Primary aim

To compare the performance of the local vascular unit against UK NICE guidance relating to SSIs (supplementary material 1) in the groin incisions of vascular patients.

Secondary aims

The secondary aims are:

• (i)  to calculate a contemporaneous rate of groin incision infections in the UK and abroad;
• (ii)  to identify patient, surgical and theatre risk factors associated with groin SSIs;
• (iii)  to examine the value of previously published tools in predicting SSIs in groin incisions;
• (iv)  to identify areas of equipoise which could be examined in future effectiveness trials.

Design

An international, multicentre audit of practice, and a prospective observational cohort study disseminated via VERN.

Setting

Hospitals providing emergency and/or elective vascular surgery in the UK and abroad, recruited via VERN. Based on current interest it is expected at least 25 centres will be enrolled.

Time frame

The GIVE audit was officially launched on 21st January 2019. The UK will close to new participating centres on 1st May 2019. Centres are encouraged to start data collection as soon as possible once appropriately registered. Data from consecutive groin incisions meeting the inclusion criteria will be collected prospectively for 3 months.

Follow up data will be captured at 90 days following surgery. The entire study period will span 6 months. Centres are required to return all data within 7 months of starting, which will allow 1 month from last data point for all data capture.

Participants

The audit will enrol consecutive patients undergoing a groin incision for access to the femoral arteries.

Inclusion criteria:

• •  Patients undergoing an emergency or elective groin incision(s) for arterial intervention including endarterectomy, embolectomy, thrombectomy, bypass, repair of (non-infected) traumatic injury (i.e. iatrogenic pseudoaneurysm) or exposure for an endovascular procedure.
• •  Groin incisions which are extended down the leg (i.e. for vein harvest) or above the groin (i.e. for iliac vessel exposure) are included, although outcomes will be limited to the incision overlying the femoral triangle.

Exclusion criteria:

• •  Patients undergoing groin surgery for infective complications, e.g. infected pseudoaneurysm, explantation of infected prosthetic material.
• •  Groin exposure for venous access only without arterial exposure or intervention, (e.g. vein harvest only).
• •  Groin exposure performed for cardiac procedures, e.g. Transcatheter Aortic Valve Implantation (TAVI); access for cardiac bypass.

Patient identification and data collection/management

The local study team will be responsible for contemporaneous data collection. Theatre information technology systems will be used to screen for eligible patients regularly. Datapoints consisting of key demographic, baseline variables and intra-operative variables will be collected as early as possible following surgery. Post-operative sequelae datapoints will be preferentially collected as soon as the 90-day follow up threshold is reached. Should an SSI develop, further details will be required about the extent of the infection and resultant patient outcome. Data capture tools were drafted and refined to ensure appropriate data points will be captured (supplementary material 2).

Data will be obtained using patient’s notes and electronic records – including pre-operative assessment documentation, theatre information technology systems, discharge letters, clinic letters and accident and emergency department records. No changes to normal patient follow up will be made, and the patient will not be contacted to enquire about SSIs, unless this encompasses centre specific standard care.

Data will be recorded on an anonymised online data collection tool or held electronically locally, and anonymous data transferred to the GIVE Audit team via secure encrypted email. No patient identifiable information will be transferred from participating centres. Patients will be pseudoanonymised at each participating centre. The local audit lead will organise for the local centre to hold a secure database with a minimum of three patient identifiers (e.g name/hospital number/date of birth) and a 3 digit pseudoanonymised number. The pseudoanonymised number will be used to link peri-operative data and 90-day post-operative outcome data.

Data will be collated, stored and analysed by the GIVE Audit team at the Royal Gwent Hospital, Aneurin Bevan University Health Board, Newport, UK.

Primary outcome

The primary outcome will be compliance with NICE guidelines on SSI prevention.

An SSI will be defined as per the Centres for Disease Control and Prevention (CDC) criteria (supplementary material 3) [17]. These guidelines state that superficial SSIs are those presenting within 30 days of surgery, whilst deep or organ/space SSIs are those presenting within 90 days.

SSIs will be limited to those apparent to the treating vascular team within 90 days of surgery. It is recognised that this study may therefore fail to capture milder infections which are treated in the community.

Secondary outcomes

Secondary outcomes include (all for up to 90 days):

• •  Rate of post-operative SSI
• •  Rate of additional dressings used to manage SSI
• •  Rate of vacuum dressing therapy used to manage SSI
• •  Rate of antibiotics (both oral and intra-venous) used to manage SSI
• •  Rate of surgical and radiological re-interventions to manage SSI
• •  Organisms grown from samples sent to microbiology
• •  Rate of post-operative acute kidney injury (AKI)
• •  Rate of SSI resulting in sepsis
• •  Rate of SSI resulting in unplanned admission to a critical care setting
• •  Length of stay in hospital
• •  Mortality

AKI will be defined according to the Acute Kidney Injury Network (AKIN) criteria (supplementary material 4) [18].

