and secondary outcomes, and time points reported

• Notes: sources of funding and notable conflicts of interest

We resolved any disagreements by consensus or by consulting a third review author (SBS or AN). One review author (MAMS) transferred data into the Review Manager 5 file (Review Manager 2020). We double‐checked all entered data by comparing the data presented in the systematic review with the data in the extraction form.

Assessment of risk of bias in included studies

Two review authors (MAMS and SGJS) independently assessed the risk of bias in the included trials using the Cochrane risk of bias tool RoB 1, which covers the following domains (Higgins 2011).

• Sequence generation (selection bias)

• Allocation concealment (selection bias)

• Blinding of participants and personnel (performance bias)

• Blinding of outcome assessment (detection bias)

• Incomplete outcome data (attrition bias)

• Selective outcome reporting (reporting bias)

• Other potential sources of bias (e.g. extreme baseline imbalance)

We assessed the risk of performance bias, detection bias, and attrition bias for each outcome separately. We resolved any disagreements by discussion. We presented our assessments as a risk of bias summary figure, which shows all the judgements in a cross‐tabulation of study by entry.

For cluster‐RCTs, we planned to consider the risk of bias in relation to recruitment bias, baseline imbalance, loss of clusters, incorrect analysis, and comparability with individually randomised trials (Eldridge 2016; Higgins 2022). However, no included trials used cluster randomisation.

Measures of treatment effect

We conducted our analysis using Cochrane Review Manager 5 software (Review Manager 2020). For dichotomous outcomes (e.g. reulceration), we calculated risk ratios (RRs) with 95% confidence intervals (CIs). For continuous outcomes (e.g. duration of reulceration episodes), we calculated mean differences (MDs) with 95% CIs when all studies used the same scale, and standardised mean differences (SMDs) with 95% CIs if different studies measured the same outcome with different scales.

Unit of analysis issues

We observed whether the trials had used the participant as the unit of analysis. A specific unit of analysis issue may occur if trials perform randomisation at the participant level but analyse data by leg ulcer (Higgins 2022). If trials included multiple intervention arms, we considered only the arms relevant to the scope of our review. Where there were two or more eligible intervention groups in the same meta‐analysis, we combined groups to create a single pairwise comparison. Where a study included repeated observations, we followed recommendations in Chapter 23 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022).

We addressed the level at which randomisation occurred in our analysis, considering each participant as the unit of analysis.

Had we included any cross‐over RCTs, we would have only used data from the first phase (before cross‐over) to avoid the risk of carry‐over effects, as described in Section 23.2.4 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2022).

Dealing with missing data

Where we observed incomplete data in the reports, we contacted the study authors for further information. When a study did not specify participant numbers before dropouts, we presented only complete case data. We presented all analyses as complete‐case for continuous variables, and we did not impute any missing data. Where there were missing data, such as standard deviations (SDs), we aimed to calculate these using all available data (Higgins 2021). If the calculation was not possible, we contacted the study authors. If we could not obtain or calculate the missing data, we excluded the study from any relevant meta‐analyses. We reported missing data in the Characteristics of included studies table.

Assessment of heterogeneity

First, we considered clinical heterogeneity (variability in the participants, interventions, and outcomes reported) and methodological heterogeneity (variability in study design, outcome measurement tools, and risk of bias). We then assessed statistical heterogeneity using the Chi² test (with significance set at 0.10) in conjunction with the I² statistic (Higgins 2003). We considered that I² values of 25% or less indicated no or low heterogeneity, and values of more than 75% indicated considerable heterogeneity (Deeks 2022; Higgins 2003).

Assessment of reporting biases

Reporting biases arise when the disclosure of findings is influenced by the nature and direction of the results. Publication bias is one of several possible causes of small‐study effects. Funnel plots visually assess whether small‐study effects may be present in a meta‐analysis. A funnel plot is a simple scatter plot of the intervention effect estimates from individual RCTs against some measure of each trial's size or precision (Page 2022). Funnel plots are only informative when the analysis includes many studies.

In future updates of this review, if we include more than 10 studies in any meta‐analysis, we intend to assess the presence of publication bias and other reporting biases using funnel plots. If asymmetry is present, we intend to explore possible causes, including publication bias, poor methodological quality, and true heterogeneity (Higgins 2021).

Data synthesis

We summarised the details of the studies in a narrative review according to intervention and comparator, participants, and outcome measurement, including duration of follow‐up. We considered clinical and methodological heterogeneity, and we pooled the results if the studies were sufficiently similar. Otherwise, we presented the results narratively.

We grouped findings first by comparison then by outcome, and we used a random‐effects approach for all meta‐analyses. The random‐effects model provides a more conservative estimate of effect in the presence of known and unknown potential heterogeneity (Deeks 2001). We used Chi² and I² to quantify heterogeneity but not to guide the choice of model for meta‐analysis. We exercised caution when meta‐analysed data were at risk of small‐study effects, because a random‐effects model may have been unsuitable.

Using Review Manager 5 software (Review Manager 2020), we created forest plots to present summary estimates (RR or MD/SMD) with 95% CIs. We also considered the representation of subgroups in forest plots without overall pooling.

Subgroup analysis and investigation of heterogeneity

There were insufficient data (i.e. stratified data presented in the trials) to perform subgroup analyses to determine whether the following factors influence reulceration rates.

• Funding sources (not‐for‐profit versus other/unclear)

• Follow‐up duration

When we identified significant heterogeneity, we planned to: check the data extraction and data entry for errors and possible outlying studies; if outliers existed, perform a sensitivity analysis by removing them; and, if heterogeneity was still present, perform subgroup analyses for study‐level characteristics to explain heterogeneity as far as possible (Cipriani 2013; Deeks 2022).

Sensitivity analysis

There were insufficient data to perform sensitivity analyses. We had planned to assess the robustness of effect estimates for reulceration and noncompliance to the inclusion of studies with selection bias (excluding trials with inadequate or unclear allocation concealment); detection bias (excluding trials with unclear or inadequate blinding of outcome assessors); and attrition bias (excluding trials with incomplete outcome data). However, we could not perform this combined sensitivity analysis because all studies were at unclear or high risk of bias in at least one of the three domains.

Summary of findings and assessment of the certainty of the evidence

Using GRADEpro GDT software (GRADEpro GDT), we prepared a summary of findings table for each main comparison, reporting the following outcomes.

• Reulceration

• Duration of reulceration episodes

• Proportion of follow‐up without ulcers

• Ulceration on the contralateral leg

• Noncompliance

• Comfort

• Adverse effects

We used the five GRADE considerations (risk of bias, consistency of effect, imprecision, indirectness, and publication bias) to assess the certainty of the body of evidence for each outcome. We rated the certainty of the evidence as 'high', 'moderate', 'low', or 'very low', and justified any downgrading in footnotes, including additional comments where necessary. When evaluating the risk of bias consideration, we downgraded the certainty of the evidence when a study was classified at high risk of bias for at least one domain (except performance bias), or unclear risk of selection bias (random sequence generation and allocation concealment). Since blinding of participants or personnel was unlikely due to the nature of the interventions, we did not downgrade the GRADE assessment based on risk of performance bias. We based the summary of findings tables on the methods described in Chapters 11 and 12 of the Cochrane Handbook for Systematic Reviews of Interventions (GRADE 2004; Higgins 2021).

Results

Description of studies

See the Characteristics of included studies, Characteristics of excluded studies, and Characteristics of ongoing studies tables.

Results of the search

Figure 1 illustrates the study selection process in a PRISMA diagram.

1.

Study flow diagram.

For the update of this review, the database searches returned 2744 records (1755 records after duplicate removal). We excluded 1722 records based on their titles and abstracts, and retrieved the full‐text articles of the remaining 33 records. Four new studies (five reports) met our eligibility criteria (Clarke‐Moloney 2014; Kapp 2013; Milic 2018; Milic 2019). We excluded 26 studies (27 reports), with reasons given in the Characteristics of excluded studies table. We also identified one ongoing study (NCT01680809).

Included studies

We identified four new studies in this update, bringing the total number of included studies to eight (Clarke‐Moloney 2014; Franks 1995; Kapp 2013; Milic 2010; Milic 2018; Milic 2019; Nelson 2006; Vandongen 2000). The studies were published between 1995 and 2019. Three studies were conducted in Serbia (Milic 2010; Milic 2018; Milic 2019), two in Australia (Kapp 2013; Vandongen 2000), one in Ireland (Clarke‐Moloney 2014), one in England (Franks 1995), and one in Scotland (Nelson 2006).

The Characteristics of included studies table provides details of each trial.

Study participants

The eight studies provided data for 1995 participants. The trialists in five trials calculated an appropriate sample size before beginning recruitment (Franks 1995; Milic 2018; Milic 2019; Nelson 2006; Vandongen 2000). The number of participants varied from 100 (Clarke‐Moloney 2014) to 477 (Milic 2019). All studies evaluated people with a recently healed leg ulcer of venous aetiology. The exclusion criteria varied widely, but most studies excluded people with peripheral artery disease and people considered unable to manage daily application of the stockings. Two studies did not mention the exclusion criteria (Clarke‐Moloney 2014; Milic 2019).

