She was aghast—her heart is damaged and, at any time, can start quivering instead of beating. If the device, a defibrillator, was unable to shock her heart back to normal, her life was in danger.
On the horizon is an even smarter heart device, one that detects deterioration in various heart functions and tells the patient how to adjust medications.
They are part of a new wave of smart implantable devices that is transforming the care of people with heart disease and creating a bonanza for researchers. The hope is that the devices, now being tested in clinical trials, will save lives, reduce medical expenses and nudge heart patients toward managing their symptoms much the way people with diabetes manage theirs. Patients, who often are frail or live far from their doctors, can be spared frequent office visits. Doctors can learn immediately if devices are malfunctioning or if patients' hearts are starting to fail.
"It's like having an office visit every day and a complete physical every week," said Dr. Leslie Saxon, a cardiologist at the University of Southern California.
The big leap forward came a few years ago when device companies figured out how to make transmitters that send data over a broader range, 20 or 30 feet. That meant that, with her device, Elzo did not have to wait till her doctor could put a receiver directly on her chest. Instead, she simply went near a small box, which is attached to a phone jack near her bed. Once a week, she also measures her weight and blood pressure—key indicators of heart failure— and that information is automatically transmitted to her doctor. If there are problems, the machine alerts her doctor.
"Now, every single day the device is being queried," said her doctor, James Coman of the Heart Rhythm Institute in Tulsa. "It's just a phenomenal tool."
There is a downside, though: "Information overload is a very serious problem" for the doctors, said Dr. Lynne Warner Stevenson, director of the Heart Failure Program at Brigham and Women's Hospital in Boston and a professor at Harvard Medical School, who counts herself as a proponent of smart devices. More information, she warned, is not always beneficial.
The devices transmit useful data along with data whose significance is not clear, like variations in heart rate. Large swings in heart rate can indicate risk, but it is not clear what to do about them.
Even more confusing are changes in thoracic impedance, a measurement of resistance to electric current through the lung. Impedance changes can predict future heart crises, but more often have no clinical explanation. Yet when doctors get data on impedance changes, they often feel uneasy and call patients to see how they are, making patients uneasy in turn, Stevenson said.
Stevenson likened such information to the game of "Jeopardy!" -- doctors are given answers in search of a question. It's a challenge even for the nation's 1,000 heart failure specialists. But it can be even harder for primary care doctors, who have less expertise in heart failure yet care for most of the six million patients in the country with the condition.
Dr. Richard Page, president of the Heart Rhythm Society, said doctors wonder if they can be held liable if they do not look at all the data. Still, he said, the new technology "is potentially transformative."
For researchers the information deluge leads to a different problem: how to analyze the data. A large clinical trial of a cardiac device used to involve 1,000, maybe 2,000 patients. Now, Boston Scientific, a maker of one of the smart heart devices, is following 400,000 patients.
"No one has ever done research like this before," said Saxon, who leads an independent team of academic scientists overseeing Boston Scientific research. The company has no editorial control over the papers the scientists write, Saxon said.
Boston Scientific gets data from patients' defibrillators. It also gets information on deaths from Medicare.
The data are stripped of patient identifiers and analyzed, a task requiring the company to become more like a Google or a Microsoft, handling enormous amounts of information. There are, for example, more than four million recordings of weights and blood pressures and over 60,000 instances when the defibrillators went off, shocking a patient's heart.
So far, Saxon's group has reported on the first 90,000 patients. Half of them had not been enrolled for remote monitoring and served as a control group.
Patients whose doctors looked at the data survived 5 percent to 15 percent longer than patients in earlier clinical trials of the devices, Saxon reported. And, in a paper under review, the group reports that their three-year survival was significantly greater than that of patients in the study whose doctors did not see the data.