CA 19-9 levels predict results of staging laparoscopy in pancreatic cancer

Abstract

Laparoscopy has emerged as an important staging procedure for determining resectability of pancreatic cancer. However, a small fraction of patients with pancreatic cancer benefit from its use and therefore the routine application of laparoscopy remains controversial. We hypothesized that serum CA 19-9 levels may identify patients who will or will not benefit by laparoscopy. We retrospectively reviewed our database of 63 patients with pancreatic adenocarcinoma who underwent staging laparoscopy and correlated findings with CA 19-9 levels. Overall, laparoscopy identified metastatic disease in 12 patients (19%). None of those required any further operation. The resectability rate (patients who underwent resection after laparoscopy) was 73.5%. There was one false-negative laparoscopy (1.6%). Patients with higher CA 19-9 levels had significant higher odds of having metastasis identified by laparoscopy (odds ratio, 1.83; 95% confidence interval, 1.03-3.24; P = .04). There was no patient with CA 19-9 levels below 100 U/ml in whom metastatic disease was identified during laparoscopy: 18 patients (28.6%) with CA 19-9 levels below this cutoff point had negative laparoscopy and could have avoided the procedure had this cutoff been used for screening. This would have increased the laparoscopy yield to 26.7%. In patients with adenocarcinoma of the pancreas, low CA 19-9 levels predict low probability of metastatic disease; in those patients, laparoscopy can be spared. On the contrary, patients with elevated CA 19-9 have an increased probability of metastatic disease, and these patients may benefit from diagnostic laparoscopy.