A systematic review of the impact of compression therapy on quality of life and pain among people with a venous leg ulcer

Abstract

Aim

To gain a greater understanding of how compression therapy affects quality of life, this systematic review appraised existing published studies measuring the impact of compression therapy on health quality of life (HRQoL), and pain, among people with venous leg ulcers (VLU).

Method

Five databases were searched, and two authors extracted data and appraised the quality of selected papers using the RevMan risk of bias tool. Due to heterogeneity in the types of compression and instruments used to evaluate HRQoL, meta‐analysis was not appropriate; thus, a narrative synthesis of findings was undertaken.

Results

Ten studies were included, 9 RCTs and one before‐after study. The studies employed nine different HRQoL tools to measure the impact of a variety of compression therapy systems, with or without an additional exercise programme, versus other compression systems or usual care, and the results are mixed. With the use of the Cardiff Cardiff Wound Impact Schedule, the SF‐8 and the SF‐12, study authors found no differences in QoL scores between the study groups. This is similar to one study using QUALYs (Iglesias et al., 2004). Conversely, for studies using EuroQol‐5D, VEINES‐QOL, SF‐36 and CIVIQ‐20 differences in QoL scores between the study groups were noted, in favour of the study intervention groups. Two further studies using QUALYs found results that favoured a two‐layer cohesive compression bandage and the TLCCB group, respectively. Results for the five studies that assessed pain are also mixed, with one study finding no difference between study groups, one finding that pain increased over the study period and three studies finding that pain reduced in the intervention groups. All studies were assessed as being at risk of bias in one or more domains.

Conclusion

Results were varied, reflecting uncertainty in determining the impact of compression therapy on quality of life and pain among people with a venous leg ulcer. The heterogeneity of the compression systems and the measures used to evaluate HRQoL make it a challenge to interpret the overall evidence. Further studies should strive for homogeneity in design, interventions and comparators to enhance both internal and external validity.

1. INTRODUCTION

Venous leg ulcers (VLU) occur on the lower region of the leg between the knee and ankle due to venous insufficiency and generally become chronic wounds. ¹ Venous valve dysfunction, deep vein occlusion and failure of the calf muscles represent conditions that cause the underlying venous insufficiency and associated venous hypertension that leads to ulceration. ² , ³ The incidence of VLU increases with age and the estimated incidence ranges between 1% and 3% in the adult population. ⁴ , ⁵ , ⁶ Although VLUs are more common in elderly patients, 22% of individuals develop their first VLUs by 40 years of age and 13% before 30 years of age, affecting their ability to work and participate in social activities. ⁷

The aetiology of a VLU is complex and thought to be a result of structural changes in the venous system of the legs, which lead to chronic venous hypertension as a result of venous insufficiency caused by a deficient calf muscle pump. ⁸ The standard clinical practice considers compression an essential component in the treatment and prevention of venous ulcer recurrence. ⁹ , ¹⁰ , ¹¹ Compression therapy represents the standard of care for conservative treatment of VLU. ¹² Compression therapy works by generating external pressure on the superficial veins and tissues, thereby assisting venous return. Improved venous return helps to reduce peripheral oedema and promotes lower limb wound healing. ¹³ Published healing rates of VLU managed with compression therapy (CT) vary widely from 40% to 95%. ¹⁴ Clinicians can provide compression therapy by three different techniques: bandage systems, stockings/hosiery or active intermittent compression devices. ⁷ A meta‐review by Patton, Avsar ¹⁵ appraised and synthesised the evidence from existing systematic reviews to understand the impact that compression therapy has on VLU healing. Twelve published systematic reviews were included, and the type of compression therapy varied between the reviews. A number of comparisons are made in this meta‐review: Compression system versus no compression; elastic compression versus inelastic compression, 4 layers versus <4‐layer bandage systems; comparison between different 4‐layer bandage systems; comparison between different elastic compression bandage systems; compression bandage versus compression stocking; and comparison between different types of inelastic compression systems. Compression therapy represents the gold standard for VLU management. The goal of any system is to deliver therapeutic compression during mobility and rest. ¹⁶ The key finding to emerge from the review is that there is moderate certainty evidence of the effect of compression bandages on the healing of VLU when compared with no compression. However, the identified reviews do not conclude which compression systems represent the most effective for healing VLU.