All studies evaluated women and men. There were more female participants than male participants in five studies (Clarke‐Moloney 2014; Franks 1995; Kapp 2013; Nelson 2006; Vandongen 2000) and overall. Clarke‐Moloney 2014 provided no details about participant age. The mean age of the participants in other studies ranged from 58 years to 78 years. All studies used participants as the unit of analysis.

Interventions

Studies evaluated different classes of compression and duration of treatment. Franks 1995 compared different brands of stocking with the same compression level (UK class 2, moderate) in community leg ulcer clinics. Four studies evaluated UK class 2 (moderate) versus UK class 3 (high) compression hosieries (Milic 2010; Milic 2018; Milic 2019; Nelson 2006). Clarke‐Moloney 2014 compared EU class 1 (light) and EU class 2 (moderate) compression stockings. Kapp 2013 compared German class 2 (moderate) and German class 3 (high) compression stockings. Vandongen 2000 compared EU class 3 (high) graduated stockings with no compression.

Co‐interventions also varied widely. Clarke‐Moloney 2014 and Milic 2018 administered a questionnaire about compliance and experience during study visits. Kapp 2013 and Milic 2018 provided education and assistance with stocking applications. New stockings were provided at regular intervals: every three months (Franks 1995), four months (Milic 2010; Nelson 2006), or six months (Vandongen 2000). Milic 2019 provided no information regarding co‐interventions.

Outcomes

All studies evaluated the rate of reulceration at specific time points during follow‐up. The duration of the intervention and follow‐up varied widely among studies, ranging from 26 weeks (Kapp 2013) to 10 years (Milic 2019). No publications specified how trialists had obtained the data for ulcer recurrence (whether through questionnaires or medical records).

Nelson 2006 defined reulceration as a break in the skin of the leg persisting for at least six weeks (the outcome assessor was not blinded). Milic 2018 and Vandongen 2000 defined reulceration as any break in the skin between the ankle and the knee. Franks 1995, Milic 2010, and Milic 2019 provided no definition of reulceration. Clarke‐Moloney 2014 defined reulceration as an epithelial breakdown below the knee lasting more than four weeks. Kapp 2013 defined time to recurrence as the number of days between healing of the study wound and occurrence of any wound in the same anatomical location.

Vandongen 2000 also assessed the effect of compression stockings for reducing the area of lipodermatosclerosis in people with previous venous ulceration.

Seven studies assessed noncompliance with treatment (Clarke‐Moloney 2014; Franks 1995; Kapp 2013; Milic 2010; Milic 2018; Milic 2019; Nelson 2006). Clarke‐Moloney 2014 reported the factors that influenced reulceration and noncompliance.

Milic 2010 was published as a conference abstract only.

We attempted to contact all authors of included studies to minimise missing data. Only Dr Suzane Kapp replied, confirming the data published in the paper, but without providing any other methodological information or unpublished results (Kapp 2013).

Excluded studies

See the Characteristics of excluded studies table.

In total, we excluded 26 studies (27 full‐text articles): 11 evaluated elastocompression in people with active ulcers (Adderley 2014; Franks 2000; Harrison 2011; ISRCTN95282887; Junger 2002; Junger 2006; Moffatt 2003; NCT02782689; Polignano 2004; Rocca 2012; Wilkinson 1997); 11 enroled people with active ulcers, evaluating treatments other than elastocompression as the primary intervention (Davies 2004; Gohel 2005; Gohel 2018; ISRCTN75319519; Kapp 2012; Kelechi 2017; Klonizakis 2018; Milic 2007; NCT03688841; Stubbs 2004; Van Gent 2015); two were not RCTs (Iglesias 2004; Mauck 2014); one evaluated treatment other than elastocompression as the primary intervention (NCT05239416); and one evaluated elastic compression in prevention of venous stasis in people undergoing abdominal surgery (Lewis 1976).

Ongoing studies

Previous versions of this review listed no ongoing studies. For this update, we identified one ongoing study (NCT01680809; see the Characteristics of ongoing studies table).

Risk of bias in included studies

Figure 2 and Figure 3 summarise our risk of bias assessments. See also the Characteristics of included studies table for further details.

2.

Review authors' judgements about each risk of bias item for each included study.

3.

Review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

Of the eight studies, only two were judged at low risk of selection bias (Clarke‐Moloney 2014; Kapp 2013).

Random sequence generation

Six trials used a computer‐generated randomisation sequence (Clarke‐Moloney 2014; Franks 1995; Kapp 2013; Milic 2018; Milic 2019; Nelson 2006), and the remaining two trials did not describe the randomisation method so were at unclear risk of bias (Milic 2010; Vandongen 2000).

Allocation concealment

Allocation concealment (i.e. the person recruiting people into the trial was unaware of which group they would be recruited to) was attempted using a closed envelope system in two trials (Clarke‐Moloney 2014; Vandongen 2000). Clarke‐Moloney 2014 used sequentially numbered, sealed envelopes assigned to consecutive people, and was judged at low risk of bias. However, it is unclear whether the envelopes were shuffled, numbered or logged in Vandongen 2000, so we considered the study at unclear risk.

All identifying product information was removed from the stockings in Kapp 2013. The butt of the tag was marked with red and blue markers as appropriate, and stockings were placed in Ziplock bags with care information provided on an unbranded card. We assessed Kapp 2013 at low risk of bias. Although Nelson 2006 used a remote method of allocation by telephone, there was a possibility of selection bias as the person randomising participants was occasionally informed of the allocation of subsequent people by the remote randomisation office (personal communication, Nelson). We considered Nelson 2006 at unclear risk of bias for allocation concealment. We also considered Milic 2010, Milic 2018 and Milic 2019 at unclear risk of bias because they provided limited information on allocation concealment.

Franks 1995 used an open computer‐generated randomisation list, and therefore allocation was not concealed (high risk of bias).

Blinding

In Kapp 2013, all personnel and study participants were blinded to the interventions, stocking brand, and the level of compression. Blinding participants and caregivers would not have been possible in five studies, which we rated at high risk of performance bias (Clarke‐Moloney 2014; Franks 1995; Milic 2010; Nelson 2006; Vandongen 2000). Milic 2018 and Milic 2019 provided insufficient information about blinding of participants and personnel (unclear risk of performance bias).

Five studies did not blind outcome assessors so were at high risk of detection bias (Clarke‐Moloney 2014; Kapp 2013; Milic 2010; Nelson 2006; Vandongen 2000). The remaining studies provided insufficient information about blinding of outcome assessors so were at unclear risk (Franks 1995; Milic 2018; Milic 2019).

Incomplete outcome data

Franks 1995 was at unclear risk of bias for this domain as it was unclear if all randomised participants were included in the analyses. Vandongen 2000 was at high risk of attrition bias because participants were withdrawn if they did not wear their stockings, and this is likely to have influenced the reulceration outcome; 16 participants withdrew from the stocking arm in the first six months for "stocking‐related" reasons, compared with three participants from the no compression arm (who wanted to wear stockings). More people dropped out between six and 12 months, and we did not analyse these data.

The remaining studies reported outcome data for all participants so were at low risk of attrition bias (Clarke‐Moloney 2014; Kapp 2013; Milic 2010; Milic 2018; Milic 2019; Nelson 2006).

Selective reporting

Milic 2010 was published as an abstract only and provided few details of the study methodology; we judged this study at unclear risk of reporting bias. The remaining studies reported all expected outcomes so were at low risk.

Other potential sources of bias

In Nelson 2006, the treatment groups were comparable at baseline for age, sex, ulcer location, and general medical history, so we judged the study at low risk of other bias. Four studies did not report baseline group comparability so were at unclear risk of other bias (Milic 2010; Milic 2018; Milic 2019; Vandongen 2000). In Franks 1995, participants allocated to the class 2 Medi sock had a median ulcer duration of 5.7 months, compared with 2.0 months in the Scholl group. This may reflect greater severity of ulcer disease in the Medi group. It seemed the trialists did not adjust for this difference in the analysis, so we judged Franks 1995 at high risk of other bias. The other two studies (Clarke‐Moloney 2014; Kapp 2013) were judged to have unclear risk of other bias due to source of funding.

Effects of interventions

See: Table 1; Table 2; Table 3; Table 4

We grouped the studies into the following comparisons.

• EU class 3 compression hosiery versus no compression (Vandongen 2000)

• EU class 1 compression hosiery versus EU class 2 compression hosiery (Clarke‐Moloney 2014)

• UK class 2 compression hosiery versus UK class 3 compression hosiery (Kapp 2013; Milic 2010; Milic 2018; Milic 2019; Nelson 2006)

• Different brands of UK class 2 compression hosiery (Franks 1995)

No studies compared different lengths of compression hosiery (below‐knee/above‐knee).

