Background

Chronic wounds are those that have not progressed through the ordered process of healing to yield a functional result [1]. Recently, the terminology for chronic wounds has changed. The preferred term to refer to a chronic wound is a “complex wound” [2]. One of the following characteristics is necessary for a wound to be classified as being complex: i) has not healed in 3 months, ii) infection is present, iii) compromised viability of superficial tissues, necrosis, or circulation impairment, and iv) association with systemic pathologies, impairing normal healing [2]. The main types of complex wounds include diabetic leg/foot ulcers, pressure ulcers [3], chronic venous ulcers, infected wounds [1,4,5], and those related to vasculitis and immunosuppressive therapy that have not healed using simple care [2].

Complex wounds are a significant burden on society. It has been estimated that complex wounds cost the healthcare system $10 billion annually in North America alone [6]. These estimates often fail to capture indirect costs, including patient/caregiver frustration, economic loss, and decreased quality of life.

Healthcare providers and patients have numerous regimens available for treating wounds [7], including dressings, wound cleansing agents, skin replacement therapy, biologic agents, stockings, nutritional supplementation, complementary and alternative medicine, bandages, and surgery, to name a few. Furthermore, wound care is often multi-faceted, and several interventions may be used concurrently. Some of these interventions have been examined in overviews of Cochrane reviews [8,9]. As the evidence of these interventions is recondite across the literature, we sought to elucidate optimal treatment strategies for complex wounds through an overview of all available systematic reviews, including Cochrane reviews and non-Cochrane reviews.

Methods

Protocol

A protocol for our overview of reviews was developed using the Cochrane Handbook for overviews of reviews [10]. The draft protocol was circulated for feedback from systematic review methodologists, policy-makers, and clinicians with expertise in wound care. It was revised as necessary and the final version is available from the corresponding author upon request.

Eligibility criteria

We included systematic reviews that focused on interventions to treat complex wounds (including venous and arterial ulcers due to chronic illness, diabetic ulcers, pressure ulcers, and infected surgical wounds) amongst adults aged 18 years and older. We used the definition for a systematic review put forth by the Cochrane Collaboration, “A systematic review attempts to collate all empirical evidence that fits pre-specified eligibility criteria in order to answer a specific research question. It uses explicit, systematic methods that are selected with a view to minimizing bias, thus providing more reliable findings from which conclusions can be drawn and decisions made” [10].

A list of the 14 different intervention categories can be found in Additional file 1. All comparators, such as other wound care interventions, no treatment, placebo, and usual care were eligible for inclusion. To be included, a systematic review also had to report on our outcomes of interest as identified by decision-makers, including healing (e.g., number of ulcers healed, improvement of ulcers, and time to ulcer healing) or admission to hospital (including readmissions). Systematic reviews that were published or unpublished and conducted at any point in time were included. Due to resource limitations, only systematic reviews written in English were included. However, authors were contacted to obtain translations of reviews written in languages other than English.

Literature search

Comprehensive literature searches were conducted from inception until October 2012 across MEDLINE, EMBASE, and the Cochrane Database of Systematic Reviews. The search terms included both medical subject headings (MeSH) and free text terms related to wound care interventions. Literature searches were conducted by an experienced librarian (LP) on October 26, 2012. Using validated search filters, the search strategies were limited to human participants, adults, and systematic reviews. The electronic database search was supplemented by searching for systematic review protocols in the PROSPERO database [11], contacting authors of conference proceeding abstracts for their unpublished data, and scanning the reference lists of the included systematic reviews.

The search strategy was peer reviewed by another expert librarian on our team (EC) using the Peer Review of Electronic Search Strategies checklist [12] and amended, as necessary. The final search strategy for the MEDLINE database is presented in Additional file 2. The literature search was limited to adults, reviews, and economic studies. The latter limitation was employed to identify cost-effectiveness analyses for a second paper that examines the cost-effectiveness of complex wounds [Tricco et al., unpublished paper submitted to BMC Medicine]. The MEDLINE search was modified for the other two databases, as necessary. Search strategies for the EMBASE and Cochrane Database of Systematic Reviews are available from the corresponding author, upon request.

