Purpose: This study was designed to compare our laparoscopic ultrasonography (LUS) experience in the resectability evaluation of pancreatic or periampullary cancers (PAC) in two different periods: before and after the introduction of multidetector CT (MDCT).
Methods: We prospectively enrolled 104 CT-resectable patients with PAC. During Step 1 (1995-1999), we performed LUS on all patients, whereas during Step 2 (2002-2007), LUS was performed selectively according to Pisters' criteria.
Results: LUS was satisfactorily performed in all cases. At Step 1 accuracy of LUS in predicting pancreatic resectability was high (96%) but it was markedly lower in a subgroup of patients with close contact between tumor and portal vein (sensibility of 57%). At Step 2, selective LUS was performed on 9 of 64 patients (14%). LUS confirmed the MDCT finding of unresectability in 8 of 9 cases, and allowed curative resection in 1 case. Only 1 of 55 of the patients who did not undergo LUS would have benefited from the procedure. The yield of LUS decreased from 45% before to 1.8% after MDCT.
Conclusions: In resectable-MDCT patients, routine LUS is unjustified. However, in doubtful MDCT cases, LUS has yet a good yield. In the event of close vascular contact, neither MDCT nor LUS seem to be conclusive, and laparotomy is still the only solution.