Research shows that a VLU profoundly impacts a person's quality of life (QoL) because a person has to cope with an open wound that can take years to heal. ¹⁷ To heal the wound, patients must endure the application of compression bandages and dressing changes once or twice weekly, a time‐intensive process. Research has identified VLU pain as a significant burden on other areas of a person's life. ¹⁸ , ¹⁹ For example, VLU‐related pain can affect a person's sleep and ability to mobilise. ²⁰ The duration and intensity of the pain and the lack of sleep could lead to people feeling down and depressed. ²¹ A clear interconnectedness exists between the physical, social and psychological aspects of enduring a VLU and its treatment. ²²

Given that compression therapy positively impacts wound healing in individuals with venous leg ulcers, a secondary outcome may be the impact that this treatment has on health quality of life (HRQoL). ²³ This is especially important owing to the increasing drive for inclusion of patient‐reported outcome measures in clinical research, of which QoL is one. ²⁴ , ²⁵ Thus, the review considers the impact of compression therapy on QoL and pain among people with a VLU. The review synthesises findings from existing published studies.

2. RESEARCH QUESTION

• Primary: What is the effect of compression therapy on the quality of life of people with a venous leg ulcer?

• Secondary: What is the effect of compression therapy on pain in people with a venous leg ulcer?

3. AIM

Using the PICO model, this systematic review aimed to appraise existing evidence that measured the impact that compression therapy has on the quality of life of people with a venous leg ulcer.

• Population: Patients with a venous leg ulcer

• Intervention: Compression therapy, however, specified by the review author(s).

• Comparison: Other intervention or none

• Outcome:

  • Primary—Quality of life as measured by a specific instrument;
  • Secondary—Pain

4. METHODS

The review team followed the standard approach advocated for systematic reviews by using the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines ²⁶ to guide the conduct and report findings. The PRISMA flow chart for this review is presented in Figure 1. The study authors pre‐registered the protocol with the International Prospective Register of Systematic Reviews (PROSPERO: CRD42023423776).

FIGURE 1.

PRISMA 2020 flow diagram for study selection. ²⁶

4.1. Criteria for considering studies for this review

4.1.1. Inclusion criteria

• Published papers, with no geographic restriction for study sites.

• Papers written in English.

• The quality of life measurement in included studies can be a primary or secondary outcome.

• Adults (aged over 16 years).

• Clinical setting of any type.

• Compression therapy, however, defined in the paper.

4.1.2. Exclusion criteria

• Non‐English papers

4.2. Electronic searches

The research team completed the search in May 2023 with guidance from a Health Sciences Librarian. The following electronic databases were searched: MEDLINE (Ovid), EMBASE (Ovid), EBM Reviews and Cochrane Library (Ovid), Cumulative Index to Nursing and Allied Health Literature (CINAHL), Web of Science. To identify further published papers, this search also included a:

• Review of reference lists of all identified papers;

• Search of the grey literature using OpenGrey (www.opengrey.eu);

• Search of conference proceedings and research reports.

The keywords used in the search were:

#1 Chronic Venous leg Ulcer OR Venous Leg Ulcer, OR Ulcers, OR leg ulcers.

#2 Quality of life OR wellbeing OR well‐being OR health‐related quality of life OR QOL.

#3 Wound care.

#4 #1 AND #2 AND #3.

#5 Pain.

#6 Patient satisfaction.

#7 Patient outcome.

#8: #6 AND #7 AND #8.

#9: #4 AND #8.

4.3. Screening

Using EndNote, the article titles were independently assessed by two team members (ZM & PA), and the abstracts of the reviews identified by the search strategy were screened for their eligibility, according to the inclusion and exclusion criteria. The full‐text version of potentially relevant studies was obtained and 2 team members (ZM & PA) independently screened these against the inclusion criteria. Consensus between the team members in relation to the studies and the data to be included was obtained, if disagreements had existed, a third team member (DP) would have acted as the arbitrator.

4.4. Data extraction

Two authors extracted data independently of one another using a purpose‐built and piloted data extraction table. Extracted data related to the review type, types of papers reviewed, sample, setting and population, intervention, control and the results.

4.5. Risk of bias/quality assessment

The quality of studies was assessed independently by two authors, without blinding to journal, or authorship, using the Cochrane Collaboration tool for assessing risk of bias. This tool addresses six specific domains, namely sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting and other issues (e.g., extreme baseline imbalance). ²⁷ We assessed the blinding and completeness of outcome data for each outcome separately. We completed a 'Risk of bias' table for each eligible study.

4.6. Data synthesis

We had planned to undertake a meta‐analysis a priori, however, because of the large variety of compression bandages and methods used across the included systematic reviews, quantitative pooling was not appropriate. Thus, a narrative synthesis was adopted.