Screening

Prior to commencing the screening process, a calibration exercise was conducted to ensure reliability in correctly selecting articles for inclusion. This exercise entailed screening a random sample of 50 of the included titles and abstracts by all team members, independently. The eligibility criteria were modified, as necessary, to optimize clarity. Subsequently, reviewer pairs (ACT, JA, AH, AV, PAK, CW, EC, LP) independently screened the remainder of the search results for inclusion using a pre-defined relevance criteria form for all levels of screening (e.g., title and abstract, full-text review of potentially relevant articles). Discrepancies were resolved by discussion or the involvement of a third reviewer.

Data items

Data abstraction forms were pilot-tested by all team members independently on a random sample of five articles. The data abstraction forms were revised after this exercise, as necessary. Subsequently, reviewer pairs (ACT, JA, AH, AV, PAK, CW, LP, WH) independently read each article and abstracted relevant data. Differences in abstraction were resolved by discussion or the involvement of a third reviewer. Data items included study characteristics (e.g., number of studies identified, type of study designs included, interventions and comparators examined), patient characteristics (e.g., clinical population, wound types, age category), and outcome results (e.g., healing, hospitalizations).

Methodological quality appraisal

The methodological quality of the included systematic reviews were appraised using the Assessment of Multiple Systematic Reviews (AMSTAR) tool [13]. The reliability and validity of this tool has been established [14]. Items include the use of a protocol, study selection by two reviewers, comprehensive literature search, inclusion of unpublished material, list of included and excluded studies, reporting of study characteristics, quality appraisal, appropriate pooling of data methods, assessment of publication bias, and statement of conflicts of interest. Each included systematic review was appraised by two team members (ACT, JA, AH, AV, PAK, CW, EC, LP) and conflicts were resolved by discussion or the involvement of a third reviewer.

Synthesis

Literature search results and the abstracted data were summarized descriptively. An in-depth comparison of included systematic reviews was compiled and depicted in tables and figures. Conclusion statements for systematic reviews without a meta-analysis were categorized by two reviewers (ACT, JA, AH, AV, PAK, CW, LP, WH), independently, using a pre-existing framework, as follows: positive (authors stated that there is evidence of effectiveness); neutral (no evidence of effectiveness or they reported no opinion); negative (authors advised against the use of the intervention or it was not recommended); or indeterminate (authors stated that there is insufficient evidence or that more research is required) [15]. Conflicts were resolved through discussion and a third reviewer (ACT) verified the categorizations to ensure accuracy.

Results

Literature search

The literature search resulted in 6,200 titles and abstracts, of which 5,778 were excluded for not fulfilling the eligibility criteria (Figure 1). Of the 422 full-texts retrieved and screened in duplicate, 309 articles were excluded. Ninety-nine systematic reviews of wound care interventions were included in this overview of systematic reviews; 54 were systematic reviews with meta-analysis results [16-67] and 45 were systematic reviews without a meta-analysis [68-112]. In addition, 14 companion reports were included [21,113-125], the majority of which were Cochrane updates.

Figure 1
figure 1

Study flow. Details the flow of information through the different phases of the review. The flow maps out the number of records identified, included and excluded, and the reasons for their exclusion.

Full size image

Systematic review characteristics

The reviews were conducted between 1997 and 2012, with 28% taking place after 2011 (Table 1; Additional file 3). The first authors of the systematic reviews were predominantly based in Europe (66%), North America (19%), and Australia or New Zealand (7%). The number of studies included in each review ranged from 0 to 130, with 80% including between 2 and 30 studies. Ninety-three systematic reviews included randomized clinical trials. Five systematic reviews were unpublished [43,58,83,104,106].

Table 1 Summary characteristics of included systematic reviews
Full size table

Study and patient characteristics

Thirty-four systematic reviews provided information about the patient population under study (Table 1; Additional file 3) and 21% included patients with diabetes. Six categories of complex wounds were examined: venous leg ulcers (31%), diabetic foot/leg ulcers (23%), pressure ulcers (17%), mixed arterial/venous leg wounds (14%), unspecified mixed complex wounds (9%), and infected surgical wounds (6%). The five most common interventions were adjuvant therapies (20%), dressings (16%), biologic agents (10%), other topical agents (9%), and other oral agents (7%). The five most common comparators were usual care (46%), dressings (25%), bandages (6%), support surfaces (5%), and other topical agents (4%). The duration of treatment ranged from 2 days to 160 months and the duration of follow-up ranged from 2 days to 195 months across the included studies in the systematic reviews.

