Diabetic Foot Ulcer: An Evidence-based Treatment Update

Liza R. Braun; Whitney A. Fisk; Hadar Lev-Tov; Robert S. Kirsner; Roslyn R. Isseroff

Am J Clin Dermatol. 2014;15(3):267-281.

Abstract and Introduction

Abstract

Background. Diabetic foot ulcers (DFUs) are extremely debilitating and difficult to treat. Multidisciplinary management, patient education, glucose control, debridement, offloading, infection control, and adequate perfusion are the mainstays of standard care endorsed by most practice guidelines. Adjunctive therapies represent new treatment modalities endorsed in recent years, though many lack significant high-powered studies to support their use as standard of care.

Objective. This update intends to identify recent, exclusively high level, evidence-based evaluations of DFU therapies. Furthermore, it suggests a direction for future research.

Methods PubMed, Embase, Ovid Technologies, CINAHL, Cochrane, and Web of Science databases were systematically searched for recent systematic reviews published after 2004, and randomized controlled trials published in 2012–2013 that evaluated treatment modalities for DFUs. These papers are reviewed and the quality of available evidence is discussed.

Results A total of 34 studies met inclusion criteria. Studied therapies include debridement, off-loading, negative pressure therapy, dressings, topical therapies, hyperbaric oxygen therapy, growth factors, bioengineered skin substitutes, electrophysical therapy, and alternative therapy. Good-quality evidence is lacking to justify the use of many of these therapies, with the exception of standard care (offloading, debridement) and possibly negative pressure wound therapy.

Limitations. There is an overall lack of high-level evidence in new adjunctive management of DFU. Comparison of different treatment modalities is difficult, since existing studies are not standardized.

Conclusions. Many therapeutic modalities are available to treat DFU. Quality high-level evidence exists for standard care such as off-loading. Evidence for adjunctive therapies such as negative pressure wound therapy, skin substitutes, and platelet-derived growth factor can help guide adjunctive care but limitations exist in terms of evidence quality.

1 Introduction

In the USA, diabetes mellitus (DM) afflicts 9.9% of the population over 40 years of age, of which 30% suffer from lower extremity disease.^[1] Development of a diabetic foot ulcer (DFU) is associated with staggeringly high mortality rates of 16.7% at 12 months and 50% at 5 years—rates comparable to mortality rates of colon cancer.^[2] Furthermore, patients with DM and new-onset DFU have significantly reduced survival rates compared with age- and sex-matched controls with DM but without DFUs (72 and 86% 3-year survival, respectively).^[3] In the USA, healthcare costs are estimated to be 5.4 times higher in the first year after a diagnosis of DFU than for patients with DM without an ulcer.^[2,3] Therefore, management and intervention of patients with DM and DFU must be adequately addressed before onset of severe complications. Unfortunately, DM is associated with a 15–25% lifetime risk for developing DFU,^[2] and once ulceration occurs, healing is difficult and lower extremity amputations (LEAs) common. Nearly 90% of patients undergoing LEA have a history of a DFU, and LEAs are associated with a doubling in the risk of death. In 2008, more than 15,000 LEAs were performed on Medicare beneficiaries with DM, with the annual incidence of LEA in Medicare beneficiaries with DM estimated at 5 per 1,000 in 2006 and 2007, and 4 per 1,000 in 2008. The cost to treat those who have had LEA is excessive, with an average $US52,000 reimbursed annually for a Medicare beneficiary with DM and a LEA from 2006–2008.^[4–18]

Fortunately, there are growing efforts towards international consensus on management and rapid communication on enhancing standard of care and reviewing novel therapies. These therapies address various mechanisms of DFU formation in order to achieve wound healing. DFU standard of care is critical; however, for those not responding to standard care, new adjunctive modalities may provide opportunities for healing. Yet, while treatment options have expanded in recent years, the cost effectiveness and efficacy of these modalities remain in question.

