Assessment of Tissue Viability in Hyperacute Infarction with Using the Diffusion- and Perfusion-weighted Images

PURPOSE: The presence of a perfusion-diffusion mismatch is a useful indicator for predicting the progression of acute cerebral infarction. However, not all the area of the perfusion-diffusion mismatch progresses to infarction and a large proportion survives with hypoperfusion. The purpose of this study was to assess 1) whether tissue viability can be predicted using quantitative perfusion values and 2) whether there is correlation between the perfusion value and the time that elapsed after the onset of symptoms. MATERIALS AND METHODS: Twenty-two patients with acute infarction in the middle cerebral artery territory within 12 hours after symptom onset were included in this study. We excluded those patients in whom thrombolysis was attempted or the lesion volume was less than 5 mL. Patients without perfusion-diffusion mismatch on the mean transit time (MTT) map were also excluded. We categorized the ischemic lesions into 3 areas: 1) the initial infarction, 2) the area that progressed to infarction, and 3) the hypoperfused but surviving area, based on the initial and follow up diffusion-weighted images and initial mean transit time (MTT) map. We obtained the relative cerebral blood volume (rCBV), the cerebral blood flow (rCBF) and the MTT in each area by comparing to the contralateral normal area. Statistical analysis was performed using one-way ANOVA to test whether there was a difference in perfusion values between each area. The threshold value was calculated between areas 2 and 3 using the receiver operating characteristics curve. We analyzed the correlation between the perfusion values of each area and the time that elapsed after the inset of symptoms. RESULTS: The perfusion values among each region were significantly different on the rCBV, rCBF and MTT maps. Between regions 2 and 3, the rCBV and rCBF maps showed a significant difference (Bonferroni post hoc analysis), but in case of rCBV, the mean perfusion values in each region approached to the normal level and it was difficult to differentiate between the two regions on the rCBV map. The rCBF in the regions 1, 2 and 3 was 0.40, 0.64, and 0.84, respectively. The difference of the threshold values of the rCBF between regions 2 and 3 was 0.75. There was no significant correlation between the time that elapsed after symptom onset and the perfusion values of each region on the rCBV, rCBF and MTT map. CONCLUSION: The perfusion values between the area of the initial infarction, the area that progressed to infarction and the hypoperfused but surviving area showed significant differences. The rCBF was the most useful parameter in differentiating between areas that progressed to infarction and the surviving areas. Quantitative measurement of the perfusion values may have a role in selecting the candidates for thrombolysis after they have suffered hyperacute stroke.

Usefulness of Dynamic MR Imaging for the Evaluation of the Solitary Pulmonary Nodules Smaller than 15 mm: Differentiation between Benign and Malignant Nodules

PURPOSE: We wanted to evaluate the usefulness of dynamic magnetic resonance (MR) imaging for differentiating between benign and malignant solitary pulmonary nodules (SPNs). MATERIALS AND METHODS: Sixteen patients who had an undetermined SPN (<15 mm) upon chest computed tomography (8 males and 8 females; mean age: 55 years; age range: 40-76 years) underwent dynamic MR imaging. After the bolus injection of contrast material, the arterial (20-35 seconds), portal (45-60 seconds) and equilibrium (3-5 minutes) phase T1-weighted axial images were obtained with using a volumetric interpolated breath-hold examination. For discriminating the benign from malignant SPNs, the maximum relative enhancement ratio (MER) and the slope of the enhancement (SLE) were calculated and then they were statistically compared. With varying the threshold of the two indexes, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy were calculated. RESULTS: The mean MER of the malignant SPN group was significantly higher than that of the benign SPN group (malignant; 0.56+/-0.17, benign; 0.43+/-0.17). With 0.33 as the threshold of MER for distinguishing the malignant SPN group from the benign SPN group, the sensitivity, specificity, positive predictive value and negative predictive value were 100%, 70%, 50%, and 100%, respectively. The mean SLE for the benign SPN group was higher than that for the malignant SPN group (malignant; m= 0.008+/-0.006/sec, benign; m=0.013+/-0.008/sec). With 0.025 as the threshold of the SLE, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 100%, 60%, 62.5%, 100% and 69.2%, respectively. CONCLUSION: Dynamic MRI was useful for differentiating between benign and malignant SPNs. Moreover, MER and SLE might be good indexes for distinguishing benign SPNs from malignant SPNs.

Assessment of Tissue Viability Using Diffusion- and Perfusion-Weighted MRI in Hyperacute Stroke

