Anaesthesia for pancreatic resection surgery: part 2

Learning objectives.

By reading this article, you should be able to:

• •Explain the importance of risk stratification and the challenges of preoptimisation.
• •Discuss the perioperative considerations for patients undergoing pancreatic resection.
• •Detail the management of pancreatic neuroendocrine tumours.

Key points.

• •Risk factors for increased morbidity and mortality include advancing age, cardiopulmonary comorbidities and preoperative sepsis.
• •Cardiopulmonary exercise testing may predict postoperative mortality, long-term survival and risk of pancreatic fistula.
• •Anaemia, nutrition, alcohol intake and smoking should be addressed before surgery.
• •Laparoscopic techniques have yet to become the mainstay for pancreaticoduodenectomy.
• •Enhanced recovery recommendations have been successfully implemented for pancreaticoduodenectomy.

This article discusses the perioperative management of patients undergoing pancreatic resection surgery. The presentation, diagnosis and treatment of pancreatic cancer and surgical options for benign lesions of the pancreas has been reviewed in a recent related article in this journal (Jarvis M. S., Laing R. W., James A. Anaesthesia for pancreatic resection surgery: part 1. BJA Education 25; 2: 57–64).

General considerations

Most patients presenting with pancreatic cancer have evidence of metastases at the time of diagnosis. In England, Northern Ireland and Wales, only 5.7% of patients present with resectable pancreatic cancer and 26.7% with borderline resectable or locally advanced tumours.¹

Uncontrolled pain and adverse effects from opioid analgesia can be distressing for patients and their families. Neurolytic coeliac plexus block or stereotactic ablative radiotherapy may be considered for these patients. After the decision that a lesion is unresectable, these procedures should be considered early rather than only as a rescue analgesic strategy. Psychological support may aid to address the impact of fatigue and depression whilst pancreatic enzyme replacement therapy can aid with digestive symptom control. For those with unresectable disease and actual, or impending intestinal obstruction, a percutaneous jejunostomy may be performed. In patients with resectable disease, pancreas resection is still the cornerstone of curative treatment, though it is associated with a high postoperative morbidity rate of up to 50%.

Preoperative assessment

Risk stratification scores

The peak incidence of pancreatic cancer in the UK is in those aged 85–89 yrs, and pancreatic surgery itself carries significant morbidity and mortality.¹ In patients with disease amenable to resection, suitability for surgery is determined by assessment of individual risk factors. An analysis of American College of Surgeons National Surgical Quality Improvement Programme (ACS-NSQIP) data from 7571 patients who had undergone a pancreatic resection between 2005 and 2008, identified 10 risk factors for increased mortality, serious morbidity and overall morbidity.² These are listed in Table 1.

Table 1.

Risk factors associated with increased mortality, serious morbidity or overall morbidity after pancreatic resection surgery.

Age >74 yrs	ASA grading >2
Male sex	Coronary heart disease
BMI >40 kg m−2	Dyspnoea on moderate exertion
Preoperative sepsis	Bleeding disorder
Dependent functional status	Proximal or total pancreatectomy

The ACS-NSQIP risk calculator has been modified for pancreatic resection surgery. A study of 307 patients with pancreatic ductal adenocarcinoma undergoing surgery with curative intent compared prognostic stratification with Heidelberg score, the Memorial Sloan Kettering Cancer Centre (MSKCC) nomogram, Botsis score, early mortality risk score and McGill Brisbane symptom scores. The Heidelberg score and MSKCC nomogram provided superior risk stratification.³ Further assessment of risk scoring prediction models in these patients remains an area for future research.

To date, risk prediction models have focussed on postoperative pancreatic fistulae, a major complication of pancreatic surgery.

Specialised multidisciplinary teams, including anaesthetists, in pancreatic surgical centres may be able to provide advice on broad risk stratification for individual patients. This can present challenges when consenting a patient for anaesthesia and surgery as the evidence of the impact of individual comorbidities and biochemical abnormalities is limited.