We used a random‐effects analysis throughout. Each pooled result is an average effect rather than a common effect, and should be interpreted as such.

EU class 3 compression hosiery versus no compression

One study evaluated EU class 3 compression stockings versus no compression (Vandongen 2000). See Table 1.

Primary outcomes

Reulceration

Vandongen 2000 aimed to assess the effect of compression stockings for reducing the area of lipodermatosclerosis in people with previous venous ulceration. The study also reported reulceration rates with and without compression. EU class 3 compression stockings may reduce reulceration compared with no compression over six months (RR 0.46, 95% CI 0.27 to 0.76; 153 participants; low‐certainty evidence; Analysis 1.1). We downgraded the certainty of the evidence by one level for imprecision (small sample size) and by one level for high risk of bias (lack of blinding).

1.1. Analysis.

Comparison 1: EU class 3 compression hosiery versus no compression, Outcome 1: Reulceration at 6 months

Vandongen 2000 also reported reulceration at 12 months, but with data for only 22/72 participants randomised to the stockings and 24/81 participants randomised to no compression. We considered these data insufficiently reliable for inclusion in this review.

Secondary outcomes

Duration of reulceration episodes

Vandongen 2000 did not report duration of reulceration episodes.

Proportion of follow‐up period without ulcers

Vandongen 2000 did not report proportion of follow‐up without ulcers.

Ulceration on the contralateral leg

Vandongen 2000 did not report ulceration on the contralateral leg.

Noncompliance

Sixteen of 72 participants in the intervention group refused to continue wearing stockings.

Comfort

Vandongen 2000 did not report comfort.

Adverse effects

Vandongen 2000 did not measure adverse effects.

EU class 1 compression hosiery versus EU class 2 compression hosiery

One study evaluated EU class 1 compression hosiery versus EU class 2 compression hosiery (Clarke‐Moloney 2014). See Table 2.

Primary outcomes

Reulceration

There was an overall 12‐month reulceration rate of 16.1%. EU class 1 stockings compared with EU class 2 stockings may have little or no effect on reulceration (RR 1.70, 95% CI 0.67 to 4.32; 99 participants; low‐certainty evidence; Analysis 2.1). However, Clarke‐Moloney 2014 found that the risk of reulceration was higher in noncompliant participants compared with compliant participants (RR 6.22, P < 0.001). We downgraded the certainty of the evidence by one level for imprecision (small sample size and large CI that incorporates the possibility of benefit and harm) and by one level for high risk of bias (lack of blinding).

2.1. Analysis.

Comparison 2: EU class 1 compression hosiery versus EU class 2 compression hosiery, Outcome 1: Reulceration

Secondary outcomes

Duration of reulceration episodes

Clarke‐Moloney 2014 did not report duration of reulceration episodes.

Proportion of follow‐up period without ulcers

Clarke‐Moloney 2014 did not report proportion of follow‐up without ulcers.

Ulceration on the contralateral leg

Clarke‐Moloney 2014 did not report ulceration on the contralateral leg.

Noncompliance

There may be little or no difference in rates of noncompliance at 12 months between people using EU class 1 stockings and people using EU class 2 stockings (RR 1.22, 95% CI 0.40 to 3.75; 99 participants; low‐certainty evidence; Analysis 2.2). We downgraded the evidence by one level for imprecision (small sample size and large CI that incorporates the possibility of benefit and harm) and by one level for high risk of bias (lack of blinding).

2.2. Analysis.

Comparison 2: EU class 1 compression hosiery versus EU class 2 compression hosiery, Outcome 2: Noncompliance

Comfort

Clarke‐Moloney 2014 did not report participant comfort.

Adverse effects

Clarke‐Moloney 2014 did not measure adverse effects.

UK class 2 compression hosiery versus UK class 3 compression hosiery

Five studies evaluated UK class 2 compression hosiery versus UK class 3 compression hosiery (Kapp 2013; Milic 2010; Milic 2018; Milic 2019; Nelson 2006). See Table 3.

Primary outcomes

Reulceration

All five trials reported reulceration (Kapp 2013; Milic 2010; Milic 2018; Milic 2019; Nelson 2006). UK class 2 hosiery may be associated with a higher risk of reulceration compared with UK class 3 hosiery (RR 1.55, 95% CI 1.26 to 1.91; P < 0.04, I² = 60%; 5 studies, 1314 participants; low‐certainty evidence; Analysis 3.1). We downgraded the certainty of the evidence by one level for indirectness (wide variation in follow‐up between studies) and by one level for risk of bias (all studies at unclear/high risk of bias due to lack of blinding/insufficient information about blinding).

3.1. Analysis.

Comparison 3: UK class 2 compression hosiery versus UK class 3 compression hosiery, Outcome 1: Reulceration

Two trials evaluated participants over five years (Milic 2018; Nelson 2006). Milic 2018 found reduced reulceration with high compression (UK class 3) compared with moderate compression (UK class 2). Nelson 2006 found no clear difference between the groups in reulceration rates over five years. Kapp 2013 found no clear difference between the groups in reulceration rates over two years. Milic 2010 (three‐year follow‐up) and Milic 2019 (10‐year follow‐up) found reduced reulceration with high compression (UK class 3) compared with moderate compression (UK class 2).

Secondary outcomes

Duration of reulceration episodes

No trials reported duration of reulceration episodes.

Proportion of follow‐up period without ulcers

No trials reported proportion of follow‐up without ulcers.

Ulceration on the contralateral leg

No trials reported ulceration on the contralateral leg.

Noncompliance

All five trials reported participant noncompliance with compression therapy (Kapp 2013; Milic 2010; Milic 2018; Milic 2019; Nelson 2006). Follow‐up was between 18 months and 10 years. People using UK class 2 hosiery may be more compliant with compression therapy compared with people using UK class 3 hosiery (RR for noncompliance 0.69, 95% CI 0.49 to 0.99; P = 0.07, I² = 55%; 5 studies, 1372 participants; low‐certainty evidence; Analysis 3.2). We downgraded the certainty of the evidence by one level for indirectness (wide variation in follow‐up between studies) and by one level for risk of bias (all studies at unclear/high risk of bias due to lack of blinding/insufficient information about blinding).

3.2. Analysis.

Comparison 3: UK class 2 compression hosiery versus UK class 3 compression hosiery, Outcome 2: Noncompliance

Comfort

No trials reported participant comfort.

Adverse effects

No trials reported adverse effects.

Different brands of UK class 2 compression hosiery

Franks 1995 compared two different brands of UK class 2 compression stockings (Scholl and Medi). See Table 4.

Primary outcomes

Reulceration

There may be little or no difference between Scholl UK class 2 compression stockings and Medi UK class 2 compression stockings in terms of reulceration at 18 months (RR 0.77, 95% CI 0.47 to 1.28; 166 participants; low‐certainty evidence; Analysis 4.1). We downgraded the certainty of the evidence by one level for imprecision (small sample size) and by one level for high risk of bias (lack of blinding).

4.1. Analysis.

Comparison 4: Comparison between different brands of compression hosiery (UK class 2), Outcome 1: Reulceration at 18 months

Franks 1995 also reported reulceration in compliant and noncompliant participants, though it is unclear if this comparison was prespecified. There was a higher reulceration rate in partially compliant participants (11/29 (37%) with partial compliance versus 43/136 (32%) with full compliance). The study also reported increased reulceration in people who were excluded from the trial because they could not wear compression stockings compared with participants who wore compression (11/17 (65%) in those excluded versus 58/171 (34%) with compression; RR 2.58, 95% CI 1.33 to 5.01). Therefore, this trial provided some indirect evidence that compression reduces reulceration.

Secondary outcomes

Duration of reulceration episodes

Franks 1995 did not report duration of reulceration episodes.

Proportion of follow‐up period without ulcers

Franks 1995 did not report proportion of follow‐up without ulcers.

Ulceration on the contralateral leg

Franks 1995 did not report ulceration on the contralateral leg.

Noncompliance

There may be little or no difference in rates of noncompliance between people using Scholl UK class 2 compression stockings and those using Medi UK class 2 compression stockings (RR 0.97, 95% CI 0.84 to 1.12; 1 study, 166 participants; low‐certainty evidence; Analysis 4.2). We downgraded the certainty of the evidence by one level for imprecision (small sample size) and by one level for high risk of bias (lack of blinding).

4.2. Analysis.

Comparison 4: Comparison between different brands of compression hosiery (UK class 2), Outcome 2: Noncompliance at 18 months

Comfort

Franks 1995 did not report participant comfort.

Adverse effects

Franks 1995 did not measure adverse effects.

Discussion

The use of compression after venous ulcer healing is now widespread, and further trials with a no compression comparison group are unlikely. There is evidence that people who do not comply with compression hosiery have higher reulceration rates than those who do comply, but this finding is less robust than evidence from direct comparisons within RCTs.

