The Canadian Bandaging Trial: Evidence-informed leg ulcer care and the effectiveness of two compression technologies

Community care of individuals with chronic wounds has become an important issue for home care authorities as they deal with a growing demand for wound care in the community. In Canada, the impact of caring for individuals with leg ulcers is only now being recognized due to the increased pressure on the home care environment caused by nursing shortages and tightening of healthcare budgets [1]. In the UK, Luker and Kenrick estimated that up to 13% of community nurses' time was spent in leg ulcer care [2], and more recently NHS leg ulcer treatments (bandages and dressings) were reported to cost £300-600 million per annum in England [3]. Our previous work in one Ontario region demonstrated that 192 people with leg ulcers received on average 12 homecare visits per month, and expenditures for supplies and nursing visits were estimated at $1.3 million per annum [4]. While there are no national data, it is reasonable to believe that leg ulcer care in Canada consumes hundreds of millions of dollars annually.

For individuals with leg ulcers, the condition has a profound impact on quality of life [5‐7]. It is chronic, may take years to heal, 66% of individuals experience recurring ulceration and 45% have a history dating back ten years [8]. Studies using the Short Form (36) Health Survey (SF-36^®) instrument to measure health-related quality of life (HRQL) have demonstrated that individuals with leg ulcers have lower mean scores than Canadian population norms [9‐11]. Multiple studies document the discomfort with venous leg ulcers, with the prevalence of pain estimated to be between 48-96 percent [12‐23].

Best practice recommendations, supported by high level evidence, indicate that the most effective treatment for venous leg ulcers is high compression bandaging [24‐36] applied by well-trained healthcare professionals [37‐40]. However the initial Cochrane review had little evidence about the comparative effectiveness of the available types of high compression [41, 42]. This is an issue given there are substantive clinical, practical and economic differences in the classes of compression technology. Early trials of bandaging systems were methodologically weak due to factors such as being underpowered [25]. By 2000 the VenUS I trial [43] was underway in the UK to determine the relative effectiveness of two of the most popular high compression systems, 'four-layer' (4LB: strictly speaking a four-component bandage system) and short-stretch bandaging (SSB). As the effectiveness and costs of the compression systems are likely related to service delivery and bandaging skills, we realized that the UK trial would not be able to completely answer the question for the Canadian healthcare system. In 2002, a Canadian team, working with the leads of VenUS I (Drs. E.A. Nelson and N. Cullum), set out to conduct a concurrent trial to provide more data about clinical and cost-effectiveness of the high compression technologies since the results of available published trials (Duby, Scriven, Partsch: n = 226 participants) were equivocal [44‐46]. The Canadian trial, funded by the Canadian Institutes of Health Research (CIHR), was launched in 2004. Canadian homecare authorities were interested in a comparison of these commonly available bandaging systems based on a health services outcome, i.e. a 4-week difference in healing based on known costs of delivering the technologies in Canada. Thus, the purpose of this Canadian trial was to compare healing rates between two bandaging technologies, specifically: 1) determine if healing occurred at least 4 weeks earlier with SSB vs. 4LB; 2) assess related outcomes such recurrence rates, HRQL, pain, and adverse events over a one-year follow-up; and 3) identify potential baseline factors that predict healing.

The Canadian Bandaging Trial (CBT) was a multi-centre parallel-group, randomized controlled trial conducted in 10 centres in three provinces. Recruitment involved all individuals referred for community wound care services (homecare or nurse clinic) with new, existing or recurrent venous ulceration. Ethics approval for the trial was received from Queen's University Research Ethics Board, Kingston Canada (REB# NURS-140-03).

Over a 50 month recruitment period (January 2004 - March 2008), 2820 individuals referred for community leg ulcer care were screened (see Figure 1). Many of those referred for care did not have purely venous leg ulcers, thus the trial procedures required a major screening effort. For example more than half screened had diabetic or arterial issues (hence mixed ulcer aetiology) but there was no way of knowing this prior to examination by the nurse and it was vital to maintain an evidence-informed clinical assessment consistently. Consequently only 520 were assessed as eligible with 82% of those (n = 424) agreeing to participate.

Of the 424 trial participants, 215 were randomized to receive 4LB, and 209 to receive SSB. There were no differences on admission between the participants randomized to either bandage on their socio-demographic, circumstance of living, HRQL or clinical characteristics (Table 1).

