There is a high rate of R1 resection following patients undergoing pancreaticoduodenectomy for pancreatic cancer. The most commonly positive margin is the SMA. Peri-adventitial dissection has been proposed as an effective method of achieving R0 margins. There is lack of standardisation of the proposed technique and no grade 1 evidence to support routine use of this technique.

The goal of this randomised controlled trial is to investigate the role of routine peri-adventitial dissection on the SMA margin status.

Pancreatic cancer is the 4th cause of cancer death in Europe. With the existing treatments the overall 5-year survival remains 8%. For patients with non-metastatic disease, many are conventionally judged unresectable with just 10-20% eligible for upfront surgery and a 5-year survival of 10-30%. Microscopically positive margins (R1) have been associated with poorer survival of patients undergoing pancreaticoduodenectomy. The incidence of R1 varies between 16-79% in the published literature and the median survival for patients undergoing R0 resections is between 19-21 months versus 9-13 months for patients undergoing R1. Therefore, to investigate strategies to decrease the rate of R1 resection is crucial to significantly improve the patient survival1.

The SMA margin is positive in 47-77% of these resections, even when the margin is "clear" on the pre-operative staging CT2,4. Many different techniques are in practice and have been described for this part of the pancreaticoduodenectomy procedure, including the use of diathermy and ties, energy devices such as Ligasure, and staplers. The main denominator is staying close to the SMA and removing the so called "mesopancreas"3.Peri-adventitial arterial dissection is a technique that has been a commune practise in several pancreatic centres and has been proposed as an effective surgical technique to achieve R0 margins. On the other hand, SMA dissection may theoretically increase the risk of acute haemorrhage, vascular injury, and thrombosis, development of pseudo-aneurysms or persistent chyle leak due to the extensive lymphoneural dissection. Furthermore, the published evidence reveal variability in the definition of a "positive margin", as well as the lack of accurate description of the surgical technique in the majority of the studies, with only about 25% of the studies reporting the use of the technique4. Moreover, none of the published studies have documented standardisation in the use of the technique with regards to the extent of the periadventitial dissection on the SMA, longitudinally and circumferentially.

In the University Hospitals of Birmingham NHS Trust, pancreatic surgery is centralised in Queen Elizabeth Hospital. Peri-adventitial dissection is selectively performed in cases where there is a close relation of the tumour to the arteries in an effort to achieve an R0 resection. Routine practice of peri-adventitial dissection during pancreaticoduodenectomy for resectable tumours has been performed in a subset of patients based on surgeon's preference. The results suggest a R1 rate of 20%, with SMA margin positivity 5%, compared to a rate of 44% margin positivity within the unit, with 54% SMA R1 rate. Furthermore, complication rates are comparable to the average described by the unit's results and published literature (20.8% of grade B and above complications as per the Clavien-Dindo classification). More specifically, there have been no incidences of vascular related injury or complication related to the technique; or persistent chyle leak (one episode of chyle leak grade A recorded that was managed conservatively).

The goal of this prospective study is to investigate the role of routine peri-adventitial dissection on the SMA margin status. Primary outcome is SMA margin status. Secondary outcomes are: length of stay, complications, overall survival, disease free survival and adjuvant treatment uptake.

This will be a two arm, randomised 1:1 (periadventitial dissection : no periadventitial dissection) study. A pilot phase will determine feasibility to proceed with the rest of the trial and to determine:

1. Feasibility of patient recruiting and randomisation

2. Feasibility and reproducibility of the technique

3. Feasibility of use of intraoperative photography

4. Intraoperative and postoperative complication rate and severity (Clavien-Dindo classification)

5. Positive SMA margin status rate

   Feasibility criteria to be met in order to proceed with the rest of the trial:

6. Recruitment of at least 20 patients in 6 months (both arms)

7. Periadventitial dissection technique to be performed in at least 80% of the randomised patients to that arm as documented by intraoperative photography

8. Increase in the complication rate no more than 10%

For the full trial a sample size of 51 patients per arm has been based on a margin positivity of 44%, based on the unit's results, and expected 60% reduction (expected R1 18%) at power 80% (type A error 0.05).

Inclusion Criteria:

• All adult (≥ 18 years old) patients with a non-metastatic (based on NCCN definition) on imaging pancreatic tumour of the head or uncinate process treated with a pancreaticoduodenectomy

Exclusion Criteria:

1. All paediatric patients (< 18 years old)
2. Patients that cannot provide consent
3. All borderline, locally advanced and metastatic pancreatic tumours on imaging (based on NCCN criteria)
4. All patients with a cytological or histological diagnosis of cholangiocarcinoma, ampullary and duodenal carcinoma
5. All patients with benign disease or dysplasia with no evidence of malignancy

Sealed envelope system for allocation of intervention Master list held by third party, researcher and surgical team aware of intervention Intervention NOT documented in notes making clinical follow up team, investigator and outcomes assessor blind to arm allocated