Effects of physical activity as an adjunct treatment on healing outcomes and recurrence of venous leg ulcers: A scoping review

Abstract Healing time is protracted and ulcer recurrence is common in patients with venous leg ulcers. Although compression is the mainstay treatment, many patients do not heal timely. Physical activity may be a clinically effective adjunct treatment to compression to improve healing outcomes. This scoping review provides a broad overview of the effect of physical activity as an adjunct treatment to compression on wound healing and recurrence. We followed the six‐step framework developed by Arksey and O'Malley. We searched electronic databases and trial registration websites for relevant studies and ongoing trials. Two authors independently screened and selected articles. Findings were presented in a descriptive statistical narrative summary. We consulted and presented our findings to the wound consumer group to ensure the relevance of our study. Physical activity interventions in 12 out of the 16 eligible studies consisted of only one component, eight studies were resistance exercises, three studies reported ankle and/or foot range of motion exercises, and one study reported aerobic/walking exercises. The remaining four studies involved multicomponent exercise interventions. Resistance exercise combined with ankle and/or foot range of motion exercise minimised ulcer size on day 12 (intervention group: 4.55 ± 1.14 cm2 vs. control group: 7.43 ± 0.56 cm2) and improved calf muscle pump performance on day 8 (ejection fraction: 40%–65%; residual volume fraction: 56%–40%). We identified one study that reported ulcer recurrence rate with no clinical difference in the intervention group versus the control group (i.e., 12% in intervention vs. 5% in control). Our review identified that resistance exercise was the most common type of physical activity intervention trialled in the published literature. Resistance exercise combined with ankle and/or foot range of motion exercise appears to be effective adjunct treatments; however, the overall evidence is still relatively weak as most programmes had a short intervention period which limited clinical outcomes.

relatively weak as most programmes had a short intervention period which limited clinical outcomes.

K E Y W O R D S
adjunct treatment, compression, healing, physical activity, recurrence, venous leg ulcer 1 | BACKGROUND Venous leg ulceration (VLU) is commonly caused by chronic venous insufficiency which causes hypertension in the lower limbs. 1 The prevalence of VLU is estimated to be 1% of adults worldwide 2,3 ; however, this figure rises with age, reaching 1.69%-2% among the population aged 65 and over. 4,5 Current best clinical practice is application and adherence to compression bandages or stockings to reduce the hydrostatic pressure in the lower legs and aid venous return. 6,7 Healing rates vary from 17%-67% and the healing period ranges from 45 to 112 days. [8][9][10] Further, up to one third of VLUs reoccur when treated with compression therapy alone. 11 The recurrence rate in the first three months post healing has been reported at 36%, 12 and more than one in four patients experience multiple episodes of recurrent ulceration and recurrence in their lifetime. 13 VLUs cause physical pain, discomfort and mobility issues. 14,15 Repeated cycles of VLU healing and recurrence cause emotional distress and lead to poor quality of life (QoL). 16,17 Adjunct physical activity (PA) treatments, such as exercise, may improve healing rates and reduce VLU recurrence when used in conjunction with compression therapy. 18,19 Deterioration of calf muscle pump function in patients with VLUs is reported to be associated with delayed healing. 20 Past evidence identified that PA stimulates the calf muscle pump function, and this increases venous pressure, promotes blood return and subsequently improves ulcer healing. 21 Previous reviews have examined various types of PA interventions as adjunct treatments to compression therapy for VLU management, including (progressive) resistance exercise, aerobic exercise, walking and ankle exercise. 18,22,23 Although these reviews indicate that PA interventions improve calf muscle pump function, it does not define which type of PA may increase calf muscle pump function. 18,22,23 One review suggests that progressive resistance and aerobic exercise would improve ulcer healing. 18 This review, however, does not clarify how the exercise intervention improves ulcer healing, whether it accelerates ulcer healing or reduces ulcer size. 18 The current clinical practice guidelines also state that PA interventions aim to enhance calf muscle strength, and recommend to incorporate exercise into VLUs management plan. [24][25][26][27] Nevertheless, the type of PA intervention improving ulcer healing remains unknown, and the evidence base for this recommendation, as acknowledged by the guideline authors, is very limited. [24][25][26][27] We sought to scope and synthesise available evidence to better understand the types of PA that have been studied, as well as the effects of PA as an adjunct treatment to complement compression therapy in people with VLUs.

