Michael Shaffer lay on an examining table staring at a screen that displayed an X-ray image of his coronary arteries -- the blood vessels that supply oxygen and nutrients to his heart. He could see that one of the large arteries ended abruptly.

A blood clot had clogged it shut at a point where fatty deposits had narrowed the vessel. Downstream, heart muscle cells deprived of their blood supply were dying, creating crushing chest pain that caused the 62-year-old electrical engineer to collapse.

Michael Shaffer was having a heart attack.

As Shaffer watched the screen, doctors at George Washington University Medical Center began injecting tissue-type plasminogen activator (tPA), an experimental drug that causes blood clots to dissolve, into his arm. Minutes later, a trickle of blood began to ooze past the blockage. Then a bit more, and then much more. Soon, the screen showed blood filling the artery below the blockage, restoring the vital, oxygen-carrying flow.

"It was a very, very, very nice feeling," Shaffer recalls in his soft British accent. "I don't know if it was the pain that went away or my elation of seeing the clot dissolve, but things started to take a rosier picture."

The picture, too, could one day turn rosier for more than half a million heart attack victims each year. They may be helped by clot-dissolving agents such as tPA that enable physicians to interrupt heart attacks as they occur and limit the amount of damage done to the heart. That could save lives and reduce disability in those who survive.

At least three kinds of clot-dissolving proteins have been identified since the late 1950s -- tPA, streptokinase and urokinase. TPA only recently became available, and preliminary results of its effectiveness were so good that they were uncharacteristically rushed into print by The New England Journal of Medicine last month because, wrote journal editor Dr. Arnold S. Relman, the initial findings indicated that a drug like tPA "might be of immense clinical value."

TPA is "probably the most exciting single development so far this decade," says Dr. Allan Ross, chief of cardiology at George Washington. "It is even more important than the artificial heart" because it has the potential to help far more people.

Heart attacks kill more Americans than any other cause. About 1.5 million will have a heart attack this year, the American Heart Association estimates. More than half a million will die -- 350,000 before they reach the hospital.

"Here we are talking about an event so common that before a reader is done with the article," Ross says, "there will be 50 deaths in the U.S."

But tPA "is not going to be a panacea," says Dr. Burton E. Sobel, director of cardiology at Washington University Medical Center in St. Louis. "We know that the coronary artery disease is still there after the clot is dissolved . To say this is going to cure heart disease and render people immortal is very naive."

While tPA may reduce the damage done to the heart by a heart attack, it won't prevent the heart attack from occurring or recurring, nor will it prevent all damage.

"TPA dissolves blood clots in about 40 minutes on average," says Dr. Alan Guerci, director of coronary care at Johns Hopkins Hospital in Baltimore. "It takes about 20 minutes from the time the clot forms to get some permanent damage . . . So, yes, there are problems with the strategy that are very serious."

Still, if tPA works as well as it did in the initial tests, and if its use becomes widespread, "the risk of dying from a heart attack could be reduced by up to 30 percent," says Dr. Eugene Braunwald of the Harvard Medical School and chairman of the federally sponsored Thrombolysis in Myocardial Infarction (TIMI) study, in which Schaffer became a participant when he walked into the emergency room.

Braunwald also believes that the chances of surviving a year or longer after a heart attack also would improve.

The TIMI study showed that tPA is safe and able to unclog coronary arteries in two out of three patients when given intravenously (i.v.). Streptokinase, another clot-dissolving chemical cardiologists have been using since the early 1980s to open coronary arteries, could reopen the arteries in only one out of three patients when given intravenously.

"That drug tPA , per se, is a major development," Braunwald says. "It is a major step forward."

Heart attacks happen because atherosclerosis -- the slow, chronic process of building up fatty deposits inside blood vessels -- narrows regions of the coronary arteries. These arteries form a network that envelops the heart's surface and carries oxygen and nutrients to the muscular cells of this constantly pumping organ.

A blockage in the coronary artery, however, does not interfere with the flow of blood through the pumping chambers of the heart itself. As long as the heart continues to beat, blood will be pumped through the lungs and then back out to the rest of the body.

But if a clot forms in a vital area in the coronary artery and is not quickly removed, then heart muscle served by that vessel will die and stop pumping blood.

"If 40 percent of the left ventricle the main pumping chamber of the heart is lost, the patient almost always dies," says Dr. Eugene R. Passamani, associate director for cardiology of the National Heart Lung and Blood Institute, a part of the National Institutes of Health.