Cardiopulmonary exercise testing

At the time of diagnosis, patients with pancreatic cancer often suffer from severe weight loss and loss of physical fitness, known as (cancer) cachexia. In addition to clinical investigations targeted to clinical findings or medical history (e.g. echocardiography in the presence of a murmur), functional assessment with use of cardiopulmonary exercise testing (CPET) may aid decision-making, risk stratification and potentially preoperative optimisation.

When analysing CPET results, the $\dot{V}$ o_2max (or $\dot{V}$ o_2peak) is defined as the highest oxygen uptake attained and anaerobic threshold (AT) is the rate of oxygen uptake above which anaerobic metabolism takes over and arterial lactate concentrations start to increase. Ventilatory equivalent of carbon dioxide (Ve/Vco₂) is the ratio of minute ventilation to carbon dioxide output.⁴ Outcomes specific to pancreatic surgery: AT <10 is demonstrated to predict extended length of stay, higher rate of pancreatic leak and higher morbidity after pancreatic surgery.⁵ A single study also demonstrated a Ve/Vco₂ >41 was associated with poorer long-term survival.⁶

Preoperative optimisation and prehabilitation

As pancreatic cancer often at advanced age, many patients have several preexisting comorbidities that require investigation and optimisation. This may be limited because of late presentation and short timeframes before surgery. However, identification and treatment of confounding factors such as anaemia can be achieved in centres with established pathways. Separate interventions are increasingly integrated into bundles to improve efficiency. Trials to investigate the benefit of stepped interventions before pancreatic resection are ongoing such as the PROMISE-P study investigating a stepped cluster of interventions before pancreatic resection with the aim to reduce time to functional recovery.⁷

Anaemia

Iron deficiency anaemia is common amongst patients with cancer and is associated with worse postoperative outcomes in major non-cardiac surgery.⁸ Perioperative allogenic blood transfusion is associated with decreased disease-free survival and overall survival after surgery for pancreatic cancer with curative intent.^9,10 Whether blood transfusion is an independent risk factor for poor prognosis remains controversial. Blood transfusion may simply reflect a more complex and advanced cancer or patients with more complex comorbidities.¹¹

A single-centre retrospective study showed that preoperative anaemia is associated with higher mortality, major complication rate, blood transfusion rate and overall complication rate for proximal pancreatic resections. An increased incidence of delayed gastric emptying was also noted. No increase in morbidity was noted for distal pancreatectomy.¹²

Early identification of anaemia, determination of the cause and treatment with i.v. iron therapy, if appropriate, can reduce the requirement for perioperative transfusion. International consensus statement on the perioperative management of anaemia and iron deficiency suggest that in the presence of inflammation, a serum ferritin concentration <100 mg ml⁻¹, transferrin saturation <20%, or both, iron deficiency is likely present. It also suggests that i.v. iron should be first-line therapy if surgery is planned within 6 weeks of diagnosis.¹³

Nutrition

Many cancers are associated with anorexia, weight loss, tissue wasting and malnutrition. Pancreatic cancer has the highest incidence of cachexia of all cancer types. Up to 80% of patients have cachexia at the time of diagnosis and this is associated with a worse outcome after resection surgery.¹⁴

The European Society for Clinical Nutrition and Metabolism (ESPEN) guidelines recommend to regularly evaluate nutritional intake, weight change and BMI after a diagnosis of cancer. They recommend brief, inexpensive, sensitive tests such as the malnutrition universal screening tool (MUST) or Nutrition Risk Screening 2002 (NRS-2002).¹⁵ After abnormal screening, this should be followed by a more detailed assessment to provide a ‘malnutrition score’ with assessments such as the subjective global assessment (SGA) or patient-generated SGA.^16,17 Departmental preference dictates which assessments are preferred and exact thresholds for intervention. However pancreatic cancer patients, or those with recurrent acute or chronic pancreatitis, often score highly enough to justify early nutritional support.

Poor nutritional status in patients with pancreatic cancer may be related to reduced caloric intake, the effects of pancreatic insufficiency on effective digestion or the catabolic effects of the tumour on the host. Multidisciplinary teams frequently involve dietitian input to address these nutritional challenges at an early stage with a view to improving conditions at the time of surgery. Nutritional support may comprise high protein intake (1.5 g kg⁻¹ day⁻¹), additional vitamins and supplementation of pancreatic enzymes.

