Surgical site infection (SSI) is a significant problem for patients and the health service, but is potentially preventable.

It has been estimated that SSIs currently cost the NHS around £700 million per year, largely through prolonged postoperative inpatient stay and additional inpatient and outpatient treatment costs.

It is increasingly recognised that SSI incidence is widely underreported. Traditional monitoring relies heavily on passive surveillance with minimal review after discharge, but at least 60% of SSIs present in the community after discharge. Out of hospital SSI events are therefore often unaccounted for. With the increase of enhanced recovery programmes and shorter lengths of stay, the proportion of SSIs presenting outside of hospital has increased further. Rates of SSI vary significantly between different types of surgery, but is particularly prevalent in abdominal operations; as many as one in four patients get an SSI when the operation involves the bowel. 

Preventing SSI is a complex process which is affected by interventions throughout the surgical care pathway. SSI reduction measures when bundled, or poorly implemented can be ineffective or even increase SSI risk. A large proportion of SSIs are known to be caused by wound contamination by endogenous bacteria from the patient’s skin, or cross contamination from mucous membranes, hollow viscera, free pus or bowel contents. This has resulted in the development of many intraoperative interventions to try to decrease this contamination and thereby decrease SSI rates. Unfortunately, clinical studies exploring the efficacy of many of these interventions are often underpowered or poorly designed, or used in low risk groups leaving uncertainty if they are clinically and cost-effective. ROSSINI 2 aims to study simple biologically plausible interventions that may decrease SSI rates after abdominal surgery, but currently lack evidence in controlled studies.

There is a significant health need for research to address the problem of SSI, with benefit for both patients and the NHS.

Primary objective

To determine whether several specific in-theatre interventions, used alone or in combination, result in decreased rates of SSI up to 30 days post-operation in adult patients undergoing abdominal surgery.

Clinical Objectives

- Do the intraoperative interventions used alone, or in various combinations (within seven possible intervention arms and one control arm) reduce the overall rate of SSI after abdominal surgery?

- Is the efficacy of the intervention/treatment arm dependent upon:

• degree of wound contamination (clean, clean-contaminated, contaminated, dirty)?

• patient comorbidity (e.g. diabetes, smoking, obesity)?

• duration of operation?

• stoma formation?

• operation approach?

• SARS-CoV-2 virus status on day of surgery?

• SARS-CoV-2 antibody status on day of surgery?

- Do the intraoperative interventions:

• have an acceptable safety profile?

• reduce the rates of wound complications?

• reduce the rates of mortality?

- What is the impact of SARS-CoV-2 virus infection status on overall SSI rate, and in those who previously had COVID-19 (proven or suspected), is there a time-dependent impact upon overall SSI rates? 

Economic

- Does the use of the interventions:

• improve health-related QoL?

• reduce the length of stay in hospital?

• reduce wound complication related hospital re-admissions?

• reduce the occurrence of unplanned wound reopening and/ or re-operations?

• are cost-effective?

Primary Outcome Measures

In ROSSINI 2 the primary outcome measure is the SSI rate up to 30 days after surgery as defined according to the 2017 Centers for Disease Control (CDC) and Prevention criteria. Whilst several systems exist to define SSI, the internationally recognised CDC definitions are the current gold standard for SSI assessment and have been used in a number of multicentre randomised trials. 

The following CDC definition will be used in ROSSINI 2 to identify deep incisional or superficial incisional SSIs:

Secondary Outcome Measures

• 30-day postoperative mortality rate (POMR).

• 30-day postoperative wound complication rate (Clavien-Dindo classification).

• Serious Adverse Events up to 30 days (wound or intervention-related only).

• Length of hospital stay after surgery as measured from the date of surgery to the date of discharge.

• Hospital re-admission for wound related complications within 30 days.

• Occurrence of unplanned wound reopening and/or re-operations within 30 days post-operation.

• Preference-based QoL measure (EQ-5D-5L) at Baseline, Day 7 (or discharge) and Day 30.

• Cost effectiveness (using Resource Use questionnaire).

ROSSINI 2 is a phase III, multi-arm, multi-stage (MAMS), pragmatic, blinded (patient and outcome assesor), multicentre, randomised controlled trial, with an internal pilot, to evaluate the use of several in-theatre interventions, use alone or in combination, to reduce the rates of SSIs in patients undergoing surgery with an abdominal incision. A non-factorial superiority design with allocation to various combinations of the trial interventions to be used during the same operation.

SSI prevention lends itself well to an adaptive or multi-arm, multi-stage (MAMS) trial design, because the primary outcome result is, by definition, available 30 days after the intraoperative intervention is applied. Interim analyses can exploit this short timeline to create an efficient trial that will evaluate several interventions (both individually and in combination) under a single umbrella structure. This decreases both the time and cost investment necessary to simultaneously determine if several non-bundled interventions are effective. Determining small incremental benefits will be slow and difficult to achieve. This trial design allows exploration of interactions between interventions, which each have a different biological mechanism in this multifactorial process.

