Abstract
The main goal of neurosurgical intensive care in comatose patients with head injuries is to prevent secondary neuronal damage resulting from ischemia and tissue hypoxia. In his thesis, Olaf Cremer critically examines the rationale and risk-to-benefit ratio of aggressive manipulation of blood pressure, intracranial pressure, cerebral perfusion pressure, and temperature
... read more
in head trauma patients. He concludes that there is little evidence to manage these variables at strictly controlled values in all patients at all times, and stresses the necessity to individualize treatment in the future, so that optimal therapy can be offered with a reduced risk of complications.
The thesis first describes two clinical experiments that sought to determine the effects of varying temperature and blood pressure on the regulation of cerebral blood flow and brain tissue oxygenation in humans. A trial in 19 patients, who were experimentally heated to 41.8°C for two hours, showed that profound hyperthermia causes transient cerebral vasoparalysis and metabolic uncoupling, but does not impair brain oxygenation. Another experiment in 13 patients, designed to find optimal cerebral perfusion pressure, showed that the effects of blood pressure augmentation by vasopressors on intracranial pressure, autoregulation capacity, and brain tissue oxygenation varies considerably between patients and from day to day. However, the intervention generally seemed to be beneficial only when intracranial hypertension was present.
Subsequently, the results of a large cohort study, involving 333 patients in two trauma centers with very different approaches to the management of head-injured patients, are reported. Aggressive management of intracranial pressure and cerebral perfusion pressure at strictly controlled values in one of the two centers resulted in increased levels of therapy intensity and a much-prolonged stay in the intensive care unit, but did not improve long-term neurological outcome. Furthermore, aggressive therapy that included the use of both vasopressors and high-dose propofol infusions to suppress cerebral metabolic demand for oxygen was associated with a rare, but lethal cardiac complication (which has since been named 'propofol-infusion syndrome').
The thesis concludes that, due to a large heterogeneity within the head trauma population, clinical algorithms that are designed to retain cerebral and systemic physiological variables within rigid and arbitrary limits at all times, are likely unnecessary or unrealistic for many patients with head injuries. Therefore, to reduce the risk of iatrogenic injury, future management should be targeted more at specific subgroups of head-injured patients with a particular cerebral pathology and an intermediate prognosis. To enable the early selection of patients for more specific treatment, statistical models for outcome prediction can be helpful. Thus, in a cohort of 304 patients, a prediction rule was developed for individuals who remain comatose for more than 24 hours following blunt head trauma. The model was externally validated in another 122 subjects and proved to be robust, providing accurate predictions of the probabilities for death, survival with major disability, and functional recovery
show less