And usually, dying heart muscle hurts.

"It is a pain," Shaffer recalls of his heart attack. "You know pretty well what it is because it is in the area of the chest and it hurts, and it is accompanied by some sweating. You feel overwhelmed. You knew something serious is happening.

"I suppose I was scared."

For physicians, too, heart attacks can be quite agonizing. Until recently, they could treat only the complications, not prevent some of the damage.

"We used to stand fairly helpless next to our patients with heart attacks," says Dr. K. Peter Rentrop, director of interventional cardiology at Mount Sinai School of Medicine in New York. "We could not get to the basic process, the loss of heart muscle. We were helpless."

Before the early 1960s, a doctor could only order a heart attack survivor to bed for six to 10 weeks.

By the early '60s, doctors recognized that heart attack survivors often died later from a disturbance in the rhythm of the heart's contractions. But if a patient could be kept alive through these irregular heart beats, they often subsided as the heart healed. Doctors used electric shocks to restore normal rhythms and developed drugs such as lidocaine to prevent the arrhythmias.

In the late '60s, Harvard's Braunwald suggested that not all of the heart muscle jeopardized by a heart attack died right away because it continued to receive some small amount of blood from other arteries (so-called collateral blood flow).

"There is some tissue which stops contracting, but is not irreversibly damaged . . .it is in a twilight zone," Braunwald says. "It is neither dead nor alive. It is stunned, like it has been clobbered very hard. But it has the capacity to return to life."

Researchers tried many different drugs to protect these twilight zone cells but, says NIH's Passamani, "we really haven't demonstrated that we can limit the amount of heart damage by going that route."

A new approach to rescuing these cells arose by accident in 1978 when the coronary artery of a 41-year-old West German truck driver abruptly clotted shut during a diagnostic procedure called a coronary angiogram.

Physicians use coronary angiograms, an X-ray study, to see the heart's blood vessels and determine how badly they are narrowed. To do the study, a catheter -- a long, thin tube -- is threaded into an artery in the groin or the arm and guided through the body until it enters the coronary arteries. There it can squirt an iodine-based dye directly into the artery, outlining the vessel while X-ray movies are taken.

Cardiologist Peter Rentrop was called into the catheterization lab where the study was being performed by a junior colleague when the patient on the table suddenly began having a heart attack.

"After about a half an hour of agonizing, fumbling and deciding, I pushed a guide wire used to get the catheter into the artery through the obstruction," Rentrop recalls. "This caused an immediate improvement of chest pain, and the EKG electrocardiogram became normal within two to three minutes. We had never seen anything like that with any therapy."

Rentrop, then at the University of Gottingen, West Germany, began experimenting to see if guide wires could routinely open blocked coronary arteries. About a year later, he decided to see if the clot-dissolving agent streptokinase could be used instead of guide wires.

Streptokinase had been used since the late 1950s to dissolve clots in the legs and lungs. This bacterial enzyme converts an inactive molecule called plasminogen into a clot-dissolving molecule called plasmin. Plasminogen normally circulates in the blood, and, once converted to its active form, can break up the fibrin proteins that form a clot.

Doctors tried to use streptokinase intravenously in the 1950s to dissolve clots in the coronary arteries, but the results were inconclusive and interest waned.

Rentrop decided to modify the approach taken by those doctors. Instead of trying to inject streptokinase intravenously, he decided to inject streptokinase directly into the coronary artery through a catheter. He hoped that concentrating the drug directly next to the clot would make a difference. It did.

After Rentrop published the first positive results of streptokinase use in 12 heart attack patients, "everybody and his brother started with streptokinase by intracoronary injection," says Washington University's Sobel. "Then as people became sensitized to the delays that using the catheter entailed, people began to use it i.v., injudiciously. There was no proof of benefit and there was risk."

"We have not been successful in preventing blind enthusiasm," Rentrop admits.

By the early 1980s, it was clear that a carefully controlled study was needed to determine whether intravenous streptokinase could indeed open clotted coronary arteries quickly enough and at an acceptable risk, so the NIH-sponsored TIMI trial was devised.

But by now, discoveries in this field were moving fast. Initially, TIMI was to compare streptokinase to a placebo. By the time the trial was ready to begin, large quantities of tPA, a new, rare, native protein produced by the body, had become available through the use of gene engineering.

