David – A Standard Whipple Procedure Results in a Debilitating Complication and Recurrent Cancer
Most pancreatic cancers occur in the head of the pancreas. The classic operation to remove a pancreatic head tumor is the pancreatico-duodenectomy, first described in 1946 by Dr. Allen O. Whipple, the Chief of Surgery at Columbia Presbyterian Hospital in New York City, and therefore referred to as the Whipple procedure or just the Whipple. As originally described, the operation includes removing the lower third of the stomach, the entire duodenum, the head of the pancreas, the gallbladder, and the lower part of the duct that carries bile from the liver to the duodenum (Figure 1). New connections, or anastomoses, of the stomach, bile duct, and the body of the pancreas are made to the upper intestine. The trickiest of these connections is the anastomosis of the pancreas to the intestine, which can leak. In addition, because a new opening is made between the remaining portion of the stomach and the intestine downstream from the inflow of bile and pancreatic juice, there is no barrier to bile spilling back into the upper stomach or even backing up into the esophagus.
Figure 1. The Whipple Operation. Two forms of the Whipple are shown. A. The “standard” operation includes removal of the lower part of the stomach and formation of a gastrojejunostomy. B. The “pylorus-sparing” version of the operation preserves the entire stomach, the pylorus valve, and the first centimeter of the duodenum and re-establishes gastro-intestinal flow with a duodenojejunostomy. The head of the pancreas is removed, and the body of the pancreas is connected to the intestine with a pancreaticojejunostomy. The gallbladder is removed, and the bile duct is connected to the intestine with a choledochojejunostomy. (From Schwartz’s Principles of Surgery, 11th Ed., McGraw-Hill, 2019, with permission)
A Standard Whipple Results in Problems
David was a musician in his 60s whose wife noticed one morning that his eyes were yellow. He felt well otherwise but immediately called his doctor to ask what he should do. Painless jaundice (yellow skin and eyes without any abdominal symptoms) is one of the most common presentations of pancreatic cancer. He was found to have a mass in the head of the pancreas on the CT scan, which appeared to compress or pinch off and obstruct the lower common bile duct. This blocked the flow of bile into the intestine and was causing the jaundice. A needle biopsy of the mass was positive for cancer, but no tumor spread was detected. He was referred to me with the hope that his cancer was curable.
I performed a “standard” Whipple procedure. The cancer was removed entirely with “clean margins,” meaning there was no visual evidence of cancer at the edges where the tissues were removed. The pancreas’ tissue is drained by a fluid surrounding the cells called lymph, and the lymph flows through a series of lymph glands, or lymph nodes, surrounding the gland (Figure 2). Subsequent microscopic examination revealed that three lymph nodes adjacent to the pancreas contained cancer cells.
Figure 2. The lymphatic drainage of the pancreas. Lymph glands or nodes act as reservoirs of the fluid which bathes all of the tissues of the pancreas. The flow of lymph is directed up into the chest, where it eventually drains into the venous system. (From Schwartz’s Principles of Surgery, 11th Ed. McGraw-Hill, 2019, with permission)
David recovered from the surgery without complications and six weeks later began a course of radiation treatment to the region of the surgery combined with chemotherapy. He tolerated his treatment very well at first but then developed severe heartburn due to gastro-esophageal reflux. Endoscopy revealed a deep ulcer in the lower esophagus. Laboratory studies showed that his inflamed esophagus was due to regurgitation of bile, or “bile reflux,” back into his stomach and up into his esophagus. Despite treatment with multiple medications, David continued to have terrible discomfort and was unable to eat. A tube was placed through his abdominal wall into his stomach (a tube gastrostomy), and he was fed a liquid diet through the gastrostomy tube. Eighteen months after his surgery, he was found to have metastatic cancer in his liver, and he died four months later.
The Cause of the Problem and a Change in Technique
Bile is just as caustic to the esophagus as acid is, and bile reflux esophagitis is difficult to treat with medication. It is a rare complication of the standard Whipple operation, and as in David’s case, it can be debilitating. Despite the fact that the tumor was removed in its entirety, David’s cancer recurred, and he developed metastatic disease from which he died. His quality of life after the Whipple procedure was poor due to his surgery complications, and his condition deteriorated progressively until his death. Unfortunately, David’s cancer recurrence less than two years after his operation was typical of many patients undergoing the Whipple procedure.
