Object. The availability of magnetic resonance (MR) imaging data obtained i
n comatose patients after head injury is scarce, because MR imaging is some
what cumbersome to perform in patients requiring ventilation and because, i
n the first hours after injury, its relevance is clearly inferior to comput
erized tomography (CT) scanning. The authors assessed the value of MR imagi
ng in the early postinjury period.
Methods. In this prospective study MR imaging was performed in 61 consecuti
ve patients within 7 days after they suffered a severe head injury. An init
ial CT scan had already been obtained. To understand the clinical significa
nce of the lesions whose morphological appearance was identified with MR im
aging, brainstem function was assessed by registration of somatosensory and
auditory evoked potentials.
Brainstem lesions were visualized in 39 patients (64%). Bilateral pontine l
esions proved to be 100% fatal and nonbrainstem lesions carried a mortality
rate of 9%.
In singular cases circumstances allowed for a clear clinical distinction be
tween primary and secondary brainstem lesions. On MR imaging all lesions we
re hyper- and hypointense after intervals longer than 2 days. Within shorte
r intervals (< 2 days) after the injury, primary lesions appeared isointens
e on MR imaging. in one secondary brainstem lesion there were no traces of
blood.
Conclusions. Because mean intracranial pressure (ICP) levels in patients wi
thout brainstem lesions were similar to those in patients with brainstem le
sions, the authors conclude that it was not mainly increased ICP that accou
nted for the high mortality rates in patients with brainstem lesions.
The authors also conclude that brainstem lesions are more frequently found
in severe head injury than previously reported in studies based on neuropat
hological or CT scanning data. Early MR imaging after head injury has a hig
her predictive value than CT scanning.