OBJECTIVE: The aim of this study was to investigate the relationship between the diffusion and perfusion parameters in hyperacute infarction, and we wanted to determine the viability threshold for the ischemic penumbra using diffusion- and perfusion-weighted imaging (DWI and PWI, respectively). MATERIALS AND METHODS: Both DWI and PWI were performed within six hours from the onset of symptoms for 12 patients who had suffered from acute stroke. Three regions of interest (ROIs) were identified: ROI 1 was the initial lesion on DWI; ROI 2 was the DWI/PWI mismatch area (the penumbra) that progressed onward to the infarct; and ROI 3 was the mismatch area that recovered to normal on the follow-up scans. The ratios of apparent diffusion coefficient (ADC), the relative cerebral blood volume (rCBV), and the time to peak (TTP) were calculated as the lesions' ROIs divided by the contralateral mirror ROIs, and these values were then correlated with each other. The viability threshold was determined by using the receiver operating characteristic (ROC) curves. RESULTS: For all three ROIs, the ADC ratios had significant linear correlation with the TTP ratios (p < 0.001), but not with the rCBV ratios (p = 0.280). There was no significant difference for the ADC and rCBV ratios within the ROIs. The mean TTP ratio/TTP delay between the penumbras' two ROIs showed a significant statistical difference (p < 0.001). The cutoff value between ROI 2 and ROI 3, as the viability threshold, was a TTP ratio of 1.29 (with a sensitivity and specificity of 86% and 73%, respectively) and a TTP delay of 7.8 sec (with a sensitivity and specificity of 84% and 72%, respectively). CONCLUSION: Determining the viability thresholds for the TTP ratio/delay on the PWI may be helpful for selecting those patients who would benefit from the various therapeutic interventions that can be used during the acute phase of ischemic stroke.

Diagnosis and Prediction of Clinical Outcomes in Patients with Acute Lacunar Infarction: Usefulness of Perfusion MR Imaging

PURPOSE: To correlate the findings of perfusion-weighted imaging (PWI) with clinical outcomes in patients with acute lacunar infarction. MATERIALS AND METHODS: Eleven patients (7 males and 4 females) with acute lacunar infarction who were examined within 50 (mean, 29) hours of the onset of symptoms underwent conventional MRI, diffusion-weighted imaging (DWI) and PWI. Gadolinium (0.2 mmol/kg) was injected at a rate of 2 ml/sec, and PWI was performed using a gradient-echo EPI pulse sequence and the following parameters: TR/TE, 2000/60; flip angle, 90 degree; matrix size, 128X128. Relative cerebral blood volume (rCBV) maps were derived from gadolinium bolus perfusion-weighted images where rCBV ratios between infarcted areas were detected by DWI, and contralateral control areas were obtained. In each case, the resulting rCBV ratio at a lesion site was compared with the clinical outcome determined on the basis of the difference between National Institute Health Stroke Scale (NIHSS) scores at admission and discharge. RESULTS: With the aid of the time-intensity curve obtained at PWI, the rCBV maps revealed a hypoperfused area in 10 of 11 patients, and there was positive correlation (r=0.81) with clinical outcome. CONCLUSION: Although PWI has a lower detection rate than DWI, it may be a useful modality for helping determine prognosis in cases of acute lacunar infarction.

Perfusion MR Imaging: Clinical Utility for the Differential Diagnosis of Various Brain Tumors

OBJECTIVE: To determine the utility of perfusion MR imaging in the differential diagnosis of brain tumors. MATERIALS AND METHODS: Fifty-seven patients with pathologically proven brain tumors (21 high-grade gliomas, 8 low-grade gliomas, 8 lymphomas, 6 hemangioblastomas, 7 metastases, and 7 various other tumors) were included in this study. Relative cerebral blood volume (rCBV) and time-to-peak (TTP) ratios were quantitatively analyzed and the rCBV grade of each tumor was also visually assessed on an rCBV map. RESULTS: The highest rCBV ratios were seen in hemangioblastomas, followed by high-grade gliomas, metastases, low-grade gliomas, and lymphomas. There was no significant difference in TTP ratios between each tumor group (p<0.05). At visual assessment, rCBV was high in 17 (81%) of 21 high-grade gliomas and in 4 (50%) of 8 low-grade gliomas. Hemangioblastomas showed the highest rCBV and lymphomas the lowest. CONCLUSION: Perfusion MR imaging may be helpful in the differentiation of thevarious solid tumors found in the brain, and in assessing the grade of the various glial tumors occurring there.

Assessment of Neoplastic Angiogenesis Using Perfusion-Weighted MR Imaging: Experimental Study in VX2 Carcinoma in Rabbits

PURPOSE: To evaluate the perfusion-weighted MR imaging findings of hepatic VX2 carcinoma in rabbits and to explain the perfusion characteristics of this condition by correlation with the histopathological findings. MATERIALS AND METHODS: Twelve New Zealand white rabbits, each weighing between 2.5 and 3.5 (mean) 3.1kg, were used in this study. Perfusion MRI using single-shot gradient-echo EPI was performed 7 -21 days after the injection of tumor cell suspension into the hepatic parenchyma by laparotomy. On the basis of the calculated enhancement ratio, the time-intensity perfusion curves for VX2 tumor and normal liver parenchyma were created, and the shapes of these curves, the time to maximum SI decrease, and the maximum enhancement ratio in each, were evaluated. To assess microvessel density in each VX2 carcinoma and in normal liver parenchyma, immunohistochenical study using factor VIII-related antigen was performed. RESULTS: A total of 15 tumors 1 -3 cm in diameter were revealed by MR imaging. The perfusion curve showed rapid decrement and immediate recovery of the signal intensity of VX2 carcinoma during the early arterial perfusion phase and slower decrement and gradual recovery of that of normal liver parenchyma during the late portal perfusion phase. In all cases, these were constant findings. The time to maximum signal intensity decrease was 13 -16 (mean, 15) secs in VX2 carcinoma and 28 -36 (mean, 32) secs in normal liver parenchyma (p<0.01). The maximum enhancement ratio of VX2 carcinoma and normal liver ranged from 27 to 84% (mean 47%) and from 36 to 82% (mean, 56%), respectively. Immunohistochemical study showed that the MVD of VX2 carcinoma was significantly greater than that of normal liver parenchyma(75 vs 17, p<0.01). CONCLUSION: Perfusion-weighted MR imaging appears to be a useful tool for the diagnosis of neoplastic angio-genesis, and thus holds promise differentiating liver tumors.