5. RESULTS

5.1. Overview of all included studies

Figure 1 outlines the flow of articles through the reviews. Following a review of titles and abstracts from 1091 possible papers, 1052 were excluded. Then, following a review of the full papers of the remaining 14 references, four papers were rejected (Table 1). Ultimately, 10 articles met the inclusion criteria. ²³ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶

TABLE 1.

Overview of excluded studies with reasons.

Excluded studies with reasons for exclusion
Author	Study	Reason for exclusion
Franks, Moody 37	Randomised trial of cohesive short‐stretch versus four‐layer bandaging in the management of venous ulceration	Not included primary outcome
Blecken, Villavicencio 38	Comparison of elastic versus nonelastic compression in bilateral venous ulcers: A randomised trial	Not included primary outcome
Lazareth, Moffatt 39	Efficacy of two compression systems in the management of VLUs: results of a European RCT	Not included primary outcome
Douglas 40	Living with a chronic leg ulcer: an insight into patients' experiences and feelings	Relates to venous leg ulcers rather than compression bandages

5.2. Study characteristics

Table 2 provides an overview of the included studies. This review comprised 10 studies published between 2010 and 2023, 9 RCTs and one before‐after study. ²³ A total of 2333 participants were included in the studies (mean: 233; SD: 243; min 24 ³⁶ ; max 675 ²⁶ ). The majority were conducted in the UK (44%; n = 4/9) or across multi‐geographical sites including the UK and other countries (22%; n = 2/9). The mean follow‐up time for participants in all studies was 23 weeks (SD: 17 weeks, min 6 weeks ³³ ; max 12 months). ³⁴ , ³⁵

TABLE 2.

Characteristics of included studies.

Author, year	Country	Sample Size	Intervention	Comparator 1	Comparator 2
Iglesias et al., (2004) 30	UK	387 Mean age: 71.45 years; 41% male	4‐layer bandage (4LB) (n = 195)	Short‐stretch bandage (SSB) (n = 192).	N/A
Clarke‐Moloney et al., (2005) 29	Ireland	200 (no participant demographics)	4‐layer compression bandaging (n = 100)	Usual system of care (n = 100)	N/A
Moffatt et al., (2008) 24	USA, UK, Canada	81 Mean age 63 years; 58% male	Two‐layer compression bandage system (n = 39)	4‐layer system (n = 42)	N/A
Vanscheidt et al., (2012) 25	Belgium, Germany, Netherlands, UK	234 Mean age 68.4 years; 56% females	Coban 2 layer compression system (n = 117)	Short‐stretch bandage (n = 117)	N/A
Guest et al., (2015) 26	UK	675 Mean age 75 years; 57% female	Two‐layer cohesive compression bandage (TLCCB; Coban 2) (n = 250)	2‐layer compression system (TLCS; Ktwo) (n = 250	4‐layer compression system (FLCS; Profore) (n = 175)
Guest et al., (2017) 27	UK	600 Mean age 72 years; 58% female.	Two‐layer cohesive compression bandage (TLCCB; Coban 2) (n = 200)	2‐layer compression system (TLCS; KTwo) (n = 200)	4‐layer compression system (FLCS; Profore) (n = 200)
O'Brien et al., (2016) 28	Australia	63 Mean age: 71·5 years; 51.6% male	Progressive resistance exercise and walking in addition to compression. Compression type: not specified (n = 30)	Compression alone. Compression type: not specified (n = 29)	N/A
Klonizakis et al., (2018) 31	UK	39 Mean age: 63.5 years; 59% male	Exercise plus compression therapy or compression only. Compression type: not specified (n = 18)	Compression alone. Compression type: not specified (n = 21)	N/A
Berszakiewicz et al., (2021) 19	Poland	30 (only those with active VU) Mean age: 58.53 years 63% female	Ready‐made 2‐in‐1 compression system, with a liner worn for 24 h	N/A	N/A
Rerkasem et al., (2023) 32	Thailand	24 Mean age: 65.6 years; 54.2% males	Compression plus progressive tailored exercise training. Compression type: not specified (n = 12)	Compression alone. Compression type: not specified (n = 12)	N/A

5.3. Interventions and comparisons

The type of compression therapy varied among the studies. One study compared 4‐layer bandage with a short‐stretch bandage, ³⁴ one compared a 2‐layer compression bandage system with a 4‐layer system, ²⁸ one compared a 2‐layer compression bandage system with a short‐stretch bandage ²⁹ and one study compared a 4‐layer bandage system with the usual system of care (no compression). ³³ Two studies included 3 groups of participants, comparing a 2‐layer cohesive compression bandage (TLCCB), to a different 2‐layer compression system (TLCS), and a 4‐layer compression system (FLCS). ²⁶ , ²⁷ Three studies compared compression and exercise, to compression alone. ²⁸ , ³¹ , ³² Finally, Berszakiewicz, Kasperczyk ²³ conducted a pre‐post test study, before and after the use of a ready‐made 2‐in‐1 compression system.

