MARLIN is a stratified, multi-arm, multi-stage factorial randomised platform trial aiming to reduce the incidence of post-operative surgical site infection (SSI).

MARLIN's primary objective is assess whether different interventions used at three different time points during the participant's operative care (pre-, intra- and post-operatively) can reduce SSI at 30 days in patients undergoing abdominal surgery. MARLIN will assess the clinical effectiveness in both clean-contaminated wounds and contaminated or dirty wounds and will assess the interventions being assessed individually and in combination.

Patients (adults and children) undergoing surgery with abdominal incision of at least ≥5cm, with an anticipated clean-contaminated or contaminated or dirty wound, undergoing emergency or elective, and open or laparoscopic surgery are eligible. Participants will be recruited from hospitals within the National Institute for Health and Care Research Global Surgery Unit (NIHR GSU) network (i.e. hospitals in Benin, Rwanda, Ghana, India, Mexico, Nigeria, South Africa).

Interventions will be introduced at the preoperative window, intra-operatively and postoperatively as it is not yet known which if any of these timed intervention periods are most effective. It is planned that only one intervention will be introduced in each window. This format will allow testing for interaction between interventions including cumulated benefit. It will also enable unsuccessful interventions to be withdrawn and potentially substituted. This is a pragmatic design and wherever possible, standard care will be provided in the control arm.

After patient eligibility has been confirmed and informed consent has been obtained, patients will be randomised into the MARLIN trial by a member of the MARLIN research team at the site. Randomisation will occur within 72 hours prior to the patient's operation. Patients will be randomised dependent on which arms are available at the randomising site, and in which they consent to participate (i.e. it is possible for patients to 'opt out' of one or more of the intervention timepoints).

Patients may be randomised to receive:

• Chlorhexidine Gluconate preoperative shower 2.0% - 4.0% (e.g. hibiscrub) or standard preoperative preparation
• Topical wound wash before closure (e.g. granudacyn) or standard intraoperative preparation
• Dialkylcarbamoyl chloride (DACC) wound dressing (e.g. Leukomed sorbact) or standard post-operative wound care

The randomisation ratio at each timepoint will be 1:1 between the intervention and control in the first instance, although if and when multiple intervention arms are open, the ratio may change with agreement of the data monitoring committee (DMC), with reduced randomisation into the control arm due to pooling. Eligible participants will be randomised to receive some or all of the trial interventions; sites will opt-in to each treatment depending on local deliverability, which will be reflected in the randomisation system. Randomisation will occur around the time of pre-operative assessment.

Sample size:

An internal pilot will be carried out for each intervention within the study. For each pilot, Internal pilots of 100 participants per intervention, will be undertaken to assess deliverability, safety, compliance and participant acceptability. Internal pilots will be undertaken in at least two countries for any given intervention to assess generalisability. Interventions will then be made available across the wider Global Surgery network.

For the clean-contaminated stratum, the maximum sample size required for a 3-stage design, with control SSI rate of 16%, two-sided overall alpha of 0.043, overall power of 0.86, and 20% relative reduction (3.2% absolute reduction) with 5% lost to follow up, would be 6006 participants for all 3 questions as a factorial design. As the second and third treatment will start after the first stage interim analysis of the first intervention, 7544 participants are required. 8196 participants are required if the first intervention works and leads to lower SSI rate.

For the contaminated/dirty stratum, the maximum sample size required for a 3-stage design, with control SSI rate of 30%, two-sided overall alpha of 0.043, power of 0.86 and 30% relative reduction (9% absolute reduction) with 10% lost to follow up, would require 1342 participants for all 3 questions as a factorial design. As the second and third treatment will start after first stage interim analysis of the first intervention, 1596 are required participants. 1896 participants are required if the first intervention works and leads to lower SSI rate.

Multistage analysis:

For the clean-contaminated group there will be two interim analyses, the first after 1824 participants and the second at interim analysis at 3126 for each of the questions. For the contaminated/dirty group, the first interim will happen at 372 participants, and the second at 638 participants.

Inclusion Criteria

• Patients with at least one abdominal incision that is ≥5cm (open or laparoscopic extraction site), with an anticipated clean-contaminated, contaminated, or dirty surgical wound.

• Patients undergoing emergency (surgery on an unplanned admission) or elective (surgery on a planned admission) operations.

• Any operative indication for abdominal surgery (excluding caesarean section; see exclusion criteria).

• Patient able and willing to provide written informed consent (signature or a fingerprint) prior to surgery (including emergency cases).

• Patients aged 5 years and over. ( Each country will decide the lower (and upper, if applicable according to local regulations age limit for the trial. This will be dependent on country-specific regulatory approvals. Age eligibility will vary by country.) Patient Exclusion Criteria

  • Patient unable to complete post-operative follow-up (i.e., will not be contactable after discharge).
  • Patients undergoing clean surgical procedures.

Inclusion Criteria:

• Patients with at least one abdominal incision that is ≥5cm (open or laparoscopic extraction site), with an anticipated clean-contaminated, contaminated, or dirty surgical wound. Definitions and examples of contamination are given in Table 1.
• Patients undergoing emergency (surgery on an unplanned admission) or elective (surgery on a planned admission) operations.
• Any operative indication for abdominal surgery (excluding caesarean section; see exclusion criteria).
• Patient able and willing to provide written informed consent (signature or a fingerprint) prior to surgery (including emergency cases).
• Patients aged 5 years and over. (This criteria MUST be made country-specific. Each country will decide the lower (and upper, if applicable according to local regulations age limit for the trial. This will be dependent on country-specific regulatory approvals. Age eligibility will vary by country.)

Exclusion Criteria:

• Patient unable to complete post-operative follow-up (i.e., will not be contactable after discharge).
• Patients undergoing clean surgical procedures.
• Patients undergoing an obstetrics procedure, including caesarean sections.

Hospital-level data will not be released or published. Country-level analyses will only be conducted with permission of lead investigators from each participating country. Local investigators may access their data across their country to perform country-level analyses (all participating hospitals should consent to their data being used in this way).

Participants should be randomised by a member of the research team who will not be performing the any aftercare, including trial-specific follow-up assessment(s). The follow-up assessor at 30 days will therefore be blinded to treatment allocation until after the completion of follow-up. Unless otherwise instructed by a research ethics committee, the operative notes should not include details of the specific intervention used.