Share this story

COVID critical care experience leads UMass Chan surgery resident to study CPR survival

Research finds no survival benefit among sickest patients

Daniel J. Baldor, MD, MPH

A new controlled study from UMass Chan Medical School provides evidence for the first time that could inform physicians about which very sick or injured patients would be most likely to benefit from cardiopulmonary resuscitation (CPR).

The study by Daniel J. Baldor, MD, MPH, chief resident in general surgery, and his colleagues found that “full code” status, including aggressive CPR, provided no survival benefit over do-not-resuscitate (DNR) status among the most severely ill individuals, those with greater than 75 percent predicted mortality as determine by Apache IV score, a severity of illness rating. Among less severely ill patients, full code status was associated with improved survival. The article was published in Journal of Intensive Care Medicine.

Dr. Baldor was profoundly affected by his experiences performing CPR on patients in the COVID intensive care unit whom the medical team knew would not survive. He spent hours with patient families trying to help them come to accept the inevitable death.

“Part of this,” he explained, “is due to misconceptions from the media, where there is a 60 percent to 100 percent survival rate for in-hospital CPR on TV shows. In real life, it is usually around 20 to 30 percent, with half of those having some form of remaining disability.”

Such scenarios in which health care providers perform “futile CPR,” meaning resuscitations that have an extremely low likelihood of survival, do not accomplish the patient’s broader goals and are performed against the best judgment of the health care team because of patient or family demands, are not rare, according to Baldor. Futile CPR can be physically traumatic to the patient and adversely affect clinicians’ mental health.  

“CPR is an intervention that’s been placed in a unique category,” said Baldor. “Procedures are otherwise not indicated when they have no effect, especially if they are harmful or costly. CPR is lifesaving for certain patients, but for many it is considered futile.”

Baldor wanted to know why the practice of futile CPR was often the norm but could find no controlled studies on which to guide patient selection for CPR candidacy. So, he undertook a research project that   he hoped could directly affect a problem he saw.

“No one wants to do futile CPR, but the problem with saying something is nonindicated is that you need really good evidence to say that,” Baldor said. “There’s never been a controlled study on CPR before this. It represents a higher level of evidence to help guide the decision-making process.”

The research team hypothesized that the protective effects of CPR decrease as illness severity increases. They looked at 17,710 ICU encounters from five hospitals, stratifying by predicted mortality quartiles using Apache IV scores and analyzing survival between full-code and DNR patients within each quartile. Patients were followed for a median of 760 days.

In addition to finding that CPR intervention was associated with prolonged life in patients in the first three quartiles of illness severity, but not the sickest or most traumatically injured, researchers identified a stepwise decrease in survival benefit for full-code patients when compared to DNR patients as illness severity increases. The researchers suggested this pattern shows a possible ceiling effect of full-code CPR for improving survival.

Baldor said the study needs to be repeated with a larger data set and in multiple health systems. More importantly, advanced directives and conversations about end-of-life care ne