Search

The objective of this study was to investigate the distribution of 11C-methionine (MET) and F-18 fluorodeoxyglucose (FDG) uptake in positron emission tomography (PET) imaging and the hyperintense area in T2 weighted imaging (T2WI) in glioma with no or poor gadolinium enhancement in magnetic resonance imaging (GdMRI). Cases were also analyzed pathologically. We prospectively investigated 16 patients with non- or minimally enhancing (< 10% volume) glioma. All patients underwent MET-PET and FDG-PET scans preoperatively. After delineating the tumor based on MET uptake, integrated 3D images from FDG-PET and MRI (GdMRI, T2WI or FLAIR) were generated and the final resection plane was planned. This resection plane was determined intraoperatively using the navigation-guided fencepost method. The delineation obtained by MET-PET imaging was larger than that with GdMRI in all cases with an enhanced effect. In contrast, the T2WI-abnormal signal area (T2WI+) tended to be larger than the MET uptake area (MET+). Tumor resection was > 95% in the non-eloquent area in 4/5 cases (80%), whereas 10 of 11 cases (90.9%) had partial resection in the eloquent area. In a case including the language area, 92% resection was achieved based on the MET-uptake area, in contrast to T2WI-based partial resection (65%), because the T2WI+/MET- area defined the language area. Pathological findings showed that the T2WI+/MET+ area is glioma, whereas 6 of 9 T2WI+/MET- lesions included normal tissues. Tissue from T2W1+/MET+/FDG+/GdMRI+ lesions gave an accurate diagnosis of grade in six cases. Non- or minimally enhancing gliomas were classified as having a MET uptake area that totally or partially overlapped with the T2WI hyperintense area. Resection planning with or without a metabolically active area in non- or minimally enhancing gliomas may be useful for accurate diagnosis, malignancy grading, and particularly for eloquent area although further study is needed to analyze the T2WI+/MET- area.

Purpose: To evaluate the ability of noncontrast electrocardiogram (ECG)-gated fast-spin-echo (FSE) perfusion MR images for defining regional lung perfusion impairment, as compared with technetium (Tc)-99m macroaggregated albumin (MAA) single-photon emission computed tomography (SPECT) images., Materials and Methods: After acquisition of ECG-gated multiphase FSE MR images during cardiac cycles at selected lung levels in nine healthy volunteers, 11 patients with pulmonary artery-occlusive diseases, and 15 patients with chronic obstructive pulmonary diseases (COPD), the subtracted perfusion-weighted (PW) MR images were obtained from the two-phase images of the minimum lung signal intensity (SI) during systole and the maximum SI during diastole, and were compared with SPECT images., Results: ECG-gated PW images showed uniform but posture-dependent perfusion gradient in normal lungs and visualized the various sizes of perfusion defects in affected lungs. These defect sites were nearly consistent with those on SPECT images, with a significant correlation for the affected-to-unaffected perfusion contrast (r = 0.753; P < 0.0001). These MR images revealed that the pulmonary arterial blood flow in the affected areas of COPD was relatively preserved as compared with pulmonary artery-occlusive diseases, and also showed significant decrease in blood flow, even in the areas with homogeneous perfusion on SPECT images in patients with focal pulmonary emphysema., Conclusion: This noninvasive MR technique allows qualitative and quantitative assessment of lung perfusion, and may better characterize regional perfusion impairment in pulmonary artery-occlusive diseases and COPD., (Copyright 2004 Wiley-Liss, Inc.)

Purpose: To define regional function impairment in airway obstruction (AO) and pulmonary embolic (PE) dogs with a combination study of noncontrast electrocardiogram (ECG)-gated perfusion and gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA) aerosol magnetic resonance (MR) images., Methods: After acquisition of multiphase fast-spin-echo (FSE) MR images during cardiac cycles in 14 AO dogs and 19 PE dogs, ECG-gated perfusion-weighted (PW) images were obtained by subtraction between two-phase images of the minimum lung signal intensity (SI) during systole and maximum SI during diastole. Each dog subsequently inhaled Gd-DTPA aerosol for 20 minutes, and subtracted Gd-DTPA aerosol images were obtained from precontrast and maximally enhanced images. ECG-gated PW images were compared with intravenous Gd-DTPA-enhanced pulmonary arterial perfusion phase (PAPP) images., Results: ECG-gated PW images were consistent with Gd-DTPA-enhanced PAPP images in all dogs, with significant correlations in the affected-to-unaffected lung perfusion ratios (P < 0.005). Gd-DTPA aerosol images showed sufficient and uniform enhancement in the unaffected lungs. In all the AO areas, these combined images showed the matched perfusion and aerosol deposition defects. These images showed perfusion defects without aerosol deposition defects in the relatively small embolized areas, but showed the matched defects in the widely embolized areas probably due to hypoxic bronchial constriction., Conclusion: The combination MR studies may be acceptable for noninvasively defining regionally impaired lung function in AO and PE., (Copyright 2004 Wiley-Liss, Inc.)