Sarcopenia

A progressive loss and decline in quantity and quality of skeletal muscle mass is normal with ageing. However the process is accelerated by cancer and by chronic inflammatory conditions such as pancreatitis. Sarcopenia is associated with worse outcomes in pancreatic cancer. Studies show conflicting results on the impact of sarcopenia on morbidity after pancreatic surgery. However long-term survival after pancreatic cancer surgery is affected.¹⁸ Sarcopenic obesity carries an additional association with increased rates of postoperative pancreatic fistulae.¹⁹ Although nutrition impacts on sarcopenia, exercise therapy to increase the quality of skeletal muscle mass may be challenging to implement in the short timeframes associated with pancreatic cancer surgery. Preoperative chemotherapy accelerates the rate of loss of muscle mass and serves to further exacerbate the level of sarcopenia.

Smoking cessation

Smoking is an independent risk factor for postoperative pulmonary complications. When combined with prolonged surgery close to the diaphragm, it places these patients at particularly high risk. Intervention early in preparation for potential pancreatic cancer surgery provides the maximum amount of time for the effects of tobacco smoking to be minimised. Brief advice from a physician, such as in the preoperative clinic, increases the rate of stopping smoking by up to 3%, with more intensive interventions being more effective.²⁰

Alcohol

Alcohol abuse is associated with a two-to three-fold increase in postoperative morbidity after surgery.²¹ In addition to the malnutrition associated with chronic alcohol misuse, there is also an increased risk of infection, bleeding, postoperative delirium and subclinical cardiac insufficiency. Sudden, prolonged abstinence after surgery also places patients at risk of withdrawal syndrome. This can be anticipated in some patients and appropriate withdrawal prophylaxis started, such as benzodiazepines to reduce symptoms of withdrawal, and vitamin B or thiamine complexes to prevent and treat Wernicke's encephalopathy.

Glucose management

Pancreatic disease may cause abnormalities of glucose control requiring careful perioperative management. For patients with preexisting diabetes, haemoglobin A_1c can provide a useful screen of blood sugar control over the previous 2–3 months. It can also be a useful screen for new diagnoses of diabetes. Many patients with chronic pancreatitis will progress to eventually require insulin therapy.

Patients requiring insulin therapy before surgery should be managed with a variable-rate infusion of insulin in the perioperative period until normal oral intake and insulin dosing is resumed. Blood sugar concentrations may be more difficult to control with insulinomas and glucagonomas; specific advice from a neuroendocrinologist may be sought.

Direct consequence of the tumour

Painless obstructive jaundice is the textbook presentation of pancreatic cancer. A mass compressing the biliary system as it passes through the head of the pancreas causes obstructive jaundice. In addition to local effects on the liver and biliary tree, jaundice can also cause systemic effects such as haemodynamic instability, acute renal failure, myocardial depression, coagulation abnormalities, immunosuppression, nutritional impairment and poor wound healing. Routine endoscopic stenting does not reduce mortality; and although it reduces morbidity, the increase in procedure-related complications (infection, sepsis, failure, damage to surrounding tissues) is likely to limit these benefits.²² While the obstruction may not be relieved before surgery, these sequelae must be addressed before surgery before the decision to operate is taken.²³ Indications for preoperative biliary stenting are discussed in our previous related article.

Gastric outlet obstruction may be caused by direct compression of a pancreatic mass on the stomach or duodenum, or abnormalities in gut motility may occur by invasion into the coeliac plexus. This should be thoroughly investigated before surgery as it may indicate that the tumour has grown further or spread and is no longer resectable. Gastric outlet obstruction primarily presents with nausea and vomiting and if prolonged, dehydration, electrolyte abnormalities and alkalaemia. If a patient with gastric outlet obstruction presents for surgery then they should be anaesthetised as though they have a full stomach regardless of recent oral intake, and a rapid sequence induction performed.