5.4. Outcomes measured

Five studies ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ reported quality of life, and five studies reported quality of life and pain. ¹⁹ , ²⁹ , ³⁰ , ³¹ , ³²

5.5. Results for the outcome: Health‐related quality of life

5.5.1. Health‐related quality of life instruments employed

Table 3 provides an overview of the HRQoL instruments employed within the studies including the results for these measures. The studies used nine HRQoL instruments, including the Cardiff Wound Impact Scale ²⁴ , ²⁵ (scores from 0 to 100; higher scores indicate a higher quality of life); quality‐adjusted life years ²⁶ , ²⁷ (one quality‐adjusted life year is equal to 1 year of life in perfect health); the SF‐8 ²⁸ (scores range from 0 to 100 with higher scores indicating better health status); the EuroQol‐5D³¹ (scores range from 0 to 100 with higher scores indicating better health status); VEINES‐QOL ³¹ (higher scores are better); the SF‐36 ¹⁹ , ²⁹ (scores range from 0 to 100 with higher scores indicating better health status); the CIVIQ‐14 ³² (scores range from 0 to 100 with higher scores indicating better health status); the CIVIQ‐20 ¹⁹ , ²⁹ (items on are scored from 1 to 5, with a low score corresponding to greater patient comfort); and the SF‐12 ³⁰ (scores range from 0 to 100, with higher scores indicating better physical and mental health functioning) (see Table 3).

TABLE 3.

Quality of life measurement tools and results.

Measurement tool	Included study in SR	Result
Cardiff wound impact schedule (Higher score is better)	Moffatt et al., (2008) 24	Changes in overall HRQoL scores during the pre‐crossover period (0–4 week) Overall HRQoL: Coban 2 Layer: 0.6 ± 2.0; Profore: 0.2 ± 1.8; p = 0.860 Changes in overall HRQoL scores during the post‐crossover period (4–8 weeks) Overall HRQoL: Coban 2 Layer: 0.6 ± 1.2; Profore: 0.3 ± 1.9; p = 0.494
	Vanscheidt et al., (2012) 25	3 months of follow‐up. Coban 2 Layer Compression System: Mean:1.060; Short‐Stretch Bandage: Mean: 0.820; p = 0.46
Quality‐adjusted life years	Guest et al., (2015) 26	6 months of follow‐up Two‐layer cohesive compression bandage group (0.374 quality‐adjusted life years (QALYs) per patient), two‐layer compression system (0.368 QALYs per patient) and four‐layer compression system (0.353 QALYs per patient). p = 0 < 00001 in favour of the two‐layer cohesive compression bandage
	Guest et al., (2017) 27	6 months of follow‐up TLCCB group (0.413 QALYs per patient), TLCS (0.404 QALYs per patient), FLCS group (0.396 QALYs per patient). p = 0 < 00001 in favour of the TLCCB group
	Iglesias et al., (2004) 30	4th Quarter QUALYS (12 months) Intervention: 0.2 (0.06); Control: 0.2 (0.06)
SF‐8 (Higher score is better)	O'Brien et al., (2016) 28	Physical and Mental Component Summary Scale scores: 12 weeks (higher sore is better) Physical Component Summary Scale, (usual care group: 43 (±8.9); intervention group: 46 (±10.2) Mental Component Summary Scale (usual care group: 51 (32–66); intervention group: 53 (16–65)
EuroQol‐5D (Higher score is better)	Klonizakis et al., (2018) 31	12‐month follow‐up Overall score: Intervention: Mean: 75.53 ± 20.37; control: 56.20 ± 27.58
VEINES‐QOL (Higher score is better)	Klonizakis et al., (2018) 31	12‐month follow‐up Overall score: Intervention: 67.23 ± 29.86, Control group: 52.46 ± 34.81
SF‐36 (Higher score is better)	Clarke‐Moloney et al., (2005) 29	6‐week follow‐up Statistically significant differences in the physical function and the role physical components of the SF‐36 in favour of the 4‐layer bandage group: Physical function: median (IQR): intervention 70: 45–85; control: 50:25–80; p = 0.001; Role physical: median (IQR): intervention: 100 (0–100); control: 25 (0–100); p = 0.006, All other elements there were no different between the study groups.
	Berszakiewicz et al., 2021 19	6‐month follow‐up; results presented for the CEAP group 6 (active ulcer). Before compression/after compression Physical functioning: 21.33 (±17.36)/78.99 (±12.82) Role physical: 10.62 (±13.24)/ 81.25 (±19.14) General health: 1.00 (±3.8)/66.43 (±13.99) Vitality: 7.29 (±9.99)/64.68 (±10.8) Social functioning: 0.41 (±2.28)/72.91 (±15.77) Role emotional: 14.16 (±15.18)/82.22 (±18.66) Mental health: 9.5 (±10.2)/73.9 (±9.61)
CIVIQ‐14 (Higher score is better)	Rerkasem et al., (2023) 32	12‐week follow‐up Global: Control: 73.0 (3.8); Intervention: 84.7 (3.8)
CIVIQ‐20 (lower score is better)	Clarke‐Moloney et al., (2005) 29	6‐week follow‐up Global: Control 28.8 (18.8–43. 8); 4LB: 18.8 (12.5–31.3); p = 0.006
	Berszakiewicz et al., (2021) 19	6‐month follow‐up; results presented for the CEAP group 6 (active ulcer). Before compression/after compression Physical dimension: 17.3 (±1.7)/7.83 (±2.18) Social dimension: 13.63 (±1.06)/5.66 (±1.88) Psychological dimension: 40.33 (±4.16)/15.9 (±3.36)
SF‐12 (Higher score is better)	Iglesias et al., (2004) 30	4th quarter follow‐up (12 months) Physical component: 4‐layer bandage: Mean: 35.0 (8.03); SSB: 34.2 (7.60) Mental component 4‐later bandage: Mean: 51.7 (11.76); SSB: 49.1 (11.43)

