|Vol. 4, Issue 2, Article 2||Sanelli, P., Shetty, S. & Lev, M.|
There are several important reasons for imaging acute stroke patients:
* CTA/CTP may assist in selecting patients that are most likely to benefit from thrombolytic treatment, and excluding those most likely to hemorrhage.
* CTP has the potential to quantitatively distinguish infarct "core" from the "ischemic penumbra", which may provide a more rational basis for establishing the maximum safe time window for administering thrombolytic agents (21,22).
* CTP has prognostic value in predicting both final infarct size and clinical outcome in embolic stroke patients (23).
* CTP may assist in monitoring patients that do not receive thrombolysis for hypertensive therapy (Fig. 7).
* CTP may be used to define enrollment criteria for patients in clinical trials, and can serve as a surrogate marker for treatment response (21,24).
|Four Key Questions in Acute Stroke Imaging: CT versus MRI (Table 2)
CTA-SI: In the acute stroke patient, the admission NCCT is used to identify the extent of the hypodensity from early edema, as well as to exclude hemorrhage prior to treatment. CTA can quickly identify both circle-of-Willis occlusion or stenosis, and carotid artery stenosis. With appropriate window and level settings, the CTA source images (CTA-SI) can be initially reviewed at the scanner for detection of proximal large vessel thrombus or occlusion. For more detailed review, the CTA-SI can also be rapidly reformatted at the scanner console - in less than a minute - into axial, coronal, and sagittal collapsed MIP views (3 cm thick with 5 mm spacing) of the middle cerebral artery, anterior cerebral artery, and basilar artery. The accuracy of circle-of-Willis CTA for detection of proximal, large vessel thrombus approaches 99% (25). The CTA-source images (CTA-SI) are also useful for evaluating ischemic hypodensities that, like MR-DWI, represent tissue likely to be irreversibly infarcted. It has been shown that coregistration and subtraction of conventional pre-contrast head CT images from the "whole-brain" contrast-enhanced CTA-SI (Fig. 9) can provide quantitative maps of perfused-blood-volume (5,6). This technique requires the assumption of a steady-state level of contrast during image acquisition (6). The sensitivity and specificity of CTA-SI for detection of ischemia have been estimated to be approximately 95 and 100%, respectively, when small DWI lesions (< 15 mL, typical of lacunar and small brainstem infarcts) are excluded from consideration (26). In addition, final infarct size roughly equals the admission CTA-SI lesion in the presence of early, complete recanalization. The CBF and MTT lesions can be used to define operational ischemic penumbra; in the absence of early complete recanalization, the progression of ischemic tissue to infarction is therefore proportional to the degree of CBF/CBV mismatch. There has been good agreement of CTA-SI with PET-derived measurements of CBV (5).
ISCHEMIC PENUMBRA and INFARCT CORE: Quantitative CT-CBF and MTT maps can be used to describe regions of ischemic tissue "at risk" for infarction, aka "ischemic penumbra". Interpretation of CTP maps using color scale may be more optimal than gray scale to delineate precise degrees of ischemia (Fig. 1). Indeed, new software programs are being developed that will enable the user to select color-coded threshold values for the CTP maps, to facilitate segmentation of ischemic regions. The "core" infarct, representing irreversibly damaged tissue, demonstrates perfusion characteristics of decreased CBF and CBV with elevated MTT (Table 3). However, the "ischemic penumbra" will have decreased CBF and elevated MTT with normal or elevated CBV, due to activation of cerebral autoregulatory mechanisms (Fig. 1) (7,9). The transition from ischemia to infarction depends not only on CBF values, but also on the duration of diminished blood flow; with worsened histological damage occurring with increasing severity and duration of ischemia (27,28,29). The extent of regional abnormalities on perfusion maps is greatest for MTT, compared with CBF and CBV. Although MTT may be a more sensitive indicator of disturbed circulation, it does not necessarily imply the presence of ischemia, with changes in CBF and CBV being more specific for distinguishing infarct "core" from "ischemic penumbra" (30). Some authors recommend to initially review MTT maps for the presence of any flow abnormalities in the setting of acute stroke, and to use the CBF and CBV maps to try to distinguish ischemia from infarction (31). In summary, like DWI, CT-CBV (and CTA-SI) tend to define a lower limit to final infarct size, whereas MTT tends to define an upper limit to final infarct size.
PREDICTING OUTCOME: In addition to facilitating more accurate detection of early stroke, CTP may also be used to predict both final infarct size and patient outcome (23). Patients with acute MCA stem occlusions that had whole-brain perfused-blood-volume lesions >100 mL (approximately equal to 1/3 the MCA vascular territory) on admission had poor clinical outcomes, regardless of recanalization status.
