The use of hydrogen peroxide in the treatment of burn wound infection: a systematic review, and survey of current clinical practice in the United Kingdom

Abstract

Background: Current treatment of burn wound infection (BWI) is with antibiotics and/or wound cleaning/superficial debridement. The overuse of antibiotics has contributed to antibiotic resistance. One possible solution is the use of hydrogen peroxide (H₂O₂). The aim of this study is to investigate the current use of H₂O₂ in the treatment of BWI through a comprehensive review of published evidence and a survey of current clinical practice. Methods: A systematic review was performed on the clinical use of H₂O₂ in the treatment of BWI using four major search engines from inception to 1^st July 2018. English-written full-text publications of any study design were included and data extraction was conducted in duplicate. An 11-question survey on the use of H₂O₂ in the treatment of BWI was sent to all burn services in the United Kingdom (UK). Results: The systematic review generated 1,168 papers, with only one fulfilling inclusion criteria. This was a randomised control trial (RCT) which demonstrated that soaking grafts with 2% H₂O₂ prior to grafting improved graft take rate in infected burn wounds compared with grafts treated with saline prior to grafting, concluding that H₂O₂ can be recommended in the treatment of BWI intraoperatively. A 72.7% (16 burns services) response rate was achieved to the survey. Of these, 75% of burn services (n = 12) do not currently use H₂O₂ in clinical practice. Of the 25% (n = 4) which do use H₂O₂, no service had a protocol for its use. The most common reasons for not using H₂O₂ were a lack of published evidence and fear of side-effects. Conclusion: Only 1 paper suggests H₂O₂ to be effective in BWI treatment and there is no national consistency or protocol for the use of H₂O₂ in the treatment of BWI in the UK. More large-scale research is required to determine whether H₂O₂ may offer a solution to the need to use antibiotics to treat BWI.

Background

Burn wound infection (BWI) usually occurs when bacterial load reaches a critical colonisation threshold, provoking a host immune response and stagnating the wound healing process [1,2]. There is limited published data on the incidence of BWI, but evidence estimates that it is 10-20%, with the most frequent infectious organisms being Pseudomonas aeruginosa and Staphylococcus aureus [3-6]. BWI may lead to delayed healing, and results in reduced immune responses, predisposing to burn wound sepsis, and mortality rates from sepsis related to BWI of up to 75% have been reported [1,7-10]. Immune deficit, along with loss of the protective skin barrier, results in increased susceptibility to BWI, with the size of the wound proportionally related to the risk of infection [11,12].

Current treatment of BWI is with antibiotics and/or cleansing/superficial debridement [13]. While antibiotics are the mainstay of treatment, overuse will increase the incidence of bacterial resistance, especially that of multi-drug resistant (MDR) bacteria. Broad-spectrum antibiotics are prescribed upon suspicion of BWI and narrowed after microbiological diagnosis. Thus broad-spectrum antibiotics are necessarily overused to treat and prevent complications from BWI early [14]. When gram-negative MDR infection is present in a burn wound, even the most effective narrow-spectrum antibiotics are based on over 20-year-old molecular structures and are becoming gradually less effective, with low likelihood of newer products in the near future [6]. This is worrying given that patients whose BWIs are colonized with MDR organisms undergo more aggressive antibiotic courses, longer hospital stays, and more surgical operations as compared to patients whose burn wounds are infected with susceptible bacteria [15]. For these reasons, and due to the risk of antibiotic use promoting resistance, there is a need for alternative antimicrobial treatments.