5.5.2. Health‐related quality of life results

Results are presented here according to the quality of life instrument employed in the study (see Table 3).

6. CARDIFF WOUND IMPACT SCHEDULE

Two studies employed the Cardiff Wound Impact Schedule. ²⁴ , ²⁵ Moffatt, Edwards, ²⁸ evaluating four‐layer bandage versus 2 layer bandage, found no statistically significant difference in overall quality of life scores for both the pre‐ and post‐crossover periods (pre: 2 Layer: 0.6 ± 2.0; 4 layer: 0.2 ± 1.8; p = 0.860; post: Coban 2 Layer: 0.6 ± 1.2; Profore: 0.3 ± 1.9; p = 0.494). In Vanscheidt, ²⁹ at 3 months of follow‐up, for the 2 Layer compression system: the overall mean score was: 1.060; and for the short‐stretch bandage group, the overall mean score was: 0.820; p = 0.46, indicating no statistical difference between the study groups.

7. QUALITY‐ADJUSTED LIFE YEARS

Three studies employed measurement using quality‐adjusted life years (QUALYs). ²⁶ , ²⁷ , ³⁰ Iglesias, Nelson ³⁴ found no difference in QUALYs between the study groups (4‐layer bandage versus short‐stretch bandage) (4th Quarter QUALYS: Intervention: mean: 0.2; SD: 0.06; Control: mean: 0.2; SD: 0.06). Conversely, the two studies of Guest, Gerrish, ³⁰ Guest, Fuller, ³¹ found a statistically significant difference in QUALYs in favour of a two‐layer cohesive compression bandage (Guest, Gerrish ³⁰ : TLCCB: 0.374 QALYs per patient, TLCS: 0.368 QALYs per patient; FLCS 0.353 QALYs per patient; p = 0 < 00001; Guest, Fuller ³¹ : TLCCB group: 0.413 QALYs per patient; TLCS: 0.404 QALYs per patient; FLCS group: 0.396 QALYs per patient; p = 0 < 00001).

8. SF‐8

One study employed the SF‐8 instrument. ²⁸ In this study, the scores were very similar among the study groups (progressive resistance exercise and walking in addition to compression, versus compression alone), in terms of physical summary scale scores (usual care group: 43 [±8.9]; intervention group: 46 [±10.2]), and mental component summary scale scores (usual care group: 51 [32–66]; intervention group: 53 [16–65]). ²⁸

9. EuroQol‐5D

One study employed the EuroQol‐5D. ³¹ In this study, exercise plus compression therapy was compared with compression only. At 12 months, the score was higher in the intervention group (Intervention: mean: 75.53 ± 20.37; control: 56.20 ± 27.58).

10. VEINES‐QOL

Klonizakis, Tew ³⁵ also employed the VEINES‐QOL, and at 12‐month follow‐up, there was a higher score in the intervention group (intervention: 67.23 ± 29.86, control group: 52.46 ± 34.81).