The trial would be the first of its kind in a surgical setting. In addition to generating new knowledge in our primary research area, by utilising this advanced design in the context of our relatively simple primary endpoint of SSI, it will also pave the way for future efficient and rapid trials in other aspects of surgical care.

The interventions chosen in ROSSINI 2 impact different phases of perioperative case and as such can be used either in isolation or in conjunction with each other and although there may potentially be interaction between the interventions (positive or negative) they appear to be mechanistically disparate.

The current interventions are:

1. 2% alcoholic chlorhexidine skin prep

2. Iodophor-impregnated incise drape

3. Gentamicin-impregnated sponge

Please note: In January 2022, following the first interim analysis, arms including intervention 2 – Iodophor-impregnated incise drape (arms C, E, G and H) were closed to recruitment.

At least 60 NHS hospitals in the UK will participant in ROSSINI 2.

Approximately 6610 patients will be required to detect a 5% absolute risk reduction in the interventions arm(s) (15% to 10%; 33% relative reduction) with 85% power.

Pilot Stage Summary of Completion (Stage 1)

The Independent Trial Steering Committee (TSC) and Data Monitoring Committee (DMC) held a joint meeting on 30th September 2019 to review the internal pilot stage objectives and results. They were satisfied with the progress against the predefined progression criteria and recommended that ROSSINI 2 continue to the main phase of the trial without the need for any major modifications. The NIHR also reviewed trial progress and reports from the oversight committees; they confirmed progression to the main phase in October 2019. The table below details the success of the internal pilot stage:

Stage 2 Summary of Completion (First Interim Analysis)

It was agreed in an extraordinary meeting of the Trial Management Group (TMG), TSC and DMC on October 5th 2021 (and subsequently by the Funder) that it would be reasonable to bring forward the first interim analysis i.e. sooner than 2357 patients as originally planned. This decision was based on observing a slightly higher than expected overall event rate.

On December 15th 2021, the results of the first interim analysis were presented and reviewed by the independent DMC. They made the recommendation to drop the intervention arms containing Intervention 2, Iodophor-impregnated incise DRAPE. The DMC recommendations were discussed by the ROSSINI 2 TMG, independent TSC, Sponsor and Funder at a meeting on January 7th 2022. Following this meeting, it was agreed that the following arms containing Intervention 2, Iodophor-impregnated incise DRAPE, would close to recruitment:

• Arm C DRAPE

• Arm E SKIN PREP & DRAPE

• Arm G DRAPE & SPONGE

• Arm H SKIN PREP & DRAPE & SPONGE

All other aspects such as the primary objective, trial design, trial setting, outcome measures and key eligibility criteria remain unchanged.

Centre and Surgeon eligibility

Any centre performing emergency and/or elective abdominal surgery will be eligible to participate in ROSSINI 2. Sites entering the trial must not be routinely using these interventions and be willing to accept patients being randomised to receive (or not receive) each of them, including combinations thereof. Surgeons must be willing to adhere to arm allocation and be willing to be trained in a standardised application technique.

Patient Inclusion Criteria

• Patients undergoing an abdominal operation, these include colorectal, hepatobiliary, upper GI, urological, vascular, or gynaecological operations

• Patients undergoing abdominal operations (open or laparoscopic assisted) with a planned incision of at least 5cm

• Patients aged 16 years or older

• Patients able and willing to undergo a wound assessment at day 30-37 after surgery

• Patients able and willing to give written informed consent

• All contamination strata, including clean, clean-contaminated, contaminated or dirty surgery.

• Patients undergoing planned (elective or expedited) or unplanned (emergency) surgery.

Patient Exclusion Criteria

• Previous laparotomy within 3 months prior to randomisation.

• Known to be pregnant or currently breast feeding

• Operations where the wound is not anticipated to be closed primarily 

Notes of Eligibility

• SARS-COV-2 virus status and antibody status on day of surgery – all patients are eligible to enter the trial regardless of positivity, negativity or not tested status

• Patients with a new or documented allergy/ intolerance to any of the trial interventions (chlorhexidine, collagen or gentamicin) will not be randomised to an arm containing this intervention, but will still be eligible for recruitment to other arms of the trial

• Patients with end-stage renal failure where gentamicin administration would otherwise be contra-indicated (according to local policy) will not be randomised to arms containing the gentamicin-impregnated sponge

• Purely groin incisions such as inguinal hernia repair or vascular access of the femoral vessels do not count as abdominal surgery and are not eligible to enter the trial

It will be the responsibility of the Principal Investigator (or delegate if local site allows) to obtain written informed consent for each participant prior to performing any trial related procedures. Research nurses and other allied health professionals (GCP trained) can also obtain consent if local practice allows and this responsibility has been delegated by the PI and captured on the ROSSINI 2 Site Signature and Delegation Log.