| METHODS
This scoping review was guided by the six-step framework developed by Arksey and O'Malley 28 and updated by Levac et al. 29 and the Joanna Briggs Institute (JBI). 30 The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 31 extension for scoping reviews checklist was also followed to ensure all essential components have been covered. The protocol for this review has been published in July 2021. 32

| Stage 1: Identifying the research question
The scoping review aimed to provide a comprehensive overview of the effect of PA as an adjunct treatment to compression therapy on the healing and recurrence of VLUs. To address this aim, the following questions were identified.

| Stage 3: Study selection
Citations from each database were uploaded into EndNote, deduplicated using default settings and exported to Covidence 33 for screening. Two review authors independently screened the titles and abstracts of all identified studies and rated them as exclude, include or unclear. All citations rated as include or unclear progressed to full text review by two independent review authors. Any disagreements at any stage of screening were resolved via discussion and mutual agreement or via assistance of a third independent assessor.

| Stage 4: Charting the data
Data from each included study were extracted by one author independently using data extraction tables created and piloted prior to the study by two authors. All extracted data were checked for accuracy by a second author, with any discrepancies resolved via discussion. Where available, the following data were extracted from each included study: • Characteristics of studies

• Population characteristics
• PA/exercise interventions characteristics • Primary outcomes, including time to heal, proportion of ulcers healed during the trial period, rate of changes in the area of the ulcer during the trial period and incidence of recurrence of healed VLUs • Secondary outcomes, including changes in calf muscle pump function, QoL, pain, adverse events and economic outcomes. Adverse events were defined as any unfavourable sign, symptom or occurrence that occurred during treatment that may or may not have been related to treatment. 34 These included, but were not limited to, increased exudate from ulcer sites, infection or hospitalisation.

| Stage 5: Collating, summarising and reporting results
The findings of this review were reported via descriptive analysis and a narrative summary of findings using text and tabulations. We reported the characteristics of the PA interventions against the Consensus on Exercise Reporting Template (CERT) criteria. 35 This instrument details the core elements that must be provided in order to bearing heel raises) aiming to improve aspects of strength.
2. Aerobic exercise/walking: Aerobic exercise included cycling and walking interventions that involved a clearly defined frequency, intensity, type and/or timing (duration) that aimed to enhance blood circulation by increasing one's heart rate. 36 In contrast, walking interventions that involved daily step count targets (e.g., 10,000 steps/day goal) but lacked details regarding bout length (i.e., how long participants were to walk for at any given time) were classified as 'walking'.  This suggestion aligned with the research questions of this review.
We presented a summary of the review findings to the wound consumer advisory group members, and asked the group to provide their comments in regards to clarity of the included information.
We also asked the group to identify any additional issues relating to PA as adjunct treatment to VLU management that were not addressed in the review findings to ensure the relevance of our findings to this group.

| Study characteristics
Detailed characteristics of each eligible study are presented in the Supporting Information S3. Included studies were published between March 1999 and February 2021, with 71% (n = 12) published after 2012. Thirteen studies were randomised control trials (RCTs), two were single-armed cohort studies, one was a prospective controlled study.
Intervention duration ranged from 7 days to 52 weeks, with 12 weeks being the most common (n = 10). The sample size of RCTs varied from 11 participants to 224 participants per study.

| Participants
Most participants in the included studies were over 60 years old and 330 of 522 (63.2%) were female. Five studies reported the body mass index (BMI), and the mean BMI of participants were 30 kg/m 2 or above. 22,24,[37][38][39] More than 80% (n = 319) of participants had VLUs for over 3 months, and the mean/median size of the ulcers in most participants was less than 10 cm 2. A summary of exercise components involved in PA/exercise interventions in eligible studies is outlined in Table 2.

