The Fire Down Below

A hot stone embedded in the middle of my belly. That’s what the nagging ache felt like when it asserted itself abruptly one evening—a distinctive new addition to the repertoire of usual gastric phenomena I’d experienced over decades of stomach ownership. Like anyone else’s, at one time or another the old tummy has felt queasy, nauseated, acidic, cramped, gassy, growlingly empty, pleasantly full, and Thanksgivingly overstuffed. But not until that particular evening had it felt hot-stone-achy. This worrisome ache—along with volcanic belching—waxed and waned over a few weeks.

When it finally became too intense, I visited my primary-care physician. He put me on a blessedly effective acid reducer that seemed to give relief, and he referred me to a Duke gastroenterologist. A subsequent test showing a faint sign of blood in a stool sample prompted her to order the Full Monty of tests on my middle-aged, and therefore suspect, gastrointestinal system. The resulting medical odyssey led me through three of Duke’s GI diagnostic labs, where, fortunately, the kind ministrations of some very sympathetic professionals made the experience about as pleasant as such procedures can probably be.

In the upper GI lab, I gulped chalky liquid to make my stomach opaque to X-rays and swallowed fizzy crystals to inflate it. Barely holding back the mother of all belches, I watched with utter fascination the sloshings of my honest-to-God actual stomach on the video screen—for the first time really witnessing that heretofore invisible, and sometimes rebellious, pouch. The jovial technician and I also discussed my intestines. She volunteered that she’d seen lots of intestines in her time, and not one set looked like the neat assemblage portrayed in textbooks. She’d seen intestinal kinks, loops, knots, and all manner of other unruly arrangements.
A visit to the ultrasound lab brought me to another nice technician, who showed me the shadowy outlines of my gallbladder and pancreas. We were both gratified to find them in their place—although I recall with some chagrin that my liver was pronounced slightly fatty.

A week or so later came the piéce de résistance—the colonoscopy. The day before, I had undergone the rigors of overnight fasting, of drinking only clear liquids, and of ingesting a thoroughly pipe-clearing physic. That morning found me lying woozily on a table in the colonoscopy lab watching on a video monitor a Fantastic Voyage through my glistening, beige nether passage.
As I waited for the final report on the various explorations of my internal labyrinth, I launched an exploration of my own into the impressive array of other diagnostic and treatment techniques that Duke gastroenterologists employ to understand and cure GI disorders. Such a broad capability is one reason the gastroenterology division has been ranked among the best in the country.

McGrath and endoscope: examining via EUS (endoscopic ultrasound)

Gastroenterologists can diagnose and treat the rising tide of GI disorders using ever more effective and benign tools, including ultrasound endoscopes that can explore the GI tract from the inside, “photodynamic therapy” techniques that use laser light to activate chemicals that selectively destroy dysplastic cells in the esophagus, and computer-generated X-ray images, called “virtual colonoscopies,” of the intestine.
Of these, perhaps the most revolutionary-sounding is the virtual colonoscopy, which theoretically enables radiologists to explore a colon for cancers, polyps, and other pathologies without using an invasive probe. The technique involves performing an X-ray CT scan of the abdomen and pelvis and using a computer to analyze the X-ray data to reconstruct the interior topography of the colon. A new technology is not necessarily a better technology. So, associate professor Don Rockey and his colleagues are leading a $7-million study to compare the accuracy of virtual colonoscopy with the venerable barium enema X-ray and the colonoscopy.
“We’re comparing the three techniques to find which one reveals such abnormalities as cancers, polyps, colitis, and inflammation,” says Rockey, who heads the gastroenterology division’s Liver Center. “And, we’re also asking patients for their opinions on comfort, pain, and the likelihood that they would be willing to repeat the procedure.” After all, adds Rockey, even the most effective diagnostic technique is useless unless patients are willing to undergo it.

As a radiologist, associate professor Erik Paulson M.D. ’85, who is also participating in the trial, agrees with the need for patient cooperation. He and his fellow radiologists are only too aware of the problems of persuading patients to undergo the colonoscopy screening that could save their lives. Says Paulson, “We know that if we can catch colon polyps early and remove them, we can greatly reduce or eliminate colon cancer. But despite the fact that screening works, lots of patients don’t get screened. Either their family doctor or internist doesn’t recommend it, or they don’t want to because of the discomfort.” Thus, says Paulson, virtual colonoscopy—more correctly called “CT colonography”—seems to offer a way to screen patients that is not, literally, a pain in the butt.

