When is it OK to do a pancreaticoduodenectomy without a histologic diagnosis of cancer?

By James E. Barone MD

NEW YORK (Reuters Health) - A new consensus statement from the International Study Group of Pancreatic Surgery (ISGPS) says surgeons may proceed with a pancreaticoduodenectomy without biopsy proof when a solid mass suspicious for malignancy is present.

In a telephone interview, lead author Dr. Horatio J. Asbun said, "Pancreatic carcinoma is an area where there isn't solid level I evidence, and it's also hard to do randomized controlled trials on patients. We were happy to reach an agreement at the consensus meeting with surgeons from so many different countries."

Dr. Giuseppe Garcea, a pancreatic surgeon at the University Hospitals of Leicester, England, who has written about pancreaticoduodenectomy and whether histologic proof of cancer is necessary, told Reuters Health, "I think the paper is a reasonable attempt to come up with some evidence-based guidelines on what is a very difficult clinical scenario."

The ISGPS panel also said that if neo-adjuvant chemotherapy is to be given prior to exploration, a positive biopsy for malignancy is mandatory. When autoimmune pancreatitis (AIP) is suspected, a biopsy and serum levels of IgG4 should be done. If no malignancy is found, a short course of corticosteroids might obviate the need for surgery.

"If AIP responds to steroids, you will see decrease in size of the pancreas within four to six weeks," said Dr. Asbun.

For many years, the need for a biopsy of a clinically resectable pancreatic head mass suspicious for cancer has been debated. Now that pancreaticoduodenectomy is safer and imaging yields better diagnostic accuracy, the need for confirmation of malignancy with a biopsy may be less.

The centralization of these operations to high-volume centers has resulted in a dramatic decrease in mortality, to below 1%, but there is still considerable morbidity associated with the procedure, said Dr. Garcea. He added, "It is always comforting to know when embarking on these major six-hour surgeries that a tissue diagnosis of malignancy has been obtained; but frequently the pancreatic surgeon is not afforded this luxury and has to commit himself to an operation on imaging and clinical suspicion alone."

The consensus statement, released online January 6 in the journal Surgery, was developed from a review of relevant English-language studies published between January 2008 and February 2013.

The first draft of the statement was written by a subgroup of the ISGPS and discussed at a meeting of the entire group in April 2013. The final version of the document was reviewed and agreed upon by all 25 co-authors.

When pancreaticoduodenectomy is performed without biopsy of a suspicious mass, benign pathology is found in 5% to 13% of cases, most of which are AIP. Similarly, 5% to 9% of operations for chronic pancreatitis will yield carcinoma.

The utility of various diagnostic tests was discussed. Although imaging studies have improved over the years, reviewing benign cases shows high rates of false positive diagnoses with CT scans, MRIs, endoscopic ultrasonography, and endoscopic cholangiopancreatography, including those done with brush biopsies.

Percutaneous fine needle aspiration can be highly specific but not very sensitive, and seeding of the needle tract is a potential complication. Fine needle aspiration under endoscopic ultrasound guidance eliminates the seeding problem but its negative predictive value is highly variable.

"I'm pleasantly surprised that the authors do not recommend the routine pursuit of a preoperative diagnosis--for example via endoscopic ultrasound--prior to resection," said Dr. Garcea, who was not a participant in the ISGPS panel. Some national guidelines, such as those from the British Society of Gastroenterology, suggest that this is desirable.

Dr. Asbun said, "It was interesting to find that despite all the advances in imaging and endoscopic ultrasound we still are in a position to have to go ahead with the pancreatic resection without a tissue diagnosis."

Positron emission tomography-CT scans and serum tumor markers such as CA 19-9 and CEA can be useful if positive, but when negative are of no value, and CA 19-9 elevations have been found in patients with benign obstructive jaundice.

Intraoperative frozen section also has false-negative rates ranging from 7% to 26%. A negative frozen section which later showed carcinoma would mandate a second operation to complete the pancreaticoduodenectomy.

But Dr. Asbun said he still does frozen sections in the operating room because they are useful to determine if the bile duct and neck of the pancreas margins are negative.

The consensus statement also strongly recommended going ahead with a pancreaticoduodenectomy in cases of chronic pancreatitis if histologic proof of malignancy was lacking.

Dr. Garcea points out that as the paper concludes, many surgeons - and he says he is among them - "would prefer to take the risk of undertaking a pancreaticoduodenectomy in a patient with benign disease rather than miss a resectable cancer."

SOURCE: http://bit.ly/1d1DnkX

Surgery 2014.

(c) Copyright Thomson Reuters 2014. Click For Restrictions - http://about.reuters.com/fulllegal.asp