Secretory neuroendocrine tumours

Neuroendocrine tumours account for <5% of pancreatic tumours and only 10–30% of these are secretory (functional) in nature. European Neuroendocrine Tumour Society (ENETS) consensus guidelines for the perioperative management of patients with neuroendocrine tumours suggest the following measures:²⁴

• •Gastrinoma: give a high-dose proton pump inhibitor to control gastric acid hypersecretion and limit the risk of gastrointestinal perforation and haemorrhage, including after surgery.
• •Insulinoma: diazoxide may be started in patients with significant hypoglycaemic episodes. However it is recommended that this is stopped 24 h before surgery because it may cause severe fluid retention and oedema. Intravenous dextrose infusion may be required to avoid hypoglycaemic episodes and close monitoring of blood glucose concentrations intraoperatively is essential.
• •Glucagonoma: cachexia and skin manifestations may be limited by amino acid infusion, somatostatin analogue and antibiotics before surgery. There is an increased risk of venous thromboembolic events and so high-dose prophylaxis should be initiated.
• •VIPoma: associated diarrhoea and electrolyte abnormalities must be corrected before surgery and somatostatin analogue therapy may be required.

Intraoperative management

Surgical approach

Knowledge of the surgical approach to pancreatic surgery is vital to plan anaesthesia. Pancreaticoduodenectomy and total pancreatectomy are most commonly undertaken as an open laparotomy via an upper midline or subcostal incision. A variety of laparoscopic techniques—including laparoscopically assisted, fully laparoscopic and robotic surgery—have been described.

Laparoscopic pancreaticoduodenectomy is a very challenging procedure which requires advanced laparoscopic skills and initial attempts were of very long duration. The LEOPARD-2 trial was stopped early after 99 of the projected 136 patients were included, because of a five-times higher 90-day complication-related mortality rate with laparoscopic compared with open technique (5/50 [10%] vs 1/49 [2%], p=0.20).²⁵ Surgical experience and skills have improved and, although still of longer duration than the open approach, laparoscopic surgery is associated with reduced blood loss, intraoperative transfusion and postoperative pain and infection.²⁶ Some centres have adopted a robotic approach and have performed >1500 robotic hepatopancreatobiliary (HPB) cases claiming equivalent if not superior outcomes.²⁷

By contrast, laparoscopy with or without robot, is the first-choice approach for distal pancreatectomy. This approach benefits from reduced intraoperative blood loss, shorter hospital stay, higher rate of splenic preservation, earlier oral intake and reduction in surgical site infections over open resection.²⁸ Where open resection is required, for example for bulky tumours or those with vascular involvement, an upper midline laparotomy with or without left subcostal incision is usually preferred.

Analgesia

Open pancreatic surgery presents a challenge to provide adequate postoperative analgesia. In many centres, epidural analgesia remains the gold standard analgesic technique for open procedures such as pancreaticoduodenectomy. It is associated with lower pain scores, and reduction in postoperative pulmonary, cardiac, renal and venous thromboembolic complications.29, 30, 31 However, epidural analgesia has a failure rate even in experienced hands and is associated with longer intensive care admissions and need for vasopressor and higher rates of postoperative pancreatic fistulae.^32,33

Other approaches are in common use, including intrathecal opioids and local anaesthetic wound or abdominal wall infusions. The PAKMAN trial demonstrated no difference in overall survival at 3–5 yrs, recurrence-free survival, long-term effects on quality of life or pain severity between epidural analgesia and patient-controlled i.v. analgesia (PCIA) after pancreatic surgery.³⁴ A second study demonstrated no difference in pancreatic fistula, bile leakage, delayed gastric emptying, gastrointestinal bleeding or postoperative ileus within 30 days between epidural analgesia and PCIA.³⁵

Anaesthesia and cancer survival

Evidence of differences in outcomes between total i.v. and volatile anaesthetic techniques is currently limited to retrospective studies. However the impact of anaesthetic technique on disease-free survival after oncologic surgery is an area of current research within anaesthesia.

