Three wound-dressing strategies to reduce surgical site infection after abdominal surgery: the Bluebelle feasibility study and pilot RCT

Barnaby Reeves, Leila Rooshenas, Rhiannon Macefield, Mark Woodward, Nicky Welton, Benjamin Waterhouse, Andrew Torrance, Sean Strong, Dimitrios Siassakos, William Seligman, Chris Rogers, Lloyd Rickard, A Pullybank, Caroline Pope, Thomas Pinkey, Samir Pathak, Anwar Owais, Jamie Callaghan, Stephen O'brien, Dmitri NepogodievKhaldoun Nadi, Charlotte Murkin, Tonia Munder, Tom Milne, David Messenger, Christel McMullan, Jonathan Mathers, Matthew Mason, Morwena Marshall, Richard Lovegrove, Robert Longman, Jessica Lloyd, Kathryn Lee, Vijay Korwar, Daniel Hughes, George Hill, Rosie Harris, Mohammed F. Hamdan, Hannah Gould Brown, Rachael Gooberman-Hill, James Glasbey, Caroline Fryer, Lucy Ellis, Daisy Elliott, Jo C. Dumville, Timothy Draycott, Jenny Donovan, David Cotton, Joanna Coast, Maddie Clout, Melanie Calvert, Benjamin Byrne, Oliver Brown, Natalie Blencowe, Kasia Bera, Joanne Bennett, Richard Bamford, Danya Bakhbakhi, Muhammad Atif, Kate Ashton, Elizabeth Armstrong, Lazaros Andronis, Piriyankan Ananthavarathan, Jane Blazeby

Research output: Contribution to journalArticle (Academic Journal)peer-review

6 Citations (Scopus)
326 Downloads (Pure)


Surgical site infection (SSI) affects up to 20% of people with a primary closed wound after surgery. Wound dressings may reduce SSI.
To assess the feasibility of a multicentre randomised controlled trial (RCT) to evaluate the effectiveness and cost-effectiveness of dressing types or no dressing to reduce SSI in primary surgical wounds.
DesignPhase A – semistructured interviews, outcome measure development, practice survey, literature reviews and value-of-information analysis. 
Phase B – pilot RCT with qualitative research and questionnaire validation. Patients and the public were involved.SettingUsual NHS care.ParticipantsPatients undergoing elective/non-elective abdominal surgery, including caesarean section.
Phase A – none. 
Phase B – simple dressing, glue-as-a-dressing (tissue adhesive) or ‘no dressing’.

Main outcome measures
Phase A – pilot RCT design; SSI, patient experience and wound management questionnaires; dressing practices; and value-of-information of a RCT. 
Phase B – participants screened, proportions consented/randomised; acceptability of interventions; adherence; retention; validity and reliability of SSI measure; and cost drivers.
Data sources
Phase A – interviews with patients and health-care professionals (HCPs), narrative data from published RCTs and data about dressing practices. 
Phase B – participants and HCPs in five hospitals.
Phase A – we interviewed 102 participants. HCPs interpreted ‘dressing’ variably and reported using available products. HCPs suggested practical/clinical reasons for dressing use, acknowledged the weak evidence base and felt that a RCT including a ‘no dressing’ group was acceptable. A survey showed that 68% of 1769 wounds (727 participants) had simple dressings and 27% had glue-as-a-dressing. Dressings were used similarly in elective and non-elective surgery. The SSI questionnaire was developed from a content analysis of existing SSI tools and interviews, yielding 19 domains and 16 items. A main RCT would be valuable to the NHS at a willingness to pay of £20,000 per quality-adjusted life-year. 
Phase B – from 4 March 2016 to 30 November 2016, we approached 862 patients for the pilot RCT; 81.1% were eligible, 59.4% consented and 394 were randomised (simple, n = 133; glue, n = 129; no dressing, n = 132); non-adherence was 3 out of 133, 8 out of 129 and 20 out of 132, respectively. SSI occurred in 51 out of 281 participants. We interviewed 55 participants. All dressing strategies were acceptable to stakeholders, with no indication that adherence was problematic. Adherence aids and patients’ understanding of their allocated dressing appeared to be key. The SSI questionnaire response rate overall was 67.2%. Items in the SSI questionnaire fitted a single scale, which had good reliability (test–retest and Cronbach’s alpha of > 0.7) and diagnostic accuracy (c-statistic = 0.906). The key cost drivers were hospital appointments, dressings and redressings, use of new medicines and primary care appointments.
Multiple activities, often in parallel, were challenging to co-ordinate. An amendment took 4 months, restricting recruitment to the pilot RCT. Only 67% of participants completed the SSI questionnaire. We could not implement photography in theatres.
A main RCT of dressing strategies is feasible and would be valuable to the NHS. The SSI questionnaire is sufficiently accurate to be used as the primary outcome. A main trial with three groups (as in the pilot) would be valuable to the NHS, using a primary outcome of SSI at discharge and patient-reported SSI symptoms at 4–8 weeks.
Trial registration
Phase A – Current Controlled Trials ISRCTN06792113; 
Phase B – Current Controlled Trials ISRCTN49328913.
This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 23, No. 39. See the NIHR Journals Library website for further project information. Funding was also provided by the Medical Research Council ConDuCT-II Hub (reference number MR/K025643/1).
Original languageEnglish
Number of pages204
JournalHealth Technology Assessment
Issue number39
Publication statusPublished - 1 Aug 2019

Structured keywords

  • BTC (Bristol Trials Centre)
  • Centre for Surgical Research


  • Feasibility study
  • wound dressings
  • surgery
  • surgical site infection
  • pilot randomised controlled trial
  • qualitative research
  • value of information


Dive into the research topics of 'Three wound-dressing strategies to reduce surgical site infection after abdominal surgery: the Bluebelle feasibility study and pilot RCT'. Together they form a unique fingerprint.
  • ConDuCT-II

    Blazeby, J.


    Project: Research

Cite this