Research from the Johns Hopkins Children's Center exposes alarming gaps in training hospital
residents in "first response" emergency treatment of staged cardiorespiratory arrests in children,
while at the same time offering a potent recipe for fixing the problem.
The research was conducted just before the release of the 2005 American Heart Association's
practice guidelines focusing on strengthening first-response skills, timing that suggests that at least
some of the findings in the study may paint a grimmer picture than current reality, researchers say.
Changes already made to the Johns Hopkins resident training program beginning in 2005 have resulted
in significant improvement, they add.
The Johns Hopkins study, now available online and to be published in the July print issue of the
journal Resuscitation, revealed critical mistakes during lifesaving maneuvers like chest compressions
and defibrillations in children undergoing arrests, or "codes" as they are medically known.
Staging mock cardiopulmonary arrests with life-size dummies, researchers observed that of the
70 residents participating in the drills, one-third (24) never started chest compressions, while two-
thirds (46) did so with a delay of more than one minute, the critical cutoff time to initiate
compressions in a child without a pulse. Nearly half the residents (46 percent) failed to restore heart
rhythm using a defibrillator within the recommended three minutes. Timely resuscitation of a child
whose breathing or heartbeat has stopped is, of course, critical to prevent permanent brain damage
Because most arrests in children are caused by respiratory rather than cardiac problems,
pediatric life-support training in most teaching hospitals traditionally has emphasized airway rather
than heart maneuvers to resuscitate a lifeless child. But in a patient without a pulse, airway maneuvers
will work only if used together with chest compressions to circulate the blood, investigators say.
Therefore, the Johns Hopkins team calls for a shift in focus that would equally emphasize cardiac
maneuvers along with airway ventilation.
The findings, even though not necessarily applicable to other teaching hospitals, suggest the
need for an honest examination of the way academic programs across the country train pediatric
residents to deliver life support during cardiopulmonary arrests.
"We're firm believers in the idea that only by identifying our weaknesses can we know exactly
how and when we can improve care," said lead investigator Elizabeth Hunt, a critical-care specialist at
"This has been a sobering experience," she said, noting that no one likes to have problems
exposed, but without the courage to gather evidence about what really is working and what is not,
change won't happen.
Hunt said the solution to the problem has so far proved relatively simple: Practice, practice,
practice with simulated arrests, and strict measurement of results to increase skills and speed of
Hands-on training including monthly mock drills on pediatric units and simulations with child
manikins--like those staged by the Johns Hopkins researchers--appear to dramatically improve
fledgling doctors' performance, according to preliminary and not-yet-published reports.
While length of residency training (first, second or third year) did not make much difference in
performance in the study, experience in performing resuscitation did. The results show that residents
who had even once used a defibrillator--either during a drill or in a real patient--were 87 percent
more likely to successfully restore heartbeat during the exercise than those who had never used the
lifesaving device. Making the residents practice all the steps required to defibrillate, rather than just
watching a training video of someone else doing so, was the key to success. "There's no substitute for
practice," said Hunt, who is also the director of the Johns Hopkins Medicine Simulation Center.
The study also demonstrates the importance of monitoring performance, the researchers say.
Senior investigator Peter Pronovost, a critical care specialist at Johns Hopkins and an
internationally renowned expert on patient safety, said, "Simply taking a course is likely not enough to
ensure adequate performance. We must couple hands-on training with monitoring. After all, patients'
lives depend on it."
Past research shows that 14 percent of all arrests in hospitalized children are cardiac in origin,
and many respiratory arrests quickly evolve into cardiac arrests. More than one-quarter of all arrests
in children involve heart rhythm abnormalities that require use of a defibrillator to shock the heart
into normal rhythm.
"The prevailing wisdom of focusing on ventilation rather than circulation during pediatric
arrests is well-founded, but it may have caused the pendulum to swing too far," Pronovost said. "We
must restore the balance and start paying attention to circulation and heart rhythm maneuvers and
teach future pediatricians [that] these are equally important."
The Johns Hopkins curriculum has already augmented its basic life-support courses, and
advanced life-support courses are required for all residents, with monthly mock codes on pediatric
units and monthly resuscitation sessions using simulator life-size dummies, training that provides
hands-on experience and helps residents learn how to communicate during a crisis.
Previous studies have shown that only 14 percent to 36 percent of children who suffer an arrest
in the hospital survive. Even though the absolute number of deaths is quite small, the few deaths that
do occur can be averted by strengthening first-responder instincts in residents, as well as in other
Co-investigators in the study are Kimberly Vera, Marie Diener-West, Jamie Haggerty, Kristen
Nelson and Donald Shaffner, all of Johns Hopkins.