Evidence Corner

Evidence Corner:
Evidence Corner
- Laura Bolton, PhD, FAPWCA

Dear Readers:

Laura Bolton
Venous ulcers (VU) can be slow to heal reducing patients’ economic well being and quality of life.¹ The multitude of therapies and variations in application techniques can be baffling. For example, one may discover too late that required compression was insufficient, was applied incorrectly, or was not worn as prescribed.² Wound care professionals learning how to optimize VU compression, primary dressings, surgical methods, and alternative modalities often turn to evidence summaries and systematic reviews to identify safe and effective protocols.³ This Evidence Corner discusses 2 systematic reviews with different conclusions about whether the primary dressing affects VU healing when used under effective compression. How does one resolve such an apparent discrepancy?

Laura Bolton, PhD, FAPWCA

Cochrane Systematic Review: Venous Ulcer Dressings

       Reference: Palfreyman SJ, Nelson EA, Lochiel R, Michaels JA. Dressings for healing venous leg ulcers. Cochrane Database Syst Rev. 2006(3):CD001103.
       Rationale: The main treatment for VUs is compression. Primary dressings are usually applied to aid healing, improve patient comfort, or manage exudate.
       Objective: This systematic review explored the effects of wound dressings on healing VUs.
       Methods: The authors retrospectively reviewed prospective randomized controlled trials (RCTs) in all languages published before 2005 comparing dressing effects on time to healing, proportion healed, or rate of healing of patients with VU. RCTs evaluating topical agents and skin grafting were excluded as subjects of prior Cochrane systematic reviews. Searches included the Cochrane, MEDLINE, EMBASE, and CINAHL databases, derivative references, and queries to authors, wound professionals, and manufacturers. A different author checked data extraction. Study quality was assessed on group comparability at baseline, intent-to-treat analyses, follow-up completeness, allocation concealment and objective, and blinded outcome measures. Meta-analyses were performed on statistically homogeneous data of multiple studies addressing each question reviewed. Subgroup analyses tested for effects of study size, compression use, dressing type, and blinded assignment.
       Results: Of 254 citations, 42 RCTs of 4- to 40-week duration involving 3,001 patients with a VU qualified for inclusion in the analyses. There were sufficient data for meta-analyses comparing hydrocolloid dressings with foams (4 trials, 311 patients), alginates (2 trials, 80 patients), other hydrocolloids (3 trials, 98 patients) or low-adhering dressings (9 trials, 928 patients); foam dressings with low-adherent (3 trials; 253 patients) or other foams (2 trials; 136 patients); and hydrogels with low-adherent dressings (2 trials; 134 patients). Many studies were underpowered. None of the meta-analyses revealed significant differences between healing effects of any 2 primary dressing categories. Effects of patient variables, such as ambulation and adherence to compression protocols, were not analyzed owing to insufficient data reporting.
       Conclusion: Primary wound dressings applied beneath compression have not been shown to affect VU healing.

Journal of Vascular Surgery Review: Venous Ulcer Dressings

       Reference: O’Donnell TF Jr, Lau J. A systematic review of randomized controlled trials of wound dressings for chronic venous ulcer. J Vasc Surg. 2006;44(5):1118–1125.
       Rationale: Annual direct costs of VU treatment are about $30,000 per patient in the United States with 60% to 80% healing in 6 months. Any treatment modality improving VU healing time or proportion healed would reduce the burden of VU care. A 1999 systematic review of prior research concluded that methodological flaws reduced the validity of most studies, preventing firm conclusions.⁴
       Objective: Update the prior review to determine whether more “modern” complex wound dressings improve VU healing.
       Methods: A retrospective search of the Cochrane Controlled Trials Registry, MEDLINE, and CINAHL databases identified RCTs reported from October 1997 to September 2005 that compared topical modalities. The search included studies of only VUs of at least 30-day duration managed with compression that reported the primary outcome proportions of VUs as healed or a secondary outcome of Kaplan-Meier projected time to complete healing. Study design and methodological quality was assessed according to the US Food and Drug Administration (FDA) Draft Guidance.⁵ Meta-analysis determined the statistical significance of modality effects on VU healing if the trials were not heterogeneous.
       Results: Twenty RCTs involving 1,820 VU patients qualified for analysis. All used “maximal” compression except for 1 study that used the Unna Boot alone as compression. The only meta-analysis conducted revealed a significantly higher likelihood of healing of a VU randomized to receive growth factor (GF) as compared to mainly gauze controls (686 VUs in 8 studies). Meta-analyses were not performed on dressings (D: 8 studies; 687 patients) or human skin equivalents (SE: 5 studies; 447 patients) due to heterogeneity of data between studies. Among the individual studies 2 GF studies, 1 SE study and 2 D studies reported effects (P < 0.05), increasing percents healed during intervals ranging from 8 to 26 weeks and reducing healing time in several of the studies. Quality improvements were seen compared to the 1999 review⁴ on all study design parameters assessed, with 6 studies reporting at least 6 of the 7 factors required for good study design, and the seventh, a study comparing hydrocolloid dressings, reporting 2 of the 7 factors.
       Conclusions: Certain modalities can improve the percent of VUs healed and time to healing.

