Poor recruitment

Patients suffering from pancreatic cancer are associated with a poor prognosis and survival of less than one year is expected in inoperable tumours. Management of these patients would be towards palliation of symptoms. Severe pain occurs in 50 to 70% of the patients and this "intractable" pain is often difficult to treat. Different pharmacological agents have been used in the past to control this pain and these include non-steroidal anti-inflammatory drugs and narcotic agents. However, patients' responses are often variable and difficult to predict. Furthermore, these agents are associated with their own adverse effects and may further impair quality of life.

Celiac plexus neurolysis (CPN) was first described in 1919, since then, different approaches of performing the procedure have been described. The standard technique involves a percutaneous approach but CPN can also be performed by an intra-operative approach with open or laparoscopic means. Results from meta-analysis have shown that CPN was associated with superior pain relief as compared to analgesic therapy alone and reduces the need for opioids analgesics in patients with inoperable pancreatic cancer. Furthermore, CPN causes fewer adverse effects than opioid analgesics and it is the preferred method of improving pain relief in these patients.

Recently, endoscopic ultrasonography (EUS) - guided CPN has become popular. The approach is safe and effective and was shown to be associated with long lasting pain relieve in patients suffering from chronic pancreatitis or pancreatic cancer. Serious complications are uncommon and are less than 2% in these series. Transient diarrhoea and hypotension are common after CPN and is seen up to 30% to 40% of the patients, regardless of whether the procedure is being done by the EUS or percutaneous approach. The EUS approach offers several theoretical advantages over the percutaneous option. Most notably is the visualization of the celiac ganglia situated anterior to the aorta, allowing direct injection of the ganglia with alcohol resulting in celiac ganglion neurolysis (CGN). This increases the accuracy of CPN and may result in improved pain control. Furthermore, it could reduce complications associated with the percutaneous approach that includes lower extremity paresthesia and paralysis.

Hence, the aim of the study is to compare the efficacy and safety of endoscopic ultrasound (EUS)-guided celiac ganglion neurolysis (CGN) versus percutaneous celiac plexus neurolysis (CPN) in reducing cancer pain in patients suffering from inoperable cancer. With direct visualization and injection of the celiac ganglion, the investigators hypothesis that EUS-guided CGN is more advantageous on improving pain relief and decreasing the need for opioid analgesics in patients with inoperable cancer as compared to percutaneous CPN.

Inclusion Criteria:

1. All patients ≥ 18 years old with cytology or histology confirmed pancreatic cancer, or radiologically suggestive of pancreatic cancer ( for patients whom biopsy is impossible)
2. Pain associated with inoperable cancer ( including abdominal pain or back pain, managed according to WHO analgesic ladder, with VAS score ≥4 despite simple analgesics (first 2 steps of WHO analgesic ladder)
3. Inoperability of cancer as demonstrated by EUS, computed tomography (CT) or Positive emission tomography
4. Informed consent available

Exclusion Criteria:

1. Unable to safely undergo EUS for any reason
2. Patient is unable to lie prone for procedure
3. Coagulopathy (prolongation of prothrombin time > 18 sec) or thrombocytopenia <80,000 platelets/ml)
4. Previous CPN or other neurolytic block that could affect pancreatic cancer-related pain or had implanted epidural or intrathecal analgesic therapy
5. Another cause for abdominal pain such as pseudocyst, ulcer or other intra-abdominal disorder
6. Allergy to local anaesthesia, contrast, or alcohol
7. Potential patient noncompliance (refusing to follow schedule of events)
8. Active alcohol or other drug use or significant psychiatric illness
9. Expected survival less than 6 weeks