11. SF‐36

Two studies employed the SF‐36. ¹⁹ , ²⁹ Clarke‐Moloney, O'Brien ³³ found statistically significant differences in the physical function and the role physical components of the SF‐36 in favour of the 4‐layer bandage group (physical function: median (IQR): intervention 70: 45–85; control: 50:25–80; p = 0.001; role physical: 100 (0–100); 25 (0–100); p = 0.006), all other elements were reported as not being a statistically significant difference between the study groups. In Berszakiewicz, Kasperczyk, ²³ we report results for the CEAP group 6 (active ulcer) before the use of compression and after the use of compression at 6‐month follow‐up. All domains of the SF‐36 were much higher after the use of compression as follows: physical functioning: 21.33 (±17.36)/78.99 (±12.82); role physical: 10.62 (±13.24)/81.25 (±19.14); general health: 1.00 (±3.8)/66.43 (±13.99); vitality: 7.29 (±9.99)/64.68 (±10.8); social functioning: 0.41 (±2.28)/72.91 (±15.77); role emotional: 14.16 (±15.18)/82.22 (±18.66); and mental health: 9.5 (±10.2)/73.9 (±9.61).

12. CIVIQ‐14

One study employed the CIVIQ‐14, where higher scores indicate a better health status. ¹⁹ At week 12, the global score for the control group (compression alone) was 73.0 (SE: 3.8), whereas for the intervention group (exercise and compression), the score was 84.7 (SE: 3.8).

13. CIVIQ‐20

Clarke‐Moloney, O'Brien, ³³ Berszakiewicz, Kasperczyk ²³ also employed the CIVIQ‐20, to measure HRQoL. With this instrument, lower scores are better. For Clarke‐Moloney et al., (2005), at 6‐week follow‐up, the Global: CIVIQ‐20 score in the control group was 28.8 (18.8–43. 8) and in the 4LB was 18.8 (12.5–31.3); p = 0.006. In Berszakiewicz et al., (2021), at 6‐month follow‐up, results following compression are much lower: physical dimension: 17.3 (±1.7) versus 7.83 (±2.18); social dimension: 13.63 (±1.06) versus 5.66 (±1.88); and psychological dimension: 40.33 (±4.16) vs 15.9 (±3.36).

14. SF‐12

One study employed the SF‐12. ³⁰ In this study, results are very similar between the groups (4‐layer bandage, versus short‐stretch bandage), physical component: Mean: 4‐layer bandage: 35.0 (8.03); SSB: 34.2 (7.60), Mean: mental component 4‐later bandage: 51.7 (11.76); SSB: 49.1 (11.43).

14.1. Results for the outcome: Pain

Five studies reported the outcome, of pain. Iglesias, Nelson ³⁴ found that reported levels of pain increased over the study period. Whereas, Clark Maloney ²⁹ reported no statistically significant difference in bodily pain between the study groups (Median (IQR): 4LBG: 84 (61–100); usual care: 72 (51–100); p = 0.840). Conversely, 3 studies found that pain was reduced in the intervention groups, Klonizakis, Tew ³⁵ intervention group: 7.9 ± 22.8; control group: 30.5 ± 36.6; Berszakiewicz, Kasperczyk, ²³ SF‐36: bodily pain, before compression: 0.00 (±0.00); after compression: 66.56 (±13.26), CIVIQ‐20: pain dimension: before compression: 19.73 (±0.69); after compression: 8.6 (±1.61), and Rerkasem, Nantakool ³⁶ : intervention 92.9 (SE: 4.9); control: 89.1 (SE: 4.9) (see Table 4).

TABLE 4.

Pain measurement results.

Author/Year	Result for pain assessment
Iglesias et al., (2004) 30	4th quarter follow‐up (12 months): SF‐12: Pain increased across both groups over the study period.
Clarke‐Moloney et al., (2005) 29	6‐week follow‐up: SF‐36 (Higher score is better): Bodily pain: Intervention: 84 (61–100); control: 72 (51–100); p = 0.840 CIVIQ (Lower score is better): Pain: Intervention: 18.8 (6.3–37.5): control: 31.3 (18.8–43.8); p = 0.140
Klonizakis et al., (2018) 31	12‐month follow‐up: (Higher score is better) Unclear how pain was measured: Pain scores: Intervention: 7·9 ± 22·8; Control: 30·5 ± 36·6;
Berszakiewicz et al., (2021) 19	6‐month follow‐up; results presented for the CEAP group 6 (active ulcer). SF‐36 (Higher score is better): Bodily pain: Before compression: 0.00 (±0.00); after compression: 66.56 (±13.26) CIVIQ‐20 (lower score is better): Pain dimension: Before compression: 19.73 (±0.69); after compression: 8.6 (±1.61)
Rerkasem et al., (2023) 32	12‐week follow‐up CIVIQ‐14 (higher score is better): Pain dimension: Intervention 92.9 (SE: 4.9); Control: 89.1 (SE: 4.9)

14.2. Risk of bias

The risk of bias in the included studies is summarised in Figures 2 and 3. Two review authors independently assessed the risk of bias for each study and resolved any disagreements through consensus. For all included studies, the Cochrane tool for assessing the risk of bias was used. ²⁷

FIGURE 2.

