CPR: Are we doing it wrong?

Published: March, 2014

Some say just doing chest compressions might be easier and more reliable. And there are questions about where the defibrillator fits in.

Cardiac arrest means simply that the heart has stopped beating. Without that constant thumping in our chests, the 10 or so pints of blood that usually circulate through 60,000 miles of arteries and veins stop flowing — and bad things start to happen, fast. The lungs are not delivering fresh oxygen. Metabolic waste products pile up. The blood turns dangerously acidic.

The brain and nervous system react first because they're such ravenous consumers of oxygen, but soon other organ systems start to falter. If circulation isn't restored within four or five minutes, the brain shuts down completely and permanently. That's the definition of death.

Enter cardiopulmonary resuscitation (CPR). The term is a bit misleading, because its purpose isn't to restart the heart, although that has been known to occur. The idea is to keep the person alive until they can be treated in a hospital. Rapid chest compressions push blood through the body. They must be done quickly (100 times per minute, according to guidelines) because they're no match for the pumping power of the heart, which normally beats about 70 times per minute if you're resting. The ventilations, or puffs into the mouth (2 for every 15 compressions), are surrogates for breathing, meant to resupply the blood with oxygen. Although there's less oxygen in expired air than in the atmosphere (16% versus 21%), the difference isn't important in an emergency.

A good outcome is hard to find

It's been almost 30 years since the first CPR guidelines were written. Studies have shown that when it's done right away — and correctly — CPR does save lives. Some research suggests that it more than doubles your chances of surviving a cardiac arrest outside the hospital.

The gloomy "on the other hand" is that those chances aren't very good. The statistics vary tremendously, partly because emergency response times vary, but studies in large cities have found that only about 1 in 20 people who have a cardiac arrest outside of the hospital survive — even if they receive CPR.

CPR isn't going to save everyone; after all, cardiac arrest is often not some bolt from the blue but the end of a long slide into poor health. Still, the low survival rate is troubling, especially in light of all the effort and publicity that's been put into the technique.

The advent of the automated external defibrillator (AED) has added another wrinkle. AEDs analyze the activity of the heart and if it has developed a lethal rhythm (it's fibrillating) or is not beating at all, the machine delivers an electric shock to jolt it back into a normal beat. In 2005 the FDA started allowing the sale of AEDs without a prescription. The price is about $1,500.

So, when someone collapses from what seems to be cardiac arrest, should you do CPR or, if there is one nearby, run and get an AED? So far, there's no consensus.

Even the experts do it wrong

Two studies published in the Journal of the American Medical Association in January 2005 have suggested one reason for CPR's low batting average: It isn't being done correctly, even by trained professionals. That doesn't bode well for the efforts of laypeople.

The first study included 176 adult cardiac arrest patients from three hospitals, in Stockholm, London, and Akershus (a county outside of Oslo). Paramedics or nurse anesthetists gave the patients CPR on the way to the emergency room. The ambulances carried equipment that measured the depth and frequency of chest compressions and the number of ventilations. The main finding: Chest compressions weren't given 48% of the time when the patients' hearts weren't beating. The compressions were also too shallow — less than the recommended 1½–2 inches "deep."

The other study involved 67 cardiac arrest patients given CPR by doctors or nurses at the University of Chicago Hospitals between December 2002 and April 2004. Using special monitoring equipment, the researchers divided the first five minutes of CPR into 30-second intervals. The compression rate was too slow during 28% of those intervals, and 40% of the compressions were too shallow. Professionals didn't get the ventilations right, either. During 60% of the 30-second intervals, patients were hyperventilated.

This isn't the only research that's found shortcomings in CPR. For example, in a report on 13 cases published in Circulation in 2004, doctors at the Medical College of Wisconsin found emergency service workers ventilated patients at twice the recommended rate (30 times per minute versus 12–15).

'You're going too fast,' says the computer

Dr. John Tobias Nagurney, an emergency department doctor at Massachusetts General Hospital and member of the Health Letter's editorial board, notes that brief interruptions in CPR are unavoidable, at least in the hospital, as doctors put in intravenous lines, check for a pulse, and perform other procedures. Keep in mind that although these studies have shown departures from guidelines, they haven't taken the next step and proved that those departures result in bad outcomes.

One solution is to improve CPR training, so doctors and emergency service workers get the message about keeping up with chest compressions and not hyperventilating patients. But that's just a partial answer, because research has shown that CPR training starts to "wear off" in just a few months.

Technology could help professional rescuers. Some have high hopes for computer-guided CPR that will give audio prompts: The compression rate is too slow; the ventilation rate is too fast; and so on. Early attempts at automating the chest compressions failed, but now there are improved devices that wrap around the patient and squeeze the chest at the appropriate time. It isn't clear, though, how much fire departments and ambulance companies will want to spend on this equipment.

Breathless in Seattle

Changing CPR is another possibility. Some experts are suggesting that ventilations — particularly in out-of-hospital cardiac arrests — may not be worth doing because they get in the way of the chest compressions. Studies have identified chest compressions as the critical element during the first few minutes. In most circumstances people can survive for four to five minutes (longer if the body temperature is low) without having their blood reoxygenated.