This review intends to identify recent evidence-based evaluations of all DFU therapies, focusing exclusively on high-level evidence. Furthermore, it identifies gaps in current data and suggests direction for further investigation.

1 2 3 4

Next Section

Table 1. Summary of studies and systematic reviews for the treatment of diabetic foot ulcer.
References	Study design	Number of DFU-specific trials/participants	Intervention	Major conclusions
Off-loading
Lewis and Lipp [25]	Cochrane SR	14 RCTs 709 participants	1) Removable devices 2) Non-removable devices 3) Surgical procedures	1) Non-removable devices are associated with a significantly higher proportion of healed ulcers vs. removable 2) Surgical procedures were more effective than non-surgical comparators but are costly and long-term benefits are not proven
Bus et al. [28]	SR	21 controlled studies	1) Removable devices 2) Non-removable off-loading devices	1) TCCs are more effective than removable off-loading devices but this may be due to better compliance and more limited physical activity 2) Therapeutic footwear should not be used, as more effective modalities are available
Morona et al. [29]	MA	13 (11 RCTs)	1) Removable devices 2) Non-removable devices	1) Non-removable devices are more effective than removable but this may be due to better compliance 2) Two types of non-removable devices found to be equally effective 3) Little evidence to suggest a significant difference in adverse events or infection rates between the methods
Debridement
Edwards and Stapley [32]	Cochrane SR	6 RCTs; sample size 31–172	1) Surgical debridement 2) Hydrogels 3) Larvae therapy	1) Hydrogels significantly more effective in healing DFUs (RR 1.84, 95% CI 1.3–2.61) 2) Surgical debridement showed no significant benefit over standard treatment
Lebrun et al. [33]	SR	5; sample size 20–241	Surgical debridement	No significant benefit of surgical debridement over standard wound care based on limited evidence
Hinchcliffe et al. [34]	SR	3 studies	1) Sharp debridement 2) Larvae therapy	1) Weak evidence supports the use of sharp debridement based on a subgroup analysis of a single RCT 2) Two studies of larvae therapy reported improved healing at 2 weeks and decreased incidence of major amputation in bed- or wheelchair-bound subjects with peripheral arterial disease
Game et al. [35]	SR	2 studies	1) Larvae therapy (Lucilia cuprina) 2) Versajet® hydrosurgery	1) Larvae therapy did not significantly improve healing or amputation risk 2) Versajet® decreased wound debridement time but did increase proportion of wounds healed at 12 weeks
Dressings
Dumville et al. [37]	SR	15 studies	Advanced wound dressings	1) Insufficient high-quality data to support the use of more expensive or advanced dressings over basic wound dressings
Dumville et al. [38]	MA	5 studies 535 participants	Hydrocolloid dressings	1) Insufficient high-quality data to support the use of hydrocolloid dressings over other types of wound dressings
Dumville [39]	MA	6 studies 157 participants	Foam dressings	1) Insufficient high-quality data to support the use of foam dressings over other types of wound dressings
Dumville et al. [40]	MA	6 studies 375 participants	Alginate dressings	1) Insufficient high-quality data to support the use of alginate dressings over other types of wound dressings
Dumville et al. [41]	MA	5 studies 436 participants	Hydrogel dressings	1) Hydrogel dressings may be more effective in low grade DFUs 2) No studies compared hydrogel dressings vs. other advanced dressings
Holmes et al. [42]	SR	26 studies 2,386 participants	Collagen dressings	While studies suggest potential benefit for collagen dressings, there is no evidence that they should replace DFU SOC
Hinchliffe et al. [34]	SR	9 studies	Advanced wound dressings	1) There is little evidence to support silver-containing dressings 2) Hydrogels have minimal statistically significant findings, though more support is needed