Despite the technique’s patient-friendliness, CT colonography might not become widespread because of its complexity, says Paulson. “While there are many well-trained gastroenterologists and radiologists, we need to make sure that they can routinely interpret the results from CT colonography accurately. There may not be that many medical centers like Duke, in which gastroenterologists and radiologists work together so effectively.”

Baillie uses ECRP optics for noninvasive surgery on bile ducts.

While the clinical trial—funded by the National Cancer Institute through the Duke Comprehensive Cancer Center—is only in the first of its four years, Rockey suspects that the results will not be a slam-dunk for any of the techniques. “Colonoscopy is probably the most accurate test compared to the X-ray methods, but the question is whether the X-ray tests are still accurate enough to make them useful. The fact is that there are simply not enough people to perform colonoscopies on patients with colon abnormalities,” he says. “Also, colonoscopies are very expensive, especially compared with barium enema X-rays, and there is a small but significant risk of complications such as perforation of the intestine. So, I suspect that the three will be complementary in many respects, and I would be very surprised if any of them becomes the exclusive method of choice.”

Another exotic-sounding visualization technique being advanced by the Duke gastroenterologists is “endoscopic ultrasound” (EUS). Most ultrasound examinations, such as the one that revealed my gallbladder in all its glory, are conducted from the outside, with the same type of ultrasound probe used to see a future-junior in a pregnant mom’s belly. However, EUS sees from the inside. Basically, the technique involves threading an endoscope, whose tip holds an ultrasound transducer, either through the throat or rectum. Physicians can, thus, obtain an up-close sonar image of the GI tract that provides far better resolution. Says assistant professor Kevin McGrath, who participates in the division’s EUS research and applications, “We use EUS mainly for determining the stage of cancers, including esophageal, gastric, pancreatic, and rectal.”

Determining a cancer’s stage of progression is a key to guiding treatment. In particular, the fact that ultrasound penetrates tissue means that physicians can also use it to evaluate abnormalities immediately adjacent to the GI tract. The EUS system allows physicians to do more than see a cancer, says McGrath; a needle aspirator allows the scientists to use real-time ultrasound guidance to biopsy abnormal lymph nodes around the stomach or pancreas or to sample any tissue that looks abnormal on a previous endoscopy or CT scan. The system even has a “Doppler” capability that allows physicians to detect flowing blood, allowing them to avoid blood vessels and minimize the possibility of bleeding from the biopsy. 

According to associate professor Paul Jowell, who directs the division’s EUS effort, “EUS has probably become the most accurate technique for staging of tumors within the wall of the GI tract.” However, he cautions, the new technique has brought with it a demand for new skills and training. “It’s a challenge to position the probe adequately, but even once you do that there are pitfalls to interpreting the images accurately. So, there is a significant learning curve to both the technique and the interpretation.”

Jowell is leading a clinical trial in which the researchers are comparing EUS and CT scans with subsequent needle biopsies for diagnosing pancreatic cancers. “One problem with CT scans is that they require a reasonably sized mass before you can see it and biopsy it,” he says. “EUS seems to be able to detect smaller tumors. Also, with CT, there is often first a diagnostic study and then
a second study for the biopsy. With EUS,
we normally do both at the same time.”

Technological advances will continue to improve EUS as a diagnostic tool, says Jowell. These improvements include smaller, more maneuverable endoscopes and a new instrument with forward-viewing optics, as opposed to current instruments that allow visualization only at an angle. He predicts that future EUS instruments will not only visualize tumors, but also be used as a method for injecting chemotherapeutic drugs or vaccines.
Gastroenterologists are already routinely using one endoscopic treatment called “photodynamic therapy” (PDT) for abnormal GI tissues and cancers. In PDT, a light-sensitive chemical is first injected into the bloodstream. For a reason physicians still do not completely understand, cancerous or dysplastic tissues tend to accumulate more of the chemical than do normal tissues. The gastroenterologists next insinuate a fiber-optic-equipped endoscope into the GI tract near a tumor or such dysplastic tissue as found in Barrett’s esophagus. Flashing a low-power laser through the optical fiber, they expose the chemical to light, activating it to generate a highly reactive form of oxygen, which triggers a toxic chain reaction in the tissue, killing the tumor or dysplastic cells.