Prolonged surgery

Because of the complex anatomy and number of delicate anastomoses, pancreatic resection surgery is often of prolonged duration. As with all prolonged surgeries, consideration must be given to careful positioning with protection of pressure points, maintenance of normal temperature and use of prophylactic antibiotics, often with repeat intraoperative dosing or continuous infusion. Venous thromboembolism prophylaxis should be given in line with local protocols and the timing of insertion and removal of epidural catheters should be considered carefully. Graduated compression stockings and intermittent mechanical compression devices may be used in the immediate perioperative period as an adjunct to chemical prophylaxis.

Fluid balance and haemodynamic monitoring

For open anastomotic surgery, a fluid-restrictive approach is often advised to mitigate the risk of tissue oedema impairing the formation of the anastomoses—in particular the pancreaticojejunostomy. However, intraoperative fluid therapy has been the subject of little high-quality comparative research.

A higher positive postoperative fluid balance (>6795 ml at 48 h and >6999 ml at 72 h) is associated with increased rates of blood transfusions, cardiopulmonary events, overall morbidity and longer hospital and critical care admissions in pancreaticoduodenectomy surgery.³⁶ However, the analysis of fluid balance and its impact on outcomes is complex and a higher positive fluid balance may reflect a more profound systemic inflammatory response to surgery, rather than complications developing purely as a result of positive fluid balance.

Cardiac output monitoring, such as oesophageal Doppler, may be used to guide intraoperative fluid therapy but has not demonstrated a decrease in length of stay or incidence of complications after the Whipple procedure.³⁷ The OPTIMISE II trial demonstrated no reduction in postoperative infection, acute kidney injury or mortality with the use of cardiac output monitoring in major elective gastrointestinal surgery.³⁸

Monitoring of urine output and serial blood gas analysis may be used to aid monitoring of fluid status although they are nonspecific.

The close anatomical relations of the pancreas to major vessels presents a risk of sudden and brisk haemorrhage. In addition, handling around the liver, inferior vena cava (IVC) and the portal vein can impact venous return causing temporary haemodynamic instability. While periodic IVC compression may be considered a ‘normal’ part of surgery to the head of the pancreas, clear communication between the anaesthetist and surgeon is vital; it is important to note that a bleeding point may not be immediately obvious. Whilst increasing circulating volume can reduce the impact of IVC compression on venous return and therefore cardiac output, highly positive fluid balance makes pancreatic anastomosis more challenging.

Optimal blood pressure targets are yet to be defined in these patients. However in our centre, we aim to maintain a MAP >65 mmHg.

Venous resection and reconstruction may be required depending on tumour location and local spread, usually involving the portal vein or superior mesenteric vein. This presents a risk not only of intraoperative haemorrhage and reduced venous return, but also of hepatic ischaemia that may result in hepatic dysfunction in the early postoperative period.

Somatostatin analogues

Somatostatin analogues, such as octreotide, reduce pancreatic exocrine secretion, theoretically reducing the incidence of pancreatic fistulae or promoting their healing when they do occur. Meta-analysis demonstrates a reduced rate of pancreatic fistulae with its use, but the rate of clinically significant fistulae and morbidity and mortality were unchanged, even with subgroup analysis.³⁹ Somatostatin analogues are not recommended routinely.

Somatostatin analogues may still be requested by the surgical team intraoperatively. In addition to having a role in the management of secretory tumours, they have a number of other important actions, including delaying late gastric emptying, reducing portal vein pressure and an antidiuretic effect. A disturbance in glucose homeostasis may be observed, and rarely, a sinus bradycardia.

Postoperative management

Level of monitoring

After pancreaticoduodenectomy or total pancreatectomy, patients are commonly admitted to ICU or other higher dependency area to enable ongoing monitoring of analgesia, fluid status, glucose and electrolyte abnormalities and to observe for early postoperative complications.

Glucose control

Close monitoring of blood glucose concentration is required in both the intraoperative and postoperative periods because of the potential for impaired glucose homeostasis. This is particularly important for patients undergoing total pancreatectomy, which results in a unique form of diabetes, sometimes referred to as ‘pancreatic diabetes’, characterised by frequent hyper- and hypoglycaemic episodes. These patients will require a variable-rate insulin infusion. A systematic review including numerous studies demonstrated significantly decreased quality of life compared with the general population.⁴⁰ However the same review notes that diabetes-related mortality is reported only in patients treated before 2005, possibly pointing toward an improvement in diabetes control since this time.⁴⁰

Where total pancreatectomy is performed for chronic pancreatitis, islet cell autotransplantation may be considered to negate many of the problems with pancreatic diabetes.

