"I heard a female voice say, 'We have a problem, her arteries are too small,'" Reynolds said.
The conversation and other details she recounted were in the surgical records.
Dr. Karl Greene, who was on the team of surgeons, asked, "Why would she have this kind of information if she was so deeply under a barbiturate infusion of medications that should be shutting down her brain?"
In the absence of facts, some partial theories have been proposed. The body may have a self-defense mechanism that produces endorphins to create a sense of well-being in anticipation of death. Because cells never stop working all at once, Reynolds' brain might have continued to function after it had flat-lined, absorbing or envisioning details.
Reynolds, however, believes it was a distinct memory.
"It was just too in the pocket to make it up," she said.
"Whether that image came from somewhere else that she then internalized somehow, I don't think there is any way to tell," said Spetzler. "But it was sort of intriguing how well she described what she shouldn't have been able to see."
As the science of resuscitation improves and tackles unanswered questions, doctors are not only redefining what we mean by the term "clinical death," they are re-engineering the ways they can resuscitate those who have experienced it.
"I don't think you want to live at those limits, but at least you know that there is maybe a little maneuvering room out there," Waters said.