Such therapy has proven especially useful in treating Barrett’s esophagus with high-grade dysplasia, says McGrath, and that’s important to preventing cancer. “Barrett’s esophagus is a known risk factor for esophageal cancer, which is probably increasing faster than any other type of cancer,” he says. Particularly worrisome, says associate professor Scott Brazer ’77, M.H.S. ’90, director of Duke’s PDT treatment and research program, is that many people with Barrett’s don’t know it. “Many who have heartburn symptoms alleviated with antacids may have Barrett’s. And about a third of patients with the disorder have no symptoms, perhaps because their esophageal tissues have become resistant to the acid. So, there are a frightening number of patients out there at risk for adenocarcinoma that don’t have any idea that they have it.”

Brazer and McGrath are testing a new PDT technique for more effectively treating advanced cases of near-cancerous dysplasia. The method involves inflating a clear plastic balloon in the esophagus to flatten out the hills and valleys in the overgrown tissue, allowing a more uniform exposure to the laser light. The physicians have already completed the treatment phase of the clinical trial and are observing the patients to determine whether the “centering balloon”—invented by endoscopic PDT pioneer Gene Overholt at the University of Tennessee—will prove more effective at removing, or ablating, dysplastic tissue.

Such advances are making PDT an even better choice over surgery, says McGrath. “Until now, treatment for high-grade dysplasia was often surgical removal of the entire esophagus, which is a major surgical procedure for patients. And because many of the patients were elderly, there was a significant risk of complications and even death. I think this trial will show that PDT is an effective treatment and will save patients with high-grade dysplasia from esophagectomy. However, we need to learn more about safety and other issues before we can say whether it will be useful in patients with lower-grade Barrett’s.”

Among the most demanding of the gastroenterologists’ endoscopic techniques is the one with the mouthfilling title “endoscopic retrograde cholangiopancreatography” (ERCP) —a division service directed by professor John Baillie and associate director Malcolm Branch, who is an associate professor. This technique with the huge title employs one of the tiniest endoscopic devices. ERCP involves threading an endoscope down the throat, through the stomach, and into the duodenum. Unlike other forward-looking endoscopes, the ERCP optics aim sideways, and the physicians use the imaging to seek out the “ampulla,” a tiny opening where the pancreatic and bile ducts exit into the intestine. Once they find the ampulla, the gastroenterologists thread a tiny catheter through it to inject X-ray dye, or to use small baskets to snag bile duct stones, electrocautery to open the duct, or cylindrical stents to prop the duct open.

“For such procedures as retrieving bile duct stones and relieving certain types of malignant obstructive jaundice, ERCP is superior to surgery in its reduced complications and length of hospital stay,” says Jowell, who participates in ERCP procedures. “In general, it’s a one-day outpatient procedure.” He emphasizes that considerable skill is needed to master the art of threading the catheter into the ampulla in the confusing, turbulent depths of the intestine. “It sounds so straightforward because there’s this opening and you think you can just slide the catheter in. But the ampulla is surrounded by a muscle that often makes it difficult to thread the catheter into the duct. Once past that, there are curves in the duct and the pancreatic duct is often no larger than a pencil lead.” 

In fact, a study of fellows in training at Duke by Baillie, Jowell, and their colleagues revealed that ERCP training elsewhere is often inadequate. “The study showed that, to gain minimum competence in ERCP, a physician must do from a 180 to 200 procedures. But the average training number we found in most fellowships was twenty-five to fifty. We came in for a lot of heat about the study, but our data were indisputable,” says Baillie, who was awarded the 2001 “Master Endoscopist Award” by the American Society for Gastrointestinal Endoscopy. 

Many surgeons were initially resistant to ERCP, but most medical centers have now developed the same kind of productive partnership between gastroenterologists and surgeons that Duke enjoys. Baillie says, “There are surgeons depending on me and my team for what we do, and vice versa. Most diseases involving the liver, bile ducts, and pancreas need a multidisciplinary approach. For example, we’ll have patients come in with terrible pancreatitis and cholangitis—infection of the bile from a stone in the common bile duct. If such a patient were operated on during the illness, they would run a high risk of death. So, we’ll use ERCP to remove the stone and let the pancreatitis settle down; and later the surgeons can operate to remove the diseased gallbladder and the stones it contains, to prevent a recurrence of the problem.”