The trials reporting the relative benefits of medium‐ or high‐compression hosiery described only event rates at trial completion and not time to recurrence. Without this information, it is unclear whether the contradictory results for these five trials result from the difference in the follow‐up duration. As recurrence of an ulcer is likely after healing, and this risk does not reduce over time, trials with very long follow‐up may miss clinically essential differences in effectiveness. This would occur if high‐compression hosiery delayed reulceration significantly longer than moderate‐compression hosiery; however, by five years, the differences in effect size were minimal. Therefore, future studies need to report risk of reulceration at different time points and also measure time to reulceration, as this is likely to be crucial for patients and clinicians.

Given the high prevalence of venous disease and the relatively large number of compression trials for the treatment of ulceration (Cullum 2000), it is disappointing that so few trials have assessed compression for prevention of ulcer recurrence. There may be additional unpublished trials, and their impact on these results is unknown. Prospective registration of trials would reduce any potential publication bias.

Summary of main results

Non‐systematic reviews of the literature invariably state that compression hosiery reduces the recurrence of venous leg ulcers (Capeheart 1996). Our review included eight RCTs (1995 participants). The studies evaluated a variety of compression bandages and stockings; the most frequently used system was class 2 compression hosiery. No studies compared different lengths of compression (e.g. below versus above the knee).

EU class 3 compression stockings may reduce reulceration compared with no compression over six months.

There may be little or no difference between EU class 1 and EU class 2 compression stockings in terms of reulceration and noncompliance over 12 months. However, the primary outcome of the trial that provided this evidence was reduction in the area of lipodermatosclerosis, rather than reulceration, and attrition was high.

UK class 3 hosiery may be superior to UK class 2 hosiery in terms of reulceration, but inferior in terms of patient noncompliance.

There may be little or no difference in reulceration and patient noncompliance between Scholl and Medi UK class 2 compression stockings.

Overall completeness and applicability of evidence

This review included eight RCTs published between 1995 and 2019. The median study sample size was 249 participants, and all studies included people with a healed venous ulcer. The studies evaluated different classes of compression, and follow‐up ranged from 26 weeks to 10 years. Readers should bear these limitations in mind when interpreting the results. There was no information on duration of episodes of reulceration, ulceration on the contralateral leg, or adverse effects.

Certainty of the evidence

We rated the evidence for all reported outcomes as low certainty. We downgraded the certainty of the evidence for imprecision (few participants, few studies, or large CIs that included both benefit and harm), risk of bias (lack of blinding), and indirectness (wide variation between studies in duration of follow‐up).

Potential biases in the review process

We followed prespecified methods to prevent potential bias in the review process. We ran comprehensive electronic searches, searched trial registries, and checked the references of included studies and systematic reviews identified in electronic searches. We also tried to contact five trialists, but received no response (Clarke‐Moloney 2014; Milic 2010; Milic 2018; Milic 2019; Vandongen 2000).

This review has some limitations. First, we considered specific compression bandages or socks with similar levels of compression (Table 7), UK classes of compression hosiery (Table 5) and European classes of compression hosiery (Table 6) for analysis. Second, because of the limited number of included studies, we did not undertake subgroup analysis according to types of compression bandages/stockings or specific comparators. Third, the included studies measured outcomes differently and provided few usable data.

3. Comparative classification of compression stockings (UK/French/German).

Class	Support	British Standard BS 6612:1985	French Standard ASQUAL	German Standard RAL‐GZ 387:2000
1	Light	14–17 mmHg	10–15 mmHg	18–21 mmHg
2	Medium	18–24 mmHg	15–20 mmHg	23–32 mmHg
3	Strong	25–35 mmHg	20–36 mmHg	34–46 mmHg
4	Heavy	Not reported	> 36 mmHg	> 49 mmHg

Agreements and disagreements with other studies or reviews

We are unaware of any previous systematic reviews focused on compression for preventing the recurrence of venous ulcers. One systematic review and meta‐analysis considered both the treatment and prevention of ulcers using elastocompression (Mauck 2014). Another Cochrane review evaluated compression for the treatment but not the prevention of venous ulcer recurrence (Shi 2021).

Authors' conclusions

Implications for practice.

Compression with European (EU) class 3 compression stockings may reduce the risk of reulceration compared with no compression. Use of EU class 1 compression stockings compared with EU class 2 compression stockings may result in little or no difference in reulceration and noncompliance over 12 months. UK class 3 compression hosiery compared with UK class 2 compression hosiery may reduce the risk of reulceration; however, people who use the class 2 hosiery may be more compliant with compression therapy. There may be little or no difference between two brands of UK class 2 compression stockings (Scholl and Medi) in terms of reulceration and noncompliance. The evidence for all outcomes was of low certainty. We found no data on adverse effects, duration of reulceration episodes, or ulceration on the contralateral leg.

Implications for research.

More research is needed regarding acceptable modes of long‐term compression therapy for people at risk of recurrent venous ulceration. Future trials should consider interventions to help people wear compression, since higher compression may reduce reulceration rates compared with lower compression, but people may be less inclined to adhere to high‐compression therapy. In‐depth qualitative research is needed to understand patients' perspectives and explain low compliance rates.

Future trials of maintenance compression therapy and interventions to promote concordance should:

• be large enough to detect clinically important differences in recurrence;

• define ulcer recurrence clearly, as there may be small skin breaks due to varicose eczema that can be confused with a true ulcer recurrence;

• report co‐interventions thoroughly, including surgery, exercise advice, and drug therapies;

• use blinded outcome assessment;

• include harms of the use of the intervention;

• employ survival analysis methods to assess time to recurrence;

• incorporate economic evaluations;

• record information on quality of life.

What's new

Date	Event	Description
7 March 2024	New citation required and conclusions have changed	Updated. Conclusion changed.
7 March 2024	New search has been performed	Third update, 3 new authors. New search, 3 new trials identified (Clarke‐Moloney 2014; Kapp 2013; Milic 2018; Milic 2019), bringing the total number of included studies to 8 (1995 participants). Previously, there was little evidence that high compression hosiery reduces rates of reulceration. Now, there is low‐certainty evidence that high‐compression hosiery (EU class 3) may reduce reulceration compared with no compression. For the comparison between EU class 1 and EU class 2 compression hosieries, EU class 1 compression hosiery may result in little or no difference in reulceration and noncompliance. There is low‐certainty evidence that high‐compression hosiery (UK class 3) may reduce rates of reulceration compared with lower‐compression (UK class 2) hosieries.

History

Protocol first published: Issue 3, 2000
Review first published: Issue 4, 2000

Date	Event	Description
5 September 2014	Feedback has been incorporated	Second update, new search no new trials identified.
5 September 2014	New citation required but conclusions have not changed	Conclusions remain unchanged.
28 June 2012	New citation required and conclusions have changed	'Risk of bias' tables completed. Conclusions strengthened.
28 June 2012	New search has been performed	New searches, 2 trials added (Vandongen 2000; Milic 2010).
18 June 2008	Amended	Converted to new review format.
23 August 2000	New citation required and conclusions have changed	Substantive amendment.

Appendices

Appendix 1. Glossary of terms

Venous ulceration	Wound on the leg or ankle caused by abnormal or damaged veins
Venous ulcer	Wound on the leg or ankle caused by abnormal or damaged veins
Ankyloses	Joint stiffness due to abnormal adhesion and rigidity of the bones of the joint
Chronic venous disease	Medical condition in which the veins cannot pump enough blood back to the heart
Compression therapy	Application of a compression garment around the leg
Oedema	Swelling caused when fluid leaks out of the body's capillaries; excess of watery fluid collecting in the tissues
Fibrosis	Thickening and scarring of connective tissue
Hyperpigmentation	Increased pigmentation of an area of the skin
Lipodermatosclerosis	Induration caused by fibrosis of subcutaneous fat
Perforating veins	Veins that connect the deep leg veins to the surface veins
Placebo	Substance or treatment with no active therapeutic effect
Duplex ultrasound	Noninvasive evaluation of blood flow through the arteries and veins
Varicose veins	Gnarled, enlarged veins
Vascular	Relating to blood vessels
Venoactive drugs	Heterogeneous group of medicinal products that affect symptoms related to chronic venous disease
Venous	Relating to a vein
Venous eczema	Long‐term skin condition that affects the lower legs

Appendix 2. Search strategies

Cochrane Wounds Specialised Register

1 MESH DESCRIPTOR Leg Ulcer EXPLODE ALL AND INREGISTER

2 ((varicose next ulcer*) or (venous next ulcer*) or (leg next ulcer*) or (stasis next ulcer*) or (crural next ulcer*) or (ulcus next cruris) or (ulcer* next cruris)) AND INREGISTER

3 #1 OR #2 AND INREGISTER

4 MESH DESCRIPTOR Compression Bandages EXPLODE ALL AND INREGISTER

5 compression* AND INREGISTER

6 stocking* or hosiery AND INREGISTER

7 sock or socks or tights AND INREGISTER

8 bandag* AND INREGISTER

9 wrapp* AND INREGISTER

10 #4 OR #5 OR #6 OR #7 OR #8 OR #9 AND INREGISTER

11 #3 AND #10 AND INREGISTER

The Cochrane Central Register of Controlled Clinical Trials (CENTRAL)