T A B L E 2 A summary of exercise components involved in PA/exercise interventions in eligible studies
Author ( Participants were asked to take 10,000 steps per day More than half (n = 11) of PA/exercise programmes were conducted in the home environment without supervision. 37,38,41,42,[44][45][46][47][48][49]52 Three studies utilised a small group format (n = 4 participants per group) that also involved direct supervision by a qualified exercise physiologist. 43,47,51 A summary of PA interventions is presented in Warm up: 5 min low-intensity treadmill walking, cycling or both Cool down: 5 min low-intensity treadmill walking or cycling; static stretch to the point of mild discomfort for 60 s The exercise session comprised a combination of resistance, aerobic and flexibility exercises. The intensity of each exercise was increased on individual basis once the initial goal has been achieved Aerobic exercises: participants were asked to perform either treadmill walking, cycling or both for 30 min. The speed and incline level of treadmill and resistance of bike were increased via progression Resistance exercises: four exercises were involved, two exercises targeting the thigh and hip muscles (i.e., partial squats, chair sit-to-stand exercise) and two exercises targeting the calf muscle (i.e., standing calf raise). Participants performed it for 15 min, with/without using dumbbells and stability balls. The weight or type of exercise were increased/changed via progression Participants were required to perform the exercise for 10-15 reps and instruments, and air plethysmography) to measure calf muscle pump function. 38,42,43,47 Five studies used visual analogue scales or numerical rating scales (e.g., 0 or 1 corresponding to no pain and 10 corresponding to the worst possible pain). 37,41,48,49,52 Five studies reported QoL outcomes. 37,44,46,49,51 Three of these used generic tools (i.e., SF-18, SF-12, FLQAw), 44 For interventions involving foot/ankle range of motion exercise components, no data on healing outcomes were reported.

| Proportion of ulcer healed
For interventions involving resistance exercise components, five studies measured the proportion of ulcer healed at week 12 and month 12. 37,38,42,44,51 In comparison to control groups, a higher proportion of participants in intervention groups with ulcers healed at week 12 (intervention group: 67% vs. control group: 41%; intervention group: 50% vs. control group: 40%; intervention group:77% vs. control group: 53%) and month 12 (intervention group: 83% vs. control group: 60%). 37,38,44,51 For interventions involving aerobic exercise/walking components, three studies reported the proportion of ulcers healed. 44,48,51 The results of three studies were similar, intervention groups had higher proportion of ulcer healed than control groups at week 12 and month 12. 44,48,51 No data were reported for interventions involving foot/ankle range of motion exercise components.

| Rate of changes in the area of the ulcer
For interventions involving resistance exercise components, seven studies reported rate of changes in the area of the ulcer at week 12 and month 12. 37-40,42,50,51 Two studies reported the mean ulcer size in intervention groups were smaller than control groups at the end of the trials on day 12 (intervention group: 4.55 ± 1.14 vs. control group: 7.43 ± 0.56 cm 2 ; p ≤ 0.05) and week 12 (intervention group: 1.6 ± 0.62 cm 3 vs. control group: 2.39 ± 0.75 cm 3 ; p = 0.004). 40,50 For interventions involving aerobic exercise/walking components, one study reported rate of changes in the area of the ulcer at month 12. Median ulcer size in both groups was similar. 51 For interventions involving foot/ankle range of motion exercise components, four studies reported rate of changes in ulcer area. 39,40,45,49 At the end of trial period, the mean ulcer size in intervention groups were smaller than control groups in all four studies. 39 This study involved both resistance and aerobic exercise/walking components and found two participants (12%) in the intervention group reported ulcer recurrence in comparison with one participant (5%) in the control group. 51

| Calf muscle pump function
Four studies examined the effects of resistance exercise on calf muscle pump function. 38,42,43,47 The improvements were found with respect to ejection fraction and residual volume fraction in intervention groups in all four studies. 38,42,43,47 The change was noticed as early as on day 8, when the ejection fraction increased from 40% to 65% (p = 0.006), and the residual volume fraction decreased from 56% to 40% (p = 0.008). 43 No data were reported for interventions involving aerobic exercise/walking components or foot/ankle range of motion exercise components.