Purpose: To evaluate a new approach of magnetic resonance (MR) lymphography with intraalveolar injection of a conventional extracellular contrast agent (gadopentetate dimeglumine) for imaging lymphatic basin draining from specific portions of the lung., Methods: Three-dimensional T1-weighted spoiled gradient-recalled echo MR sequence images were acquired serially before and for 40 minutes after intraalveolar injection of gadopentetate dimeglumine in a total of 14 anesthetized beagle dogs. Six of these dogs received 1 mL undiluted and low-concentration (75%) contrast agent into the same portion of the right caudal lobe during a 7-day interval. In all dogs, including these 6 dogs, MR lymphography was repeated with injection of the low-concentration contrast agent into different lung regions at 7-day intervals to evaluate the differences of the visualized draining lymphatic station. Lymphatic enhancement was quantified by percent increases of signal intensity against precontrast. Postmortem examination of the lymphatic anatomy was performed in 7 of these animals., Results: In all dogs, the lymphatic station draining from the injection sites was visualized within 5 minutes after contrast injection. The maximum percent increase of signal intensity of the same middle tracheobronchial lymph nodes was significantly greater with a low-concentration (75%) contrast agent than with an undiluted one in the same 6 dogs (n = 6, 247.6 +/- 30.5% vs. 204.2 +/- 33.8%; P < 0.01). Different lymphatic stations draining from the different injection sites were visualized in all dogs. In a total of 12 MR studies that showed extended nodal enhancement after injection of the low-concentration contrast agent, the enhancement peak of the most proximal nodes (n = 12) from the injection sites appeared earlier than that of their distant nodes (n = 12), with a maximum percent increase of signal intensity of 249.8 +/- 42.4%. The visualized lymph nodes were found in the appropriate locations postmortem, with significant correlation for nodal sizes (r = 0.965; P < 0.0001)., Conclusion: MR lymphography with low-concentration gadopentetate dimeglumine can quickly and sufficiently visualize the drainage lymphatic station from specific lung portions, and may have the potential of sentinel node mapping in lung cancer.

Purpose: The capability of the interstitial magnetic resonance lymphography (MR-LG) using a widely available extracellular contrast agent, gadopentetate dimeglumine, for the visualization of normal and interrupted lymphatic drainage was tested in dog legs., Materials and Methods: With a 7-day interval, 1 mL and 2 mL undiluted gadopentetate dimeglumine were injected intradermally into the dorsal foot of both hind legs of 12 dogs. T1-weighted 3-dimensional (3D) spoiled gradient-echo (repetition time/echo time = 6.7/1.6 ms) and maximum intensity projection (MIP) images covering the legs and pelvic region were acquired at 1.5 T for 10 minutes after 30 seconds gentle massage at the injection sites. These dogs also underwent the MR-LG with 2 mL of this contrast agent 7 days after surgical ligation of the popliteal lymphatic vessels., Results: In the normal dogs, the interstitial MR-LG quickly and consistently visualized the direct connection of the lymphatic vessels and lymph nodes draining from the injection sites up to the inguinal region, without any late adverse effects. The enhancement effect was significantly greater with a 2-mL dose than with a 1-mL dose. After lymphatic obstruction, the MR-LG clearly revealed the anatomic compromises of the lymphatics, such as poor enhancement of the affected lymphatic vessels and lymph nodes, collateral lymphatic vessels, abnormal soft-tissue enhancement, and delayed lymphatic migration of the contrast agent. The topographic 3D images provided a comprehensive anatomy of these normal and compromised lymphatic drainage., Conclusion: Although gadopentetate dimeglumine is not lymphotropic, the interstitial MR-LG using this T1-contrast agent appears to have potential for quick and sufficient mapping of the lymphatic drainage from the injection sites and for the characterization of interrupted lymph flow.

Purpose: To test the capability of an indirect MR lymphography (MRLG) using a conventional extracellular gadolinium (Gd)-based contrast agent for localizing breast sentinel lymph node (SLN)., Methods: A total of 1 and 0.5 ml of undiluted gadopentetate dimeglumine were injected subcutaneously into the two periareolar areas overlying each caudal mammary gland in 10 female dogs. Contiguous 2-mm-thick fast gradient echo MR images were acquired through the upper breast and axilla before and for 60 min after gentle massage at the injection sites, yielding 3D displays. The localized SLN was resected from the living animals, followed by post mortem examinations. MRLG (1 ml contrast agent injected) was also attempted in 5 female volunteers., Results: Even with 0.5-ml contrast agent, the MRLG clearly visualized the connection of SLN and lymphatic vessels directly draining from the injection sites, and the remaining distant nodes, with the maximal enhancement peaks within 10 min after massage. The 3D images provided comprehensive anatomy of these lymphatic pathways in each axilla. Of the 20 SLN and 128 distant nodes visualized on the MRLG, all the SLN (100%) and 107 (83.5%) distant nodes could be resected pre- and post mortem, in good correlation with the locations and sizes on the MR images. MRLG also effectively localized SLN in the volunteers, without significant adverse effects., Conclusion: An indirect MRLG using small volumes of conventional Gd-based contrast agent may have potential for accurate identification and surgical biopsy of breast SLN.

