Stay Connected


Advertise With Us

Learn More

Skip to content


A Computer Glitch Turns Miracle Machine into Monster for Three Cancer Patients

Posted on

Marietta, Ga., June 4, 1985. Katy Yarbrough entered the mauve-colored treatment room at the Kennestone Regional Oncology Center and lay down for her radiation treatment on the table under the Therac 25. Katy had breast cancer, and the ultra-sophisticated machine was to irradiate her left breast and nearby lymph nodes.

“Ordinarily, the treatment was very uneventful,” she says. “You just went in and got it and you left.” But on this day something terrible happened. “When they turned on the machine,” says Yarbrough, “I felt this tremendous force of heat going into the left side of my body, this red-hot sensation.”

When the technician re-entered the room, Katy was crying and trembling. “You burned me,” Yarbrough told the woman. “I’m sorry,” she recalls the technician said, “but that’s not possible. It’s just not possible.” The Therac had been operated at the cancer center 18,000 times without incident. But Katy insisted, “I know something happened.”

She was back home when a triangular swelling began to rise in the skin above her left breast. Within days the pain had become excruciating, so Sheila Yarbrough, Katy’s housemate and former daughter-in-law, took her to West Paces Ferry Hospital in Atlanta. She was admitted with what her oncologist believed to be muscle spasms and a frozen shoulder.

The first week she was hospitalized her doctors continued to send Yarbrough back to Kennestone to go under the Therac 25. Finally, when the welt above her breast began to break down and slough off layers of skin, Katy rebelled. “I have had my last treatment,” she told Sheila. “I can’t tolerate any more.” She said she would rather die than face another radiation treatment.

None of her doctors appeared to grasp what Yarbrough had suspected from the instant it occurred: She had suffered radiation burns from the Therac 25. Not until two more patients fell victim to the machine, almost a year later and nearly a thousand miles away, did Katy’s cancer specialists learn what had happened: The Therac 25, miracle of modern medicine, therapeutic wonder, had become an instrument of destruction. Instead of 200 rads of radiation, Katy had received at least 20,000—the most anyone is known to have sustained in a medical accident and survived.

Tyler, Texas, March 21, 1986. Oilfield worker Voyne Ray Cox, 33, who’d had a tumor removed from his back, was undergoing his ninth treatment with the Therac 25 at the East Texas Cancer Center. Suddenly he felt a jolt of heat. As he flinched he got hit again. “Hey, are you pushing the wrong button?” he called out, half-humorously, to the technician. The third jolt was a cure for humor. Cox felt as though he’d been hit by a punch harder than any man could throw.

When Ray asked his radiological oncologist, Lee Schlichtemeier, if he’d received too much radiation, Schlichtemeier said the machine indicated that he had not even gotten his prescribed dosage. The machine had turned itself off, registering on its computer screen the message “Malfunction 54″—a computerese red flag indicating only that something was wrong.

Schlichtemeier called in Fritz Hager, the cancer center radiological physicist in charge of the Therac. Hager inspected the machine, then called its manufacturer, Atomic Energy of Canada, Ltd. Following the suggestions of AECL, Hager checked various functions but could find nothing wrong. The machine was put back in use.

As Ray’s condition rapidly deteriorated—spitting blood, he checked into the Good Shepherd Medical Center in Longview—Hager called AECL again. The Canadian firm sent its senior engineer and a service technician to Tyler to examine the machine with Hager. Hager also had an independent testing firm check the Therac for possible electric-shock problems. Like a car that refuses to stall when it’s at the repair shop, the Therac could not be induced to go into a Malfunction 54, and the experts remained in the dark as to the cause or nature of the problem.

“We did not know that the patient had been injured by radiation,” says Hager. “There was no clue, other than the patient, that anything was wrong with the machine. The machine wasn’t telling us what it had done.” The Tyler clinic resumed use of the Therac 25.

Then, 22 days after Ray Cox’s injury, the machine struck again. Verdon Kidd, 66, had been irradiated more than 30 times already and had just four treatments left for a skin cancer on his ear. As the machine switched on, he heard a frying sound, saw a flash of light and felt something hit him in the ear. He screamed in pain. The technician ran to tell Hager that there’d been another Malfunction 54.

Hager immediately called AECL and later informed the radiology physicists at the four other U.S. hospitals where the Therac 25 was in use, including the Kennestone Regional Oncology Center in Marietta. “I wasn’t going to let the machine be used until we figured it out,” he says. Then he went back and had the technician recount step-by-step what she had done that had led to Malfunction 54.

It turned out that she had caught herself mistakenly typing an X-ray mode into the machine’s computer. She had used the “up” arrow to get into edit function, where she changed the “x” mode for X-ray to the “e” mode for electron beam and the energy from 25 million electron volts down to 10 mev. She used the return key to go to the bottom of her screen and waited for the “beam ready” command. When she typed in “b” to turn the beam on, the machine showed a Malfunction 54 and shut itself off, but not before jolting the patient with an excessive dose of radiation. Hager believed he had found the problem: The machine had scrambled the X-ray and electron functions.

The Tyler machine is now shut down. The initial fix on the Kennestone machine has been to disable the “up” arrow button on the keyboard, so that the operator may not move into the edit function and start the chain of events leading to the malfunction.

The solution came too late for Verdon Kidd, who died in early May, and for Ray Cox, who died four months later. Ray, according to his wife, Debbie, maintained his sense of humor right up to the end, claiming, for example, that his injury was caused by “Captain Kirk forgetting to put the machine on stun.”

Katy Yarbrough has shown similar pluck. Radiation damage deprived her of her left breast and the use of her left arm. “I could not take a bath by myself,” she says. “I was in pain.” Then one Sunday seven months after the accident and following reconstructive surgery, Katy made up her mind to drive her car. It was a luminous moment for her—”the happiest of my life,” she says. Now she works four or five hours a day as a receptionist at the beauty shop where she once designed the best fake nails in town. Her doctors tell her to enjoy each day as much as she can. “That’s what I try to do,” she says with her usual cheerfulness, adding, “I know that I am lucky.”

Responsibility for the tragedy of Malfunction 54 is a question that will be determined by ongoing lawsuits. But there is no question that the toll could have been even worse. Says Sam Bishop, director of risk management for the Kennestone Regional Health Care Systems: “We are lucky that we know what happened to Mrs. Yarbrough. Had they not been able to recreate this accident in Tyler, it is conceivable that we could still not know.”