Postoperative complications

Pancreatic fistula formation is a serious complication of pancreatic surgery. In 2016, the International Study Group of Pancreatic Fistula developed a definition of clinically relevant pancreatic fistula as ‘a drain output of any measurable volume of fluid with an amylase concentration more than three times the upper limit of institutional normal serum amylase activity, associated with a clinically relevant development/condition related directly to the postoperative pancreatic fistula’.⁴¹ This definition recognises that most leaks resolve with time and without further intervention and only persistent leaks form fistulae.

Leaks into the retroperitoneum may cause sepsis or erode through tissues and blood vessels with the risk of major haemorrhage. Exact incidence of this complication varies widely, mainly because of a history of varying definitions. Pancreatic leaks are typically treated by conservative management in the absence of clinical signs of peritonitis, sepsis or organ failure.

Whilst distal pancreatectomy, especially if the spleen is preserved, may be considered a less substantial procedure, morbidity remains high with some observers noting a higher pancreatic leak rate than is associated with pancreaticoduodenectomy.⁴²

Intra-abdominal abscess formation is often a result of an anastomotic leak and typically collects under the liver or diaphragm.

Haemorrhage rates vary with different types of pancreatic resection. Early haemorrhage typically presents within the first 24–48 h. Inadequate haemostasis at the end of surgery or a slipped tie are common causes. However in those patients with obstructive jaundice, diffuse retroperitoneal haemorrhage may be observed. Jaundice is a significant risk factor for postoperative haemorrhage, likely indicating an underlying coagulation abnormality.⁴³

Late haemorrhage, typically 1–3 weeks after surgery, is often more challenging to manage. Causes include an anastomotic leak eroding into retroperitoneal blood vessels and pseudoaneurysm formation. Should the haemorrhage be localised to the pancreaticojejunostomy (in the case of pancreaticoduodenectomy), the only surgical options may be refashioning of the anastomosis or a completion pancreatectomy.

Delayed gastric emptying is the leading cause of procedure-related morbidity after pancreaticoduodenectomy.¹⁶ Whilst there remains no widely accepted definition of delayed gastric emptying, it causes significant patient discomfort, requires prolonged gastric decompression, longer hospital stay and higher healthcare costs. Whilst there is evidence that i.v. erythromycin reduces the incidence of delayed gastric emptying after pancreaticoduodenectomy, persistent gastric stasis must prompt investigation for more sinister causes such as an undetected anastomotic leak.⁴⁴

Postoperative respiratory failure presents a significant challenge after major intra-abdominal surgery. Common causes after pancreatic surgery include pain, aspiration, ileus and intra-abdominal collection. Patients undergoing pancreatic resection surgery are at particular risk because of the prolonged nature of their surgery, the location and length of open incisions (if used), poor nutrition, significant delay in mobilisation after surgery, the acute nature of many patients presenting for surgery, the need for prolonged gastric decompression and the frequency of anastomotic leaks. Typically, patients present with respiratory distress on postoperative days 3–7.

Enhanced recovery

Enhanced recovery pathways

Enhanced recovery programmes are now widespread throughout many surgical specialties. Enhanced recovery after surgery (ERAS) guidelines for pancreaticoduodenectomy were updated in 2020 and include many of the measures included in this text.⁴⁵ Some of these recommendations are based on data from other major abdominal surgeries where data specific to pancreatic surgery are limited. These recommendations are summarised in Table 2.

Table 2.

Enhanced Recovery After Surgery (ERAS) guidelines for pancreaticoduodenectomy modified in 2020 from original guidance produced by the ERAS® society. Summary recommendations modified from those presented in Melloul and colleagues,⁴⁵ Guidelines for Perioperative Care for Pancreatoduodenectomy: Enhanced Recovery After Surgery (ERAS) Recommendations 2019. World J Surg 2020.

