Resectable pancreatic head cancer presenting with grade III cholangitis
Mrs D, 64, presents with a three-week history of painless jaundice, dark urine, pale stools and 6 kg of weight loss. Bilirubin 312 µmol/L, ALP 480 U/L, ALT 90 U/L, CA19-9 840 U/mL, haemoglobin 104 g/L, albumin 31 g/L. Contrast CT shows a 2.8 cm hypodense mass in the pancreatic head; the superior mesenteric artery, coeliac axis and portal vein are clear, and there are no hepatic or pulmonary metastases. On the second evening she develops rigors, temperature 39.1 °C, right upper quadrant pain, heart rate 118, blood pressure 96/58 after 1 L of crystalloid, respiratory rate 24 and lactate 2.4 mmol/L; she is drowsy but rousable. White cell count 19.6 × 10⁹/L and creatinine 156 µmol/L. She takes no anticoagulants. The ERCP list is full tomorrow and the next slot is in 48 hours; percutaneous transhepatic drainage is available today, and the pancreatic multidisciplinary meeting is on Friday.
1. Classify the cholangitis and set out your management for the next 12 hours, including the route and timing of biliary drainage.
Read the worked answer
This is grade III (severe) acute cholangitis. Grade III requires organ dysfunction, and she has at least two domains: cardiovascular (hypotension and a lactate of 2.4 mmol/L) and neurological (drowsiness). The creatinine of 156 µmol/L is abnormal but does not meet the renal criterion of above 2 mg/dL.
First six hours: high-flow oxygen, two large-bore cannulas, balanced crystalloid boluses with reassessment after each, urinary catheter and hourly output, blood cultures before antibiotics, and broad-spectrum cover for Enterobacterales and anaerobes (for example piperacillin–tazobactam, or ceftriaxone plus metronidazole), with dose adjustment for renal function. Check a full blood count, coagulation, blood gas and group and save. Nil by mouth, analgesia, antiemetics, glucose monitoring. If she remains hypotensive after 30 mL/kg of crystalloid, involve critical care for vasopressor support.
Drainage: the source is the biliary tree, so antibiotics alone will fail. Grade III disease needs drainage within 12–24 hours; mortality climbs steeply when drainage is delayed. ERCP with sphincterotomy and stent is first line where it can be delivered in that window. It cannot be delivered here for 48 hours, so percutaneous transhepatic drainage today is the correct choice and is far preferable to waiting. EUS-guided drainage is an alternative where local expertise exists.
Stent choice: a plastic stent is adequate if curative surgery is planned within about six weeks; a covered self-expandable metal stent is preferred if neoadjuvant chemotherapy or palliation is likely, because of longer patency, accepting a higher risk of cholecystitis and some added difficulty at surgery.
2. After successful drainage she asks whether she needs 'the big operation'. The MDT debates upfront pancreaticoduodenectomy against neoadjuvant chemotherapy. What determines the answer, and how do the recent cholangitis and the stent affect timing?
Read the worked answer
Resectability: on the described CT the tumour abuts neither the SMA, coeliac axis nor portal vein, so this is anatomically resectable. But resectability is not the same as an indication for immediate surgery. Very high CA19-9, significant weight loss, a low albumin and a recent severe infection all raise the risk of early recurrence and postoperative complications; many MDTs would offer neoadjuvant FOLFIRINOX or gemcitabine–nab-paclitaxel in this setting, and would consider staging laparoscopy to exclude occult peritoneal disease before committing to a Whipple.
If upfront surgery is chosen, timing matters. Operate only once the infection has resolved, the bilirubin is falling, nutrition has been optimised (dietitian, pancreatic enzyme replacement, diabetes assessment) and she has recovered physiologically - typically four to six weeks after the cholangitis. A plastic stent left in situ for months blocks, migrates and seeds the bile duct, increasing infective complications and surgical site infection after pancreatoduodenectomy; if surgery is deferred more than about six weeks the stent should be exchanged, and it should be removed at operation.
Counselling: frame it honestly - in a high-volume centre, pancreatoduodenectomy carries roughly 2–5% mortality but 40–50% morbidity (pancreatic fistula, delayed gastric emptying, bleeding, infection), plus new diabetes and exocrine insufficiency. The recent cholangitis is not a reason to abandon surgery; it is a reason to sequence it correctly.
3. The ERCP attempt fails because of a tight distal stricture. How do you choose the next drainage route?
Read the worked answer
Failed ERCP in grade III cholangitis is an emergency, not a scheduling problem; the decision must be made within hours.
Options: percutaneous transhepatic biliary drainage, EUS-guided choledochoduodenostomy or hepaticogastrostomy, or surgical drainage (rarely appropriate in this setting).
Choosing between them:
- Percutaneous drainage is widely available, can be done today under local anaesthetic and sedation, and decompresses reliably even with a tight hilar or distal stricture. Its drawbacks are external drainage with electrolyte and fluid losses, blockage or dislodgement, bleeding and bile leak, and the need for internalisation later. In an unstable patient on a district general site, it wins on availability.
- EUS-guided drainage gives internal drainage with better patient comfort and avoids an external tube, but requires an experienced endosonographer and carries a risk of perforation, stent migration and cholangitis if the system is not fully drained. Its advantage is that a metal stent can be placed for a longer interval if neoadjuvant therapy follows.
- Surgical drainage is reserved for cases where both fail and the patient is already undergoing laparotomy.
Given grade III disease with hypotension, my priority is the fastest, most reliable decompression with local expertise - percutaneous drainage today - with a plan to internalise or convert before any neoadjuvant treatment. The key error to avoid is repeated endoscopic attempts that consume the drainage window.