| Quality of life
Two studies reported participants' QoL at week 12 44 and month 12 51 after receiving PA interventions, respectively. These two studies involved both resistance and aerobic/walking exercise components. 44,51 One study used a generic instrument (SF-8) to measure QoL score, 44 while the other used both generic (EQ-5D-5L) and disease specific QoL instruments (VEINES-QOL). 51 Both studies reported that the mean or median QoL score in intervention groups were higher than control groups at week 12 (physical component score: For interventions involving foot/ankle range of motion exercise components, two studies reported changes in health-related QoL at the end of study period. 46,49 One study found a 'significant difference' in QoL scores between the intervention group and the control group on day 90 (p = 0.03); however, supportive data were not provided. 49 One other study reported increased QoL scores in intervention group after receiving foot/ankle range of motion exercises. 46

| Pain
For interventions involving resistance exercise components, pain scores were reported by four studies. 37,39,41,51 Two studies assessed pain scores at week 12, 37,39 while the others reported it at week 24 41 and month 12, 51 respectively. Improvements in pain were reportedly much higher in intervention groups than in control groups. 37,39 For interventions involving aerobic exercise/walking components, one study reported changes in pain scores. 51 By month 12, mean pain score was 7.9 in the intervention group and 30.5 in the control group. 51 However, this study did not provide sufficient information to interpret the clinical relevance of these data. 51 For interventions involving foot/ankle range of motion exercise components, three studies reported the changes in pain levels. 39,41,49 One study reported improved (lower) pain scores in the intervention group compared to the control group by day 90, but did not supply raw data. 49 Sallam et al. 39 reported the mean pain score in the intervention group was 2.8/10 comparing with 3.6/10 in the control group by week 12. Davies et al. 41 carried out a single-group clinical trial in which the intervention group's median pain score decreased from 5.2/10 to 2/10 by week 24.

| Adverse events
For interventions involving resistance exercise components, five studies recorded adverse events during the trial period. [37][38][39]42,51 Two studies reported no adverse events, 38,39 while two studies reported the discharge/exudate level from VLU increased after receiving PA interventions at first month and month 12. 38,39 One study found that 19 participants in the intervention group compared to three participants in the control group experienced one or more adverse events including ulcer deterioration, surrounding skin deterioration, development of new ulcer, pain, infection and hospitalisation. 42 For interventions involving aerobic exercise/walking components, one study reported adverse events at month 12. 51 In this study, two participants in the intervention group reported an increased discharge/exudate level from VLUs after receiving PA interventions. 51 No adverse events were reported in the two studies that involved foot/ankle range of motion exercise components. 38,39 3.6.9 | Cost Only one study reported on cost to health services and individuals at month 12. 51 This study involved both resistance and aerobic/walking exercise components and found the total cost for the intervention group was £1931.76, compared to £3666.12 for the control group. 51 The total cost to the healthcare service was £13825.60 for the intervention group, which is less than half of the cost for the control group (£48270.0). 51 No data were reported for interventions involving foot/ankle range of motion exercise components.

