Conducting an imaging study of the gland is often the first step in the pancreatic cancer diagnosis process. This can be achieved through an ultrasound or CT scan of the abdomen that produces a detailed picture of the organ.
The helical "spiral" CT scan is considered to be state-of-the-art in this regard, as it provides information about the nature and location of tumors, as well as their resectability (degree to which they are capable of being surgically removed). Additional diagnostic procedures may be ordered if CT scan results are inconclusive. Among such procedures are:
Endoscopic ultrasound (EUS)
An endoscope (a long, thin, illuminated optic instrument) is passed through the patient's mouth into the stomach. By using this instrument in combination with ultrasound images, it can be determined if the cancer has invaded adjacent veins or lymph glands. This procedure can also uncover smaller tumors than those currently detectable by CT scanning techniques. (more)
Endoscopic retrograde cholangiopancreatography (ERCP)
As in EUS (see above), an endoscope is passed through the patient's mouth. Then, through the endoscope, a narrower length of tubing called a cannula is inserted into the pancreatic duct. In this way, pancreatic cells can be removed for analysis, and opaque dye can be injected to determine, by way of x-rays, if the duct has narrowed or become blocked. ERCP also can be used therapeutically for the placement of a biliary stent to relieve jaundice. (more)
Laparoscopy
A diagnostic procedure in which the abdominal cavity is examined by inserting a laparoscope (a miniature camera) through small incisions in the abdominal wall. For some surgeons, laparoscopy is a routine component of pre-operative workups. Others, however, believe laparoscopy is unnecessary in those cases where tumors have been deemed operable. (more)
Biopsy
This is a test to confirm the diagnosis of pancreatic cancer. One approach is fine needle aspiration (FNA) in which a long, thin needle is passed through the abdomen, into the pancreas, to remove cells. Another method involves surgically removing a piece of pancreatic tissue. Many surgeons don't recommend these procedures for those patients with operable tumors. However, in cases where a non-operative approach is appropriate, tissue diagnosis through biopsy is generally advised. (more)
The Helical ("Spiral") CT Scan [back to top]
The helical ("spiral") CT scan is now the best overall study for diagnosis and preoperative staging. It provides information about the nature and site of the lesion (e.g., pancreatic vs. other periampullary tumors, bile duct tumors), its resectability (e.g., liver metastases, vascular invasion), and vascular anatomy. It is important to stress that the helical CT is the current state of the art. When the helical CT is compared with the conventional technique, and especially when imaging of both the pancreas and liver is done separately according to published protocols, the differences in the sensitivity and specificity of the two approaches is dramatic. The primary lesion itself as well as liver metastases may be evident with the helical CT, and these may not be seen with the standard scan. Extraordinary detail is provided about vascular anatomy and vascular invasion by tumor. In our experience at UCLA, if the radiologist indicates that the tumor is unequivocally "unresectable" because of major vascular invasion, we have been able to resect in only about 2% of cases. These were patients in whom the superior mesenteric and portal veins were narrowed by the tumor, which was interpreted wrongly as invasion. At surgery, the vessels were merely displaced by the lesion. If the tumor is judged to be "resectable" on the basis of helical CT, we have been able to resect in about 85% of cases.
Endoscopic Ultrasound [back to top]
Endoscopic ultrasound (EUS) is not likely to be used as a first line diagnostic study in most patients with pancreatic cancer. However, once the diagnosis is suspected, it can provide valuable information about whether the pancreatic cancer has invaded adjacent vessels like the superior mesenteric vein. This is important in decisions about resection, and whether the patient is a candidate for operation. EUS also may detect smaller tumors than those currently detectable by CT scanning techniques, and it permits accurate placement of biopsy needles (e.g., fine needle aspiration cytology) if there is need to establish the diagnosis. The eventual role of EUS in the workup of these patients is evolving. It requires considerable expertise to perform, and this is still not widely available. It is still unclear whether EUS will provide more reliable information about resectability than helical CT scans.
Endoscopic Retrograde Cholangiopancreatography [back to top]
Diagnostic endoscopic retrograde cholangiopancreatography (ERCP), has been an overused study in the workup of patients with pancreatic tumors. In a patient who presents with obstructive jaundice, and some weight loss, and in whom the abdominal CT scan shows a mass in the head of the pancreas, ERCP provides little additional information that affects decisions about management.
We are not critical of cases in which an ERCP may be done before the CT scan, perhaps because of the manner in which the disease manifested itself, or because of established referral patterns. And, in patients where the CT fails to display a mass or raises other questions about the diagnosis, ERCP is certainly indicated. In patients who will not be operated upon, therapeutic ERCP with the placement of a biliary stent to relieve the jaundice is also a useful technique.
Should jaundice be relieved before surgery?
Obstructive jaundice can cause defects in hepatic, renal, and immune function. It was hoped that preoperative relief of the jaundice would correct these defects, and decrease postoperative morbidity and mortality rates. However, there are still no randomized controlled studies that have examined the value of preoperative biliary drainage using endoscopically placed stents.