One method for reducing the use of antibiotics to treat BWI is the use of hydrogen peroxide (H₂O₂). H₂O₂ is a reactive oxygen species produced via the respiratory chain cascade and intracellular enzyme reactions, such as those involved in protein folding [16]. It is involved in hypoxic signal transduction and mediates immune responses [17]. The mechanism of its antiseptic action is through oxidative stress: the formation of free hydroxy radicals which lead to the oxidation of thiol groups in bacterial proteins and enzymes [18]. In the aforementioned commonly occurring organisms in BWI, Pseudomonas aeruginosa and Staphylococcus aureus, de-repression of glycolytic genes and subsequent glycolysis induction by H₂O₂ leads to their destruction in vitro [19]. The effects of H₂O₂ are dose-dependent, with low concentrations of H₂O₂ supporting wound healing whilst high concentrations impair wound healing [20]. Whilst in-vitro research had investigated the efficacy of H₂O₂ as a wound healer and cleanser, at present there are no published surveys or systematic reviews on the current use of H₂O₂ in the treatment of BWI.

The purpose of this paper is to investigate the current use of H₂O₂ in the treatment of BWI.

Methods

Survey

A comprehensive list of burn services in the UK was obtained from the British Burns Association (BBA). The survey was comprised of 11 questions (Appendix A). The questions aimed to determine how and by whom H₂O₂ is used clinically, and, where it is not used, why it is not used. Each service was contacted by phone to obtain the name of the lead specialist for burns management. In 80% (n = 4) of services which do use H₂O₂, this is by a consultant plastic surgeon. In 20% (n = 1), this is either a plastic surgeon or lead specialist nurse. The lead, or their secretary, was emailed the survey, which was completed on a Microsoft Word document and returned via email.

Systematic review

This review focuses on the current use of H₂O₂ in the treatment of BWI. It follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.

Study eligibility

Types of studies

Due to the limited literature published on this topic, any English-written full-text publications of any study design which focused specifically on the clinical use of H₂O₂ in the treatment of BWI in human participants were included.

Types of participants

All studies recording outcomes for patients of any age, irrespective of background or comorbidity, who required treatment for cutaneous BWI as assessed by clinical or objective investigation and who were treated with H₂O₂ were included. Any type of burn wound (thermal, mechanical, chemical, electrical, or radiation) of any depth (full or partial thickness) of any size were included. As H₂O₂ could offer a potential alternative to the use of antibiotics, only bacterial organisms treated with H₂O₂ were included, however no specific bacterial organism infecting the burn wound was required to meet inclusion criteria.

Type of intervention

Any intervention, surgical (e.g. Versajet^TM) or non-surgical, which utilised pure H₂O₂ were included.

Identification of studies

A systematic search was conducted using Medline 1946-1^st July 2018, EMBASE 1974-1^st July 2018, CINAHL 1937-1^st July 2018, and The Cochrane Database of Systematic Reviews. This period was chosen as a prior literature search showed very little evidence on the clinical use of H₂O₂ for the treatment of BWI and thus a broad timescale would more likely generate relevant publications, if any. The starting period was the inception of each respective search engine. Abstracts and titles were the fields selected to be searched in all data bases. A full search strategy is found in Appendix B.

Study selection

All titles derived from the four databases were compiled. Screening of all titles and abstracts against eligibility criteria was undertaken independently by one author (THM) trained in systematic review methodology. Studies were then excluded according to reasons outlined in the PRISMA Flow Diagram illustrated below in Figure 1. All articles were assessed according to the PICO method.

Figure 1.

A PRISMA flow diagram of study selection.

Data extraction

Data was extracted into an Excel spreadsheet using Microsoft Excel. Author, publication year, study design, number of participants per study, method of H₂O₂ application, and outcome in each study were recorded. There was no prescribed limit with regards to amount of time after injury that treatment was received or that outcome was measured.

Results

Survey

From the burn services across the UK which responded to the survey (Table 1), a number of important conclusions are drawn. 16 out of 22 burns services (72.7%) responded to the survey. Of these, 75% do not currently use H₂O₂ in the treatment of BWI and 25% do. Antibiotics, debridement, and/or Versajet^TM with 0.9% normal saline are the mainstay of hospital protocol in those services which responded.

Table 1.