Preoperative recommendations

3–6 Week prehabilitation programme before major surgery

Avoid preoperative biliary drainage unless bilirubin >250 μmol L−1

At least 4 weeks abstinence from smoking and abstinence from heavy alcohol consumption

Preoperative nutritional intervention if >15% weight loss or BMI <18.5 kg m−2

Perioperative oral immunonutrition is not recommended

Perioperative

Preoperative fasting 6 h for solids, 2 h for liquids if no specific risk factors Carbohydrate loading

Anxiolytic drugs should be avoided Premedication with paracetamol and single dose of gabapentinoid NSAIDs can be used after surgery if renal function allows

Venous thromboembolism prophylaxis with heparin whilst in hospital, and for 4 weeks in patients with cancer Mechanical measures in addition to chemical prophylaxis

Single-dose antibiotics <60 min before skin incision, repeat doses may be necessary

Thoracic epidural analgesia for open pancreaticoduodenectomy Continuous wound infiltration may be an alternative Postoperative multimodal opioid-sparing analgesia strategy

All patients should receive postoperative nausea and vomiting prophylaxis

Postoperative

Optimal glycaemic target remains unclear. However glucose concentrations should be maintained as close as possible to normal without compromising safety

Nasogastric tubes should not be maintained after surgery

Avoid fluid overload

Early drain removal at 72 h, if amylase content in drain <5000 U L−1 on day 1 after surgery

Somatostatin analogues cannot be recommended

Urinary catheters can be removed on postoperative day 1 or as soon as the patient is ambulant

In patients with prolonged delayed gastric emptying, artificial nutrition may improve outcome

Stimulation of bowel movement: •Chewing gum is safe and may accelerate bowel recovery •Alvimopan and mosapride improve ileus

Patients should be allowed a normal diet after surgery Artificial nutrition should be considered based on an individualised approach and nutritional assessment. The enteral route is preferred

Early and scheduled mobilisation from postoperative day 0

Processes

Laparoscopic pancreaticoduodenectomy should only be performed in highly experienced, high-volume centres

Regular audit and feedback to improve compliance and outcome

No such ERAS guidelines exist for other pancreatic surgery including distal pancreatectomy.

Conclusions

Anaesthesia for pancreatic resection surgery is complex and despite surgical advances, remains high risk. Given the short timeframe from diagnosis of pancreatic cancer to surgery, opportunities for prehabilitation are limited unless highly efficient pathways are in place. Anaesthetists have a key role to play in the counselling and perioperative care of these patients.

Complications are common and remain a challenge for pancreatic surgeons and perioperative physicians. Enhanced recovery after surgery guidelines for pancreaticoduodenectomy provide a framework for audit and quality improvement work focussing on improving outcomes after pancreatic resection. However they are limited in their scope.

Endocrine pancreatic neoplasms remain a rare subtype of pancreatic cancer but can present extra challenges for the anaesthetist and perioperative management.

MCQs

The associated MCQs (to support CME/CPD activity) will be accessible at www.bjaed.org/cme/home by subscribers to BJA Education.

Declaration of interests

The authors declare that they have no conflicts of interest.

Biographies

Michael Jarvis BSc FRCA is a consultant anaesthetist at University Hospitals of Derby and Burton NHS Foundation Trust. His interests include regional anaesthesia, anaesthesia for vascular surgery and upper gastrointestinal surgery.

Richard Laing BSc Med Sci PhD FRCS is a consultant hepatobiliary and pancreatic surgeon at University Hospitals of North Midlands NHS Trust. His interests include diagnostic pathways and management of pancreas cancer. He is the West Midlands upper gastrointestinal surgery lead for the faster diagnosis standard.

Alex James BMedSci Dip MC FRCA is a consultant anaesthetist at University Hospitals of North Midlands NHS Trust. His areas of interest are anaesthesia for upper gastrointestinal surgery, trauma resuscitation and prehospital emergency medicine.

Matrix codes: 1A01, 2A03, 3A03