Clinical Perspective

       Though these 2 systematic reviews seem to differ in their conclusions, they differed in their scope and methods of review. The Cochrane review analyzed all qualifying evidence on topical dressings only—the JVS review focused on studies of GF, SE, and dressings after October 1997. Heterogeneity and wide variability in study parameters or measures prevented meta-analysis of all but GF data in the JVS review. Meta-analyses were not possible for several dressing categories in the Cochrane review, which only featured 1 study. One should interpret with caution the 2 studies in the JVS review supporting the conclusion that dressings make a difference in VU healing outcomes. The study comparing 2 hydrocolloids reported only 2 of the 7 recommended elements of quality study design and was underpowered with insufficient proof of baseline comparability as cited in Palfreyman et al. The other study compared an alginate dressing or a zinc-impregnated stockingette to a zinc paste bandage that may augment compression applied, depending on application technique. It is not clear that the dressing alone caused the VU healing effect observed. These artifacts do not disprove the conclusion that there may be dressing effects on VU healing. They underscore the need for higher quality study design and methods and call for better data to support those effects, echoing the conclusion of Palfreyman et al.
       Fundamental flaws in the wound healing literature diverted the authors’ careful, hard work from accomplishing their stated goals of analyzing dressing effects on VU healing. First, by classifying dressings according to their material content, one mixes categories that perform different functions.⁶ In addition to compression, VUs require a moist environment, management of excess exudate, and debridement of necrotic tissue for optimal healing.³ Both reviews classified dressings by their material content rather than their capacity to meet these clinical wound needs. What would a meta-analysis based on functional dressing categories, such as moisture retention or absorbency find? A second issue not addressed by either review is that a properly compressed VU may need a protocol of care, not a single dressing to meet its clinical needs. For example, management of a fibrin-covered VU may begin with autolytic debridement followed by excess exudate management until surrounding edema is gone, then a moist environment until the VU is healed. Rarely is a single wound dressed uniformly throughout its course of care. How does one rationally compare dressings in this context? Finally, using a validated protocol of care in a real-world cohort, full-thickness VUs took twice as long to heal as partial-thickness VUs.⁷ Neither review above controlled for VU depth or explored it as a covariate. Mixing full- and partial-thickness VUs in an analysis expands variability of healing results and obscures the effects of therapeutic modalities on healing.
       Implications for future studies and clinical practice are to continue to improve study quality, report dressing effects by functional rather than manufacturing categories, explore clinically relevant protocols of care, and meticulously report and stratify VU depth.

References
1. Phillips TJ. Successful methods of treating leg ulcers. The tried and true, plus the novel and new. Postgrad Med. 1999;105(5):159–174.
2. Ham S, Padmore J. Two-layer compression hosiery for patients with venous leg ulceration. Nurs Stand. 2006;20(45):68–76.
3. Bolton L, Corbett L, Bernato L, et al. Development of a content-validated venous ulcer guideline. Ostomy Wound Manage. 2006;52(11):32–48.
4. Bradley M, Cullum N, Nelson EA, Petticrew M, Sheldon T, Torgerson D. Systematic reviews of wound care management: (2). Dressings and topical agents used in healing of chronic wounds. Health Technol Assess. 1999;3(17 Pt 2):1–35.
5. US Food and Drug Administration. Chronic Cutaneous Ulcer and Burn Wounds—Developing Products for Treatment. Available at: http://www.fda.gov/cber/gdlns/ulcburn.htm. Accessed December 8, 2006.
6. van Rijswijk L, Beitz J. The traditions and terminology of wound dressings: food for thought. J Wound Ostomy Continence Nurs. 1998;25(3):116–122.
7. Bolton L, McNees P, van Rijswijk L, et al. Wound healing outcomes using standardized care. J Wound Ostomy Continence Nurs. 2004;31(2):65–71.

Wounds - ISSN: 1044-7946 - Volume 19 - Issue 2 - February 2007 - Pages: A16 - A21