Almost an emergency: W. Bosseau Murray and residents bring a flat-lined patient simulator back to "life".

On screen those surgical pincers look big as battery clamps, the "blood vessel" like a length of sturdy rope. Even so, I'm biting my lip. Again I nudge a little forward with the left hand, close the grip, and come up empty. "That 2-D camera takes a little getting used to," the technician concedes. Nudge, close, empty. Nudge, close—and at last I've got a hold of that strand of red-striped twine. Look, Ma! I'm a cardiac surgeon!

The real resident, Edward Stephenson—he's the one in the surgical greens—stands back with arms folded and a small smile on his lips as the next wannabe sits at the console.

Ten feet away, at the operating table, three robotic arms are poised over the thoracic trainer, a plastic chest-shaped shell with a cut-away rib cage. At the end of each arm, a metal tube thin as a pencil penetrates the shell. On the inside, where the heart would be, two of these tubes are tipped with tiny pincers, grasping for twine. The third is an endoscope, the tiny camera that feeds that magnified image back to the console monitor.

The console and arms, Stephenson had told us moments earlier, are components of the ZEUS robotically assisted microsurgery system, the latest thing in "minimally invasive" cardiac surgery. Developed by Computer Motion, Inc., of Goleta, CA, ZEUS is currently being clinically tested and refined at four U.S. sites, including here at Penn State's Hershey Medical Center.

"There are many possible applications for this system," Stephenson said, but the focus right now is on coronary artery bypass grafting, or heart bypass.

Open-heart surgery is performed 450,000 times a year in the United States alone, he said, and as the name suggests, it is a drastically invasive procedure. To get at the heart a surgeon has to make a 12- to 15-inch incision the length of the chest, then split the breastbone; recovery from such trauma can take a patient months. Endoscopic surgery, by contrast, requires only three small "ports:" one for the lighted endoscope, and two for the surgical tools. This gentler approach is already routine for scraping knee cartilage (where it's called arthroscopy) and removing gall bladders (laparoscopy); but when it comes to the heart, and stitching coronary vessels that are only 1 to 2 millimeters in diameter, until now the precision just hasn't been there.

"You've got these long instruments, and very small needles," Stephenson explained. And at 20 times magnification, every little tremor is an earthquake. With hand-held instruments, you very quickly get beyond human capability.

"ZEUS obviates these problems," he said. The robot arms themselves are rock-steady. Computer-controlled tremor elimination and motion scaling, Stephenson said, provide "virtual stillness" even if the human operator is a little nervous. The endoscope, too, is voice-controlled, so that both hands can remain always on the instruments. "This particular system recognizes over 20 words," Stephenson said.

Over the past three years, a Penn State team led by Stephenson's boss, Ralph Damiano, chief of cardiothoracic surgery, has been both testing ZEUS and developing techniques for its use. The first trials were completed on mechanical trainers like the one we're playing on today. Next the team moved to cadaver hearts, and then to live dairy calves—several of which we had earlier seen, post-procedure, gamboling in a nearby pasture. In December 1998, Damiano performed the first robotically assisted heart bypass surgery in North America, on a 70-year-old patient from Lebanon, PA.

"We're still in the investigative phase with this," said Stephenson. But ZEUS's early success, he added, in addition to the promise of reduced patient trauma and faster recovery, "has significant implications for the way surgery is done and the way it is taught."

Endoscopic surgery, Stephenson acknowledged, "Takes a little adjustment," for surgeons used to conventional methods. Fortunately, there's a place down the hall for honing the necessary skills. In the Simulation Development and Cognitive Science Laboratory, we encounter a virtual trainer, a sort of high-end video game which connects endoscopic instruments to force feedback devices to allow the sensation of manipulating objects in space. On the screen, the current task is to grasp a small colored ball and deposit it into a series of transparent boxes. The computer keeps time, counts errors, and scores for efficiency.

Across the room, W. Bosseau Murray, professor of anesthesiology and director of the lab, introduces the human patient simulator, a hard-wired medical mannikin that looks like a life-size G.I. Joe. Lying there on the operating table, this nameless patient does everything but whistle through his nose.

Robotic surgery system that makes cardiac procedures far less invasive.

His chest rises and falls. His pupils dilate. Grasp his ankle and you can feel his femoral pulse. A nearby monitor keeps track of his vital signs—heart rate, blood pressure, and blood-oxygen level—which soar and dip as Murray tweaks the system by tapping instructions on a nearby keyboard.

A shot in the arm of epinephrine, and his heart rate surges. A muscle relaxant, and his thumb ceases to twitch. "There's a bar code on the syringe that the computer recognizes," Murray explains.

Time for a little test. Murray hits a few strokes, and in a second or two ("It takes a while for the disease process to occur," he says), the monitor starts to beep, beep, beep. Looks like respiratory distress. No problem, however. A visitor who watches ER on television knows what to do. Tube him and bag him!

Murray passes the breathing tube; nervous hands fumble tube into mouth. After a short struggle, the tube slides past the teeth, and into the larynx. "I'm in," the rookie sputters. But the noise from the monitor is not reassuring. BEEP BEEP BEEP, it now reports. And now BEEPBEEPBEEPBEEPBEEP. . . .

Eyes widen around the table. Hands scramble, but to no avail. "He's dying!" the rookie blurts at last. "He's just starting to die," Murray murmurs pleasantly. He pulls the tube back a few centimeters; the beeping slows and then ceases. "The tube should measure 19 to 21 centimeters at the teeth," he says, pointing to the marks on the plastic. "You were in a bit too far."

A second induced emergency calls for a tracheotomy; a second visitor probes the patient's neck warily with her fingers before successfully sinking the needle. By now Murray has gauged his audience, and the third problem comes with a giveaway clue: "When something goes wrong on the monitor, look at the patient," he says. Sure enough, only one side of the chest is rising and falling. Collapsed lung.

The crisis management sessions that Murray conducts with Hershey's medical staff sound considerably more rigorous. "We put them through various scenarios," he says, describing a series of progressively worsening cardio-respiratory problems. "We bring in nurses first for a few minutes, then comes the surgeon, and finally the anesthesiologists. Just like in an emergency, not everyone arrives at the same time." The two-hour sessions are videotaped for later review.

"Overwhelmingly," he says, "the staff has said that this training has improved their performance in a crisis. They have an increased ability to step back both physically and mentally and analyze the entire situation."

Ralph J. Damiano, M.D., is professor of surgery and chief of cardiothoracic surgery in the College of Medicine, Hershey Medical Center, 500 University Drive, Box 850, Hershey, PA 17033; 717-531-8330; rdamiano@psu.edu. W. Bosseau Murray, M.D., is professor of anesthesiology and director of the Simulation Development and Cognitive Science Laboratory; 531-4265; wbmurray@psu.edu.