#1 MeSH descriptor: [Leg Ulcer] explode all trees

#2 ((varicose next ulcer*) or (venous next ulcer*) or (leg next ulcer*) or (stasis next ulcer*) or (crural next ulcer*) or (ulcus next cruris) or (ulcer* next cruris)):ti,ab,kw

#3 #1 or #2

#4 MeSH descriptor: [Compression Bandages] explode all trees

#5 compression*:ti,ab,kw

#6 stocking* or hosiery:ti,ab,kw

#7 sock or socks or tights:ti,ab,kw

#8 bandag*:ti,ab,kw

#9 wrapp*:ti,ab,kw

#10 #4 or #5 or #6 or #7 or #8 or #9

#11 #3 and #10

The Cochrane Central Register of Controlled Clinical Trials (CENTRAL) Trial Registry specific search combined with main CENTRAL search

12 (NCT0* or ACTRN* or ChiCTR* or DRKS* or EUCTR* or eudract* or IRCT* or ISRCTN* or JapicCTI* or JPRN* or NTR0* or NTR1* or NTR2* or NTR3* or NTR4* or NTR5* or NTR6* or NTR7* or NTR8* or NTR9* or SRCTN* or UMIN0*):AU AND CENTRAL:TARGET

13 http*:SO AND CENTRAL:TARGET

14 #12 OR #13

15 #11 AND #14

Ovid MEDLINE

1 exp Leg Ulcer/

2 (varicose ulcer* or venous ulcer* or leg ulcer* or stasis ulcer* or (lower extremit* adj ulcer*) or crural ulcer* or ulcus cruris).tw.

3 1 or 2

4 exp Compression Bandages/

5 compression*.tw.

6 (stocking* or hosiery).tw.

7 (sock or socks or tights).tw.

8 bandag*.tw.

9 wrapp*.tw.

10 or/4‐9

11 3 and 10

12 randomized controlled trial.pt.

13 controlled clinical trial.pt.

14 randomi?ed.ab.

15 placebo.ab.

16 clinical trials as topic.sh.

17 randomly.ab.

18 trial.ti.

19 or/12‐18

20 exp animals/ not humans.sh.

21 19 not 20

22 11 and 21

Ovid Embase

1 exp Leg Ulcer/

2 (varicose ulcer* or venous ulcer* or leg ulcer* or stasis ulcer* or (lower extremit* adj ulcer*) or crural ulcer* or ulcus cruris).tw.

3 1 or 2

4 exp Compression Therapy/

5 exp Compression Bandage/

6 exp Compression Garment/

7 compression*.tw.

8 (stocking* or hosiery).tw.

9 (sock or socks or tights).tw.

10 bandag*.tw.

11 wrapp*.tw.

12 or/4‐11

13 3 and 12

14 Randomized controlled trial/

15 Controlled clinical study/

16 Random$.ti,ab.

17 randomization/

18 intermethod comparison/

19 placebo.ti,ab.

20 (compare or compared or comparison).ti.

21 ((evaluated or evaluate or evaluating or assessed or assess) and (compare or compared or comparing or comparison)).ab.

22 (open adj label).ti,ab.

23 ((double or single or doubly or singly) adj (blind or blinded or blindly)).ti,ab.

24 double blind procedure/

25 parallel group$1.ti,ab.

26 (crossover or cross over).ti,ab.

27 ((assign$ or match or matched or allocation) adj5 (alternate or group$1 orintervention$1 or patient$1 or subject$1 or participant$1)).ti,ab.

28 (assigned or allocated).ti,ab.

29 (controlled adj7 (study or design or trial)).ti,ab.

30 (volunteer or volunteers).ti,ab.

31 trial.ti.

32 or/14‐31

33 (exp animal/ or animal.hw. or nonhuman/) not (exp human/ or human cell/ or (human or humans).ti.)

34 32 not 33

35 13 and 34

EBSCO CINAHL Plus

S36 S12 AND S35

S35 S34 NOT S33

S34 S13 OR S14 OR S15 OR S16 OR S17 OR S18 OR S19 OR S20 OR S21 OR S22 OR S23 OR S24 OR S25 OR S26 OR S27

S33 S31 NOT S32

S32 MH (human)

S31 S28 OR S29 OR S30

S30 TI (animal model*)

S29 MH (animal studies)

S28 MH animals+

S27 AB (cluster W3 RCT)

S26 MH (crossover design) OR MH (comparative studies)

S25 AB (control W5 group)

S24 PT (randomized controlled trial)

S23 MH (placebos)

S22 MH (sample size) AND AB (assigned OR allocated OR control)

S21 TI (trial)

S20 AB (random*)

S19 TI (randomised OR randomized)

S18 MH cluster sample

S17 MH pretest‐posttest design

S16 MH random assignment

S15 MH single‐blind studies

S14 MH double‐blind studies

S13 MH randomized controlled trials

S12 S3 AND S11

S11 S4 OR S5 OR S6 OR S7 OR S8 OR S9 OR S10

S10 TI wrapp* OR AB wrapp*

S9 TI bandag* OR AB bandag*

S8 TI ( sock or socks or tights ) OR AB ( sock or socks or tights )

S7 TI ( stocking* or hosiery ) OR AB ( stocking* or hosiery )

S6 TI compression* OR AB compression*

S5 (MH "Compression Therapy")

S4 (MH "Compression Garments")

S3 S1 OR S2

S2 TI ( ((varicose ulcer*) or (venous ulcer*) or (leg ulcer*) or (stasis ulcer*) or (crural ulcer*) or (ulcus cruris) or (ulcer* cruris)) ) OR AB ( ((varicose ulcer*) or (venous ulcer*) or (leg ulcer*) or (stasis ulcer*) or (crural ulcer*) or (ulcus cruris) or (ulcer* cruris)) )

S1 (MH "Leg Ulcer+")

US National Institutes of Health Ongoing Trials Register (ClinicalTrials.gov)

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Venous Leg Ulcer

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Ulcer Venous

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Varicose Ulcer

Top of Form

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Venous Stasis UlcerBottom of Form

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Venous Ulceration

World Health Organization International Clinical Trials Registry Platform

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Venous Leg Ulcer

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Ulcer Venous

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Varicose Ulcer

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Venous Stasis Ulcer

compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping | Venous Ulceration

LILACS and IBECS search strategy

("bandagens compressivas" OR bandagens OR "curativos oclusivos" OR compression OR stockings OR tights OR hosiery OR sock OR socks OR bandage OR wrap OR wrapping) AND ("ulcera de perna" OR "venous leg ulcer" OR "varicose ulcer" OR "ulcer venous" OR "ulcera varicosa") AND ( db:("LILACS"))

Data and analyses

Comparison 1. EU class 3 compression hosiery versus no compression.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Reulceration at 6 months	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected

Comparison 2. EU class 1 compression hosiery versus EU class 2 compression hosiery.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Reulceration	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
2.2 Noncompliance	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected

Comparison 3. UK class 2 compression hosiery versus UK class 3 compression hosiery.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Reulceration	5	1314	Risk Ratio (M‐H, Random, 95% CI)	1.55 [1.26, 1.91]
3.2 Noncompliance	5	1372	Risk Ratio (M‐H, Random, 95% CI)	0.69 [0.49, 0.99]

Comparison 4. Comparison between different brands of compression hosiery (UK class 2).

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
4.1 Reulceration at 18 months	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected
4.2 Noncompliance at 18 months	1		Risk Ratio (M‐H, Random, 95% CI)	Totals not selected

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Clarke‐Moloney 2014.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Ethics and informed consent: ethics approval and consent obtained Total number: 100 participants (50 in group A, 50 in group B) Start date: no details given Duration of participation: 12 months End date: no details given Method of randomisation: computer‐generated randomisation list Blinding: participants: no; investigators: no; outcome assessors: no Power calculation: sample size not calculated a priori Losses to follow‐up: 1 participant randomised to class 1 was lost to follow‐up and withdrawn from the study analysis Unit of allocation: participants Source of funding: no details given
Participants	Setting: outpatients Country: Ireland Sex: 19 men, 40 women Age: no details given Inclusion criteria Healed venous leg ulcers Exclusion criteria: no details given
Interventions	Group A: EU class 1 compression stockings (18–21 mmHg) Group B: EU class 2 compression stockings (23–32 mmHg) Follow‐up: 1 week, 3 months, 6 months, 9 months, and 12 months. Participants completed a questionnaire regarding compliance and experience of wearing the stockings. Co‐interventions: education on stocking wear before randomisation
Outcomes	Outcomes of interest for this review (collected) Rate of ulcer recurrence in both stocking classes (ulcer recurrence was defined as an epithelial breakdown below the knee lasting > 4 weeks) Noncompliance (not properly defined) Other reported outcomes Factors that influence ulcer recurrence Factors that affected compliance Negative effect on daily activities of patients wearing compression stockings
Notes	Daily activities were not limited in either group.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisation list.
Allocation concealment (selection bias)	Low risk	Results of randomisation were entered sequentially into numbered, sealed envelopes, assigned to consecutive participants.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Personnel and participants were not blinded.
Blinding of outcome assessment (detection bias) All outcomes	High risk	No blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Only 1 participant lost to follow‐up and excluded from analysis.
Selective reporting (reporting bias)	Low risk	Trial registration and protocol not available; however, all expected outcomes were reported.
Other bias	Unclear risk	No information on who paid for the stockings.