What’s more, Baillie says, ERCP will become even more important as researchers learn to better detect early malignancies of the bile duct and pancreas—an especially virulent cancer—by testing the cells that can be retrieved with the technique. “There are promising techniques that will allow us to progress from what used to be 25 percent sensitivity for picking up cancer, to 70 to 80 percent. Ultimately, if we can identify people at high risk really early, we may be able to change the natural history of the disease, by diagnosing the precursor lesion and finding a way to reverse it.”

While the endoscopic techniques used by the gastroenterologists are perhaps the most visible of their efforts, the division’s faculty are also conducting pioneering basic research to understand GI disorders including pancreatitis, hepatitis C, cancer, and Crohn’s disease. And they are operating specialized services such as the Inflammatory Bowel Disease Clinic not only to treat patients’ medical disorders, but also to counsel them on how to cope with their disease.

So with all this expertise available, what of the “hot rock” lodged in my belly? The colonoscopy report pronounced my colon an “excellent preparation,” which made me rather proud. Annoyingly, however, the report’s characterization of my dynamic self as a “fifty-four-year-old man with dyspepsia” did make me sound like a cranky old goat, which I did not particularly appreciate. But gradually, as I took my gastroenterologist’s advice to lay off caffeine and aspirin, that hot rock shrank away to almost nothing. And, when I quit taking a joint-fortifying glucosamine supplement, the rock disappeared altogether.

Virtual colonoscopy: doctors Rockey, left, and Paulson are conducting trials on a new probe-free technique for screening cancer.

According to associate professor Paul Jowell, who directs the division’s EUS effort, “EUS has probably become the most accurate technique for staging of tumors within the wall of the GI tract.” However, he cautions, the new technique has brought with it a demand for new skills and training. “It’s a challenge to position the probe adequately, but even once you do that there are pitfalls to interpreting the images accurately. So, there is a significant learning curve to both the technique and the interpretation.” 

Jowell is leading a clinical trial in which the researchers are comparing EUS and CT scans with subsequent needle biopsies for diagnosing pancreatic cancers. “One problem with CT scans is that they require a reasonably sized mass before you can see it and biopsy it,” he says. “EUS seems to be able to detect smaller tumors. Also, with CT, there is often first a diagnostic study and then a second study for the biopsy. With EUS, we normally do both at the same time.” 

Technological advances will continue to improve EUS as a diagnostic tool, says Jowell. These improvements include smaller, more maneuverable endoscopes and a new instrument with forward-viewing optics, as opposed to current instruments that allow visualization only at an angle. He predicts that future EUS instruments will not only visualize tumors, but also be used as a method for injecting chemotherapeutic drugs or vaccines.

Gastroenterologists are already routinely using one endoscopic treatment called “photodynamic therapy” (PDT) for abnormal GI tissues and cancers. In PDT, a light-sensitive chemical is first injected into the bloodstream. For a reason physicians still do not completely understand, cancerous or dysplastic tissues tend to accumulate more of the chemical than do normal tissues. The gastroenterologists next insinuate a fiber-optic-equipped endoscope into the GI tract near a tumor or such dysplastic tissue as found in Barrett’s esophagus. Flashing a low-power laser through the optical fiber, they expose the chemical to light, activating it to generate a highly reactive form of oxygen, which triggers a toxic chain reaction in the tissue, killing the tumor or dysplastic cells.

Such therapy has proven especially useful in treating Barrett’s esophagus with high-grade dysplasia, says McGrath, and that’s important to preventing cancer. “Barrett’s esophagus is a known risk factor for esophageal cancer, which is probably increasing faster than any other type of cancer,” he says. Particularly worrisome, says associate professor Scott Brazer ’77, M.H.S. ’90, director of Duke’s PDT treatment and research program, is that many people with Barrett’s don’t know it. “Many who have heartburn symptoms alleviated with antacids may have Barrett’s. And about a third of patients with the disorder have no symptoms, perhaps because their esophageal tissues have become resistant to the acid. So, there are a frightening number of patients out there at risk for adenocarcinoma that don’t have any idea that they have it.”

Brazer and McGrath are testing a new PDT technique for more effectively treating advanced cases of near-cancerous dysplasia. The method involves inflating a clear plastic balloon in the esophagus to flatten out the hills and valleys in the overgrown tissue, allowing a more uniform exposure to the laser light. The physicians have already completed the treatment phase of the clinical trial and are observing the patients to determine whether the “centering balloon”—invented by endoscopic PDT pioneer Gene Overholt at the University of Tennessee—will prove more effective at removing, or ablating, dysplastic tissue. 

