Read Beyond the Abstract - Results of Poorly Designed Studies are Irrelevant

As busy wound clinicians, it is easy for us to base decisions upon evidence obtained by reading abstracts summarizing research, rather than by reading the entire article. This can be a dangerous shortcut, because sometimes abstracts do not accurately reflect the content of the accompanying article. For example, this article's abstract seems to imply that polymeric membrane dressings (brand name, PolyMem?) were the least preferred of the three test dressings, and yet the full-length article concludes with this statement, "In the meantime, we will continue using polymeric membrane dressings as our first line choice deliberating their lower costs and good wound healing performance."(1)

A letter to the editor(2) warned that the article revealed many clear study design flaws. These include: donor site wound closure was assessed only at 6 months post-op (unsurprisingly, all were equally closed by then); pain was assessed by an unblinded physician palpating each wound area; infection was not distinguished from inflammation by the authors and was defined simply as increased secretions; fully 50% of patients had a wound which was identified as infected and removed from the study protocol, but their results were averaged in with the dressing group to which they were initially assigned; dressings were left stapled in place far longer than the study protocol permitted, leaving the wounds dressed with the only study dressing designed to promote moist healing dramatically oversaturated... the list goes on. Any one of these problems should be reason enough to reject a manuscript. However, even though the editorial board was presented with an analysis prior to publication detailing 6 CONSORT guideline violations, 12 methodological flaws which ensured bias, 12 contradictions and factual errors within the report itself, and 8 citation errors (authors' assertions were contradicted by their references, or the reference was in no way related to the statement it supposedly supported), the article was published in a prestigious wound-oriented journal.(1) Although a cursory reading of the article quickly revealed that the study was poorly designed and implemented, none of these serious flaws could be discerned from reading the abstract.(1)

Fairly often, in-vitro studies are conducted to evaluate silver-containing dressings. PolyMem Silver contains a powerful continuous wound cleansing system that breaks the bonds adhering contaminants to the wound bed and pulls them up into the dressing.(3) This unique design feature allows PolyMem Silver to avoid silver toxicity issues: microbes are killed when they are drawn into contact with silver locked into the dressing.(4,5,6) Unsurprisingly, PolyMem Silver is consistently shown to be by far the least toxic of all antimicrobial dressings.(4,5,6) However, some researchers claim that PolyMem Silver is ineffective, simply because appreciable quantities of silver are not released into bacteria-laden petri dishes or broth. These studies are often designed inappropriately out of ignorance, but in any case, reading only the abstract will not alert the reader to the flawed study design.

At least three separate wound research studies have been formally criticized because their study designs were clearly biased against one or more of the test dressings.(7,8,9) Two of the studies, published in WOUNDS in 2015 and 2018, were retracted because the researchers reported the results erroneously and their study designs' flaws could not by explained by ignorance. My criticism of the third, published in a peer-reviewed Letter to the Editor of ASTR, includes this statement, "Study results will not translate into real world settings when a comparator is used inappropriately. For example, when negative pressure is compared with conventional dressings, researchers would not omit suction from both arms of the study and use the same dressing change intervals, because adding uniformity in these aspects of the study methods would yield meaningless results."(9)

Evidence-based practice demands more than mere lip service. If the field of wound care is to advance scientifically, we must all take responsibility and examine the evidence carefully. When researchers use dressings contrary to the Instructions for Use, or clearly disregard the way the dressings are intended to function, it is our responsibility to let the journal's editors and readers know that the results are not reliable. If journals regularly receive letters to the editor criticizing articles with clearly flawed study designs, they will instruct peer reviewers to be more attentive to the relevance of the study design when evaluating manuscripts. And, hopefully more editorial committees will have the moral courage to retract the worst offenders, as WOUNDS did.(10,11)

References:

(1) Schulz A, Depner C, Lefering R, Kricheldorff J, K?stner S, Fuchs PC, et al. A prospective clinical trial comparing Biobrane? Dressilk? and PolyMem? dressings on partial-thickness skin graft donor sites. Burns. 2016 Mar 1;42(2):345–55.

(2) Benskin LLL. A prospective clinical trial comparing Biobrane(?)Dressilk(?), and PolyMem(?)dressings on partial-thickness skin graft donor sites raises concerns. Burns. 2016 Aug;42(5):1151–2.

(3) Benskin LL. Polymeric Membrane Dressings for Topical Wound Management of Patients With Infected Wounds in a Challenging Environment: A Protocol With 3 Case Examples. Ostomy Wound Manage. 2016 Jun;62(6):42–50.

(4) Burd A, Kwok CH, Hung SC, Chan HS, Gu H, Lam WK, et al. A comparative study of the cytotoxicity of silver-based dressings in monolayer cell, tissue explant, and animal models. Wound Repair Regen. 2007 Feb;15(1):94–104.

(5) Lee YK. Author Response: Degree of the hazards of silver-containing dressings on MRSA-infected wounds in Sprague-Dawley and streptozotocin-induced diabetic rats. Wounds. 2015 Jul;27(7):1 p following 198.

(6) Kempf M, Kimble RM, Cuttle L. Cytotoxicity testing of burn wound dressings, ointments and creams: A method using polycarbonate cell culture inserts on a cell culture system. Burns. 2011 Sep 1;37(6):994–1000.

(7) Benskin L. Letter to the Editor: Degree of the hazards of silver-containing dressings on MRSA-infected wounds in Sprague-Dawley and streptozotocin-induced diabetic rats. Wounds. 2015 Jul;27(7):A10.

(8) Benskin L. Study on Comparison of DNA Damage and Apoptosis is too Flawed to be Clinically Relevant. Wounds. 2018;30(9):282.

(9) Benskin LL. Methods used in the study, Evaluation of a polyurethane foam dressing impregnated with 3% povidone-iodine (Betafoam) in a rat wound model, led to unreliable results. Annals of Surgical Treatment and Research. 2018 Oct 1;95(4):230–2.

(10) Yeo ED, Yoon SA, Oh SR, Choi YS, Lee YK. Degree of the hazards of silver- containing dressings on MRSAInfected wounds in Sprague-Dawley and streptozotocin-induced diabetic rats. Wounds. 2015 Apr;27(4):95–102.

11. Choi YS, Gwak H-C, Park JK, Lim JY, Yeo ED, Park E, et al. Comparison of DNA Damage and Apoptosis Induced By Silver Nanoparticle-containing Dressing Materials During Wound Healing. Wounds. 2018 Apr 13;