Franks 1995.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Ethics and informed consent: no details given Total number: 188 participants (166 randomised: 92 in group A, 74 in group B) Start date: no details given Duration of participation: 18 months End date: no details given Method of randomisation: computer‐generated randomisation list Blinding: participants: no; investigators: no; outcome assessors: unclear Power calculation: a priori sample size calculation based on estimates of rate of reactions to stockings (20% versus 40%; power = 0.80, P = 0.05) Losses to follow‐up: 5 participants required made‐to‐measure stockings and 17 could not tolerate stockings owing to friable skin. Unit of allocation: participants Source of funding: Medi
Participants	Setting: outpatients (within a community leg ulcer service) Country: England Sex: 59 men, 129 women Age: mean 72 years Inclusion criteria Newly healed venous leg ulcers Ability to apply a compression sock Exclusion criteria "mild" arterial disease
Interventions	Group A: below knee (Medi) UK class 2 compression stockings Group B: below knee (Scholl) UK class 2 compression stockings Follow‐up: at 1 week, then at 3‐month intervals until 18 months. Participants received new stockings of the allocated type at the 3‐monthly visits. Co‐interventions: participants who experienced a reaction put a cotton sock on the leg before the stocking.
Outcomes	Outcomes of interest for this review (collected) Ulcer recurrence Noncompliance (participant‐reported) Other reported outcomes Adverse reactions
Notes	Duration of ulcer before healing was longer in the Medi group than in the Scholl group (5.7 months versus 2.0 months).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated (personal communication, Franks).
Allocation concealment (selection bias)	High risk	Open randomisation list (personal communication, Franks).
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of participants and personnel not possible due to nature of intervention.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Unclear who conducted the outcome assessment or if they were blinded to the allocation group.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Data reported as percentages. Unclear if all randomised participants were included in the analyses.
Selective reporting (reporting bias)	Low risk	Trial registration and protocol not available; however, all expected outcomes were reported.
Other bias	High risk	Baseline imbalance: median duration of ulcer before the study was almost 3 times longer in the Medi group. Unclear why there were 20% fewer participants in the Scholl group.

Kapp 2013.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Ethics and informed consent: ethics approval and consent obtained Total number: 100 participants (53 in group A, 47 in group B) Start date: no details given Duration of participation: 26 weeks End date: no details given Method of randomisation: computer‐generated randomisation list Blinding: participants: yes; investigators: yes; outcome assessors: no Power calculation: a priori sample size calculated based on effect size of 0.15 and attrition over a 26‐week study period Losses to follow‐up: 4 participants lost to follow‐up (2 each group). 3 participants excluded from analyses (2 from group A, 1 from group B) Unit of allocation: participants Source of funding: Department of Health, Victoria; Angior Family Foundation; Victorian and Australian governments
Participants	Setting: outpatients (home nursing service) Country: Australia Sex: 25 men, 68 women Age: mean age 78.7 (SD 10.40) years Inclusion criteria ≤ 1 week since complete healing of all lower leg ulcers Last healed wound (on the study leg) was a medically diagnosed venous leg ulcer (study ulcer) Ankle‐brachial pressure index of study leg between 0.8 and 1.2 Willingness to wear 4‐layer bandaging (4LB; Profore, Smith & Nephew, Hull, UK) for 2 weeks after healing Nursing care subsidised via government or private funding programmes Exclusion criteria Fragile skin assessed as being at high risk of tearing Peripheral vascular arterial disease (of the study leg) based on previous diagnosis or signs of distal critical ischaemias (e.g. dead toes) or clear symptoms of intermittent claudication (e.g. calf pain from exercise) Peripheral neuropathy (of the study leg) according to a 10‐g monofilament test Lymphoedema as primary cause of the last healed wound Last healed wound from malignant ulcer Ankle circumference (of the study leg) < 19 cm or calf circumference > 47 cm Terminal illness
Interventions	Group A: UK class 2 compression stocking (23–32 mmHg) below knee (Venosan 5000) Group B: UK class 3 compression stocking (34–46 mmHg) below knee (Venosan 5000) Follow‐up: at 13 weeks and 26 weeks from recruitment Co‐interventions: education about the intervention and formal assistance with stocking application and removal was offered and provided to participants and carers as needed.
Outcomes	Outcomes of interest for this review (collected) Ulcer recurrence (number of days since healing of the study wound for any wounds occurring in the same anatomical location) Adherence (use of compression stocking on > 50% of the days analysed) Other reported outcomes Number of recurring study wounds Number of other wounds occurring and recurring Nonadherence
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization lists (stratified by geographical site) had been generated using the random number function in Microsoft Excel and randomisation was conducted remotely by researchers." Comment: randomisation sequence was computer‐generated using Microsoft Excel and was conducted remotely by researchers.
Allocation concealment (selection bias)	Low risk	Quote: "All identifying product information was removed from the tags of the stockings and the butt of the tag was marked with red and blue markers as appropriate. Stockings were placed in Ziplock bags with care information provided on an unbranded card." Comment: product information was removed from the tags, and stockings were provided in bags without brand.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Nursing staff and study participants were blinded to the interventions." Comment: all personnel and study participants were blinded to the interventions, stocking brand, and level of compression.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "Analysis was undertaken non blinded." Comment: personnel involved in data analysis were non blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	All outcomes reported. Paper mentioned exclusions and missing data.
Selective reporting (reporting bias)	Low risk	Trial registration and protocol not available; however, all expected outcomes were reported.
Other bias	Unclear risk	Quote: "Department of Health, Victoria and the Angior Family Foundation who provided funding for the compression stockings and stocking applicators provided to participants of this study. Home and Community Care services provided by the organization are jointly funded by the Victorian and Australian governments." Comment: funding by Department of Health (Victoria), the Angior Family Foundation, and the Victorian and Australian governments.

Milic 2010.

Study characteristics
Methods	Study design: RCT Study grouping: no details given Ethics and informed consent: no details given Total number: 338 participants (173 in group A and 165 in group B). Data analysed for 327 people who complied with randomisation. Start date: no details given Duration of participation: 3 years End date: no details given Method of randomisation: no details given Blinding: no details given Power calculation: sample size not calculated a priori Losses to follow‐up: 11 participants withdrew during the study (8 (4.6%) in group A, 3 (1.8%) in group B) Unit of allocation: participants Source of funding: no details given
Participants	Setting: outpatients Country: Serbia Sex: 192 men, 146 women Age: average 58 years Inclusion criteria Healed venous leg ulcers Exclusion criteria Significant arterial disease Rheumatoid disease Diabetes mellitus
Interventions	Group A: UK class 3 compression stockings Group B: UK class 2 compression stockings Follow‐up: duration 3 year. Participants were instructed to wear compression stockings during the first year during day and night. In the second and the third years, participants were instructed to wear elastic stockings only during the day. Co‐interventions: new compression stockings every 4 months
Outcomes	Outcomes of interest for this review (collected) Rate of leg ulcer recurrence (not properly defined) Noncompliance (not properly defined) Other reported outcomes Noncompliance at 3 years
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Randomized into two groups were 338 patients ..." Comment: No information about sequence generation; abstract only.
Allocation concealment (selection bias)	Unclear risk	Quote: "Randomized into two groups were 338 patients ..." Comment: No information about allocation concealment; abstract only.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of intervention not possible.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "An open, prospective, randomised, single‐centre study ..." Comment: no blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Eleven patients did not comply with their randomised compression class, eight (4.6%) in class 3 and three (1.8%) in class 2." Comment: whilst not explicitly stated, 11 participants were excluded from the analysis (3.3% of randomised participants).
Selective reporting (reporting bias)	Unclear risk	Abstract only. Limited details of study methodology, although expected outcomes are reported.
Other bias	Unclear risk	Abstract only. Limited details of study methodology.