Although our experience is anecdotal, we generally have advised against the use of endoscopically placed stents. In the majority of cases, the bile is sterile before the stent is placed, which introduces bacteria. We have seen cholangitis occur, which has required antibiotic treatment, multiple stent replacements, and delay of the surgery. When surgery is performed, the tissues around the bile duct are often inflamed, which makes the dissection more tedious. The incidence of postoperative septic complications also may be higher.
However, in cases where the bile duct is obstructed and the operation must be postponed for several weeks or more, preoperative decompression may be indicated. When a biliary stent is placed, it should be at least 10 French in diameter, and should be of the polyethylene variety. Metal stents should never be placed in patients who are candidates for resection! These stents incite a severe inflammatory reaction and are eventually incorporated into the bile duct wall. Thus, if this has occurred and they cannot be removed from the duct at operation, the duct must be transected above the stent. If the stent is placed too high, this can complicate the procedure.
Laparoscopy [back to top]
The place of laparoscopy (limited examination of the abdomen using several small incisions and a TV camera to see the organs) in the management of patients with pancreatic tumors continues to evolve. Some surgeons have favored its use in the majority of cases, as a routine part of the workup before major surgery (laparotomy).
In some instances, laparoscopy was done as a separate procedure in the outpatient setting, and the findings guided the subsequent workup. More often, it was performed immediately before laparotomy under the same anesthetic; if evidence of unresectability was found, the operation was concluded. There is some older evidence that it was a cost-effective approach and that it spared many patients the discomfort of an unnecessary laparotomy.
Warshaw found that 40% of patients without apparent extrapancreatic involvement on conventional CT, and thus judged to have resectable lesions, had hepatic or peritoneal metastases at laparoscopy. However, the data that supported that philosophy was published between 1986 and 1992, and helical CT scans were not used in those cases.
Our experience suggests that we should not use laparoscopy in patients who appear to have resectable lesions on helical CT, and who are good candidates for resection. It is not that the helical CT scans show the very small hepatic or peritoneal metastases (2-3 mm) that may be present. But the newer technique shows vascular involvement and liver metastases more reliably than the conventional scans did, and we have found that most patients who have these small metastases, already were considered unresectable because of vascular involvement.
As stated previously, resection was possible in 85% of our patients who were judged to have "resectable" cancers of the head of the pancreas after CT scan. Of the remaining 15% of the total who did not have resectable cancers, small liver or peritoneal metastases, which might have been seen laparoscopically, were the reason in only half. Thus, if all of our patients had undergone laparoscopy as a routine, at best, only 7.5% might have been spared subsequent laparotomy. This seems to us to be too low a yield to justify the additional operative time and expense of the procedure. The remaining patients who were falsely designated as resectable on the basis of helical CT evaluation, had locally extensive tumor which was assessed as unresectable only after mobilization of the pancreas from the superior mesenteric and portal veins at laparotomy. This would not have been safe to do laparoscopically.
We do use laparoscopy, selectively, in certain circumstances. Examples include some patients with pancreatic cancer and CT evidence of liver or other metastases, body or tail cancers, and some patients with ascites who probably have peritoneal metastases.
Biopsy [back to top]
Preoperative cytological confirmation of the diagnosis using fine needle aspiration (FNA) is often attempted in patients with suspected pancreatic cancer. However, we advise against it in most individuals who are operative candidates. It is uncomfortable for the patient, and it adds additional time and expense to the workup. It is of little value to an experienced pancreatic surgeon who will usually perform a pancreatic resection if the clinical picture and operative findings suggest the diagnosis of cancer.
In contrast, FNA in order to confirm the diagnosis of pancreatic cancer is valuable when that knowledge will affect treatment, e.g., a non-operative approach will be recommended. Thus, we often will obtain it in cases where unresectable disease appears to be present (e.g., to confirm CT evidence of liver metastases), in many cases of cancer in the body or tail of the pancreas where the likelihood of resection for cure is very low, or when the patient is an unacceptable risk for operation and a tissue diagnosis is required. The technique is highly specific (99%) in the hands of an experienced cytopathologist. Although severe pancreatitis and seeding of the needle tract with cancer have been reported after FNA of pancreatic or other intra-abdominal tumors, this is quite rare. The overall complication rate is less than 1%, so our reluctance to obtain an FNA routinely in these patients is not because of the concern about complications.
Should the tumor be biopsied before the operation?
Experienced pancreatic surgeons usually do not require histological proof of malignancy before they are willing to perform a pancreatic resection. There are several reasons for this:
1. If the clinical picture and findings at operation are consistent with the diagnosis of cancer, then the chances that cancer is present are at least 90%.
2. In those cases where cancer is not present, the most likely diagnosis is chronic pancreatitis. Pancreaticoduodenectomy is an excellent procedure to treat this, as well.
3. The operation can be done safely, with operative mortality rates less than 5%, often between 1% - 2%. Nevertheless, it is important that the patient understands and agrees with this approach, and a full discussion should occur preoperatively, which should be documented in the records.
|