A table documenting the 16 responses of adult and paediatric burn services in the United Kingdom as obtained from the british burns association to questions of the survey

Centre	Use H2O2?	Adjunct?	Time Limit?	Wound Swab Prior?	Who?	Where?	Concentration?	Volume
1	×	N/A	N/A	N/A-routine	N/A	N/A	N/A	N/A
2	√	√	At initial presentation	√	Plastic Surgeon	Theatre, GA*	Diluted 2:1 with N/S**	Minimal
3	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
4	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
5	√	√	×	N/A	Plastic Surgeon	N/A	Diluted 2:1 with N/S	Minimal
6	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
7	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
8	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
9	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
10	√	√	×	√	Plastic Surgeons and Lead Nurses		N/A
11	Not routine	N/A	N/A	N/A	Plastic Surgeon	Theatre, GA	3% diluted in saline or betadine	200 or 400 ml
12	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
13	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
14	√	√	×	×	Plastic Surgeon	Theatre, GA	Diluted 50:50 with N/S	500 ml max.
15	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A
16	×	N/A	N/A	N/A	N/A	N/A	N/A	N/A

NB: details of individual services have been removed to preserve anonymity;

^*GA = general anaesthetic;

^**N/S = normal saline.

Amongst services which do not use H₂O₂, an important reason as to why not is for fear of gas embolism, a rare, but fatal, side effect, well documented in the current evidence base on the use of H₂O₂ in wounds in general (but not in cutaneous BWI; see discussion). Other centres use alternative effective antiseptics in clinical practice, such as Flaminal, Prontosan or sliver dressings and hypochlorite, which are used on wards or in out-patient settings, negating the need and risks of general anaesthetic (GA). This is important as one burns service deemed H₂O₂ to be too painful for use if not under GA, with less painful alternatives available for the non-anaesthetised patient which, once more, negates the risks associated with GA. One service stated H₂O₂ is not utilised in its department due to the documented potential detrimental effects of H₂O₂ on keratinocytes, which may impede healing. In those patients who are anaesthetised and on whom H₂O₂ is used, it is not used for deeper infected burn wounds as excision is deemed superior to any other form of cleaning. Rather, H₂O₂ is used for disinfection of more superficial wounds. Finally, one service remarked that they do not use H₂O₂ as there is very little published clinical evidence available on its use.

In those services which do use H₂O₂ for BWI, information was collated regarding the way in which it is utilised. In response to question 4 which asked on which patients H₂O₂ is used and whether or not there was an established protocol, answers highlight that there is no established/specific protocol and the use of H₂O₂ is nurse/surgeon-dependent. Amongst services which use H₂O₂, it is used by plastic surgeons or trained lead nurses under GA as an adjunct to treatment (such as with antibiotics/debridement). However, one service uses Flaminal which, while it itself is not pure H₂O₂, contains an enzyme system which subsequently leads to the production of minute quantities of H₂O₂ on the wound surface. Wound swabs are not performed prior to use of H₂O₂ and therefore its use is not based on specific burn wound microbiology (swabs are performed routinely at an early stage where BWI is suspected).

This survey also highlighted that H₂O₂ is used differently across different services. In one service, it is used as an adjunct for debridement of heavily contaminated wounds only. There is no fixed limit on the amount of time for which H₂O₂ is used. In services where H₂O₂ is used, the volume used is up to a maximum of 500 ml and this is administered with abdominal gauze.

Systematic review

Included studies

The search strategy generated 416 papers from Medline, 663 from EMBASE, 57 from CINAHL, and 32 from The Cochrane Database of Systematic Reviews-a total of 1,168.744 articles remained following removal of 424 duplicates. 629 irrelevant titles were excluded which did not refer to either H₂O₂ or wounds. The remaining 115 underwent full-text review and 114 were excluded as they did not meet the study inclusion criteria. Formal statistical analysis could not be performed due to the small number of studies generated: only one study answered answer the research question (Appendix C). This was a randomised control trial (RCT) conducted by Mohammadi et al. [21].