Game et al. [35]	SR	2 RCTs	Advanced wound dressings	1) No new studies to reinforce evidence from prior review on use of hydrogels in DFUs 2) Two methodologically sound RCTs were unable to demonstrate value in hydrofiber dressings vs. other dressings
Topical therapies
Jan et al. [44]	RCT	100 patients	Honey	1) Honey topical therapy reduced recovery time (p=0.0001) in DFUs vs. povidone iodine topical therapy. Healing and amputation rates were not statistically significant between treatment groups
Shaw et al. [45]	SR	14 studies 2 DFU-specific studies	Phenytoin	1) Only one of the two DFU studies demonstrated statistically significant reduction in wound size with phenytoin vs. controls
Balangit et al. [46]	RCT (phase II)	77 patients	Angiotensin analogue (NorLeu-A[1–7]	0.03% NorLeu3-A [1–7] therapy healed DFUs on a median of 8.5 weeks vs. placebo at a median of 22 weeks (p = 0.04)
Voigt and Driver [47]	MA	9 studies 4 DFU-specific studies	HA scaffolding plus keratinocytes	1) In DFU-specific studies, HA scaffolding plus keratinocytes vs. SOC at 12 weeks did not show statistically significant improvement, though data demonstrated a trend towards healing 2) HA matrix vs. SOC at 12 weeks showed improved healing rates with fewer non-healed ulcers in the HA group
Hinchliffe et al. [34]	SR	6 studies	Topical applications	1) Evaluating use of phenytoin, trans-retinoic acid, collagen, and iodide in various studies did not provide high-quality evidence due to poor methodology
Game et al. [35]	SR	2 studies	Topical applications	Two studies evaluating honey and salicylic acid were insufficiently designed to determine benefit to wound healing
Electrophysical therapy
Kwan et al. [49]	MA	8 RCTs 325 participants	Electrical stimulation, phototherapy, ultrasound	1) Primary outcome of healed ulcers showed statistically significant benefit with electrical stimulation 2) Harmful side effects were reported in ultrasound studies 3) Larger studies are warranted before these therapies become SOC
Hinchliffe et al. [34]	SR	6 studies	Electrical stimulation, ultrasound, normothermic, magnetic, laser	Only one RCT on electrical stimulation was well powered enough to show healing at 12 weeks
Game et al. [35]	SR	4 studies	Electrical stimulation, shockwave therapy	1) Two studies evaluating electrical stimulation effect on ulcer reduction at 4 weeks were found to be methodologically weak 2) Two studies on shockwave therapy showed statistically significant findings; however, re-analysis of raw data and small sample size places study results in question
Negative pressure therapy
Medical advisory secretariat [50]	SR	2 RCTs Participants: 342 (DFU); 162 (amputation wounds)	NPWT (KCI system)	1) Proportion of patients who achieved complete ulcer closure in NPWT was significantly greater than in controls in both studies 2) Time to complete healing was significantly shorter in NPWT groups vs. controls
Noble-Bell and Forbes [51]	SR	4 RCTs 206 participants; sample size range 10–162	NPWT	1) Proportion of patients who achieved complete ulcer closure in NPWT was significantly greater than in controls in the two studies that evaluated this outcome 2) Time to complete healing was significantly shorter in NPWT groups vs. controls
Vig et al. [52]	SR and EPR	9 studies (excluding nonanalytic studies)	NPWT	1) NPWT should be considered in an attempt to prevent amputation or re-amputation 2) NPWT must be considered as an advanced wound care therapy for postoperative Texas grade 2 and 3 diabetic feet without ischemia 3) NPWT must be considered to achieve healing by secondary intention 4) NPWT should be stopped when wound has progressed suitably to be closed by surgical means
Hinchliffe et al. [34]	SR	3 RCTs	NPWT	Significant improvements in healing rate and healing time associated with NPWT (included chronic DFUs and post-amputation wounds in diabetic patients)