Now, 70-80% of all the pancreatico-duodenectomies are performed using a technique in which the stomach is left intact. The pylorus, which is the valve that controls the emptying of the stomach contents into the first portion of the small intestine, is preserved. This “pylorus-sparing” Whipple doesn’t seem to affect the rate of recurrence of the tumor, which is 80-90% regardless of which type of Whipple is performed, but preserving the pylorus valve helps to prevent bile reflux into the stomach. Most surgeons believe that the quality of life after the Whipple operation is better after the pylorus-preserving version of the procedure. This is important, as the majority of patients will have a limited amount of time to enjoy life before the cancer recurs, and the pylorus-sparing Whipple appears to improve the likelihood of a quicker recovery and better nutrition after the surgery.
The Whipple procedure is one of the most complex operations in surgery. It takes 6 to 8 hours, or longer, to perform, and the risk of complications is high. In the mid-1970s, more than one in ten patients did not survive the surgery, and, not surprisingly, relatively few patients were referred for the surgery. Some physicians advocated not performing Whipple operations at all due to the high risk of the surgery and the low likelihood of cure of the cancer. As the safety of the operation improved, due to improvements in anesthesia and post-operative care, the mortality rate (defined as the risk of dying within 30 days of the surgery) gradually fell to its current level of 1-2% in major centers. Today, 8,000 to 9,000 Whipple procedures are performed each year in the United States, but the rate of post-operative complications (the morbidity rate) has remained about 50%. Most complications are treatable or resolve spontaneously, but some cause continuing or worsening problems and can result in complete disability.
The Whipple procedure’s safety or success rate is directly related to the number of procedures performed by a given surgeon or at any given hospital. Major centers, including cancer centers at major academic institutions, have the lowest mortality rate, but the death rate at all hospitals combined is about 8%. This means that a patient with operable pancreatic cancer is more likely to survive the procedure when it is performed at a major center. Still, most patients remain at local or regional hospitals for their surgery due to the expense, inconvenience, or other concerns about transferring their care to physicians at distant medical centers. Unlike many European countries, patients in the US are not required to undergo Whipple procedures at specialized centers, with the result that outcomes vary greatly from one hospital to another. All we can do is strongly recommend that a patient with pancreatic cancer seek care at a major academic center, not their local hospital.
Another aspect of the problem is that it takes extensive experience before a surgeon is truly skilled at performing the procedure. All general surgery residents dream of when they will do their first Whipple. It is the most complex surgery a general surgeon can perform – the tour-de-force of a young surgeon’s training – but most young surgeons finishing their training, even at otherwise excellent training programs, will have performed fewer than four of these operations. Once in surgical practice, it is unusual for a general surgeon to perform more than a handful of these operations per year. But studies show that hospitals need to have had at least 12 Whipple procedures performed per year in order for the procedure to reliably meet an acceptable standard of safety. This is related to the surgeon’s repetitive experience and the availability and expertise of the anesthesia and intensive care specialists and the other physicians and staff who learn to routinely provide expert care during the complex post-operative protocol. About a third of all the Whipple procedures in the US are performed at hospitals that do fewer than 12 per year.
Improving the Chances of a Cure
Despite advances in surgical methods and post-surgical care, only a fraction of patients with Stage I cancers, or those with a real chance of being cured by surgery, actually undergo an operation. The reasons for this “lost opportunity” for so many patients are disturbing. Patient and physician nihilism about the diagnosis (the belief that the disease is hopeless) may contribute, or it may be that many people in their 70s or 80s (when the majority of pancreatic cancers occur) would rather opt for comfort and time with their families rather than the arduous course of surgery. Some patients are poor candidates for surgery because of frailty or the presence of other serious conditions. But in a 2007 study of over 9000 patients assessed to have Stage I cancers, over 70% never underwent an attempt at surgical resection, and more than half of that number were never offered the surgery. The bottom line is that only a fraction of resectable and possibly curable pancreatic cancer patients actually undergo surgery (Figure 3).