A total of 327 treatment comparisons were included in the 99 systematic reviews. As such, only statistically significant results from systematic reviews with a meta-analysis are reported in our outcome results section below in the text. Specific results for all treatment comparisons can be found in Table 2. To facilitate the summary and comparison of a large number of reviews, we have presented the review results using positive, negative, or neutral conclusions (Tables 2,3,4,5,6,7); however, we have also included the statistical effect sizes from each of the included meta-analyses in Additional file 4.

Table 2 Summary of evidence for venous ulcer management
Full size table
Table 3 Summary of evidence for mixed arterial/venous ulcer management
Full size table
Table 4 Summary of evidence for diabetic ulcer management
Full size table
Table 5 Summary of evidence for pressure ulcer management
Full size table
Table 6 Summary of evidence for mixed complex wounds management
Full size table
Table 7 Summary of evidence for infected surgical wounds management
Full size table

Methodological quality appraisal

Almost half (45%) of the systematic reviews were deemed high quality with an AMSTAR score ≥8 out of a possible 11 (Figure 2; Additional file 5). Many systematic reviews did not provide a list of excluded studies from screening potentially relevant full-text articles (60%) or address publication bias (65%). Conversely, 95% searched at least two electronic databases, 91% provided the characteristics of included studies, and 89% adequately used the quality appraisal results in formulating conclusions. Half of the systematic reviews including a meta-analysis had an AMSTAR score ≥8, and 40% of the systematic reviews without a meta-analysis had an AMSTAR score ≥8.

Figure 2
figure 2

AMSTAR methodological quality results. NA, Not applicable. 1. A priori design. 2. Duplicate selection/DA. 3. Literature search. 4. Publication status. 5. List of studies. 6. Study characteristics. 7. Quality assessed. 8. Quality used. 9. Methods appropriate. 10. Publication bias assessed. 11. Conflicts stated.

Full size image

Outcome results for venous leg ulcers

Wound area/size reduction

Two systematic reviews including two meta-analyses examined venous leg ulcer area/size reduction [56] (Table 2; Additional file 4), one of which had an AMSTAR score ≥8 [36]. Topical cadexomer iodine [36] and oral micronized purified flavonoid fraction [56] were more effective than placebo in each meta-analysis comparing these interventions.

Time to healing or rate of healing

Five systematic reviews including six meta-analyses [34,39,44,45,56] and one systematic review without a meta-analysis [77] (Table 2; Additional files 4 and 6) examined the time to healing for venous leg ulcers. Two of these had an AMSTAR score ≥8 [44,45].

Two systematic reviews (AMSTAR score ≥8) including two meta-analyses found that four-layer bandages were more effective than short stretch bandages [44] and compression systems [45]. One systematic review including two meta-analyses found conflicting results for bandages versus stockings [39]. Oral micronized purified flavonoid fraction was more effective than placebo in one meta-analysis [56].

Ulcer healing

Five systematic reviews including eight meta-analyses [29,30,39,51,62] and five systematic reviews without a meta-analysis [72,73,77,89,103] (Table 2; Additional files 4 and 6) examined venous leg ulcer healing. Four of these had an AMSTAR score ≥8 [30,62,73,89].

Elastic high compression was more effective than multi-layer inelastic compression in a systematic review (AMSTAR score ≥8) with a meta-analysis [62] and stockings were more effective than bandages in another meta-analysis [39]. Tissue engineered skin was more effective than dressings in one meta-analysis [51], topical granulocyte-macrophage colony stimulating factor was more effective than placebo in another meta-analysis [29], and oral pentoxifylline was more effective with (or without) compression than placebo in another meta-analysis (AMSTAR score ≥8) [30].

Proportion of patients with healed wounds

Nineteen systematic reviews with 39 meta-analyses [21,23,25,30,31,36,40,45,46,48,52,53,55,58,62,64-67] and two systematic reviews without a meta-analysis [77,102] (Table 2; Additional files 4 and 6) examined the proportion of patients with healed venous leg ulcers. Twelve of these had an AMSTAR score ≥8 [23,25,30,31,36,40,45,52,55,62,64,67].