University of Arizona researchers published a study in 2004 in Circulation that pointed out the problem with current guidelines. About 50 first-year medical students were trained in basic CPR, then tested on mannequins soon afterward and six months later. Even right after the training, two quick breaths took an average of 14 seconds — time not available for chest compressions.

But could eliminating the ventilations cost lives? Not according to a study done in Seattle. Fire department dispatchers there instructed rescuers at the scene of an apparent cardiac arrest to perform, at random, either compression-only CPR or standard CPR. Among hundreds of cardiac arrests, survival rates were the same in both groups.

Dr. Mickey Eisenberg at the University of Washington, a leading expert on CPR, thinks the jury is still out. The success in Seattle depended on quick response times by the fire department, he says, asking "What happens to the patient after four or five minutes" if there's no ventilation? Dr. Eisenberg is undertaking a compression-only CPR study in suburban Seattle. One possibility is keeping the ventilations but cutting the rate down to, say, once or twice every 100 compressions.

Making it simpler

Every five years, CPR experts from around the world gather to review the science, but individual "resuscitation councils" decide how to translate scientific findings into practical guidelines. In this country, the American Heart Association (AHA) serves as the resuscitation council. The expert meeting was held in January 2005 in Dallas, but its conclusions are being kept under wraps until November 2005, when they'll be published in Circulation. The AHA is following up with new guidelines in December 2005.

The AHA has moved toward simplifying CPR guidelines before. In 2000, it eliminated the recommendation that lay rescuers check the victim's pulse. The absence of a pulse is arguably the best sign that the collapse was actually caused by cardiac arrest and not something else (see chart below). But it turns out that, as many studies show, lay rescuers (and even some health professionals!) can't accurately assess the presence of a pulse, so they are just wasting time. The AHA that year also simplified the guidelines for chest compressions and compressions-to-ventilations ratio.

Is it cardiac arrest — or something else?

It's both frightening and confusing when someone suddenly collapses, especially if it's a loved one. Don't assume it is cardiac arrest. More people lose consciousness from fainting (syncope) and seizures than from cardiac arrest. Here are some ways to distinguish between the causes of sudden collapse.

Cardiac arrest

People lose consciousness and have no pulse or a very weak one (although current guidelines say laypeople should not check for a pulse). Although breathing generally stops, it may continue for several minutes after the heart stops. There may be a few seizure-like movements as the brain becomes short on oxygen (hypoxic). Note that a heart attack rarely causes loss of consciousness unless it leads to cardiac arrest. Heart attack symptoms include heavy pressure in the chest, shortness of breath, and lightheadedness.

First aid tips: Call 911 before you do anything else. Shocks from an AED may restart the heart. But if you go to find the defibrillator and it's not nearby, you could lose precious minutes.

Fainting (syncope)

People lose consciousness and may turn ashen, but they keep on breathing. The cause is a sudden drop in blood pressure that temporarily leaves the brain short of blood, so the pulse may be very weak.

First aid tips: Don't support the person in a chair or in an upright position. Think of fainting as a protective measure: Nature makes us go to the ground to spare the heart from fighting gravity as it works to restore blood flow to the brain. Lay the person down and elevate the feet. If there is vomiting, roll the person on his or her side to prevent choking.


Seizures are caused by sudden or irregular electrical activity in the brain. Conditions like epilepsy cause seizures, but they can also be triggered in otherwise healthy people by a fever, sleep deprivation, an infection, or dehydration. In most cases, the sudden jerking motions last only a couple of minutes. Afterward, people often become groggy and may fall asleep. They usually wake up in 3–5 minutes.

First aid tips: Seizures often look much more dangerous than they actually are, so stay calm. A person who is having a seizure should not be restrained, and nothing should be put in his or her mouth. If possible, it's important to protect the person's head. If there is vomiting, roll the person on his or her side to prevent choking.


The classic symptoms include numbness on one side of the body or face, slurred speech, loss of vision in one or both eyes, and dizziness. When a stroke is caused by a hemorrhage in the brain rather than the more common kind caused by a blood clot, the symptoms can be rapidly followed by a loss of consciousness, but the symptoms almost always come first.

First aid tips: Call 911 immediately, and get treatment as quickly as possible. Chances of recovery improve tremendously with early intervention. You can't offer much in the way of first aid; just stay with the person and keep him or her comfortable.

Where do AEDs fit in?

The AED presents a tough choice. On one hand, it's theoretically better than CPR because it can restart the heart, whereas CPR is merely a stopgap. Dr. Eisenberg describes the AED as a more "definitive therapy." A study published in 2004 in the New England Journal of Medicine found that a defibrillator-CPR combination improved the survival rate over CPR alone (23% versus 14%).

On the other hand, if you run around looking for an AED while neglecting CPR, you could lose valuable lifesaving minutes.

From a cost-benefit point of view, the value of a home AED is debatable. In essence, you're plunking down $1,500 on the guess that a) someone will have a cardiac arrest at home and b) someone else will be there to use the machine. If you do decide to take this gamble, at least make sure you know how to use the defibrillator and exactly where you have stored it.

As a service to our readers, Harvard Health Publishing provides access to our library of archived content. Please note the date of last review or update on all articles. No content on this site, regardless of date, should ever be used as a substitute for direct medical advice from your doctor or other qualified clinician.