Preoperative assessment of pancreatic tumors using magnetic resonance imaging, endoscopic ultrasonography, positron emission tomography and laparoscopy.

BACKGROUND

The appropriate preoperative evaluation of a pancreatic tumor remains a matter of debate.

METHODS

We retrospectively evaluated an institutional strategy including magnetic resonance imaging (MRI), endoscopic ultrasonography (EUS), positron emission tomography (PET) and laparoscopy (LAP) for detection and staging of pancreatic tumors suspected to be malignant.

RESULTS

In a consecutive series of 59 patients screened by MRI, PET, EUS and LAP between July 1998 and November 2002, 48 patients were found to bear pancreatic adenocarcinoma and surgery was performed in 27 of them. For tumor detection, the sensitivity of EUS was superior to MRI and PET (98 vs. 87.5 and 87.5%, respectively, p = 0.13). MRI best assessed loco-regional staging, i.e. arterial involvement. For the detection of distant metastases, the sensitivity of all preoperative examinations taken separately was low. When laparotomy was performed with a curative intent according to all four examinations, occult metastasis or carcinomatosis was discovered in 7/27 patients and the overall predictive value of resectability was thus 74%. Five-year and median survival were significantly better in resected vs. non-resected patients (39% and 26 months vs. 0% and 8 months, p = 0.0006).

CONCLUSIONS

MRI can be recommended has the first examination in patients bearing pancreatic tumors, complemented by EUS if the findings of MRI are non-conclusive. For detection of distant metastasis, only the combination of all preoperative examination was proved to be more accurate than a single technique.