Game et al. [35]	SR	6 studies	1) NPWT (N = 3) 2) Compression therapies (N= 3)	1) Two methodologically sound RCTs reported reduced healing time, increased incidence of healing (at 16 weeks) and reduced risk of minor amputation in participants using NPWT 2) Three studies of compression therapies (pneumatic compression, vacuum compression) were of poor methodological quality but suggested improvement in healing of post-operative wounds
Platelet-rich plasma
Martinez-Zapata et al. [53]	MA	9 RCTs 325 participants 2 DFU-specific RCTs	PRP	1) No statistically significant difference between the PRP and control in DFUs (RR 1.16; 95% CI 0.57–2.35) 2) No significant difference of PRP vs. control by ulcer etiology or by the procedure used to obtain autologous PRP
Villela and Santos [54]	MA	18 studies 5 DFU studies	PRP	1) PRP may promote healing of DFUs but only when used as an adjuvant to other therapies in a multidisciplinary approach 2) Meta-analysis was unable to establish a reference value for PRP concentration
Hinchliffe et al. [34]	SR	5 studies	PRP	No apparent benefit from PRP therapy was discernible due to limits of sample size, poorly established endpoints, and elaborate exclusion criteria
Game et al. [35]	SR	1 study	PRP	1) Only one additional RCT found improved healing at 12 weeks, time to heal, and percent area reduction 2) Inclusion and exclusion criteria were unclear and suspicious healing rate given initial high incidence of bone exposure in pretreatment wounds
Cultured keratinocytes
You et al. [55]	RCT (phase III)	59 patients	Cultured keratinocytes	1) 100% of DFUs treated with cultured keratinocytes were completely healed vs. 69% of control group DFUs (p=0.05). 2) Experimental group also had shorter time to heal on average (35 days) vs. control (57 days)
Hinchliffe et al. [34]	SR	1 RCT	Cultured keratinocytes	A single study was found; however, missing data did not allow for interpretation of results, and study was determined to be very poorly designed
Game et al. [35]	SR	1 RCT	Cultured keratinocytes	Poor methodological value; small sample size, missing data
Growth factors and bioengineered skin substitutes
Buchberger et al. [56]	SR	14 RCTs and 2 meta-analyses (9 growth factors, 4 BSS); sample size 17–382 patients	1) Growth factors (becaplermin, rhEGF, bFGF) 2) BSS	1)Adjunctive therapy with becaplermin and rhEGF improves likelihood of ulcer healing 2) bFGF not shown to improve wound healing 3) Adjuvant therapy with BSS improves wound healing
Teng et al. [57]	MA	7 RCTs 880 participants; sample size range 17–245	BSS	1) BSS group was significantly more likely to have complete wound healing than control group (p=0.0001) 2) No significant difference in recurrence rates between treatments (OR = 0.87, 95% CI 0.50–1.52; p = 0.63)
Barber et al. [58]	SR	9 RCTs	BSS	Apligraf (n = 1), Dermagraft (n = 3), and graft jacket (n = 1) significantly more effective than standard therapy in terms of proportion of patients achieving complete wound healing. Promogran and hyalograft combined with laser skin not more effective than control therapy
Blozik and Scherer [59]	MA	7 RCTs 817 participants	BSS and allografts	Pooled analysis of grafts and skin substitutes revealed these were slightly superior to standard care (OR 1.46, 95% CI 1.21–1.76)
Hinchliffe et al. [34]	SR	8 studies (4 growth factors, 4 BSS)	1) Growth factors (bFGF, EGF) 2) BSS	1) No significant improvement in wound healing with adjunctive therapy with bFGF. EGF improved healing rates of ulcers vs. placebos but did not improve proportion of ulcers healed by 16 weeks 2) Dermal fibroblast culture and fibroblast/keratinocyte co-culture may increase healing and healing rate of DFUs but results are conflicting
