A little-known device is shaking conventional wisdom for reviving people who suffer sudden cardiac arrest: People may be able to go much longer without a pulse than the 20 minutes previously believed.

The capnograph, which measures carbon dioxide being expelled from the mouth of the patient, can tell rescuers when further efforts at cardiopulmonary resuscitation, or CPR, are futile or whether they should be continued. It is the latest effort that cardiology experts and emergency teams are devising that aim to improve a patient's odds.

The American Heart Association recently revised its guidelines for first responders, with particular emphasis on initiating hard, rapid chest compressions to keep the stricken victim's blood circulating. Rescue squads increasingly are chilling victims of cardiac arrest with ice packs and other cooling approaches, a technique known as hypothermia, in order to protect the brain from injury when blood flow is restored.

Some 300,000 Americans suffer sudden cardiac arrest each year, and fewer than 10 percent survive long enough to leave the hospital.

For Howard Snitzer, a 54-year-old chef from Goodhue, Minn., a capnograph kept his rescuers from giving up. When Snitzer collapsed from cardiac arrest outside a grocery store in January, two men at a gas station across the street sprinted to his aid. Both volunteer firemen, they were soon joined by two dozen local police, firemen and rescuers from two neighboring towns in what became a 96-minute marathon to get Snitzer's heart started again.

How a Life Was Saved

Snitzer's rescuers, who rotated every couple of minutes to avoid fatigue, kept pumping his chest. Thirty-four minutes after he went down, a medical team from the Mayo Clinic swooped in via helicopter. During the ordeal, 11 shocks with a defibrillator failed to restore his heartbeat.

What kept them going? Readings displayed on the capnograph brought by the Mayo team indicated the air coming out of Snitzer's lungs had healthy levels of carbon dioxide—strong evidence that CPR was effectively moving oxygenated blood to his brain and other organs.

"If we didn't have the CO2 readings we were getting, we would have terminated efforts much sooner," said Bruce Goodman, a flight paramedic with the Mayo Clinic's Medical Transport unit, which was summoned to Snitzer's aid.

Sudden cardiac arrest typically results from a heart attack or an electrical malfunction that causes a runaway heartbeat. Snitzer went into ventricular fibrillation, a quivering of the heart that leaves it without a pulse.

Of the 300,000 annual cases of sudden cardiac arrest outside a hospital, 125,000 victims are found too late to be helped, said Michael Sayre, associate professor of emergency medicine at Ohio State University, Columbus, and co-author of the American Heart Association's new CPR guidelines. A person who is down for 10 to 12 minutes without any assistance is almost impossible to revive. For the others a critical factor in their prospects is whether someone saw them collapse, called 911 and began effective CPR.

The big worry in sudden cardiac arrest beyond restarting the heart is protecting the brain. Other organs such as the kidneys and the lungs "can tolerate longer periods" without oxygen, but the brain is the organ that succumbs most likely early on," said Dr. Roger White, a Mayo Clinic anesthesiologist who was involved in Snitzer's case and who is an expert in management of sudden cardiac arrest.

Rescuers have long monitored patients' breathing tubes for carbon dioxide to make sure the device was correctly placed in the windpipe instead of the esophagus. Now for the first time, the recent AHA guidelines, which were published last fall, call for a more sophisticated and continuous monitoring of CO2. Just last week, the AHA published course materials to teach emergency personnel how to use capnography.

The technology records CO2 pressure in milligrams of mercury. Research by White and others shows that if the maximum CO2 pressure achieved during 20 minutes of CPR is 14 or less, resuscitation is almost certainly futile. If the level is above about 25, "you need to keep working at it until you've exhausted all of your tricks," White said.

When Goodman and his co-workers hooked Snitzer up to the capnograph upon their arrival, they were impressed with his CO2: It was in the low 30s. A normal level in healthy adults is between 35 and 45. But after the effort went on for 45 minutes, Goodman became concerned. In his 15-year career as a paramedic, the longest successful CPR case he'd been on was about 45 minutes.

Goodman consulted by cellphone four times with White. After about an hour and 15 minutes had elapsed following Snitzer's collapse, the two decided that an additional dose of an anti-arrhythmia drug and one more shock—the 12th—would be the last chance. CPR continued as those steps were taken. At the 96th minute, soon after the shock was administered, Snitzer's pulse returned.

Goodman wasn't inclined to celebrate. 

"I'll be honest, it was not a great feeling that night," he recalls. "It was a good feeling that we got pulses back, but there was nothing in history to tell me he would survive this and that he could recover" with his brain intact. "I wasn't sure we had done the right thing for him."

About five days later, Goodman learned that Snitzer, who had been treated for a heart blockage, kidney failure and other problems, was out of intensive care. Goodman and other colleagues who worked on the case went to the hospital to visit. Snitzer was sitting up in a conference room, talking to his family in a normal voice.

"It was a little bit of a shell shock to see him doing that well," Goodman said.

In a report on the case published online last month by Mayo Clinic Proceedings, White and his colleagues reported that Snitzer "experienced a complete neurologic recovery" and described the episode as the "longest duration of pulselessness in an out-of-hospital cardiac arrest with a good outcome."

Snitzer goes regularly to the Mayo Clinic for cardiac rehabilitation. He says he had hardly any idea what happened to him or how he came to survive it. But as he learned about the quick response from rescue workers and how the signals from the capnograph convinced them not to quit, he said: "I'm a regular guy. I happened to die at the right place at the right time." 

Click here to read more from The Wall Street Journal.