Objectives: To systematically review whether the use of advanced wound dressings, systemic antibiotics, or venous surgery enhanced the healing of venous ulcers over the use of adequate venous compression.
Data sources: MEDLINE®, Embase®, the Cochrane Central Register of Controlled Trials, and the Cumulative Index to Nursing and Allied Health Literature (CINAHL®) from January 1980 through July 2012.
Review methods: We included studies of patients with venous leg ulcers lasting 6 or more weeks coincident with signs of preexisting venous disease. We excluded patients with arterial ulcers, pressure ulcers, postsurgical ulcers, and neuropathic ulcers. To select articles for analysis, teams of two independent investigators reviewed titles, abstracts, and articles. Conflicts between investigators regarding inclusion were negotiated. We found insufficient data for meta-analysis but qualitatively summarized studies not amenable to pooling.
Results: Our search retrieved over 10,000 articles. We included 60 studies (62 publications). Most of the studies of advanced wound dressings that regulate moisture, facilitate debridement, include antimicrobial activity, or incorporate putative wound healing accelerants did not demonstrate a statistically higher percentage of wounds healed compared with adequate compression with simple dressings. However, the newer biological dressings containing living cells such as the cellular human skin equivalents showed more rapid healing of venous ulcers (moderate strength of evidence).We could not draw definitive conclusions regarding the effectiveness of advanced wound dressings in terms of intermediate and other final outcomes, including quality of life and pain measures. We found insufficient evidence evaluating the benefits and harms of the routine use of antibiotics. Most venous surgery may not increase the proportion of ulcers healed (low to high strength of evidence), although there was a trend toward greater durability of healing.
Conclusions: These findings do not mean that the interventions do not have value. Rather, the risk of bias and lack of adequate sample size prevented us from establishing statistically valid conclusions. Many of the studies did not report statistical analyses beyond simple healing rates, stratification or adjustment to account for potential confounding variables, or sample size calculations. Many of the studies reviewed were small and therefore had limited power. The absence of these critical design elements limited our ability to draw conclusions. We suggest that there be consensus to frame a series of commonly agreed-upon definitions, develop model clinical research approaches, consider mutually agreed-upon schemes to classify patients, quantify healing parameters, and consider the development of research wound healing networks to collect sufficient number of patients to produce valid conclusions.