Doc menu

5. Guideline methodology

The aim of the 2014 systematic review was to determine if, since 2010, there was any additional evidence which could further inform or improve current clinical practice and if so, what the impact of this additional evidence would be.

Initially a search question was formulated using the Population, Intervention, Control, Outcome (PICO) method (Table 1).

Table 1: PICO method

Population Adult patients undergoing external beam radiotherapy:  radiation therapy, irradiation
Interventtion Preventative measures eg washing practices, topical applications. deodorant guidance and/or management measures - dressings, topical and medical applications
Control Standard intervention
Outcome Skin reactions, radiation effect, adverse effect, radiation dermatitis, erythema, moist desquamation, skin care, skin reactions

The review was based on a systematic search of Medline, Pub Med, CINAHL, EBSCO, Science Direct, ISI Web of Science and Index to Thesis.

Hand searches of the Journal of Radiotherapy in Practice (JRP), The European Journal of Cancer (EJC), Radiography, Journal of Medical Imaging and Radiation Science (JMIRS), the International Journal of Radiation, Oncology, Biology, Physics (IJROBP) and Radiotherapy and Oncology were also undertaken.

In addition, a secondary evaluation of the clinical trials’ databases was searched for any ongoing research as well as a search of the ‘grey literature’, including index to theses and conference papers. Finally a broad search of Google Scholar was used as a ‘mop up’ technique to ensure no additional relevant research had been missed.

Owing to the fact that a wealth of evidence had been reviewed in 2010 and this is a continuation of that work, it was deemed appropriate to map out and replicate the initial search strategy and then include any additional resources.

The traditional pearl growing method of literature searching begins with a single document relevant to the topic under review and utilises key words or seminal text; pearl growing until more recent years has often been overlooked as a strategy for literature searching (Schlosser et al., 2006). The Comprehensive Pearl Growing (CPG) method has developed from this and uses multiple key documents rather than just one. It is considered to be more systematic in its approach and deemed an appropriate method to be used for yielding results in a systematic review  (Schlosser et al., 2006). For the purpose of this review, Comprehensive Pearl Growing is an appropriate and important method to use in the initial stages of the strategy as this is following on directly from a seminal piece of previous published work and one other key document.

Table 2 indicates the key terms used within the search strategy, drawn from the seminal articles.

Table 2: Key Terms

Aspect Key term
Radiotherapy Radiotherapy, radiation therapy, irradiation
Outcome Skin reactions, radiation effect, adverse effect, radiation dermatitis, erythema, moist desquamation, skin care, skin reactions, evidence-based practice

Those studies included initially had to fulfil the following criteria:  

  • All literature from November 2010 (when the last review by the SCoR was conducted); 
  • All papers that have an English abstract;
  • Papers that assess the use of a topical agent;
  • Papers where the primary focus is skin reaction to radiotherapy.

Studies were excluded for failure to meet the above criteria or for the following reasons: 

  • Reactions caused by a pre-existing genetic or medical disposition;
  • Case studies;
  • Rare skin reactions caused by topical agents or chemotherapy drugs;
  • Papers where the primary focus is the impact of the immobilisation device or radiotherapy planning technique on the skin reaction;
  • Late effects to the skin following radiation.

All appropriate full text articles underwent quality assessment using the Scottish Intercollegiate Guidelines Network (SIGN) quality assessment tool. Initially the Grading of Recommendation, Assessment, Development and Evaluation (GRADE) system was proposed, however upon further investigation the SIGN tool was deemed more appropriate and relevant for this particular study. To ensure the correct assessment questionnaire was used, all studies were mapped against the SIGN:  “Algorithm for classifying study design for questions of effectiveness” (www.sign.ac.uk, 2013).

5.2 Literature review

A flowchart including the number of hits obtained in the database searches, those abstracts screened for relevance, down to the final number of articles included in the review, can be found under Diagram 1.

Research is continually emerging within this area, possibly due to the lack of conclusive evidence and the disparity between the published research as highlighted earlier, therefore it was deemed appropriate to include within the results any relevant ‘grey literature’, such as research protocols, conference presentations, symposiums and ongoing research related to Randomised Controlled Trials (RCTs).

Diagram 1: Flowchart of literature review

5.3 Description of how recommendations were developed 

The core group considered the evidence and developed the recommendations for the practice guideline.

Randomised Control Trials (RCTs) and Systematic Reviews (SR) 

  • Quality assessment using the appropriate SIGN checklist was undertaken, a total of 17 articles were available for review: 2 Systematic reviews, 14 RCTs, I case control. 
  • Of the RCTs and systematic reviews (n=16):  5/15 (33%) were classed as high quality evidence; 8/15 (53%) classed as acceptable evidence; 3/15 (20%) rejected as unacceptable quality. (See  Appendix 5 for summary of articles table.)
  • The final number of studies included in the review: 2 systematic reviews, 11 RCTs and I case control. 

