Update on the Management of Subarachnoid Hemorrhage

Katja E Wartenberg


Future Neurology. 2013;8(2):205-224. 

In This Article

Delayed Cerebral Ischemia

DCI is defined as the development of new focal neurological signs and/or deterioration in level of consciousness, lasting for more than 1 h, or the appearance of new infarctions on CT or MRI. The underlying pathophysiology is thought to be vasospasm and other causes are excluded.[112,113] This definition has been found to be more meaningful than symptomatic vasospasm (new focal deficit and/or decrease in level of consciousness due to vasospasm), especially in patients with severe SAH whose neurological deterioration may be unrecognized. Arterial narrowing can be demonstrated angiographically in 50–70% of patients and leads to delayed ischemia in 19–46% after SAH (angiographic vasospasm, see Figure 7). The development of DCI starts on day 3 after SAH, is maximal at 5–14 days and resolves on day 21. The presence of thick subarachnoid blood seen on admission CT and severe intraventricular hemorrhage are strongly associated with higher risk for vasospasm (Figures 2 & 3).[37–39,114,115] Prevention and management of DCI is listed in Table 1.

Figure 7.

Cerebral angiogram demonstrates vasospasm.
(A) Vasospasm of the basilar artery (arrow), (B) the right vertebral artery (arrow) and (C) branches of the left posterior cerebral artery (arrow).
Reproduced with permission from the Oxford University Press, NY, USA.219

Monitoring for DCI

Observation in a NICU with expertise in performing frequent neurological examinations (options: Glasgow Coma Scale exam hourly, NIH Stroke Scale 6 hourly[116]) and daily transcranial Doppler ultrasonography are simple and helpful monitoring tools.[29,56] Decreased level of consciousness and focal signs such as aphasia or hemiparesis in a good-grade SAH patient should prompt the clinician to take immediate action such as a confirmatory test.[29,56]

Transcranial Doppler ultrasonography is a noninvasive method used to diagnose vasospasm in the larger cerebral arteries with high specificity and variable sensitivity, dependent on the operator and other systemic conditions.[117,118] A mean flow velocity (Vm) greater than 120 cm/s in the middle cerebral artery is concerning for vasospasm, Vm above 200 cm/s is considered to be predictive, but dynamic changes of the Vm, such as a twofold increase, may be more sensitive for the diagnosis of vasospasm.[117,119] The Lindegaard index (Vm of the middle cerebral artery in relation to Vm in the extracranial internal carotid artery) above 6 also indicates the presence of arterial vasospasm.[119–121] If DCI due to vasospasm is suspected, a vascular imaging study, such as CT with CTA and/or CT perfusion, MRI with MRA or the gold standard, a cerebral angiogram, should be performed.[29,56] CTA was found to have a high negative predictive value of 95–100%, a good correlation with cerebral angiography and a tendency to overestimate the degree of arterial narrowing.[122,123] CT perfusion gives additional information on cerebral perfusion status with mean transit time and CBF. Both correlate well with cerebral angiography; mean transit time >6.4 s is more sensitive and CBF is more specific for vasospasm.[124,125] These imaging tests should be repeated if the clinician is uncertain about the change in clinical status being caused by DCI, if an endovascular intervention is considered and if the risks of the planned therapy may outweigh the benefits.[56]

Poor-grade patients in stupor or coma require different monitoring techniques to identify DCI. Multimodal monitoring may be helpful in these patients by providing direct and real-time information about partial brain tissue oxygen pressure (by polarographic technique through Clark electrode) and metabolism (cerebral lactate, pyruvate, glucose, glycerol and glutamate by microdialysis), cerebral perfusion (MAP - ICP = CPP, CBF by thermal diffusion microprobe) and depression of brain activity by continuous electroencephalography or intracortical electrodes. Quantitative continuous electroencephalography analysis demonstrated sensitive and specific detection of DCI by reductions in α-variability or α/δ ratio.[126,127] Clusters of spreading depolarizations seen on cortical electroencephalography were associated with DCI.[128] These are currently being investigated in the DISCHARGE-1 Phase III study. Partial brain tissue oxygen pressure monitoring allows for early detection of DCI showing a decrease in the cerebral oxygenation.[129] Elevations of glycerol, glutamate and lactate/pyruvate ratios as markers of ischemia were correlated with reductions of CBF on PET and DCI.[130–132] When used, these parameters should be interpreted taking into account their limited region of capture and their location in relation to blood clots and other pathology. Moreover, screening for perfusion deficits and arterial narrowing with CTA and CT perfusion may be reasonable in poor-grade patients.[56]