A review of cerebral vasospasm. Part IV. Post-traumatic vasospasm.

Cerebral vasospasm is increasingly recognized as contributing to ischaemia after head injury. The reported incidence of post-traumatic vasospasm (PTV) varies between 10% and 90%, probably largely because of differences in patient selection, in definitions of vasospasm and in methods of detecting spasm. In severe head injury, based on studies with similar criteria, the incidence is approximately 40%. PTV is often associated with traumatic subarachnoid haemorrhage (tSAH), but has been reported without tSAH. These two factors are independently associated with poor outcome, but the direct links between tSAH, vasospasm and outcome are uncertain. There is evidence that calcium antagonists improve outcome in patients with head injury and tSAH; aminosteroids may also be effective here. Other strategies such as maintaining normocapnia and control of blood volume and pressure may also be useful. Further investigation of large cohorts is required to clarify fully the significance of PTV, its relationship with tSAH and outcome and possible treatment modalities.

Incidence and effects of increased cerebral blood flow velocity after severe head injury: a transcranial Doppler ultrasound study I. Prediction of post-traumatic vasospasm and hyperemia.

Cerebral blood flow velocity was monitored in 50 patients with severe head injury (GCS 8 or less) within 24 h of injury and at least once daily thereafter, using transcranial Doppler ultrasonography (TCD). Delayed post-traumatic vasospasm occurred in 20, and cerebral hyperemia in another 15. Doppler velocities were higher in vasospasm, which also lasted significantly longer than hyperemia. The presence of visible blood on an early CT scan was of some value in predicting vasospasm only (accuracy 59%). Xenon-133 cerebral blood flow (CBF) was also measured within 24 h of injury. An abnormal early cerebral blood flow level, either above or below a narrow central range, was more effective than CT in predicting vasospasm or hyperemia (accuracy 80%), while the combination of an abnormal blood flow and hemorrhagic findings on CT scan helped to determine which of these would occur later. Patients with an abnormal early cerebral blood flow and hemorrhagic findings on CT were more likely to develop vasospasm--accuracy for prediction of vasospasm 73%. This distinction may be of great importance, since these different groups of patients may well need different management.

Incidence and effects of increased cerebral blood flow velocity after severe head injury: a transcranial Doppler ultrasound study II. Effect of vasospasm and hyperemia on outcome.

Cerebral blood flow velocities were measured in 50 severely head injured (Glasgow Coma Scale (GCS) 8 or less) patients using transcranial Doppler ultrasonography (TCD). Abnormally high TCD velocities were recorded in 35 patients; in 20 this was deemed to be due to vasospasm and in the other 15 to hyperemia. Patients who developed hyperemia also had the highest intracranial pressure (ICP) and the lowest cerebral perfusion pressure (CPP) of the three groups. Outcome was assessed at six months after injury using the Glasgow outcome scale and the disability rating scale. In the normal velocity group 87% of patients had a good outcome, compared with 47% of those with hyperemia and 40% of those with vasospasm. The highest velocity (HVEL), GCS, age, ICP and CPP were entered into a logistic regression analysis. HVEL, age and CPP were found to be the most significant predictors of outcome (chi 2 = 29.5; p < 0.0001). These factors predicted outcome with 82% accuracy, 86% sensitivity and 76% specificity. Routine monitoring of TCD velocity may be useful in detecting hyperemia and vasospasm after severe head injury, allowing appropriate treatment to be started as early as possible.

The significance of traumatic intraventricular haemorrhage in severe head injury.

Since the advent of computed tomography (CT) traumatic intraventricular haemorrhage (IVH) has been diagnosed more often. It has reportedly been associated with a poor prognosis, but pure or solitary IVH is rare, suggesting that other lesions occurring concurrently with it may contribute to the poor outcome. In a series of 65 patients with severe head injury (GCS < or = 8), 14 (22%) had IVH on initial CT. Death rate in these 14 was 21%, not significantly different from that in patients without IVH (14%), although a significantly higher proportion of patients without IVH had a good outcome. These results suggest that mortality is related to other lesions associated with IVH rather than to IVH alone and that the presence of IVH does not necessarily lead to a poor outcome.