Of the RCTs (n=11)included in the final review,  nine  were studying a different topical emollient or product (Jensen et al., 2011; Kirova et al., 2011; Miller et al., 2011; Abbas and Bensadoun, 2012; Niazi et al., 2012; Graham et al., 2013; Sharp et al., 2013; Ulff et al., 2013; Herst et al., 2014,) and two studies were reporting the use of non-metallic antiperspirants (Watson et al., 2012; Lewis et al. 2014). (see Appendix 6 for full systematic review report.)

Butcher and Williamson (2012) undertook a systematic review of the literature on the management of erythema and skin preservation for patients receiving external beam radiotherapy to the breast. All literature was assessed for quality and in total 10 studies were included in the final analysis. They concluded that no one product was considered superior to another. The review advocates the safe use of non-metallic deodorants. The review also highlights the wide variety of methods and assessment scales used to report study findings thus making meaningful comparisons very difficult.   

Chan et al. (2014) undertook a systematic review and meta-analysis which included 47 RCTs from 1962-2012. This large date range is a slight limitation as studies conducted during the 1960s are likely to include orthovoltage energies and Cobalt treatments with subsequent associated skin reactions that are not relevant to the skin sparing effects achieved with modern linear accelerators. Studies examined a range of practices:

  • 6 trials investigated oral systemic therapies
  • 2 investigated washing practices
  • 4 examined deodorant use
  • 5 investigated topical steroidal therapies
  • 23 examined non-steroidal topical therapies
  • 6 investigated dressings
  • 1 investigated light emitting diode photo-modulation

Thirty-six of the included studies were considered at high risk of bias, 10 rated at unclear risk and one as low risk; confirming our own experience of quality assessment of studies in this field. Allocation concealment was only reported in 22 of the 47 studies reviewed. Blinding of assessors was only adequately described in 21 of the 47 studies. Similarly, only 21 of the 47 studies adequately reported how attrition was handled in the analysis.

A small meta-analysis of two studies investigating oral systemic therapy (oral Wobe-Mugos E vs. no medication) indicated the odds of developing a radiation induced skin reaction was 87% lower for people receiving Wobe-Mugos E (although heterogeneity for the studies was high I2=70%). A meta-analysis of 226 participants from two un-blinded studies found no difference in radiation induced skin reactions when comparing deodorant use to no deodorant use. Four trials investigated the role of topical steroidal agents on radiation induced skin reaction. Three of these studies identified no benefit while one small study (n=20) found a statistically significant benefit for using prednisolone with neomycin compared with no treatment. However, some of the topical steroid trials had small sample sizes and wide confidence intervals hence the results should be viewed with caution.

Overall the review concludes that the evidence for any intervention is ‘thin’ i.e. no strong evidence of effect for any of the included trial products to reduce radiation induced skin reactions. The study concludes that patients should be advised to wash gently and using non-metallic deodorant is not contraindicated. Recommendations for future studies include a focus on an area of promise such as oral Wobe-Mugos E and oral zinc. Future studies should also attempt to clarify which patients would benefit from corticosteroid cream, and appropriately powered RCTs comparing different dressings for those who develop moist desquamation. 

During the inclusion period of this current review there were other studies of some note, and a number of abstracts and short publications published, as well as conference presentations, which were assessed (see Appendix 6). In order to ascertain current research being undertaken in this field, a search of the clinical trials database was also undertaken (see Appendix 7).

This review has demonstrated that additional research has been published in the field. Nevertheless the scope of this research and the results are quite wide ranging, both in their methods and in the aspect of radiotherapy-induced skin reaction being researched. Many of the studies seem to concentrate on a particular topical application, which adds to the evidence-base, but still there is a lack of emergence of any one product as superior over another.  

A number of studies have been undertaken investigating the use of topical steroids and Wong, et al. (2013) make strong recommendations in their guidelines for the use of prophylactic topical steroids. In spite of this some of the published research  recommends exercising a degree of caution and that there is a need for more work to be undertaken, particularly to determine any long term implications of using steroids. 

There are two areas where a more general consensus on guidance is closer to being achieved. Firstly with respect to the use of aqueous cream:

  • this has now been reclassified in the British National Formulary (BNF) as a soap substitute and may be applicable in this usage for patients undergoing radiotherapy;
  • it should not, however, be used as a leave-on moisturiser. 

Secondly with respect to the use of deodorant, where a much stronger evidence base refutes the adverse impact that deodorants were once thought to have (Wong et al., 2013; Lewis et al., 2014). However, further research is still needed on the use of aluminium based deodorants. 

5.4 Limitations of the guideline including consideration of possible bias 

In terms of the strength of the recommendations contained within this guideline, considerable contradictory evidence was identified with often a weak level of agreement between all sources. The sources from which each recommendation was drawn are highlighted throughout the document. However, the guideline development core and stakeholder group have reached a consensus of agreement to allow for the application of this evidence within radiotherapy departments.

Content tools

Accessibility controls

Text size

AA A

Colour