Multi-layered, high compression bandages reduced ulcers compared with single layer bandages in two meta-analyses [62,67]. Elastic high compression was more effective than inelastic bandages [45] and versus inelastic compression bandages (AMSTAR score ≥8) [67] in two meta-analyses. Intermittent pneumatic compression was more effective than compression stockings or Unna’s boot in one meta-analysis (AMSTAR score ≥8) [67] and high compression stockings were more effective than compression bandages in another meta-analysis (AMSTAR score ≥8) [62]. Two-layer stockings were more effective than short-stretch bandages in another meta-analysis [44]. Ultrasound was more effective than no ultrasound in one meta-analysis [65] and cleaning wounds with cadexomer iodine plus compression therapy was more effective than standard care in another meta-analysis (AMSTAR score ≥8) [36]. Skin replacement therapy was more effective than standard compression therapy in one meta-analysis [46] and oral pentoxifylline with (or without) compression was more effective than placebo in two meta-analyses (AMSTAR score ≥8) [30]. Periulcer injection with granulocyte-macrophage colony stimulating factor was more effective than control in another meta-analysis [53].

Outcome results for mixed arterial/venous leg ulcers

Wound area/size reduction

Two systematic reviews including three meta-analyses examined wound area/size reduction for mixed arterial/venous leg ulcers [34,65] (Table 3; Additional file 4). None had an AMSTAR score ≥8. One meta-analysis found topical silver and silver dressings more effective than placebo or conservative wound care or non-silver therapies [34] and another that ultrasound was more effective than standard treatment or placebo [65].

Time to healing or rate of healing

Four systematic reviews including five meta-analyses examined the time to healing for mixed arterial/venous leg ulcers [45,49,55,56] (Table 3; Additional file 4). Two had an AMSTAR score ≥8 [45,55]. One meta-analysis found topical negative pressure more effective than conventional therapy [49], a second found oral micronized purified flavonoid more effective than placebo or standard compression [56], and a third found that four-layer bandage was more effective than compression systems [45].

Ulcer healing

Six systematic reviews without a meta-analysis examined ulcer healing for mixed arterial/venous leg ulcers [80,82-84,101,110] (Table 3; Additional file 6). Two had an AMSTAR score ≥8 [80,84]. Details of these study results can be found in Additional file 6.

Proportion of patients with healed wounds

Two systematic reviews including three meta-analyses examined the proportion of patients with healed mixed arterial/venous wounds [49,50] (Table 3; Additional file 4). None had an AMSTAR score ≥8. Topical negative pressure was more effective than conventional therapy in one meta-analysis [49].

Outcome results for diabetic foot/leg ulcers

The following outcome results were only reported in systematic reviews without meta-analysis: wound area/size reduction (n = 2) [79,88] and time to healing or rate of healing (n = 3) [79,88,95]. Details of these study results can be found in Additional file 6.

Ulcer healing

Three systematic reviews including four meta-analyses [24,35,57] and 10 systematic reviews without a meta-analysis [69,70,74,79,81,86,88,93,98,112] (Table 4; Additional files 4 and 6) examined healing improvements of diabetic foot/leg ulcers. Three had an AMSTAR score ≥8 [24,74,98].

One meta-analysis found that subcutaneous or intravenous granulocyte colony-stimulating factor plus oral or intravenous antibiotics was more effective than control (high quality) [24]. Another meta-analysis found that Chinese herbal medicine (see Table 4 and Additional file 4 for the exact preparation) plus unspecified standard therapy was more effective than standard therapy alone [35].

No healing improvement or non-healed wounds

Three systematic reviews including five meta-analyses [22,33,35] examined no healing improvement for diabetic wounds (Table 4; Additional file 4). Two of these had an AMSTAR score ≥8 [22,33].

Hyaluronic acid derivative was more effective than standard care in one meta-analysis (AMSTAR score ≥8) [22]. Low frequency low intensity noncontact ultrasound was more effective than sharps debridement in two meta-analyses (AMSTAR score ≥8) [33] and Chinese herbal medicine (see Additional file 4 for the exact preparation) plus unspecified standard therapy was more effective than standard therapy alone in another meta-analysis [35].

Proportion of patients with healed wounds

Fifteen systematic reviews including 18 meta-analyses [17-20,22,24,27,28,35,38,46,47,57-59] (Table 4; Additional file 4) examined the proportion of patients with healed diabetic foot and leg ulcers. Eight of these had an AMSTAR score ≥8 [17,18,20,22,24,27,28,59].