| Multicomponent interventions
Five studies described interventions involving more than one PA component (multicomponent) with their key findings summarised in Table 4. [39][40][41]44,51 As shown in Table 4, there appeared a greater tendency for studies involving multicomponent interventions to yield clinically meaningful benefits compared to interventions involving single component. VLUs healing, measured by wound size, was greatly enhanced on day 12 and week 12 in participants who received interventions comprising resistance exercise and ankle/foot range of motion exercise. 39,40 The marked reduction in the pain score was also noticed in participants who received interventions comprising resistance exercise and ankle/foot range of motion exercise. 41  The predominance of resistance training interventions (11/16 studies) for this patient group is consistent with findings from earlier reviews. 18,23 The reason for this apparent preference is not entirely clear. Improvements in lower limb blood flow may be achieved following exercise training involving either walking/ aerobic training, resistance training or foot/ankle exercises due to their common involvement of the lower limbs and feet. Their mechanisms of action are, however, different. Aerobic training involving prolonged bouts of high repetition, low load activity focuses on improving endurance which may promote better cardiovascular health outcomes (e.g., microvascular epithelial changes, reduced peripheral resistance etc.). 36,53 Foot and ankle range of motion exercises aim to optimise ankle mobility and promote the calf muscle 'pumping' mechanism but are unlikely to confer many enduring effects once ceased. 21 Resistance training involving shorter bouts of low repetition, higher load activity focuses on improving strength. 54 This type of training may associate with sustained effects vascular re-capillarisation, calf muscle hypertrophy and improved blood flow. 54 Our review identified benefits on calf muscle pump function following resistance training but failed to identify studies involving other intervention types. It is important this 'lack of evidence' is not equated to 'evidence of a lack of effect'.

T A B L E 4 Synthesis of findings between interventions
The majority of interventions in our review involved home-based, unsupervised PA components. Individuals with VLUs frequently required travel assistance, making home-based PA interventions more viable for them. 55 The results showed that studies with supervision had higher compliance rate and better clinical outcomes than those without supervision. 43 46 The lack of evidence relating to VLU recurrence has been previously identified 18,22,23 and remains a topic of unmet need. Uncertainty also remains regarding the precise role of PA interventions on longer term health outcomes. While a lack of long-term follow-up in studies of people with VLUs has been previously described, 22 we know very little of the relative contribution of PA to important outcomes relating to healing and recurrence, including their potential enduring effects following treatment cessation.
Resistance exercise has been proposed as potentially having such a role due to its potential impact on calf muscle pump function, 61

| Future research and implications
Our findings suggest interventions involving a combination of training components may confer greater benefit than single elements alone.
The effectiveness of such an approach, however, requires further attention integrating perspectives of multiple stakeholders and sufficient sample sizes to ensure adequate statistical power for key outcomes such as those used within this review. Future research should also examine the long-term impact of this combination of PA therapies on calf muscle pump function. Furthermore, it is strongly advised that future clinical trials adhere to a well-structured template (i.e., CERT) when reporting the characteristics of PA/exercise therapies. The results of this review identify a gap existing in the webbased approach to deliver exercise programmes. Given that this patient population frequently require assistance when travelling, and that researchers face existing challenges such as COVID-19 when doing clinical trials, future clinical studies could explore the feasibility and acceptability of incorporating web-based elements into PA programmes. A systematic review focusing on qualitative evidence is also necessary to investigate the barriers and enablers to PA involvement in patients with VLUs, 64 and the findings of this review will aid in the design of feasible and acceptable PA interventions for future clinical trials.

| Strengths and Limitations
Key strengths of this review included the incorporation of a variety of study designs and strong emphasis on critical components of PA/exercise interventions (e.g., CERT). Furthermore, the choice of review outcomes provides an important framework that should guide future studies involving PA interventions for people with VLUs.
There are some limitations in this review. Even though we conducted an extensive literature search using a wide range of electronic databases combined with hand-searching of grey literature sources, it is possible we may have missed some studies. We encourage readers or other researchers who may have reported in this space to contact us. Further, we had little opportunity to consult with external healthcare professionals as a result of the unanticipated pandemic.
We also omitted PA levels as an outcome of this review in our protocol despite its obvious relevance; however, no studies reported such data. It remains clinically valuable to understand whether PA interventions influence PA behaviours in people affected by VLU; however, the focus of this review was the potential for PA interventions to affect outcomes primarily related to VLU healing rather than PA levels.