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

FIGURE 3.

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

14.2.1. Selection bias

Generation of the randomisation sequence

Moffatt, Edwards, ²⁸ Guest, Gerrish, ³⁰ Guest, Fuller, ³¹ O'Brien, ²⁸ Clarke‐Moloney, O'Brien, ³³ Iglesias, Nelson, ³⁴ Rerkasem, Nantakool ³⁶ used a computer‐generated randomisation schedule with permutated blocks of random sizes, therefore have a low risk of bias in this domain. Vanscheidt ²⁹ ; Klonizakis, Tew ³⁵ did not provide information about the generation of randomisation sequence, and therefore, these studies were deemed to be at unclear risk of bias in this domain. For the remaining study, the methodology of the study was a before‐after design, and randomisation was not employed; therefore, this study was judged to be at high risk of bias in this domain. ¹⁹

Allocation concealment

Moffatt, Edwards, ²⁸ Guest, Gerrish, ³⁰ Guest, Fuller, ³¹ O'Brien, ²⁸ Clarke‐Moloney, O'Brien, ³³ Iglesias, Nelson, ³⁴ Rerkasem, Nantakool ³⁶ indicated that allocation concealment was used in their study. Meanwhile, Vanscheidt ²⁹ ; Klonizakis, Tew ³⁵ did not provide information about the generation of randomisation sequence and therefore were at unclear risk of bias. For the remaining study, the methodology of the study was a before‐after design, and randomisation was not employed; therefore, this study was judged to be at high risk of bias in this domain. ¹⁹

14.2.2. Performance bias

Blinding participants and personnel

All studies were at high risk of bias in this domain because it was impossible for participants and nurses to be blinded to the study allocation due to the nature of the interventions under evaluation within the studies, which were clearly different from each other. ¹⁹ , ²⁴ , ³²

14.2.3. Detection bias

Blinding of outcome assessment

Only one study was at low risk of detection bias as they used independent reviewers to assess the study outcomes. ³⁰ Eight of the studies did not mention blinding of outcome assessment ¹⁹ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³¹ and were therefore assessed as having an unclear risk of bias in this domain. Rerkasem, Nantakool ³⁶ did not blind outcome assessors and were therefore judged to be at high risk of bias in this domain.

14.2.4. Attrition bias

Incomplete outcome data

Nine of the studies have a low risk of attrition bias since the number of participants was retained from the beginning of the study or included in the analysis by intention‐to‐treat studies. Vanscheidt ²⁹ had an unclear risk of bias in this domain as insufficient information was provided permitting understanding of whether all participants were included in the final analysis.

14.2.5. Reporting bias

Selective reporting

Most of the studies have a low risk of selective reporting bias because all outcomes were reported. Clarke‐Moloney, O'Brien ³³ had not registered the protocol on a clinical trials registry, and insufficient information was provided to adequately judge this risk of bias. Therefore, this study was at unclear risk of bias in this domain.

14.2.6. Other bias

Most of the studies were at low risk of bias in this domain, as other potential sources of bias were not identified. The remaining study was judged as unclear risk of bias in this domain due to the study funding source. ²⁹

15. DISCUSSION

We set out to determine the impact of compression therapy on the HRQoL of people with a VLU by synthesizing the data from existing published studies. This review comprises 10 studies published between 2010 and 2023 with a total of 2333 participants included in the studies (mean: 233; SD: 243). The majority were conducted in the UK (44%; n = 4/9) or across multi‐geographical sites including the UK and other countries (22%; n = 2/9). The type of compression therapy varied among the studies and included 4‐layer compared with short‐stretch, 2‐layer compared with 4‐layer, 2‐layer compared with short‐stretch, 4‐layer compared with no compression, compression and exercise compared with compression alone, and before and after use of compression. Two studies included 3 groups of participants, comparing one 2‐layer with a different 2‐layer, and with a 4‐layer system.