The aim of their study was to investigate the efficacy of debridement and wound cleansing with H₂O₂ of infected burn wounds on skin graft take. 49 patients with infected burn wounds were included. The burns of these patients were symmetrical and on two limbs (either arms or legs). Infected, right-sided arm or leg burn wounds were regarded as the intervention limb, whilst infected, left-sided limbs were the control. In the intervention limbs, the wounds were soaked with 2% H₂O₂ and washed with saline prior to receiving a graft. This group was thus termed “the hydrogen peroxide group”. The control group received the standard treatment of saline lavage, debridement, followed by skin grafting, but were not soaked in H₂O₂. Graft take was significantly higher at 82.85% in the intervention limb as compared with 65.61% in the control limb. The authors concluded that, given its antiseptic properties, H₂O₂ significantly improves graft take rate in infected burn wounds and that it should thus be used in the treatment of BWI intraoperatively.

Discussion

The survey aimed to establish how and by whom H₂O₂ is used currently, and, where it is not used, why it is not used. This is the first published survey to do so. From its findings, only 25% of the 16 burns services which use H₂O₂ responded and none have an established protocol on its use. This finding corroborates the findings of the systematic review which revealed a lack of clinical research investigating H₂O₂ in BWI treatment. There is thus a gap in evidence and the need for a consensus in an established protocol for the use of H₂O₂ in BWI treatment. Limitations to the survey include only 72.7% of burns services responding. Due to the incomplete response rate, there may be knowledge that is lacking about how H₂O₂ is used in the UK. A higher response rate is required in order to achieve a complete understanding of national use. Furthermore, this survey focuses only on the use of H₂O₂ in the UK, which is not a reflection of its use elsewhere.

The systematic review aimed to establish the current use of H₂O₂ in the treatment of BWI. This is the first article to research and review the use of H₂O₂ in the treatment of BWI. This review included four major databases, rendering review of existing literature thorough and comprehensive. The study dates were kept broad so as to not exclude any old publications. Using few and non-specific search terms led to high volumes of papers requiring to be checked manually but resulted in a more thorough review of existing articles e.g. (wound* OR burn* OR “burn wound*”) instead of (“thermal injur*” OR scald OR burn* OR “burn wound*”). The transparency of reporting this search strategy as well as the inclusion of exclusion criteria is also a strength. There are limitations to the methodology of this review. Firstly, only one author undertook independent abstract and title review, which risks introduction of selection bias. Selection bias may also have been introduced through the exclusion of non-English texts, however international publications were included in screening to reduce this risk. Finally, only one publication was found to address the research question: it is not possible to draw conclusions from this small body of evidence.

Whilst extensive literature exists on its in-vitro properties, little focuses on the clinical applicability of H₂O₂. Whilst the RCT included in this review represents high level of evidence, conclusions on the use of H₂O₂ cannot be made from this publication alone and more research is required. As a result, the relevance of studies investigating the use of H₂O₂ in infected wound treatment in general is discussed and, from this, their potential clinical applicability to the treatment of BWI is extrapolated. Suggestions for future research are proposed.

Whilst not pure liquid H₂O₂, some products utilise H₂O₂ in their mechanism of action for treating BWI. Flaminal, for example, is a gel which contains an enzymatic antibacterial complex, namely glucose oxidase and lactoperoxidase which, on contact with the wound surface, metabolises glucose into H₂O₂. Flaminal also contains guaiacol which stabilises H₂O₂, compounding bactericidal activity through cell wall degradation. In vitro and through this H₂O₂ mediation, the Flaminal complex of glucose oxidase, lactoperoxidase and guaiacol (GLG) inhibit biofilm biomass of Staphylococcus aureus, methicillin-resistant Staphylococcus aureus, and Pseudomonas aeruginosa at concentrations of < 0.5% (w/v) and inhibits existing biofilms at greater concentrations [22]. A retrospective cohort study compared 1% silver sulphadiazine (Flammazine; a topical antibiotic cream) with Flaminal in the treatment of partial thickness BWI. Infected burn wounds treated with Flaminal had shorter healing time (17 vs. 24 days) in spite of showing significantly higher bacterial load initially than those wounds treated with 1% silver sulphadiazine [23]. Thus, products relying on H₂O₂-mediated pathways have demonstrated superiority to antibiotic alternatives, a promising sign for H₂O₂ in isolation to be investigated in future clinical research for BWI treatment.

