Preoperative staging of cancer of the pancreatic head: Is there room for improvement?

Background: Despite advances in preoperative staging, cancer of the pancreatic head is frequently found to be unresectable at laparotomy. We sought to identify potential areas of improvement in preoperative staging. Methods: We performed a retrospective institutional review of patients referredfor resection of cancer of the pancreatic head over a 2-year period. The primary outcome was the rate of metastasis or unresectable disease found at laparotomy in patients who were booked for pancreaticoduodenectomy with curative intent. Results: During the study period, 133 patients were referred with suspected cancer of the pancreatic head. All underwent preoperative computed tomography scanning. Twenty-four also underwent preoperative endoscopic ultrasonography (EUS) and 23 also underwent magnetic resonance imaging (MRI). In total, 78 patients were deemed not to be candidates for surgery, leaving 55 patients with potentially resectable cancer who were scheduled for pancreaticoduodenectomy. Of these, 32 patients (58%) underwent successfulresection with curative intent, and 23 patients (42%) were found to have metastatic or locally advanced disease not identified by preoperative staging. Reasons for nonresectability were metastases (9 patients, 16%), vascularinvolvement (12 patients, 22%) and mesentery involvement (2 patients, 4%).One patient had a diagnostic laparoscopy immediately before planned open exploration and was found to have peritoneal seeding precluding curative resection. Of the patients who underwent EUS, 14 were not surgical candidates because of locally advanced tumours. Ten patients were offered surgery withcurative intent, and 5 patients (50%) were found have unresectable tumours(4 metastatic, 1 locally advanced). Of the patients who underwent MRI, 11 were offered surgery, and 5 (45%) had unresectable tumours (2 metastatic, 3 locally advanced disease). Conclusion: In our institution, preoperative staging for cancer of the pancreatic head misses a substantial number of metastatic and unresectable disease. There is clearly room for improvement, and newer technologies should be evaluated to enhance the detection of metastatic and locally advanced disease to prevent unnecessary laparotomy.