Hydrogel dressings were more effective than basic wound dressings, basic contact dressings, and gauze in three meta-analyses (two with an AMSTAR score ≥8) [17,19,28]. Artificial skin grafts were more effective than usual care in one meta-analysis [58] and skin replacement therapy was more effective than usual care in another [47]. Chinese herbal medicine (see Additional file 4 for the exact preparation) plus unspecified standard therapy was more effective than standard therapy alone in a meta-analysis [35] and subcutaneous or intravenous granulocyte colony-stimulating factor was more effective than usual care in another (AMSTAR score ≥8) [24]. Finally, a meta-analysis found platelet-rich plasma more effective than control [38].

Outcome results for pressure ulcers

The following outcome results were only reported in systematic reviews without meta-analysis: wound area/size reduction (n = 3) [78,87,94] and time to healing or rate of healing (n = 3) [68,78,94]. Details of these study results can be found in Additional file 6.

Ulcer healing

Five systematic reviews reporting on eight meta-analyses [16,43,60,64,62] and 11 systematic reviews without a meta-analysis [75,76,78,85,90,92,94,97,99,104,106] (Table 5; Additional files 4 and 6) focused on pressure ulcer healing. Of these, eight had an AMSTAR score ≥8 [16,60,62,64,75,76,90,106].

Hydrocolloid dressings were more effective than usual care in a meta-analysis (AMSTAR score ≥8) [64], electrotherapy was more effective than sham therapy in another meta-analysis (AMSTAR score ≥8) [62], and air-fluidized beds were more effective than standard care or conventional mattresses in three meta-analyses [43,60,125]. In addition, alternate foam mattresses were more effective than standard foam mattresses in a meta-analysis (AMSTAR score ≥8) [60].

Proportion of patients with healed wounds

Three systematic reviews including 25 meta-analyses [43,54,62] and two systematic reviews without a meta-analysis [78,94] (Table 5; Additional files 4 and 6) examined the proportion of patients with healed pressure ulcers. Only one had an AMSTAR score ≥8 [62].

One meta-analysis found that hydrocolloid dressings were more effective than traditional dressings and another that hydrogel dressings were more effective than hydrocolloid dressings [43]. Different brands of alternating pressure mattresses were more effective than others in a meta-analysis [43].

Outcome results for mixed complex wounds (unspecified)

Wound area/size reduction

Four systematic reviews including five meta-analyses examined the area/size reduction of mixed complex wounds (unspecified) [21,32,41,42] (Table 6; Additional file 4). Two had an AMSTAR score ≥8 [41,42]. Silver-impregnated dressings were more effective than dressings not containing silver in a meta-analysis (AMSTAR score ≥8) [41]. Topical negative pressure was more effective than standard wound care in another meta-analysis [32].

Ulcer healing

One systematic review including a meta-analysis [63] and five systematic reviews without a meta-analysis [91,100,105,107,111] (Table 6; Additional files 4 and 6) examined ulcer healing for mixed complex wounds. Three had an AMSTAR score ≥8 [91,107,111].

Proportion of patients with healed wounds

Five systematic reviews including 10 meta-analyses [21,55,58,61,62] (Table 6; Additional file 4) examined the proportion of patients with healed complex wounds. Two had an AMSTAR score ≥8 [55,62]. Hydrocolloid dressings were more effective than conventional dressings in a meta-analysis [61] and ultrasound was more effective than no ultrasound in another (AMSTAR score ≥8) [62]. In addition, artificial skin grafts were more effective than standard care in three meta-analyses [58].

Outcome results for surgical wound infections

The following outcome results were only reported in systematic reviews without meta-analysis: wound area/size reduction (n = 2) [71,108], time to healing or rate of healing (n = 5) [68,71,96,108,109], and ulcer healing (n = 2) [71,108]. Details of these study results can be found in Additional file 6.

Proportion of patients with healed wounds

One systematic review including a meta-analysis (AMSTAR <8) found that topical negative pressure was more effective than standard treatment [37] (Table 7; Additional file 4).

Length of hospital stay

One systematic review and meta-analysis (AMSTAR <8) found that vacuum-assisted closure was more effective than conventional therapy for decreasing the length of hospital stay associated with surgical wound infections [26] (Table 7; Additional file 4).

