To diagnose a patient as “brain dead,” health care providers must be able to conduct an evaluation that finds no sign of brain activity and no chance that a patient will recover after being taken off life support. Guidelines exist to help hospitals make these determinations, as patients who are brain dead continue to have a pulse and feel warm to the skin.
Though the JAMA study didn’t evaluate whether patients have been declared brain dead when they actually weren’t, it did show that updated guidelines set in 2010 by the American Academy of Neurology haven’t been built into hospitals consistently. It’s possible that health care providers are practicing above or even below what the policies require, though researchers haven’t evaluated outcomes, says Dr. David Greer, a neurologist at Yale University School of Medicine and lead author of the study.
The problem is that without standard policies, hospital staff could mispronounce or mis-determine brain death, he says. While no cases of this have been reported, the study authors caution that doctors and hospital administrators should not assume their policies are working fine and no changes are needed.
Of 492 hospitals evaluated for the study, just over a third required that a health care provider have expertise in neurology or neurosurgery to determine brain death, and 43 percent stipulated that an attending physician make the determination. But 150 policies didn’t mention who should do it, meaning that less-experienced doctors could be making decisions.
Type of health care professional performing brain death determinationCourtesy JAMA Neurology
About 66 percent of hospitals required two separate examinations to determine brain death and nearly 21 percent required one examination. In cases in which patients had a heart attack, 29 hospitals specified staff are to wait for at least 24 hours before determining brain death. When testing for brain death was being considered, 201 hospitals had policies requiring them to notify organ procurement organizations. The longer patients are brain dead, the more difficult it can be to donate their organs.
The 2010 guidelines from the American Academy of Neurology followed a study researchers conducted about hospital policies in 2008, in which the 50 U.S. News best hospitals for neurology and neurosurgery were evaluated. Researchers found hospital policies varied significantly, and the new guidelines subsequently created a detailed checklist and step-by-step instructions for determining brain death.
“That study was performed because I noticed there were different policies in the hospitals where I was practicing,” Greer says. “I thought, ‘How could there be different policies for brain death? They should be the same everywhere.’”
Monday’s study aimed to see whether hospitals have changed their policies to be in accordance with the ones set by the American Academy of Neurology. This time the study expanded to all institutions, not just those ranked as high performers by U.S. News. Researchers were able to access 508 policies of an estimated 650 hospitals with intensive care units, and 492 provided enough data to use in the study. The data analyzed were from June 26, 2012, to July 1, 2015.
“The positive finding was that people had been updating their hospital policies, but not uniformly,” Greer says. “Maybe it’s time for the Joint Commission to step in and make it part of accreditation.”
Among protocols for determining brain death, 408 hospitals required doctors to identify the cause of brain dysfunction, and 463 required that health care professionals note the possibility of medications – such as sedatives or paralytics – be taken into account. In cases like these, patients can appear brain dead but resuscitate later.
When conducting clinical examinations, 441 policies stipulated that a patient must be in a coma, 414 stipulated patients must show no reaction to pain, 456 stipulated patients show no eye reactions, and 305 stipulated patients show no spontaneous respiration while receiving oxygen through a breathing tube.
Specifics of Clinical Examination RequirementsCourtesy JAMA Neurology
To receive data from hospitals, researchers had to agree not to publish information that could identify them, so the study does not identify whether hospitals are clustered by certain geographic areas. The paper does say that the data represent hospitals in all 50 states.