Milic 2018.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Ethics and informed consent: ethics approval and consent obtained Total number: 361 participants (186 in group A, 175 in group B). Article provides details of initial exclusions before randomisation. Start date: no details given Duration of participation: 5 years End date: no details given Method of randomisation: computer‐generated randomisation list Blinding: no details given Power calculation: sample size calculated based on a 10% expected difference in recurrences between groups, a power of 80%, and a CI of 95%. Losses to follow‐up: 53 people did not complete the study (25 lost to follow‐up, 28 did not comply with treatment). Unit of allocation: participants Source of funding: no details given
Participants	Setting: hospital clinic Country: Serbia Sex: 168 men, 168 women (336 participants included in analysis) Age: mean 59 years Inclusion criteria Age > 18 years Recently healed venous leg ulcer Exclusion criteria ABI < 0.8 Heart insufficiency with EF < 35% Pregnancy Cancer Rheumatoid arthritis Diabetes
Interventions	Group A: heel‐less open‐toed elastic class 3 compression device knitted in tubular form (25–35 mmHg) Group B: UKclass 2 knee‐high elastic stocking (18–24 mmHg) Follow‐up: every 3 months for the first 2 years, then every 6 months for the third, fourth, and fifth years. Participants completed a questionnaire regarding their compliance and experience of wearing the stockings. Each patient was instructed to wear compression stockings continuously for the first 2 years of follow‐up (both during the day and at night). In the third, fourth, and fifth years of follow‐up, patients were instructed to wear elastic stockings during the day only. Co‐interventions: education on stocking wear at every visit. Stockings changed every 4 months during the 5‐year period.
Outcomes	Outcomes of interest for this review (collected) Leg ulcer recurrence (defined as skin break on the reference limb anywhere between the malleoli and the knee. Compliance with treatment (number of participants who wore the allocated hosiery > 80% of the time during the first and second years and > 50% of the time during the third, fourth, and fifth years) Time to ulcer recurrence (Kaplan‐Meier survival; ITT) Other reported outcomes Factors that influenced the ulcer recurrence rate (Cox regression)
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Randomization was computer generated, and patients were randomised into two groups ..." Comment: randomisation sequence was computer‐generated.
Allocation concealment (selection bias)	Unclear risk	No information about allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	No details about blinding.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No details about blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Twenty‐five patients were lost during the study (5 patients died, 3 moved away to live in a different area, and 17 patients stopped coming to the clinic for regular checkups)." Comment: all outcomes were reported. Article mentioned exclusions and missing data.
Selective reporting (reporting bias)	Low risk	Quote: "Twenty‐five patients were lost during the study (5 patients died, 3 moved away to live in a different area, and 17 patients stopped coming to the clinic for regular checkups)." Comment: trial registration and protocol not available; however, all expected outcomes were reported.
Other bias	Unclear risk	Insufficient information.

Milic 2019.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Ethics and informed consent: ethics approval obtained Total number: 477 participants (149in group A 167in group B, 161 in group C). Article provided details of exclusions before randomisation. Start date: no details given Duration of participation: 10 years End date: no details given Method of randomisation: computer‐generated randomisation list Blinding: no details given Power calculation: sample size calculated a priori based on ulcer recurrence rates of ≥ 20% in group C, 30% in group B, and 40% in group A Losses to follow‐up: 668 people screened, 530 people accepted and randomised. 53 participants were lost to follow‐up and 117 did not comply with the compression treatment. Overall, 360 participants (186 men, 174 women; mean age 60 years) completed the study. Unit of allocation: participants Source of funding: no financial support for the research
Participants	Setting: hospital clinic Country: Serbia Sex: 240 men, 237 women Age: mean 59 years Inclusion criteria Recently healed venous ulcers No significant arterial disease, rheumatoid disease, or diabetes mellitus Exclusion criteria: no details given
Interventions	Group A: UK class 2 compression hosiery Group B: UK class 3 compression hosiery Group C: UK class 4 compression hosiery Follow‐up: duration 10 years Co‐interventions: no details given
Outcomes	Outcomes of interest for this review (collected) Recurrence Noncompliance Other reported outcomes: none
Notes	We analysed data from group A and group B only.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "randomization was computer generated, and patients were randomized into three groups ..." Comment: No information about sequence generation.
Allocation concealment (selection bias)	Unclear risk	No information about allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	No information about blinding.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No information about blinding.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "fifty‐three patients were lost during the study (11 patients died, seven moved away to live in a different area, and 35 patients stopped coming to the clinic for regular checkups)." Comment: all outcomes were reported. Article mentioned exclusions and missing data.
Selective reporting (reporting bias)	Low risk	Trial registration and protocol not available; however, all expected outcomes were reported.
Other bias	Unclear risk	Insufficient information.

Nelson 2006.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Ethics and informed consent: ethics committees at both hospitals approved the protocol. Total number: 300 participants (151 in group A, 149 in group B). Article provided details of initial exclusions. Start date: no details given Duration of participation: 5 years End date: no details given Method of randomisation: telephone randomisation service in the hospital Blinding: participants: no; investigators: no; outcome assessors: no Power calculation: sample size calculated assuming ulcer recurrence would be 40% in 1 group and a 15% difference in reulceration rates was clinically important. 150 people were required in each group to detect this difference at the conventional 5% level of significance (80% power). Losses to follow‐up: 50 participants did not complete the 5‐year follow‐up for the following reasons: death (n = 36), moving elsewhere (n = 9), refusal (n = 2), intractability (n = 2), and illness (n = 1). Unit of allocation: participants Source of funding: stockings were prescribed and dispensed in the NHS hospital clinic. Attendance at the clinic was free of charge, and transport was arranged if necessary. NHS prescription hosiery was free to people with a low income or of pension age (60 or 65 years). The cost for other participants was GBP 4–6 (USD 6‐10) per stocking.
Participants	Setting: district general hospital Country: Scotland Sex: 125 men, 175 women Age: mean 65.1 years (Group A) and mean 63.6 years (group B) Inclusion criteria Recently healed venous leg ulcers Exclusion criteria Diabetes ABI < 0.8 Seropositive rheumatoid arthritis
Interventions	Group 1: UK class 2 compression hosiery. All participants measured for hosiery by an orthotist. Group 2: UK class 3 compression hosiery. All participants measured for hosiery by an orthotist. Follow‐up: duration 5 years, check‐up every 4 months Co‐interventions: new hosiery every 4 months. Telephone hot‐line to leg ulcer clinic in case of problems.
Outcomes	Outcomes of interest for this review (collected) Occurrence of ulceration, defined as a skin break in the same leg, that failed to heal after 6 weeks' treatment Compliance with the treatment (number of participants who wore the allocated class of hosiery throughout the study) Time to ulcer recurrence (Kaplan‐Meier survival; ITT) Other reported outcomes: none
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Computer‐generated" (personal communication, Nelson)
Allocation concealment (selection bias)	Unclear risk	Quote: "We used telephone allocation ... During the trial follow‐up, however, we were informed that on at least one occasion the allocation concealment was not maintained by the telephone service because both the leg ulcer clinic and the telephone service were extremely busy; therefore the allocations for two patients were supplied "in order to save time"".
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of staff and participants to the intervention not possible.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcome assessor not blinded to the allocation group (personal communication, Nelson)
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Our analyses were conducted on an intention to treat basis."
Selective reporting (reporting bias)	Low risk	Trial registration and protocol not available; however, all expected outcomes were reported.
Other bias	Low risk	No additional sources of bias identified.

Vandongen 2000.

Study characteristics
Methods	Study design: RCT Study grouping: parallel group Ethics and informed consent: ethics approval and informed consent obtained Total number: 153 participants (72 in group A, 81 in group B) Start date: no details given Duration of participation: 12 months End date: no details given Method of randomisation: closed envelope system Blinding: participants: no; investigators: no; outcome assessors: not mentioned Power calculation: sample size not calculated a priori Losses to follow‐up: withdrawals, missing data, and failure to complete the 12‐month study period: 33 participants in group A and 13 participants in group B Unit of allocation: participants Source of funding: no details given
Participants	Setting: outpatients Country: Australia Sex: 36 men, 42 women (only participants who finished the 12‐month study period) Age: median 67 years (range 37–85) Inclusion criteria Healed venous leg ulcers 2 weeks before entering the study Exclusion criteria Arterial disease (ABI < 0.9) Inability to manage daily application of the stockings History of noncompliance with compression therapy
Interventions	Group A: compression stockings (EU class 3: 35–45 mmHg Venosan 2003 below‐knee graduated stockings) Group B: no compression stockings Follow‐up: participants given 2 pairs of stockings every 6 months, until 2 years. Participants given a diary to record days when the stockings were not worn and other events. Co‐interventions: no information
Outcomes	Outcomes of interest for this review (collected) Incidence of ulcer recurrence Other reported outcomes: none
Notes	12‐month follow‐up data not analysed due to attrition bias.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "Randomisation was by a closed envelope system." Comment: No description of method of sequence generation; if envelopes were shuffled, sequentially numbered, etc.
Allocation concealment (selection bias)	Unclear risk	Quote: "Randomisation was by a closed envelope system. The envelopes were only opened after all inclusion criteria had been met." Comment: security of this method is unclear from the reporting. Low risk if envelopes were sequentially numbered and a log kept; high risk if not.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Personnel and participants could not be blinded to treatment allocation.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Process of outcome assessment (for reulceration) not described. Blinding of outcome assessors not mentioned but unlikely.
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "Patients were withdrawn from the study if they were non‐compliant with the treatment ..."; "A total of 16 patients in the stocking group refused to continue wearing stockings whereas 3 patients in the no stocking group withdrew so that they could wear stockings". Comment: no reulceration data for withdrawn participants. A much larger proportion of participants withdrew from the stocking arm; their reulceration rates are likely to be high but are not known.
Selective reporting (reporting bias)	Low risk	Trial registration and protocol not available; however, all expected outcomes were reported.
Other bias	Unclear risk	Participants were part of a randomised trial of longer follow‐up, assessing the role of elastic compression stockings for preventing recurrent venous ulceration. Only those with healed ulcers who consented were randomised for this study.