Discussion

We conducted a comprehensive overview of systematic reviews to identify optimal interventions for complex wounds. Data from 99 systematic reviews were scrutinized and interventions that are likely optimal were identified. These reviews examined numerous treatments and comparators and used different outcomes to assess effectiveness. Frequently, treatments considered as the intervention in one review were administered to the control in another. This renders the interpretation of our findings difficult.

We found that some interventions are likely to be effective based on data from systematic reviews including a meta-analysis with an AMSTAR score ≥8. For venous leg ulcers, four-layer bandages [44,45], elastic high compression [62], oral pentoxifylline with (or without) compression [30], compression bandages (multi-layer, elastic) [62,67], high compression [62] or multi-layer stockings [45], and wound cleansing with cadexomer iodine plus compression therapy [36] were effective compared with usual care. Only four-layer bandages [45] were effective in healing mixed arterial/venous leg ulcers versus compression systems. For diabetic foot/leg ulcers, subcutaneous or intravenous granulocyte colony-stimulating factor [24], hyaluronic acid derivative [22], low frequency, low intensity noncontact ultrasound [33], and hydrogel dressings [17,28] were effective interventions compared with usual care. For pressure ulcers, hydrocolloid dressings [64], electrotherapy [62], air-fluidized beds [60], and alternate foam mattresses [60] were effective compared with usual care. For mixed complex wounds, silver dressings [41] and ultrasound [62] were found to be more effective than no treatment. Finally, effective treatments were not identified for surgical wound infections amongst those with an AMSTAR score ≥8 including a meta-analysis. It is important to note that many of these interventions had conflicting results versus other comparators or were based on meta-analyses including few studies with a small number of patients. As such, these results should be interpreted with caution.

There are some limitations to our overview of systematic reviews. An inherent drawback of including systematic reviews is that the studies included in each of the reviews will have been published well before the search date. The inclusion of close to 100 systematic reviews, however, provides a breadth of information that is unlikely to significantly change with the inclusion of recently published studies. Although we appraised the methodological quality of the included reviews, we did not assess the risk of bias in the included studies in the systematic reviews. This is because a risk of bias tool for systematic reviews currently does not exist, but we are aware of one being developed by the Cochrane Collaboration [126]. Furthermore, we only included systematic reviews that were disseminated in English due to resource constraints. However, we attempted contacting authors to receive English translations. In addition, although we included five unpublished systematic reviews, we attempted to obtain unpublished data from a further 10 systematic reviews that were available as conference abstracts, yet we didn’t receive a response from the review authors. As such, our findings are likely representative of published literature written in English. Since we did not conduct a meta-analysis, we were unable to formally test for publication bias.

Our results suggest the need for a network meta-analysis [127], given the numerous interventions and comparators available and examined across the literature. Policy-makers focus their decisions at the systems level so require information on all treatment comparisons available. Patients and their healthcare providers need to know if the treatment they are prescribed or recommending is the most effective and safest compared with all others available. Conducting a high-quality, comprehensive systematic review and network meta-analysis is the only feasible tool available to examine multiple treatment comparisons. As the healthcare system shifts towards more complex problems and a resource-scarce environment, systematic reviews and meta-analysis of only one treatment comparison become obsolete. This is indeed the case for complex wound care interventions; despite the availability of almost 100 systematic reviews, optimal management is still unclear. Network meta-analysis also allows the ranking of all treatments for each effectiveness and safety outcome examined, making it a particularly attractive tool for decision-makers.

Almost half of the included systematic reviews were rated as being of high methodological quality according to the AMSTAR tool [13]. Consistent methodological shortcomings include not using a protocol to guide their conduct, not including a list of excluded studies at the full-text level of screening, and not addressing or referring to publication bias. Results reported in systematic reviews with higher scores on the AMSTAR tool are likely the most reliable. Furthermore, some studies only gave wound care patients two days of treatment or followed patients for two days. The utility of these short studies is questionable and studies of longer duration are recommended.

Conclusions

In conclusion, our results confirm that there are numerous interventions that can be utilized for patients with complex wounds. However, few treatments were consistently effective throughout the literature. Clinicians and patients can use our results as a guide towards tailoring effective treatment according to type of wound. Planned future analysis of this data through network meta-analysis will further assist decision-makers as it will permit multiple treatment comparisons as well as the ranking of the effectiveness of all available wound care interventions.