AHA Forum: Pediatric Cardiopulmonary Arrest Mortality No Better with Earlier ICU Care

May 11, 2007

WASHINGTON -- Getting hospitalized children who show signs of cardiopulmonary deterioration off the floors and into the ICU before they arrest may improve treatment but it doesn't save lives, researchers found.

WASHINGTON, May 11 -- Getting hospitalized children who show signs of cardiopulmonary deterioration off the floors and into the ICU before they arrest may improve treatment but it doesn't save lives, researchers found.

Early initiation of a rapid response team to get pediatric patients to the ICU before they arrested yielded a 63% drop in the number of arrests occurring outside the ICU, said Tia Tortoriello, M.D., of the Children's Medical Center Dallas, and colleagues.

"Patients who experience cardiopulmonary arrest often exhibit signs and symptoms of prolonged deterioration," they said, and hypothesized that "if these patients are identified and managed appropriately, cardiopulmonary arrest can decrease, and death can be prevented."

But study results showed that mortality rates remained unchanged, they reported here at the American Heart Association's Scientific Forum on Quality of Care and Outcomes Research in Cardiovascular Disease and Stroke.

When Dr. Tortoriello joined the hospital's cardiopulmonary resuscitation committee in 2003, she became convinced that it needed to shift from an emphasis on reaction and resuscitation to earlier intervention and prevention.

She and colleagues compared their hospital's data to other pediatric hospitals of similar size and patient type using the AHA's National Registry of Cardiopulmonary Resuscitation.

They found that their "hotspot" for cardiopulmonary arrest was the chronic mechanical ventilation floor.

Children's hearts stop most frequently because of respiratory problems, Dr. Tortoriello said.

Armed with these data and the hospital's support, Dr. Tortoriello and colleagues started a campaign educating hospital staffers about early warning signs of potential cardiopulmonary arrest. Those included fast or labored breathing, increased oxygen use, excessive sleepiness, and irritability.

The staffers were instructed to call for the rapid response team "at their earliest concern for the patient."

The investigators prospectively compared outcomes from October 2003 through March of 2005 (before the protocol change) with outcomes from April 2005 through November 2006 (after the change) using a code blue database and the National Registry of Cardiopulmonary Resuscitation database.

Before the rapid response team approach was adopted, there were 137 cardiopulmonary arrest events -- 169 in the ICU and 68 in non-critical care areas.

After the rapid response protocols were put in place, there were 304 events, of which 274 occurred in the ICU and 30 in non-critical care areas.

Among the findings, the investigators reported:

  • A 63% decrease in events outside critical care areas (1.0 versus 2.7 events per 1,000 patient discharges, P