Research has also been conducted on non-thermal cutaneous infected wounds. Whilst the Versajet^TM hydro-surgery system has been trialled in the treatment of wounds, only one study by retrospectively compared the use of Versajet^TM with and without H₂O₂ in the treatment of infected subacute and chronic (non-thermal) wounds [24]. The study arm (n = 60) underwent debridement of their wounds with Versajet^TM with H₂O₂ whilst the control arm (n = 70) underwent standard care (Versajet^TM debridement with normal saline) prior to grafting of the wound. The study showed lower hospital stay time (7.83 vs. 9.86 mean) and significantly improved graft take in the study arm as compared with the control. The authors concluded that a Versajet^TM system used with H₂O₂ holds promising future use in the treatment of infected subacute and chronic wounds. The use of Versajet^TM with H₂O₂ has also been shown to prevent accumulation of bacteria in the Versajet^TM tubing [25]. As highlighted from the survey in this paper, Versajet^TM is currently used as part of BWI treatment in some burns services, but is used with normal saline, not H₂O₂. Could this therefore be extrapolated from the treatment of infected, non-thermal wounds to the treatment of BWI? Large-scale RCTs are required.

In addition to surgery, dressings are an important element in the healing of burn wounds, and one dressing used clinically in wound treatment is Granuflex/DuoDERM. Focusing on not only the antiseptic properties of H₂O₂, the mechanism of action of Granuflex was researched in a murine model by measuring the proliferation rate of firboblasts [26]. When the hydrocolloid granule component of the dressing is applied to fibroblasts in vitro, a significant increase in proliferation was observed. When catalase, an enzyme which inhibits H₂O₂, was added to the cultures, proliferation of fibroblasts was inhibited, indicating that the effects of the Granuflex dressing in wound healing are mediated through H₂O₂. There is thus potential for this to be used in aiding the repair of BWI and large-scale clinical trials in this area are also warranted.

Whilst the amount of research focusing directly on H₂O₂ in BWI is limited, a significant amount of research has been conducted on the use of honey in the treatment of BWI. This is relevant as the effects of honey are largely mediated through generation of H₂O₂ [27]. Using mean healing time and the number of infected burn wounds rendered sterile as the primary outcome measure, a Cochrane systematic review of RCTs found honey to be a more effective therapy for treating BWI as compared to other therapies (silver sulfadiazine) [28]. Similar findings have been found in in vitro studies in which Manuka honey inhibited the growth of 17 strains of Pseudomonas aeruginosa isolated from BWI [29]. Following on from the aforementioned papers, this warrants further clinical research on the use of H₂O₂ as a therapy for BWI in and of its own right.

Whilst preclinical and clinical data available is promising, there have been potential risks to the use of H₂O₂ in wound disinfection, the main risk posed being that of fatal oxygen embolism [30-33]. As highlighted by the survey, this is why certain services did not use H₂O₂. Consequent to several case studies and a literature review having documented this risk, H₂O₂ has been recommended to be avoided as a sole treatment in the irrigation of infected wound treatment in orthopaedic surgery [34]. However, these side effects usually occur in deep or very large wounds, or in the disinfection of closed body cavities, suggesting its use may be safer in cutaneous, more superficial or partial thickness burns [35]. No such side effects were documented in the aforementioned RCT conducted by Mohammadi which concluded that intraoperative use of H₂O₂ appears to be safe for cutaneous BWI. Furthermore, no such side effects were noted in the Versajet^TM study with H₂O₂, and all treatments have potential for harm: determining the safety of H₂O₂ is a reason in and of itself to further research its potential clinical (and preclinical) applicability as a treatment for BWI.