Figure 3. Survival of patients with Stage I or Stage II pancreatic cancer who either refused surgery or underwent surgery. The average survival of the surgical patients was 36 months, while the average survival of those who refused surgery was eight months. (From Chari ST et al. Pancreas 2015; 44: 693, with permission)
The question currently being studied in clinical trials is whether chemotherapy or chemo-radiation therapy before the surgery, called neo-adjuvant therapy, can improve surgery outcome by killing cancer cells that have already been shed by the tumor into local tissues or to distant locations. Pancreatic cancer is a particularly difficult cancer to treat because it begins to shed cancer cells, or metastasize, when the cancer is still relatively early or small. Therefore, a local surgical attack on the cancer usually leaves untreated large numbers of cancer cells already nested elsewhere or circulating in the bloodstream. A combination of local removal of the primary tumor and systemic treatment, where drugs are delivered to the whole body, is the most logical approach, but the type of systemic treatment and timing of this course of treatment remains the subject of research. At the time of David’s surgery, chemotherapy and radiation were used only after surgery. This adjuvant therapy was recommended in the hope of preventing a recurrence of the tumor or the development of the metastatic disease, but the drugs had minimal benefit on long-term survival.
Neo-adjuvant therapy has been studied in clinical trials in which patients with advanced-stage disease are randomized to neo-adjuvant treatment before surgery or to surgery first, followed by adjuvant treatment. Many of these studies are in progress, and some have been inconclusive or contradictory so far. This may merely reflect the fact that patients with advanced-stage disease are being studied, as many doctors are reluctant to enlist patients with Stage I disease into a study that delays potentially curative surgery. But more recent trials have examined the value of neo-adjuvant treatment in patients with early (stage I or II) disease, and it appears to significantly improve the survival of these patients. Many centers are now recommending neo-adjuvant treatment for all patients with pancreatic cancer, even for those who appear to have Stage I tumors.
Since David’s surgery, progress has also been made in developing more effective drugs that kill cancer cells. The combined use of two or more of these drugs, which attack the cancer cells through different mechanisms, has become standard therapy. Improvements in survival after a Whipple procedure have been very modest, however. The latest research approach has been to target the recruitment of the body’s own lymphocytes, cells that govern the tumor’s immune response against the tumor. Pancreatic cancer has the ability to block the recognition of the cancer by the “killer” lymphocytes that would otherwise attack the tumor. This recognition-blocking process is called a check-point, and drugs that prevent this effect of the cancer to block the lymphocytes’ ability to “see” the cancer, called checkpoint inhibitors, are now being studied in clinical trials. These new drugs have had a major positive effect on some tumors, such as certain forms of lung cancer and melanoma. So far, the results have been less impressive in pancreatic cancer.
It has been suspected for years that one of the reasons that some patients seem to have a much better or slower course of a malignant disease is that their body’s immune system manages to somehow keep the tumor “at bay” or prevent its further spread, or actually destroy small numbers of cancer cells that are released by the tumor. My first experience with this phenomenon happened years ago when a man was referred to me who had undergone surgery for a colon cancer and had been found to have a single focus of metastatic cancer in his liver. The liver cancer did not increase in size throughout months of repeated scans. When I eventually operated on him to remove the solitary focus of tumor metastasis, I found the metastatic tumor to be completely surrounded or “walled off” by a thick shell of fibrosis or scar, which had been induced by his body’s own immune attack on the tumor. With the exception of the solitary focus of walled-off tumor, he was otherwise cancer-free. The cellular signals which control the immune attack on a tumor are the focus of much current research, and several strategies for provoking or enhancing the body’s own immune cells to assault a pancreatic cancer are being studied in animals as well as in early clinical trials in patients. We need to invest very heavily in this research.
One of the obstacles to this research is the reluctance of patients to enroll in clinical studies in which the outcome or benefit of a new drug is unknown. In addition to developing new drugs, we need to encourage patients to enroll in clinical trials where these new drugs are being tested. Patients may be reluctant to subject themselves to a new drug with only a suggestion that it might benefit them, but it’s the only way we can learn what is a better treatment. One of the benefits of enrolling in a clinical trial is that patients in these research trials actually get better care than those who decline to participate. This is because the subjects enrolled are being watched carefully by multiple professionals, not just their doctor. This increased surveillance means that problems will be discovered more quickly if they occur.