Major reports of outcomes after surgery to remove the pancreas may not reflect reality at average hospitals across the US, investigators say.

Recent years have brought an increase in the number of people undergoing surgical removal of the pancreas, or “pancreatic resections.”

The increase is due to better detection of cancerous and precancerous lesions as well as “surgeons’ willingness to operate on older and higher risk patients,” Dr. Brett C. Sheppard and co-investigators explain in a report in the journal Archives of Surgery.

To assess risk factors for complications related to these operations, the investigators - all from the Oregon Health and Science University in Portland - analyzed data on 103,222 patients who had major pancreatic surgery between 1988 and 2003; roughly half were 65 or older, and half were male.

The most common type of surgery was so-called “pancreatoduodenectomy” (66.5 percent) - a procedure in which the duodenum is removed along with all or part of the pancreas.

The next most common type of surgery was “distal pancreatectomy” (27.4 percent), in which only the body and tail of the pancreas are removed, leaving the head of the organ attached.

Total pancreatectomy, in which the entire organ is removed, often along with the spleen, gallbladder, common bile duct, and portions of the small intestine and stomach, made up 6.1 percent of the procedures and was the third most common type performed.

Overall, the researchers found that 6.5 percent of patients died while still in the hospital, 35.6 percent suffered complications around the time of surgery, and 15.6 percent died as a result of these complications.

In-hospital mortality was significantly higher for patients age 65 and older, the researchers note.

They also found that death rates in “high-volume” hospitals were significantly lower, compared to hospitals performing fewer of these procedures.

But, the investigators point out, roughly half of all distal pancreatectomies and pancreatoduodenectomies were done in small-volume hospitals that did fewer than 6 cases each year.

The favorable outcomes being reported by big US centers with large experience might therefore not be accurately reflecting outcomes from other hospitals across the country and may just be a benchmark, the researchers suggest.

On the other hand, “volume-based referrals for pancreatic surgery undermine local expertise,” they warn, “and may leave patients far from supportive networks.” Instead, they recommend that hospital outcomes be the basis for referral.

They add that low-risk patients might safely be cared for at hospitals that met benchmark standards, regardless of volume, whereas high-risk patients would best be treated at high-volume centers.

While further study of outcomes- versus volume-based referrals for pancreatic surgery are needed, at present, “candidates for major pancreatic resection are best served by referral to high-volume centers with the resources and expertise…to optimize outcome,” wrote Dr. Michael B. Farnell from the Mayo Clinic in Rochester, Minnesota, in an invited critique.

SOURCE: Archives of Surgery, August 2009.