Visual Grading System for Vasospasm Based on Perfusion CT Imaging: Comparisons with Conventional Angiography and Quantitative Perfusion CT

Department of Radiology, University of California, San Francisco, CA 94143, USA.
Cerebrovascular Diseases (Impact Factor: 3.75). 06/2008; 26(2):163-70. DOI: 10.1159/000139664
Source: PubMed


The purpose of this study was to compare simple visual grading of perfusion CT (PCT) maps to a more quantitative, threshold-based interpretation of PCT parameters in the characterization of presence and severity of vasospasm.
Thirty-three patients with acute subarachnoid hemorrhage were enrolled in a prospective study and underwent a total of 40 paired PCT and digital subtraction angiography (DSA) examinations. A neuroradiologist and a neurologist reviewed the PCT mean transit time (MTT), cerebral blood flow (CBF), and cerebral blood volume maps independently; they evaluated five anatomical regions (frontal, temporal, parietal, occipital/thalami, and basal ganglia/insula) and graded them for abnormality (0 if normal, 1 if abnormal in <50% of the region, and 2 if abnormal in >or=50% of the region). A third neuroradiologist blinded to the PCT results reviewed the DSA examinations and assessed 19 segments for the presence or absence of vasospasm. Correlation between PCT and DSA scores was assessed, as well as the sensitivity and specificity of PCT compared to DSA used as a gold standard.
MTT (R(2) = 0.939) and CBF (R(2) = 0.907) scores correlated best with DSA scores (p < 0.001). MTT scoring had a sensitivity of 92% and a specificity of 86% compared to DSA; CBF scoring had a sensitivity of 75% and a specificity of 95%. The interobserver agreement between neuroradiologist and neurologist was found to have kappa = 0.789 for MTT and 0.658 for CBF.
We propose a user-friendly visual grading system for PCT maps in patients with suspected vasospasm. This visual approach compares favorably to the results of DSA. Sensitive MTT maps should be used for screening, and specific CBF maps for confirmation of vasospasm.

1 Follower
11 Reads
  • Source
    • "The acquisition of PCT maps is, as well, dependent on the examiner. However, PCT maps are color-coded and detection of asymmetries or bilateral defects compared to other vessel territory is rather reliable and of acceptable interobserver variability [7] Wintermark et al. proposed a user-friendly visual grading system which was used in this study [8]. The authors found that the Mean Transit Time (MTT), a parameter similar to the Time to Peak obtained in our study had the highest accuracy to predict angiographic vasospasm [8]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background. If detected in time, delayed cerebral vasospasm after aneurysmal subarachnoid hemorrhage (SAH) may be treated by balloon angioplasty or chemical vasospasmolysis in order to enhance cerebral blood flow (CBF) and protect the brain from ischemic damage. This study was conceived to compare the diagnostic accuracy of detailed neurological examination, Transcranial Doppler Sonography (TCD), and Perfusion-CT (PCT) to detect angiographic vasospasm. Methods. The sensitivity, specificity, positive and negative predictive values of delayed ischemic neurological deterioration (DIND), pathological findings on PCT-maps, and accelerations of the mean flow velocity (MVF) were calculated. Results. The accuracy of DIND to predict angiographic vasospasm was 0.88. An acceleration of MFV in TCD (>140 cm/s) had an accuracy of 0.64, positive PCT-findings of 0.69 with a higher sensitivity, and negative predictive value than TCD. Interpretation. Neurological assessment at close intervals is the most sensitive and specific parameter for cerebral vasospasm. PCT has a higher accuracy, sensitivity and negative predictive value than TCD. If detailed neurological evaluation is possible, it should be the leading parameter in the management and treatment decisions. If patients are not amenable to detailed neurological examination, PCT at regular intervals is a helpful tool to diagnose secondary vasospasm after aneurysmal SAH.1. IntroductionAmong other variables, cerebral infarction and symptomatic vasospasm are the most important postoperative risk factors for poor outcome after aneurysmal SAH [1] Both are consequences of a decreasing and insufficient brain perfusion that eventually causes a loss of neurological function and, finally, structural damage of brain