Such advances are making PDT an even better choice over surgery, says McGrath. “Until now, treatment for high-grade dysplasia was often surgical removal of the entire esophagus, which is a major surgical procedure for patients. And because many of the patients were elderly, there was a significant risk of complications and even death. I think this trial will show that PDT is an effective treatment and will save patients with high-grade dysplasia from esophagectomy. However, we need to learn more about safety and other issues before we can say whether it will be useful in patients with lower-grade Barrett’s.” 

Among the most demanding of the gastroenterologists’ endoscopic techniques is the one with the mouthfilling title “endoscopic retrograde cholangiopancreatography” (ERCP)
—a division service directed by professor John Baillie and associate director Malcolm Branch, who is an associate professor. This technique with the huge title employs one of the tiniest endoscopic devices. ERCP involves threading an endoscope down the throat, through the stomach, and into the duodenum. Unlike other forward-looking endoscopes, the ERCP optics aim sideways, and the physicians use the imaging to seek out the “ampulla,” a tiny opening where the pancreatic and bile ducts exit into the intestine. Once they find the ampulla, the gastroenterologists thread a tiny catheter through it to inject X-ray dye, or to use small baskets to snag bile duct stones, electrocautery to open the duct, or cylindrical stents to prop the duct open. 

“For such procedures as retrieving bile duct stones and relieving certain types of malignant obstructive jaundice, ERCP is superior to surgery in its reduced complications and length of hospital stay,” says Jowell, who participates in ERCP procedures. “In general, it’s a one-day outpatient procedure.” He emphasizes that considerable skill is needed to master the art of threading the catheter into the ampulla in the confusing, turbulent depths of the intestine. “It sounds so straightforward because there’s this opening and you think you can just slide the catheter in. But the ampulla is surrounded by a muscle that often makes it difficult to thread the catheter into the duct.

Once past that, there are curves in the duct and the pancreatic duct is often no larger than a pencil lead.” In fact, a study of fellows in training at Duke by Baillie, Jowell, and their colleagues revealed that ERCP training elsewhere is often inadequate. “The study showed that, to gain minimum competence in ERCP, a physician must do from a 180 to 200 procedures. But the average training number we found in most fellowships was twenty-five to fifty. We came in for a lot of heat about the study, but our data were indisputable,” says Baillie, who was awarded the 2001 “Master Endoscopist Award” by the American Society for Gastrointestinal Endoscopy. 

Many surgeons were initially resistant to ERCP, but most medical centers have now developed the same kind of productive partnership between gastroenterologists and surgeons that Duke enjoys. Baillie says, “There are surgeons depending on me and my team for what we do, and vice versa. Most diseases involving the liver, bile ducts, and pancreas need a multidisciplinary approach. For example, we’ll have patients come in with terrible pancreatitis and cholangitis—infection of the bile from a stone in the common bile duct. If such a patient were operated on during the illness, they would run a high risk of death. So, we’ll use ERCP to remove the stone and let the pancreatitis settle down; and later the surgeons can operate to remove the diseased gallbladder and the stones it contains, to prevent a recurrence of the problem.”

What’s more, Baillie says, ERCP will become even more important as researchers learn to better detect early malignancies of the bile duct and pancreas—an especially virulent cancer—by testing the cells that can be retrieved with the technique. “There are promising techniques that will allow us to progress from what used to be 25 percent sensitivity for picking up cancer, to 70 to 80 percent. Ultimately, if we can identify people at high risk really early, we may be able to change the natural history of the disease, by diagnosing the precursor lesion and finding a way to reverse it.”

While the endoscopic techniques used by the gastroenterologists are perhaps the most visible of their efforts, the division’s faculty are also conducting pioneering basic research to understand GI disorders including pancreatitis, hepatitis C, cancer, and Crohn’s disease. And they are operating specialized services such as the Inflammatory Bowel Disease Clinic not only to treat patients’ medical disorders, but also to counsel them on how to cope with their disease.

So with all this expertise available, what of the “hot rock” lodged in my belly? The colonoscopy report pronounced my colon an “excellent preparation,” which made me rather proud. Annoyingly, however, the report’s characterization of my dynamic self as a “fifty-four-year-old man with dyspepsia” did make me sound like a cranky old goat, which I did not particularly appreciate. But gradually, as I took my gastroenterologist’s advice to lay off caffeine and aspirin, that hot rock shrank away to almost nothing. And, when I quit taking a joint-fortifying glucosamine supplement, the rock disappeared altogether.