Perfusion MR Imaging in Patients with Acute Cerebral Infarction:Comparison with T2-Weighted and Diffusion-Weighted MR Imaging

PURPOSE: To evaluate the clinical usefulness of perfusion MR imaging by comparing with T2-weighted and dif-fusion weighted MR imaging in patients with acute cerebral ischemic infarction. MATERIALS AND METHODS: Conventional, diffusion weighted, and perfusion MR images were obtained within one week of clinical onset in 14 cases of acute ischemic infarction. For perfusion MRI, the gradient-echo EPI technique after IV bolus injection of 15 cc of contrast media was used. Four kinds of perfusion MR images (rCBV, rCBF, mean transit time[MTT], time to peak concentration [TTP]) were generated by home-made soft-ware from the raw data. T2-weighted, diffusion-weighted, and perfusion images of each patient were retro-spectively analyzed, with attention to the number, signal intensity, and size of lesions. RESULTS: T2-weighted and diffusion-weighted images demonstrated 21 acute ischemic lesions in 14 patients. Six lesions had a long diameter of more than 3 cm, while the other 15 were smaller than 3 cm. On T2-weighted images, 17 lesions showed high signal intensity and four showed subtle high signal intensity. On diffusion-weighted images, all lesions showed bright high signal intensity. The six lesions larger than 3 cm were all delin-eated by all four kinds of perfusion MR imaging, but among the 15 smaller than 3 cm, only four (26.7%), five (33.3%) and six (40%) were delineated on rCBV and rCBF maps, the MTT map, and the TTP map, respective-ly. As compared with T2-weighted and diffusion-weighted imaging, the rCBV and rCBF maps showed that four lesions were smaller and six were the same size. On the MTT map, three lesions were seen to be larger, four were smaller, and the other four were the same size as they appeared on diffusion-weighted images, while on the In TTP map, seven were larger and five were smaller than they appeared on these images. CONCLUSION: In all cases, diffusion-weighted images most clearly delineated acute ischemic lesions, regardless of lesion size. Many such lesions smaller than 3 cm were not apparent on perfusion MR images. Among the four kinds of perfusion MR imaging, TTP and MTT maps may be clinically useful for evaluation of the penum-bral zone in cases of acute cerebral ischemic infarct.

The Usefulness of Perfusion MR Imaging in Patients with Brain Tumors

PURPOSE: To determine the usefulness of perfusion weighted MR imaging in the assessment of relative cerebral blood volume(rCBV) in brain tumors. MATERIALS AND METHODS: Twenty-three patients with primary or metastatic brain tumors [nine gliomas (6 high grade and 3 low), six metastatic tumors, five meningiomas (4 benign and 1 atypical), two neurilemmomas and one hemangioblastoma] underwent perfusion-weighted and conventional MR imaging. A total of 240 perfusion MR images were obtained from four axial slices after rapid injection of contrast media using a gradient echo planar imaging pulse sequence, and this was followed by postprocessing of these images to give CBV maps. In order to calculate the rCBV of tumor to normal white matter, ROIs were defined on the CBV map of a tumor and its contralateral normal white matter. RESULTS: The rCBV ratio of tumors to contralateral normal side was as follows: high-grade glioma, 0.40-5.64(mean +/-SD = 2.91 0.95); low grade astrocytoma, 0.77 -1.66 (mean +/-SD = 1.15 +/-0.28); benign menin-gioma,2.06 -4.90 (mean +/-SD = 3.59 +/-0.84); atypical meningioma, 0.46 -1.18 (mean +/-SD = 0.72 +/-0.25); neurilemmoma, 1.45 -3.85 (mean +/-SD = 2.56 +/-0.92); and hemangioblastoma, 6.16 -8.35 (mean +/-SD = 7.02 +/-1.12). High grade gliomas were more hypervascular than low grade astrocytomas, and showed a variable range of relative cerebral blood volume. In metaststic cancer, CBV maps showed a relatively high and variable blood volume. Benign meningiomas exhibited high relative cerebral blood volume, while in the atypical meningioma with cystic degeneration, this volume was low. In neurilemmomas, a variable range of relative cerebral blood volume, was noted, while in the mural nodule of the hemangioblastoma, this volume was the highest. CONCLUSION: Perfusion-weighted MRI indicated the rCBV of various brain tumor lesions, and this suggests that the modality can provide a very useful means of assessing brain tumor vascularity.