Conclusion

BWI can lead to burn wound sepsis which has significant impact on morbidity and mortality, and new methods of treating BWI are required to reduce the use of antibiotics in light of increasing resistance. One possible solution is H₂O₂. This article has identified how H₂O₂ is currently used clinically in the UK and the lack of evidence-base on its current use. Whilst there is only one publication on its use in BWI treatment, and very limited research still on the use of H₂O₂ to treat infected non-thermal wounds in general, the published data that is available holds promising results. It is recommended that more large-scale research providing high-level evidence in this field is required: this may help to determine whether H₂O₂ offers a solution to the need to use antibiotics to treat patients with BWI.

Appendix A: burns H₂O₂ survey

The purpose of this survey is to establish current clinical practice in treating burn wounds with H₂O₂ in the United Kingdom.

1. Does your burns service use hydrogen peroxide (H₂O₂) in the treatment of infected burns wounds?

2. Is H₂O₂ used by itself in the treatment of infected burn wounds or in conjunction with other treatments, such as antibiotics?

3. Is there a limit on the amount of time for which H₂O₂ is used in the treatment of infected burn wounds?

4. For which patients does your department use H₂O₂ in the treatment of infected burn wounds? Are there specific criteria or is there an established departmental protocol for its use (if so, please provide a brief outline of these criteria e.g. specific wound depth/size)?

5. Do you perform a wound swab for culture prior to the use of H₂O₂? If so, why is this: is H₂O₂ used in the treatment of specific organisms only (e.g. multi-drug resistant organisms)?

6. Who in the department uses H₂O₂ (e.g. specialised nurses, anaesthetists, plastic surgeons etc.)?

7. If your department does use H₂O₂, where in your department is it used (e.g. in theatre under general or local anaesthetic or on the wards)?

8. At what concentration do you use H₂O₂? Does this vary according to clinical indication?

9. At what volume do you use H₂O₂? Does this vary according to clinical indication? Is there a limit on the volume used?

10. Do you use H₂O₂ for any other purposes in the management of burn wounds other than treating infection (e.g. infection prevention, general wound healing, haemostasis etc.)?

11. If your department does not use H₂O₂ in the treatment of burn wounds, is there a specific reason as to why not?

Thank you very much for taking the time to complete this survey.

Appendix B: full search strategy

Number	Database	Search	Results
1	Medline	(wound* OR burn* OR “burn wound*”)	249,747
2	Medline	((“Hydrogen peroxide” OR H2O2) AND (treat* OR Manag* OR therap*))	23,788
3	Medline	1 AND 2	416
4	EMBASE	(wound* OR burn* OR “burn wound*”)	319,996
5	EMBASE	((“Hydrogen peroxide” OR H2O2) AND (treat* OR Manag* OR therap*))	34,193
6	EMBASE	1 AND 2	663
7	CINAHL	(wound* OR burn* OR “burn wound*”)	61,208
8	CINAHL	((“Hydrogen peroxide” OR H2O2) AND (treat* OR Manag* OR therap*))	868
9	CINAHL	1 AND 2	57
10	The Cochrane Database of Systematic Reviews	(wound* OR burn* OR “burn wound*”) AND ((“Hydrogen peroxide” OR H2O2) AND (treat* OR Manag* OR therap*))	32

Appendix C: included studies for analysis

[1] Mohammadi A, Seyed Jafari S, Kiasat M, Pakyari M and Ahrari I. Efficacy of debridement and wound cleansing with 2% hydrogen peroxide on graft take in the chronic-colonized burn wounds; a randomized controlled clinical trial. Burns 2013; 39: 1131-1136.

Disclosure of conflict of interest

None.

Abbreviation

BWI

Burn wound infection