Prognostication in Neurocritical Care: Just Crystal Ball Gazing?

Abstract

Karina Melchior, a 19-year-old Danish girl, was driving her car when she came off the road and crashed. She sustained severe traumatic brain injury. She was admitted to the intensive care unit in the Aarhus University Hospital, Denmark in a comatose state. Three days after admission, the physician determines that there was no hope for recovery and even informed the relatives that the girl may become brain dead. The CT scan images showed severe traumatic brain injury. She asked the parents’ consent for organ donation. They consented, but Carina did not progress to brain dead. Then, the doctors decided that all life support should be withdrawn in order to let the young girl die. The ventriculotomy was removed, mechanical ventilation was stopped, and the endotracheal tube was removed. But, to the surprise of the parents, sister, and brother of the girl and the medical and nursing staff, the girl opened her eyes and started moving her arms and legs. In the following months, she made a good strides at a rehabilitation center. She has fully recovered; she walks, talks, is independent, rides her horse, and continued high school again. This all can be seen in the shocking documentary Piggen der ikke ville do (The girl who wouldn’t die) [1]. The documentary, that followed Carina’s family from the moment she was admitted in the hospital, sparked a public debate in Denmark about organ donation, brain death, and withdrawal of life-sustaining treatment in cases of severe brain injury. Over 500 registered donors withdrew their consent in fear of being declared dead too soon. To their credit, the doctors in the Aarhus Hospital have admitted that they made a grave error in prognostication. A similar situation occurred in 2002, 34-year-old Jesper Bendixen was also wrongly assessed brain dead in the same hospital, after he was admitted for a severe traumatic brain injury. His relatives were also asked for organ donation, but Jesper also recovered. This documentary was also broadcasted in many other european countries. What happened? Were there just incompetent doctors in Aarhus Hospital? Did they make a too early judgment? Is prognostication in neurocritical care still very difficult? The report by Edlow et al. shows some of the challenges [2] with the unexpected recovery of a 19-year-old man who sustained a severe traumatic brain injury. The early MRI findings, suggested a poor prognosis. Despite the presence of severe axonal injury on early MRI, he regained the ability to communicate and perform activities of daily living independently a year after his surgery. The authors conclude in their report that MRI data should be interpreted with caution when prognosticating for patients with traumatic brain injury. What if the staff, based on the MRI findings, had decided to withdraw all life-sustaining measures when the patient was still in critical condition and dependent on mechanical ventilation? Most probably, he would have died. And, as this was to be expected seeing the severity of the injury as judged from the MRI images, nobody would have doubted the decision to withdraw support. We now know that this was nothing more than a self-fulfilling prophecy. A self-fulfilling prophecy is a prediction that directly causes itself to become true, by the very terms of the prophecy itself. In critical care, predictions of poor prognosis may become self-fulfilling if life-sustaining measures as mechanical ventilation are withheld or withdrawn on the basis of that prediction. Patients are believed to have a high E. J. O. Kompanje (&) Department of Intensive Care Medicine, Erasmus MC University Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands e-mail: e.j.o.kompanje@erasmusmc.nl