Mean age of the participants was 65 years with slightly more women (54%) than men and the majority being English-speaking (91%). Most participants were fully mobile and lived with others. For more than half (55%), this was their first episode of ulceration and the median ulcer duration at baseline was nearly 12 weeks. PCS baseline scores were much lower than the age- and sex-adjusted Canadian norms (39.1 vs. 51.7); indicating that those with venous leg ulcer had much poorer physically oriented HRQL. However, mental HRQL was better, with MCS scores comparable to Canadian norms (51.4 vs. 50.5) [68]. There were no differences in key aspects of the clinical care received by the two groups (Table 2).

The Canadian Bandaging Trial was adequately powered to detect a minimally important difference of 4-weeks in median healing time between 4LBs and SSBs. We put much emphasis on testing the compression technologies within an evidence-informed approach to care i.e. a protocol for venous leg ulcer assessment and management, well developed training procedures, and well prepared RNs. With these elements in place, results from our trial suggest that the two compression bandaging technologies tested are not different in terms of time-to-healing, recurrence rates, HRQL, or pain. However, individuals were less satisfied with 4LB, with more reporting problems with the bandaging system.

These healing results should be viewed in the context of all the trials making this comparison. O'Meara et al. [33] conducted a meta-analysis of individual patient data to evaluate the intervention effect of 4LBs relative to SSBs. They concluded that venous leg ulcers in patients treated with 4LBs heal faster, on average, than those treated with SSBs; and the benefits were consistent across patients with differing prognostic profiles. In contrast, our trial (the largest to date), put much emphasis on other aspects of venous leg ulcer care, namely an evidence-informed protocol of care and well-prepared RNs in administering these bandaging systems. With these elements of care in place, there were no differences between the technologies except that individuals may have more problems with 4LBs.

In the O'Meara meta-analysis, the intervention effect estimate from the UK VenUS I was weighted heavily towards the pooled effect estimate (i.e. 63% from this trial among five others). As the only other large scale trial comparable to this Canadian RCT, the UK VenUS I showed that with 4LBs there was a shorter healing time. We were unable to replicate this finding despite comparable sample sizes. Several factors may be contributing to this difference. Nurse wound-care teams in Canada have more training and experience in both technologies. The SSB was newly introduced to UK primary care just prior to their trial and there was a higher withdrawal rate in the SSB arm of the VenUS I trial. The UK authors [43] state they "investigated the introduction (their emphasis) of a high-level compression bandage system into a setting where an acceptable mode of compression was already in use" and a reasonable conclusion is that short stretch is "a useful alternative" (pg. 1298). Based on the experience from the UK trial, our Canadian sites had initial and ongoing training including quality checks to ensure adherence to the evidence-informed assessment and management protocol for either bandage. In the Canadian trial, the nurses may have been more confident and competent in delivery of both bandage systems thus reducing operator experience/preference as a source of bias. Another context factor may be important. Canadian site teams actively participated in the development of the protocol in their local settings to make it as seamless as possible within their normal delivery of care. Based on best practice recommendations, they carefully constructed an approach to adhere to key aspects of leg ulcer assessment and management. For example, the initial comprehensive assessment may have been carried out by specially trained nurses in the home with portable equipment whereas at some sites, it was more feasible at a designated clinic location where a wound expert and equipment were centralized.

In considering the site differences our team thought there were factors that should be further explored in a future study. Although we tracked key elements of the evidence-informed protocol (comprehensive assessment and use of compression), and all sites delivered these recommendations, there may be subtle differences contributing to healing rates. For example, some sites were newer and less experienced and may have been less supported regionally in the delivery of care. As well, because the trial was conducted in a large number of jurisdictions the referral and wait-times may have varied influencing the time to assessment which may have affected outcomes. Unfortunately we did not have data to drill down and explore these issues in this trial.

Lastly, healing in the Canadian trial for 4LB was 62 days compared to 77 days with SSB - both much shorter than the UK study population. This may be partly explained by their healing outcome being complete healing of all ulcers not just a reference ulcer. Nevertheless in examining healing over time compared to previous trials (Table 6), the Canadian group had higher proportions healed at 12, 16, 24 and 52 week time-points for both SSB and 4LB except SSB at 16 weeks in the Partsch study [45].

In conclusion the Canadian bandaging trial adds new information to the international literature focused on the management of leg ulcers in the community. Two commonly available technologies, short-stretch and four-layer bandages, are viable options that result in similar healing, recurrence, HRQL and pain outcomes when delivered by trained nurses using an evidence-informed protocol.