The participant will be given sufficient time to read the PIS and to discuss their participation with others outside of the site research team if they wish. The ICF will be completed using one of the following processes:

• Face-to-face (Preferred method)

• By post/ email/ telephone 

After eligibility has been confirmed and informed consent has been received, the patient can be randomised into the trial. It is essential that randomisation into the trial occurs in theatre, ideally around the time of induction of anaesthesia on the day of surgery.

The randomisation service is provided by the Centre for Healthcare Randomised Trials (CHaRT), located in the Health Services Research Unit (HSRU) at the University of Aberdeen. A patient can be randomised 24 hours a day, 7 days a week.

Randomisation will be via a secure online randomisation system or via an automated telephone randomisation system. 

Telephone Randomisation - 0800 2802 307 

You will need to know your Centre/ Site ID code and your own unique ID code to be able to randomise via telephone.

Online Randomisation - https://w3.abdn.ac.uk/hsru/ROSSINI2

You will need to know your username and password to be able to randomise online.

Your Site ID code, unique ID code and username can all be obtained by contacting the ROSSINI 2 trial team - ROSSINI2@trials.bham.ac.uk.

+It is important to also reaffirm consent at each trial visit.
* Wound assessment data collected here should be applicable and relevant between Day 0 (day of surgery) to Day 7 Wound Assessment (or up to discharge).
** Wound assessment data collected here should be applicable and relevant between Day 7 (first wound assessment) to Day 30 Wound Assessment (second wound assessment). Data should not be duplicated over forms. If no previous wound assessment has occurred, then data collected should be from Day 0 to Day 30.
++ Only to be completed if patient is required to return to theatre (within 30 days post-surgery) for any reason.
^ Quality of Life at Baseline should be completed after Consent but before Surgery.
$ RUQ will only be collected on those participants in the final stage of the trial. Sites will be notified when to implement this.

The ROSSINI 2 Trial involves several types of training:

1. Trial-specific training: An overview of ROSSINI 2, including the aims, design and logistics. This is often delivered by the ROSSINI 2 team during the Site Initiation Visit (SIV). It will be face-to-face whenever possible and ideally near the time of opening. Once open, further trial-specific training (i.e. new staff) can be delivered via teleconference by the ROSSINI 2 team or disseminated by the PI or local team.

2. Intervention application training: This will be delivered by the ROSSINI 2 team, by the CI and/or surgical trainees during the SIV. Once open, intervention application training (i.e. for new staff) is to be disseminated by the PI. Training materials will be provided.

 The above types of training must be completed BEFORE a site can open to recruitment.

1. Wound-assessment training - this is delivered online, please click on the link 

https://bctu-annex.redcap.bham.ac.uk/surveys/?s=LPJP73K3LF974YXP

The above training must be completed by anyone who will be reviewing the wound of trial participants.

It is mandatory for all staff undetaking tasks in ROSSINI 2 to have receivede the relevant training.

Trial Interventions

Intervention	Supplier
2% Alcoholic Chlorhexidine skin prep (SKIN PREP)	Becton Dickinson UK Ltd (BD
Iodophor-impregnated incise drape (DRAPE)	3M United Kingdom PLC
Gentamicin-impregnated sponge (SPONGE)	SERB

Please note: In January 2022, following the first interim analysis, arms including intervention 2 – Iodophor-impregnated incise drape (arms C, E, G and H) were closed to recruitment.

Usual care and Site requirements

All patients undergoing an abdominal operation will have their skin prepared using some form of antiseptic skin preparation. Only in participants randomised to the trial intervention of 2% Alcoholic Chlorhexidine (arms B or F) should this skin preparation be used. For all other participants, they should receive the surgeon’s choice of skin preparation as long as this is NOT 2% Alcoholic Chlorhexidine. Other commonly used agents include 0.5% chlorhexidine or povidone-iodine, in aqueous or alcoholic solution. Concentrations and volumes of preparations vary and can be mixed in theatre or used as pre-prepared solutions. No other impregnated sponges should be used.

Before opening, all sites will receive trial-specific training on the logistical and operational aspects of the trial. All investigators will undergo a training and certification process that includes:

1. Watching videos outlining proper use of study interventions

2. Training on the correct use of the interventions by a member of the TMG or the local PI

3. Access to standardised ‘training cards’ for use in theatre as an aide memoire for the application of the intervention technique

This will ensure the correct use of the various interventions to ensure a standardised and optimal method of use.

The manufacturer of each product is responsible for the free provision of trial interventions to the ROSSINI 2 Trials Team. An initial supply of the interventions will be delivered to each site prior to site opening. Please get in touch with the ROSSINI 2 Trials Team to if you require additional supplies - ROSSINI2@trials.bham.ac.uk.