Clinical assessment compared with cyst fluid analysis in the differential diagnosis of cystic lesions in the pancreas.

BACKGROUND: In adults 80% to 90% of cystic lesions in the pancreas are pseudocysts and the remainder are mostly neoplastic cysts. To choose optimal treatment for an individual patient, exact nonoperative diagnosis would be preferable. This study was done to assess the value of cyst fluid analysis, compared with clinical and radiologic findings, in the differential diagnosis of pancreatic cystic lesions. METHODS: Twenty-two patients with a cystic lesion in the pancreas underwent operation, cyst wall biopsy, and aspiration of cyst fluid. Carcinoembryonic antigen (CEA), CA 19-9, pancreatitis-associated protein (PAP), and total protein concentration, amylase activity, and cytologic findings were studied. Final diagnosis was pseudocyst in 14 patients, serous cystadenoma in two, mucinous cystadenoma in two, and mucinous cystadenocarcinoma in four patients. RESULTS: Clinical and radiologic judgment correctly differentiated pseudocysts and neoplastic cysts. Cyst fluid aspiration did not succeed in two patients with mucinous cystadenocarcinomas because of the high fluid viscosity. Cyst fluid amylase activity was high (greater than 16,000 IU/ml) in all but one pseudocyst and low (less than 83 IU/ml) in all but one neoplastic cyst. CEA level was lower in pseudocysts than in neoplastic cysts, but with an overlapping value between the groups. Mean CA 19-9 concentration was higher in pseudocysts than in neoplastic cysts, but with wide overlap between the groups. Pancreatitis-associated protein and total protein concentration and cystic fluid cytologic findings did not differ between various types of cysts. CONCLUSIONS: Clinical judgment including careful history and radiologic studies seems to be the most reliable method of differentiating neoplastic pancreatic cysts from pseudocysts. Amylase and CEA levels give suggestive information, but cyst fluid analysis may be misleading in an individual patient.

Intraoperative fine needle aspiration cytologic examination of pancreatic lesions.

Despite the possibilities in modern imaging technology and percutaneous biopsy, a surgeon may still find an undiagnosed mass in the pancreas at laparotomy. In this situation, intraoperative fine needle aspiration cytologic (IFNAC) examination has been reported to be helpful. We reviewed our experiences with IFNAC in 98 patients. Fifty patients had a malignant pancreatic tumor as verified on histologic examination. The results of IFNAC correctly suggested a malignant tumor in 35 patients, for an initial sensitivity of 70 per cent. Re-examination of the slides resulted in 81 per cent sensitivity, which was not a significant improvement. The sensitivity rate (an average of 83 per cent in the literature) does not, however, express enough the unreliability of the method in individual patients. We conclude that, although IFNAC correctly differentiates between carcinoma and benign pancreatic diseases in most instances, the justification for pancreas resection cannot always be based on cytologic findings, but rather on clinical and laparotomy findings.

Computed tomography is important in assessing resectability of pancreatic carcinoma, even after ultrasonographic demonstration of the tumour.

In Finland pancreatic tumour is often detected initially on ultrasonography (US), which is widely used as a primary examination in patients who have epigastric pain or are jaundiced. This retrospective study was performed to investigate whether computed tomography (CT) provided any essential substantial staging information to US in patients, whose primary tumour was detected in US. Both US and CT had been performed in 102 pancreatic carcinoma patients. US, being performed always prior to CT, detected the pancreatic tumour in 73 patients (sensitivity 72%). CT detected pancreatic tumour in 65 of these 73 patients (89%). US demonstrated the tumour to be unresectable in 22 patients (30%). CT demonstrated the tumour as unresectable in the same 22 patients, but also in additional 20 patients (overall 42 patients = 58%, P less than 0.001 as compared to US). Three of these 20 patients underwent biliary bypass, but in 17 patients laparotomy was considered unnecessary (including the six patients who underwent merely exploratory laparotomy). It was estimated, that the 51 CT examinations, performed in patients in whom the pancreatic tumour was detected resectable on US, with overall costs of about FIM 137,700, might obviate laparotomy in 17 patients, with operative costs of about FIM 255,000. Thus, CT appears to provide essential substantial information to US about the resectability of pancreatic carcinoma, and CT can be considered a very important preoperative examination from both medical and economical point of view in the patients whose pancreatic tumour has been demonstrated by US.

Biliary stent or surgical bypass in unresectable pancreatic cancer with obstructive jaundice.

To investigate the effects of surgical and non-surgical palliation of jaundice in unresectable pancreatic carcinoma this retrospective study was performed. Between 1980 and 1983 90 patients were treated of whom 54 (69%) were jaundiced. Of these 36 were treated with biliary bypass (67%), four underwent resection (7%), five were treated by percutaneous drainage (9%) and nine (17%) were in such poor general condition that no treatment for jaundice was possible. Ninety-eight patients were treated between 1984 and 1987 when the initial approach to palliation of jaundice was endoscopic stenting. Transhepatic drainage was used only if stenting failed, and operation only if both non-surgical methods failed. Seventy-two of the 98 patients (73%) were jaundiced, of whom 18 (25%) received a stent placed endoscopically, 11 (15%) underwent transhepatic drainage, 27 (38%) underwent biliary bypass, and 14 (19%) underwent pancreatic resection. Significantly fewer patients in the second group could not be treated because of their poor general condition (n = 2, 3%, p less than 0.02). There were no differences among the methods in overall and 30 day complication rates, or the length of hospital stay, but the late complication rate was 1/63 (2%) for biliary bypass compared with 7/29 (24%) for biliary stenting (p less than 0.001). The difference was because of the high incidence of blockage of the stents causing recurrent jaundice, but the stents could easily be replaced. There was no difference in mortality between the two periods. We conclude that stenting is an acceptable alternative to biliary decompression in the treatment of obstructive jaundice in unresectable pancreatic cancer.