Game et al. [35]	SR	4 studies (3 growth factor RCTs, 1 BSS RCT)	1) Growth factors (rhEGF, bFGF) 2) BSS	1) No high-quality evidence to support the use of rhEGF or bFGF for treatment of DFUs 2) Study of BSS Apligraf® suggested improved healing with use of BSS, but study was stopped prematurely, limiting conclusions that can be drawn from study
HBOT
Kranke et al. [61]	MA	8 RCTs 455; sample size 18–100	HBOT	1) Increased proportion of healed ulcers vs. control (RR 5.20, 95% CI 1.25–21.66; p = 0.02) in short-term follow-up, but this benefit may not persist in 1-year follow-up (RR 9.53, 95% CI 0.44–207.8; p = 0.15) 2) No statistically significant reduction in amputation rate with HBOT (RR 0.36, 95% CI 0.11–1.18; p = 0.08)
Goldman [62]	SR	13 prospective and retrospective clinical trials; sample size range 10–641	HBOT	Increased rates of complete wound healing (OR 11.64, 95% CI 3.457–39.196) and decreased risk of amputation in patients with DFU (OR 0.236, 95% CI 0.133–0.418) with HBOT vs. standard wound therapy alone
Liu et al. [63]	MA	13 prospective and retrospective clinical trials 624 participants; sample size 10–100	HBOT	1) Significantly higher proportion of healed diabetic ulcers at both short- and long-term follow-up (RR 2.33, 95% CI 1.51–3.60) 2) Reduced risk of major amputations (RR 0.29, 95% CI 0.19–0.44) but not minor amputations (p = 0.30)
Ontario Health Quality [64]	SR	4 SRs	HBOT	Insufficient evidence to determine whether HBOT improves likelihood of healing, healing rates, or risk of amputation
Roeckl-Wiedmann et al. [65]	SR	5 RCTs 175 participants	HBOT	1) HBOT decreases the risk of major amputation vs. controls by 1/3 (RR 0.31, 95% CI 0.13–0.71) 2) Trend toward greater ulcer healing with HBOT (RR 4.78, 95% CI 0.94–24.24)
Hinchliffe et al. [34]	SR	6 studies	Local (n = 2) and systemic HBOT (n = 4)	Systemic HBOT may decrease the likelihood of major amputations, but reviewed studies were methodologically flawed
Game et al. [35]	SR	3 studies	HBOT	Only one study was determined to be of high methodological quality and showed that participants receiving HBOT therapy were significantly more likely to heal within 12 months (p = 0.03)
Alternative therapy
Chen et al. [68]	MA	6 studies 439 participants	Oral Chinese herbal formulations (different herbs used in each study)	CHM combined with standard therapy significantly increased the proportion of healed ulcers vs. standard therapy alone risk ratio (RR), 0.62 [95% CI 0.39–0.97]
Game et al. [35]	SR	3 studies	1) Oral Chinese herbal formulation 2) ANGIPARS™ herbal extract	1) Oral Chinese herbal formulation did not improve healing of necrotic/gangrenous ulcers vs. placebo 2) Two studies of ANGIPARS™ suggested that topical, oral, and IV preparations may improve healing but both studies were methodologically flawed

bFGF basic fibroblast growth factor, BSS bioengineered skin substitutes, CHM Chinese herbal medicine, CI confidence interval, DFU diabetic foot ulcer, EGF epidermal growth factor, EPR expert panel recommendations, HA hyaluronic acid, HBOT hyperbaric oxygen therapy, IV intravenous, MA meta-analysis, NPWT negative pressure wound therapy, OR odds ratio, PRP platelet-rich plasma, rhEGF recombinant human epidermal growth factor, RCT randomized controlled trial, RR relative risk, SOC standard of care, SR systematic review, TCC total contact casts

Table 2. Summary of quality of evidence for the treatment of diabetic foot ulcer. ^a
Therapeutic intervention	Quality of evidence
Off-loading	Moderate quality
Debridement	Moderate quality
Dressings	Insufficient evidence
Topical therapies	Moderate quality
Electrophysical therapy	Moderate quality
Negative pressure wound therapy	Moderate quality
Platelet-rich plasma	Moderate quality
Cultured keratinocytes	Moderate quality
Growth factors	Moderate quality
Bioengineered skin substitutes	Moderate quality
Hyperbaric oxygen therapy	Moderate quality
Alternative therapy	Insufficient evidence