TPA, like streptokinase, converts the inactive plasminogen into clot-dissolving plasmin. But unlike streptokinase, tPA activates plasminogen only when the tPA itself is bound to a clot. TPA does not activate plasmin all over the body, so it doesn't cause as many bleeding complications as streptokinase, says Sobel. "The risk with streptokinase is that the patient can suffer systemic bleeding. With tPA . . . there is less risk of bleeding."

This theoretical advantage has yet to be proven. The frequency of major bleeding complications with streptokinase is less than 5 percent and the frequency of life-threatening bleeding is less than 1.5 percent, several studies have indicated. No comparable estimates are available for tPA; however, preliminary findings from the TIMI trial did not show a significant difference in bleeding complications between tPA and streptokinase.

"My own feeling is that the real virtues of tPA lie in its safety rather than any real or imagined efficacy," Sobel says. "Urokinase and streptokinase can lyse dissolve clots, but they have side effects, so the practical dose is limited."

TPA has other advantages, too. Since it's a natural body protein, it doesn't cause allergic reactions as streptokinase can. And tPA remains in the blood for only five to eight minutes, compared with hours for streptokinase. But once it is bound to a clot, Sobel says, tPA remains in the body for 60 to 90 minutes.

"It is well agreed that there is a tremendous advantage for safety for tPA," Sobel says. "Even if tPA and streptokinase were equal in clot-dissolving ability , tPA is safer."

Even more important, the TIMI trial showed that tPA opens blocked coronary arteries two times more effectively than streptokinase.

"Now the big question is," says NIH's Passamani, "if you interrupt the heart attack process, does it help and by how much and what else do you have to do to prevent it from reclosing?"

But the biggest problem is time.

"Patients all too frequently get to the hospital too late," says Johns Hopkins' Guerci. The average heart attack victim waits three hours, says the American Heart Association, before deciding to seek help.

"We frequently get patients coming in six to nine hours after the onset of chest pain," Guerci says. "If you want to save half the region of the heart that otherwise would have died , you have to establish reperfusion blood flow in two hours. If you reperfuse in four hours, you only save 15 percent. If you reperfuse at six hours, it's too late."

But, says George Washington's Ross, "we don't have to save every last cell. We just have to change big heart attacks into medium heart attacks and medium into small."

Although animals studies show that these drugs can restore blood flow, reduce the amount of damage to the heart and improve the heart's ability to pump blood, no one has proved, in a systematic way, that tPA treatments help the human heart or will lengthen a patient's life.

While the evidence so far suggests that tPA, given soon enough, will preserve portions of the heart, cardiologists caution that final proof must await the second part of the TIMI trial scheduled to begin this summer.

In the future, if tPA proves as beneficial as it seems, paramedics may be able to inject it into heart attack victims, or the patient could possibly inject himself.

"I believe that ultimately, patients should be equipped to either medicate themselves or have a paramedic or interested other be trained to do it," Sobel says. The goal is to have "the agent in someone's home, where it can be given very quickly. That is a long way from reality."

Currently, Genentech Inc. of South San Francisco, which produces tPA through gene engineering techniques, and Survival Technology Inc. of Bethesda are jointly trying to develop an automatic injection system for tPA similar to the "autoinjectors" Survival makes for the Department of Defense to inject nerve gas antidotes into soldiers in combat.

Researchers also need to decide what needs to be done after the patient's artery has been opened.

"In the overwhelming majority of patients in whom therapy is effective, a very tight narrowing remains behind in 80 to 90 percent of the arteries," says Hopkins' Guerci.

A narrowed artery can clog shut again unless the narrowing is removed or bypassed. Ross estimates that 30 percent of clogged arteries will reclot if nothing is done after tPA treatment.

For example, a second heart attack struck Michael Shaffer a week after tPA aborted his first one.

"I was just about ready to leave hospital and I had another heart attack," says Shaffer, now 63. "I would presume the coronary was so bunged in general that another clot started it off."

Three days later, he was sent for byass surgery. "I have not had a twinge since," he says. "Not a twinge."

The researchers must now learn to identify those patients who should be treated with angioplasty to widen the narrowed arteries, bypass surgery to reroute blood around the blockages or anticoagulation and other drug therapy.

Shaffer has continued to do well since surgery. "I am running two miles three times a week," he says, "plus playing tennis and golf."

The heart attack "was retribution for the sausages and eggs and the delightful things of life," Shaffer laments. "And I do miss sausages and eggs."