Sepsis will be defined according to a definition recommended by the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) (supplementary material 5) [19].

Specific considerations

A groin incision can be extended inferiorly (e.g. for vein harvest), or superiorly onto the abdomen (e.g. for iliac vessel exposure). In these instances, the groin incision will be considered as that which overlies the femoral triangle. SSIs distant to the groin, such as in the mid-thigh, will not be captured. In cases of uncertainty, the decision of the overseeing consultant will be sought.

In the event of both groins being used simultaneously, data on each incision will be captured separately. If both groins of a patient are used sequentially (at different dates, but both within the audit timeframe) they will also be captured separately.

Should a patient already enrolled onto the GIVE audit return to theatre and require re-opening of the index groin incision for a reason other than SSI (e.g. for repair of a pseudoaneurysm arising from the original surgery), this would not be captured as a new groin incision, but rather as a return to theatre.

Data completeness and validation

Data completeness is considered essential. Data points recorded as ‘unknown’, where this option is available, will count as complete data. Cases with <95% of data completeness will be returned to the participating centre for completion, or if this is impossible, will be excluded from analysis, as is standard with international collaborative audits [20], [21].

Certain centres will be asked to review a small number of patient data points (approximately 5%) to confirm case ascertainment and data accuracy, as is standard with international collaborative audits [20], [21]. This will comprise of a review of the data collection by an individual not involved with its initial collection.

Statistical analysis

Analyses will be performed to assess overall SSI rates and examine secondary outcomes. Data will be checked for normality and the appropriate parametric or non-parametric test used. Individual unit SSI rates will be presented in funnel plot form, with lines representing ±2 and ±3 standard deviations. Individual unit rates will not be identified. Univariate and multivariate regression analyses will be used to identify significant variables predictive of SSIs. Variables reaching threshold of p < 0.10 on univariate analysis will be put forward to the multivariate regression analysis. P < 0.05 will be used to define statistical significance.

Two published risk prediction tools will be evaluated to assess their accuracy in predicting SSIs in this cohort (supplementary material 6) [2], [5]. A receiver operating characteristic (ROC) curve will be plotted for each tool, with calculation of area under the curve (AUC) as the summary statistic.

Centre eligibility and team roles

The GIVE audit is open to all centres that provide an elective and/or emergency arterial vascular service. UK vascular units often comprise of a Hub and Spoke type model. A registered Hub site may be able to undertake data collection for Spoke sites, depending on practicalities and local policies, without registering the Spoke site separately. However, it is noted that the majority of arterial operations will be performed at the Hub site therefore data collection from Spoke sites is likely to be minimal.

Each centre will require the support of a supervising consultant, and a data collection team. The supporting consultant is expected to facilitate audit registration/ethical approval, provide unit support for engagement with the GIVE study, act as a guarantor to data capture and upload/transfer, provide workplace-based assessments documentation for team members, and facilitate local presentation of the audit results at an appropriate local meeting.

Ethics and registration

Decision-making tools provided by the NHS Health Research Authority (HRA) (“Is My Study Research?” [22] and “Do I need NHS Research Ethics Committee (REC) approval?” [23]) were used to determine that this study does not require approval from an NHS Research Ethics Committee.

The GIVE audit is required to be registered in each participating centre prospectively, prior to data collection. Participating centres that are situated outside of the UK will need to comply with local regulations prior to commencing, which may require prospective approval from an ethics committee. The agreement of a consultant vascular surgeon (or equivalent for non-UK centres) is required to oversee the process.

The GIVE Audit received Caldicott Guardian approval from the corresponding author’s institution to store pseudoanonymised data on secure NHS computers, and undertake appropriate statistical analysis.

Dissemination and authorship

Data will be submitted for presentation at national and international meetings. A paper, or papers of the resultant data will be submitted for peer-reviewed publication. A writing team, including those involved with the design, implementation and dissemination of the GIVE audit, and those contributing to data analysis in the cohort study will be responsible for both presentation(s) and publication(s). For both presentation(s) and publication(s) a collaborative authorship model will be used. To qualify for PubMed-citable collaborative co-authorship individuals must have either:

• 1.  had a significant role in the set up and management of the GIVE study, including audit department registration, creation of a data collection team and engagement with VERN to ensure timely upload of data (with validation as required) and completion of the questionnaire
  OR
• 2.  captured sufficient data to warrant authorship – this would be the equivalent of collecting baseline and follow up data on approximately 10 patients, although it is appreciated individuals may participate in only baseline data collection or only follow up data capture. Data collection is expected to be complete (>95% variables completed), and submitted within 7 months of starting data collection
  OR
• 3.  (for consultants) provide oversight and support as detailed in the “Centre Eligibility and Team Roles” section.

It is encouraged that data for each centre is presented locally. Summary data for each participating site will be provided to the local audit leads, with comparisons to the group average.