tissue [2]. A variety of measures is undertaken to enhance cerebral blood flow (CBF) in SAH-patients developing delayed cerebral vasospasm (DCV). These include hyperdynamic therapy, intra-arterial and intrathecal drug infusion, intra-aortic balloon counterpulsation, and new experimental methods [3]. Endovascular treatment of DCV—balloon angioplasty and chemical vasospasmolysis—has proven to effectively enhance CBF and has been increasingly used for treatment of DCV in the last years [4]. The application of these modalities or their combination may be very helpful if treatment is started in time. However, the routine use of repeated four-vessel angiography is not justifiable because of high-radiation exposure. Therefore, other monitoring tools have to indicate upcoming vasospasm after aneurysmal SAH. If DCV is missed or treatment started too late, cerebral infarction might develop in spite of maximum treatment. A number of monitoring methods are available. The use of continuous invasive brain monitoring like bedside microdialysis or intracerebral measurement of tissue oxygenation or regional cerebral blood flow (rCBF) is discussed with controversy. The ideal monitoring is noninvasive, can be repeated any time without extensive technical setup, has a high-diagnostic accuracy, and gives information about the development of DCV and the decrease of cerebral perfusion early enough to start an adequate therapy and prevent cerebral infarction. Neurological assessment and TCD are noninvasive procedures. PCT obtained at regular intervals is only little invasive regarding the exposure to radiation. This study was designed to evaluate and compare these diagnostic modalities for their prognostic value to detect angiographic vasospasm and perform endovascular treatment.2. Materials and Methods The study was approved by the local ethics committee. The data was collected in the framework of a clinical trial assessing the effectiveness of magnesium treatment in aneurysmal SAH [5]. The patients analyzed in the present study resemble the control group of this clinical trial. In order to prevent a possible contamination of results by a drug with vasodilatory and neuroprotective potential, patients receiving magnesium medication were not included in this analysis. Informed consent was obtained from the patient or a legal guardian.2.1. Inclusion- and Exclusion-CriteriaPatients were eligible for inclusion if they had suffered aneurysmal SAH no longer than 96 hours ago. Patients were not included if they were under 18 years of age, if treatment was discontinued at the time of hospital admission due to a poor clinical state on admission, if there was a history of preceding aneurysmal SAH, or if therapy was scheduled to be continued elsewhere after obliteration of the aneurysm. Further exclusion criteria were pregnancy, cancer, atrioventricular block, preexisting neuromuscular disease, and renal failure. If serum creatinine values were above 133 mol/L (1.5 mg/100 mL), values were controlled 8 hours later after fluid substitution with 1.5 liters of Ringer’s solution. If serum creatinine had declined below 133 mol/L, the patient was included in the study.All patients were admitted to the neurosurgical intensive care unit. The intervals from the presumable aneurysm rupture until hospital admission were documented as well as comorbidity and medication. The neurological examination was performed according to an examination protocol which included assessment and documentation of the Glasgow Coma Score (GCS), deficits of awareness (person, place, time, situation), deficits of cranial nerves, visual field deficits, motor and sensory deficits of the upper and lower extremities, coordination deficits, and speech abnormalities. 2.2. Assessment of Study ParametersNeurological examination: Detailed neurological examination was conducted 3 times a day (1 a.m., 9 a.m., and 5 p.m.) by the neurosurgeon on duty on the ICU following the above mentioned neurological examination protocol. Any secondary deterioration of the neurological state was considered abnormal. If secondary deterioration was detected and confirmed by the senior neurosurgeon on duty, blood counts and serum electrolytes were determined and a CT-scan was obtained. Delayed Ischemic Neurological Deficit (DIND) was defined as a secondary neurological deterioration after exclusion of rebleeding or intracerebral hemorrhage, hydrocephalus, seizure, or