The degree of early CBF reduction in acute stroke may also help predict hemorrhagic risk. Severely ischemic regions undergoing early reperfusion are at the greatest risk for hemorrhagic transformation (32,33). Severe hypoattenuation on whole-brain perfused-blood-volume images may also identify ischemic regions likely to bleed following intra-arterial thrombolysis (34).
|Using CT-CBV/CBF Mismatch to Triage Acute Stroke Patients to Thrombolytic Therapy
No CBV/CBF Mismatch (Fig. 8)
|Acute Stroke Imaging Pearls: Take-home Points
|Pearls: CTA/CTP in Acute Stroke Patients
|Pitfalls: CTP Data Acquisition and Processing
Patients with severe, long-standing carotid artery stenosis may be unable to normally augment cerebral blood flow under conditions of hypoxic stress. Under normal circumstances, hemodynamic stress results in arteriolar vasodilatation due to cerebral autoregulatory mechanisms responding to decreased perfusion pressure. Patients with maximum vasodilatation at baseline may be at increased risk of stroke, and therefore may benefit from interventions that will increase total cerebral flow, such as carotid endarterectomy (35).
CTP, in combination with acetazolamide or breath-hold challenge, can be used to assess cerebral perfusion and vascular reactivity. Acetazolamide is an intravenously administered medication that acts as a cerebral vasodilator agent (36). Normally, acetazolamide increases CBF by as much as 70-80% (37,38) (Fig. 12). However, little or no response in CBF suggests that the affected hemisphere of the brain may be at risk for stroke with any additional compromise of blood flow (39). Similarly, breath holding should result in increased CO2 and hence augmented CBF in normals, but not in those with already compromised cerebrovascular reserve (CVR). Acetazolamide challenge CTP can function as a "stress test" for the brain in evaluating cerebrovascular reserve capacity, and hence help estimate the potential risk of stroke in patients with hemodynamically significant carotid artery stenosis.
The term "cerebral vasospasm" is commonly used to refer to both the clinical picture of delayed onset of neurological deficits associated with aneurysmal SAH, and the narrowing of the cerebral arteries documented by angiography. However, this definition should be further divided into "angiographic" and "clinical" vasospasm. "Angiographic" vasospasm is the reduction of vessel size that can be detected on angiographic exams, occurring in approximately 50% of patients following aneurysmal subarachnoid hemorrhage. However, "clinical" vasospasm is the syndrome of confusion and decreased level of consciousness associated with reduced blood flow to the brain parenchyma, occurring in approximately 30% of patients. Roughly half of "clinical" spasm cases result in stroke (40). Vasospasm continues to be a significant adverse prognostic factor for outcome after subarachnoid hemorrhage, accounting for 34% of the combined morbidity and mortality (41,42).
CTA has the potential to detect "angiographic" vasospasm and CTP may aid in the diagnosis of "clinical" vasospasm. CTA has been proven to be highly accurate in the detection of proximal circle-of-Willis spasm and only slightly less accurate for the detection of more distal vessel spasm (43). MR, CT and SPECT flow studies have revealed a correlation between "angiographic" spasm, reduction in CBF, and "clinical" symptomatic spasm (44,45). Perfusion CT has been used to monitor cerebral perfusion after SAH (45) (Fig. 13). Patients with delayed infarct after SAH, presumably due to vasospasm, had lower mean CBF values compared to patients with early or no infarcts (45). Mean CBF and CBV were also significantly lower in patients with moderate - severe vasospasm. The ability to assess CBV and MTT may help in understanding the impairment of autoregulation that is believed to occur in patients after SAH (45). This information may assist in the triage of patients for urgent medical or endovascular treatment.
The degree of angiogenesis is known to be a critical factor in the growth of brain tumors (46,47). Increased angiogenic activity and neovascularization result in increased blood volume and hyperpermeability related to immature vessels (48,49,50). The technique for perfusion CT has been modified in commercially available software to account for extravasation of contrast from the intravascular space across the impaired blood-brain barrier, and for measurement of microvascular permeability-surface-area product (PS) (51). In animal models, CBF, CBV and PS have all been elevated in high-grade tumor and peritumoral areas, compared to normal tissue (51). However, results in humans have shown more variable elevations in the CBF and CBV of high-grade tumor, with more conspicuous increases in PS (52,53). The peritumoral regions may be especially accurate targets for assessment of perfusion parameters in humans. Maximal tumor CBV has been associated with high mitotic activity and vascularity (54). Other studies have also confirmed a strong correlation between tumor grade and degree of CBV elevation; typically stronger than that between tumor grade and contrast enhancement (55) (Fig. 14). Early detection of recurrent tumor has also been reported using CBV imaging, compared with conventional MR imaging, SPECT and clinical assessment (56,57). This information may provide a means to non-invasively grade tumor malignancy, guide biopsies to the most malignant portion of the tumor, assess response to treatment, and distinguish residual/recurrent tumor from post-treatment changes (52,53).
|Pearls: Advantages of CTP in Evaluation of Brain Tumors
Other Potential Clinical Applications of CTP:
Copyright © 2005. All rights reserved.
Use of this material without written consent of the author is prohibited.
Return to Neurographics Main Page