Results for HRQoL and pain measurement are mixed. With the use of the Cardiff Cardiff Wound Impact Schedule, the SF‐8 and the SF‐12, study authors ²⁴ , ²⁵ , ²⁸ , ³⁰ found no differences in QoL scores between the study groups. This is similar to one study using QUALYs. ³⁰ Conversely, authors using EuroQol‐5D, VEINES‐QOL, SF‐36, CIVIQ‐14 and CIVIQ‐20, ¹⁹ , ²⁹ , ³¹ differences in QoL scores were noted, in favour of the study intervention groups. Two authors ²⁶ , ²⁷ using QUALYs found results in favour of the two‐layer cohesive compression bandage and the TLCCB group, respectively. In terms of pain, one study found no difference between study groups, one found that pain increased over the study period and 3 studies found that pain reduced in favour of the intervention groups.

The challenges in interpreting the data here are that there are a wide variety of HRQoL tools used and the compression systems employed are also not homogenous. However, O'Meara, Cullum ⁹ have shown that regardless of the compression system employed, venous leg ulcer healing is better when compression is used versus no compression. Thus, for the studies of Clarke‐Moloney, O'Brien ³³ and Berszakiewicz, Kasperczyk, ²³ the findings of improved quality of life when compression is used, versus no compression, are unsurprising. O'Meara, Cullum ⁹ also showed, from one large RCT, that single‐component compression bandage systems were less effective than multi‐component compression for complete healing at 6 months. Despite this, the studies of Vanscheidt ²⁹ and Iglesias, Nelson, ³⁴ though both evaluating a 2 layer compression system against a short‐stretch bandage, did not find any difference in HRQoL scores between the study groups. Furthermore, in Iglesias, Nelson, ³⁴ the pain scores increased during the follow‐up period across both study groups.

Some conflicting results were also seen for studies comparing exercise and compression, to compression alone, with O'Brien ²⁸ finding no differences in QoL scores between the study groups. However, both Klonizakis, Tew ³⁵ and Rerkasem, Nantakool ³⁶ found improvements in HRQoL in favour of the exercise and compression groups. Furthermore, both authors, Klonizakis, Tew ³⁵ and Rerkasem, Nantakool, ³⁶ report improvements in pain scores also in favour of the exercise and compression groups. The follow‐up time for O'Brien ²⁸ and Rerkasem, Nantakool ³⁶ are the same, 12 weeks, whereas Klonizakis, Tew ³⁵ followed up participants for 1 year. However, overall, the sample sizes are very small 63, ²⁸ 39 ³¹ and 24. ³² Thus, it is reasonable to suggest that these studies may be underpowered ⁴¹ ; thus, results should be interpreted with caution.

The risk of bias was assessed according to six domains for all studies: sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting and other issues. All included studies were at risk of bias in one or more domains. Interpretations and thus, conclusions drawn on the effects of the interventions should be made against the background of these risk of bias findings. Blinding was poorly reported, with incomplete blinding of investigators, participants, outcome assessors and the data analyst, in most studies. Lack of blinding can introduce bias, particularly when outcomes are subjective, and may lead to potential overestimation of the effect of the intervention, resulting in a bias in favour of the intervention. ⁴² However, it is essential to acknowledge the inherent challenges in achieving blinding in wound care studies, particularly when the intervention is visibly apparent to all involved parties, including participants and caregivers. Despite these difficulties, it remains crucial to consider the potential impact of the lack of blinding on the study results. Considering the collective risk of bias findings across all domains, it is prudent to interpret the effects of the interventions cautiously. The identified limitations highlight the need for more robust research to provide greater clarity on the topic. By emphasizing the issue of lack of blinding as a primary concern, along with the other areas of bias, it becomes evident that the results should be approached with caution. Furthermore, conducting rigorous studies in the future would significantly contribute to enhancing our understanding of this matter.

16. CONCLUSION

Quality of life represents an important outcome for patients suffering from VLUs. This review aimed to appraise existing studies that evaluated the impact of compression therapy on HRQoL and pain and 10 studies met the inclusion criteria. The type of compression therapy employed varied, as did the instruments used to measure HRQoL. Results were mixed, reflecting uncertainty in determining the impact of compression therapy on quality of life and pain among people with a VLU. The heterogeneity of the compression systems and the measures used to evaluate HRQoL make it a challenge to interpret the overall evidence. Further studies should strive for homogeneity in design, interventions and comparators, to enhance both internal and external validity.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Patton D, Avsar P, Sayeh A, et al. A systematic review of the impact of compression therapy on quality of life and pain among people with a venous leg ulcer. Int Wound J. 2024;21(3):e14816. doi: 10.1111/iwj.14816

DATA AVAILABILITY STATEMENT

Data sharing not applicable to this article as this is a systematic review.