electrolyte disturbance. The examiners were not involved in the conduction and analysis of Transcranial Doppler Sonography and Perfusion-CT.Transcranial Doppler Sonography (TCD): The TCD examination was conducted daily after the neurological examination in the morning by a trained and experienced medical technician who was blinded to the result of the neurological examination. The mean flow velocities (MFV) in the vessel trunks of both middle, anterior, and posterior cerebral arteries (MCA, ACA, PCA) were determined via a temporal window, the MFV in the basilar artery was determined via the foramen magnum. Sonographic vasospasm was defined as a MFV >140 cm/s in the MCA, ACA and/or PCA or a MFV of >90 cm/s in the basilar artery.Perfusion CT: Native CT and PCT were obtained at day 3 or 4, day 6 or 7, and day 9 or 10 after admission or at any other point if it was thought to be of diagnostic relevance. If patients were hospitalized longer, further CT/PCT-scans were obtained at 3-day intervals. For the determination of irreversible brain infarction on native CT, diagnostic criteria were applied as previously described [6]. For the native CT, slice thickness was 5 mm for the posterior fossa and 8 mm for the cerebrum (Somatom Plus 4 Volume Zoom, Siemens, Erlangen, Germany). Slice positioning was above the orbital roof in the supraorbitomeatal direction. Perfusion CT was obtained as previously described [7]. In brief, two adjacent 10-mm slices were positioned at the level of the basal ganglia with the same angulation as for native CT. A bolus of 50 mL of nonionic contrast medium (Imeron 400, Bracco, Konstanz, Germany) was administered by a power injector into a central venous catheter at a flow rate of 4 mL/s followed by 30 mL of saline. Four seconds after beginning of the bolus, 40 images were collected at each slice level at a rate of two images per second (120 kV, 110 mAs, matrix ). For PCT analysis a commercially available software was used (Perfusion CT, Siemens). PCT color maps were qualitatively assessed using a visual grading scale [7, 8]. A positive visual assessment was noted for side-to-side asymmetries or clear bilateral defects suggesting a decrease in Cerebral Blood Flow (CBF), Cerebral Blood Volume (CBV), or an increase in Time To Peak (TTP) (Figure 1). CT and Perfusion-CT was analyzed by trained neuroradiologists (M.P., L.S.) and a trained neurosurgeon (C.S.) blinded to the patient data, neurological exam, and TCD values.
    Full-text · Article · Sep 2012
  • Source

    Preview · Article · Sep 2008 · Interventional Neuroradiology
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Vasospasm after aneurysmal subarachnoid hemorrhage (SAH) is thought to cause ischemia. To evaluate the contribution of vasospasm to delayed cerebral ischemia (DCI), we investigated the effect of vasospasm on cerebral perfusion and the relationship of vasospasm with DCI. We studied 37 consecutive SAH patients with CT angiography (CTA) and CT perfusion (CTP) on admission and within 14 days after admission or at time of clinical deterioration. CTP values (cerebral blood volume, cerebral blood flow (CBF) and mean transit time), degree of vasospasm on CTA, and occurrence of DCI were recorded. Vasospasm was categorized as follows: no spasm (0-25% decrease in vessel diameter), moderate spasm (25-50% decrease), and severe spasm (>50% decrease). The correspondence of the flow territory of the most spastic vessel with the least perfused region was evaluated, and differences in perfusion values and occurrence of DCI between degrees of vasospasm were calculated with 95% confidence intervals (95% CI). Fourteen patients had no vasospasm, 16 were moderate, and seven were severe. In 65% of patients with spasm, the flow territory of the most spastic vessel corresponded with the least perfused region. There was significant CBF (milliliters per 100 g per minute) difference (-21.3; 95% CI, -37 <--> -5.3) between flow territories of severe and no vasospasm. Four of seven patients with severe, six of 16 with moderate, and three of 14 patients with no vasospasm had DCI. Vasospasm decreases cerebral perfusion, but corresponds with the least perfused region in only two thirds of our patients. Furthermore, almost half of patients with severe vasospasm do not have DCI. Thus, although severe vasospasm can decrease perfusion, it may not result in DCI.
    Full-text · Article · Aug 2009 · Neuroradiology
Show more