Fine needle aspiration cytology of the pancreas. A retrospective study of 73 cases.

Retrospective review of pancreatic fine needle aspiration (FNA) biopsy specimens collected with computed tomographic guidance from 73 patients between 1980 and 1985 at the Medical Center of Delaware was performed to determine the accuracy of the procedure in our hands and to identify possible problem areas for cytologic diagnosis. When compared with clinical data or tissue diagnosis, FNA had a sensitivity for the detection of pancreatic carcinoma of 67.7%. The predictive value of a negative result was only 23.1%. When compared to the cytologic diagnosis made at the time of review, FNA had a sensitivity of 100%, but a single false-positive case was identified. In addition to the majority of probable pancreatic ductal carcinomas, a hepatoma and a lymphoma were detected. Cases of primary pancreatic carcinoma were classified by cytologic features, but all groups had dismal three-to-six-month median survivals, regardless of the degree of tumor differentiation. Survival times were similarly low for patients with negative pancreatic FNAs. The low patient survival times, regardless of FNA diagnosis, support the value of avoiding laparotomy in these patients and confirm the high false-negative rate of the procedure.