ABI: ankle brachial index; CI: confidence interval; EF: ejection fraction; ITT: intention‐to‐treat; NHS: UK National Health Service; RCT: randomised controlled trial; SD: standard deviation.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Adderley 2014	Population: people with active venous leg ulcers Intervention: compression hosiery versus compression bandages in treatment of venous leg ulcers
Davies 2004	Population: people with active venous leg ulcers Intervention: surgery versus elastocompression
Franks 2000	Population: people with active venous leg ulcers Intervention: 4‐layer bandaging versus short‐stretch bandaging in treatment of venous leg ulcers
Gohel 2005	Population: people with active and healed venous leg ulcers Intervention: surgery versus elastocompression
Gohel 2018	Population: people with active venous leg ulcers Intervention: endovenous ablation
Harrison 2011	Population: people with active venous leg ulcers
Iglesias 2004	This is an observational study with regard to recurrence.
ISRCTN75319519	Population: people with active venous leg ulcers
ISRCTN95282887	Population: people with active venous leg ulcers
Junger 2002	Population: people with active venous leg ulcers
Junger 2006	Population: presence of chronic venous disease (not mentioned if presence of healed or active ulcers)
Kapp 2012	Population: people with active venous leg ulcers Intervention: compares how provision of funding for compression bandaging and medical footwear would impact on compression therapy use, wound healing, and quality of life among people with venous leg ulcers
Kelechi 2017	Population: includes CEAP C4 and C5 Intervention: compares cryotherapy plus compression versus non‐cryotherapy sham plus compression
Klonizakis 2018	Population: people with active venous leg ulcers Intervention: feasibility of a 12‐week supervised exercise programme as an adjunct therapy to compression in people with venous leg ulcers
Lewis 1976	Population: people undergoing abdominal operation
Mauck 2014	Ineligible study design
Milic 2007	Population: people with active venous leg ulcers Intervention: multilayer bandaging system with the Tubulcus (a heelless open‐toed elastic compression device knitted in tubular form) and elastic bandages versus a multilayer bandaging system with elastic bandages only
Moffatt 2003	Population: people with active venous leg ulcers
NCT02782689	Population: people with active venous leg ulcers
NCT03688841	Population: people with active venous leg ulcers
NCT05239416	Intervention: prospective randomised trial evaluated fat grafting and compression therapy versus compression therapy only
Polignano 2004	Population: people with active venous leg ulcers
Rocca 2012	Population: presence of chronic venous disease (not mentioned if presence of healed or active ulcers)
Stubbs 2004	Population: people with active venous leg ulcers Intervention: prospective randomised multicentre trial comparing surgical treatment plus compression versus compression alone
Van Gent 2015	Population: people with active venous leg ulcers Intervention: prospective randomised multicentre trial comparing surgical treatment (combined superficial and perforating vein surgery) with ambulatory compression therapy
Wilkinson 1997	Population: RCT undertaken to compare 2 4‐layer bandaging systems in people with active venous leg ulcers

CEAP: classification based on clinical class (C), etiology (E), anatomical (A) distribution of reflux and obstruction in the superficial, deep, and perforating veins, and underlying pathophysiology (P); RCT: randomised controlled trial.

Characteristics of ongoing studies [ordered by study ID]

NCT01680809.

Study name	Compliance of compression therapy in healed venous ulcerations
Methods	RCT
Participants	People with healed venous ulcer
Interventions	Compression stocking 15–20 mmHg versus compression stocking 20–30 mmHg
Outcomes	Primary outcomes: compliance of compression therapy Secondary outcomes: ulceration recurrence
Starting date	September 2012
Contact information	Nicole E Allen‐Wilson (nicole.allenwilson@uhhospitals.org)
Notes	Lack of response from authors on trial progress

RCT: randomised controlled trial.

Differences between protocol and review

• We excluded from the objectives the ideal level of compression to avoid venous leg ulcers.

• We included different types or brands of compression hosiery and different compression regimens (e.g. long stretch, short stretch, single layer) as eligible comparisons.

• We replaced the quality of life outcome by noncompliance and comfort (separately) because we considered these outcomes could provide a more objective assessment of quality of life improvements associated with compression hosiery.

• We removed the cost outcome because cost is very time‐ and country‐specific.

• We added the outcome adverse effects. It is important for clinicians and patients alike to know about any potential adverse effects, or the lack of evidence for adverse effects.

• We updated our search terms by adding the broader MeSH term for compression bandages.

• We were unable to perform sensitivity analysis based on overall risk of bias.

• We assessed the risk of bias in included studies using the tool with seven specific domains that separated the blinding of participants and personnel from blinding of outcome assessment, rather than a tool with six domains that considered the two domains of blinding together.

• We judged the overall risk of bias using all seven domains rather than only three (sequence generation, allocation concealment, and blinding of outcome assessment) as preplanned.

• Using GRADEpro GDT software, we prepared summary of findings tables to present the key information for different types of compression for patients with healed venous ulcers.

Contributions of authors

MAMS: designed the review update; co‐ordinated the review update; extracted data; checked quality of data extraction; analysed or interpreted data; undertook quality assessment; checked quality assessment; performed statistical analysis; checked quality of statistical analysis; produced the first draft of the review update; contributed to writing or editing the review update; advised on the review update; wrote to study authors/experts/companies; approved final review update prior to submission.
AN: conceived the review; performed previous work that was the foundation of the current review update; approved final review update prior to submission.
SEMBS: advised on the review update; performed previous work that was the foundation of the current review update; approved final review update prior to submission.
SGJS: designed the review update; extracted data; checked quality of data extraction; analysed or interpreted data; undertook quality assessment; checked quality assessment; performed statistical analysis; checked quality of statistical analysis; produced the first draft of the review update; contributed to writing or editing the review update; advised on the review update; wrote to study authors/experts/companies; approved final review update prior to submission.
FMJ: designed the review update; extracted data; checked quality of data extraction; analysed or interpreted data; undertook quality assessment; checked quality assessment; performed statistical analysis; checked quality of statistical analysis; produced the first draft of the review update; contributed to writing or editing the review update; advised on the review update; wrote to study authors/experts/companies; approved final review update prior to submission.

Contributions of the editorial base

Nicky Cullum (Joint Co‐ordinating Editor): edited previous versions of the review, advised on methodology, interpretation, and review content. Approved the previous review updates prior to submission.
Gill Norman (Editor): edited this version of the review, advised on methodology, interpretation and review content, until Cochrane Wounds Review Group's closure on 31 March 2023.
Gill Rizzello (Managing Editor): co‐ordinated the editorial process; advised on content and edited this version of the review, until Cochrane Wounds Review Group's closure on 31 March 2023.
Sophie Bishop (Information Specialist): updated the search and edited the search methods section for this version of the review, until Cochrane Wounds Review Group's closure on 31 March 2023.
Tom Patterson (Editorial Assistant): edited the Plain Language Summary and edited the reference sections for this version of the review, until Cochrane Wounds Review Group's closure on 31 March 2023.

Sources of support

Internal sources

• Department of Health Sciences, University of York, UK

  This project was supported by Department of Health Sciences, University of York.

• School of Healthcare, University of Leeds, UK

  This project was supported by Department of Health Sciences, University of York.

• Universidade Federal de São Paulo, Brazil

  Division of Interdisciplinary Surgery

• Hospital de Clínicas de Itajubá, Brazil

  Division of Vascular Surgery

External sources

• NHS Health Technology Assessment Programme, England, UK

  This project was supported by NHS Health Technology Assessment Programme, England.

• NIHR/Department of Health and Social Care (England), (Cochrane Wounds Group), UK

  This project was supported by NIHR/Department of Health and Social Care (England), (Cochrane Wounds Group) until 31 March 2023.

Declarations of interest

MAMS: works as a health professional.
AN: was involved in a study that is included in this review (Nelson 2006), but was not involved in study eligibility decisions, extracting data, carrying out the risk of bias assessment for, or performing GRADE assessments of this study for this updated version. AN was an Editor with Cochrane Wounds, but was not involved in the editorial process relating to this review update.
SEMBS: none known.
SGJS: works as a health professional.
FMJ: none known.

New search for studies and content updated (conclusions changed)