Rationale and Objectives: The potential of a noncontrast, electrocardiography (ECG)-gated fast-spin-echo (FSE) MR imaging (MRI) to monitor dynamically altered regional lung perfusion was assessed in acute and temporal pulmonary embolic and airway obstruction dog models., Materials and Methods: After acquisition of ECG-gated multiphase FSE MR images during one cardiac cycle, the two phase images of the minimal lung signal intensity (SI) during systole and the maximal SI during diastole were acquired in the lower lung levels in six normal dogs, in 13 dogs before and for 35 minutes after temporal microvascular embolization in regional lungs with gradually degradable starch microspheres of spherex, and in 12 dogs before and for 45 minutes after bronchial occlusion with a balloon catheter. In three of the 13 embolic models, the opposite lung areas, however, were permanently embolized with enbucrilate. Subtraction between the diastolic and systolic images yielded a perfusion-weighted image. The results were compared with a gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA)-enhanced dynamic perfusion MRI, which was subsequently performed after the ECG-gated MRI in each animal., Results: The multiphase FSE images provided cardiac-dependent pulsatile lung SI changes, and the subtracted perfusion-weighted images provided a uniform perfusion map in normal lungs. In all the embolic models, the subtracted perfusion-weighted images showed gradual disappearance of the spherex-induced perfusion deficits, while the enbucrilate-induced perfusion deficits persistently remained in the three animals. In all airway obstruction models, these images showed gradually decreased perfusion in the hypoventilated areas. These results were consistent with the matched Gd-DTPA-enhanced pulmonary arterial perfusion phase images in each animal., Conclusion: This noncontrast perfusion MRI may have excellent potential for continuously monitoring dynamically changed regional lung perfusion within a short time on its high spatial resolution cross-sectional images.

Rationale and Objectives: A contrast-enhanced dynamic magnetic resonance (MR) study was performed experimentally and clinically to describe perfusion characteristics of radiation-injured lung according to pathologic phases., Methods: The MR study was performed before and at 0.5, 1, 2, 3, 4, and 7 months after 40 Gy-dose irradiation to the right hemithorax in 8 dogs, and clinically in 12 lung lesions of 9 patients with acute or fibrotic radiation pneumonitis. Altered Gd-DTPA kinetics in the affected lungs was assessed by time-signal intensity curves. MR findings were correlated with lung histology and CT images., Results: Within 1 month after irradiation, the irradiated animal lungs showed focal and persistent contrast enhancement relative to nonirradiated lungs. This abnormality was pronounced during the next 2 months. After 4 months, irradiated lungs conversely showed lower enhancement during the Gd-DTPA first-pass but were followed by persistently greater enhancement during Gd-DTPA redistribution phase. Similar differences in enhancement abnormalities between acute and fibrotic radiation pneumonitis were clinically observed., Conclusion: These findings indicate that Gd-DTPA kinetics can be altered according to the histopathologic change in early/acute radiation pneumonitis and radiation fibrosis and that the contrast-enhanced perfusion MRI may help differentiate the phases of radiation pneumonitis.

Rationale and Objectives: Gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA)-based aerosol ventilation and perfusion magnetic resonance (MR) images were used to define regional functional impairment in acute airway obstruction (AO) and pulmonary embolic (PE) dog models., Methods: The aerosol study was performed in 10 anesthetized normal dogs in a supine position during 20-minute spontaneous inhalation of an aerosol of 100- or 200-mmol-Gd/L Gd-DTPA solute produced by an ultrasonic nebulizer in an open-circuit delivery system, combined with a dynamic perfusion study after a 3-second intravenous bolus injection of a 0.1 mmol/kg dose of Gd-DTPA. These MR studies were also performed in the same 10 dogs approximately 30 minutes after obstructing the segmental (n = 6) or lobar (n = 4) bronchus with a balloon catheter, and in another six dogs after segmental (n = 6) and lobar (n = 4) pulmonary arterial embolization with enbucrilate. Regional lung enhancement was assessed on time-signal intensity (SI)-curves and ventilation- and perfusion-weighted images produced by a subtraction technique., Results: The normal lungs were gradually and gravity-dependently enhanced with time after Gd-DTPA aerosol inhalation regardless of the respiratory SI changes, except for three animals with the fastest breathing rate. The averaged maximal relative lung SI increase against the baseline in the successful animals was significantly greater in the slowly and deeply breathing animals than in the fast and shallow breathing animals, regardless of the difference in Gd-concentration (100 mmol Gd/L: 153.3% +/- 69.7% vs. 54.2% +/- 23%; P < 0.001; and 200 mmol Gd/L: 189.7% +/- 68.0% vs. 75.6% +/- 42.2%; P < 0.0001, respectively). There was an additional enhancement of 382% +/- 101 in the ventral lung and 722% +/- 160 in the dorsal lung on the pulmonary arterial phase perfusion image even in the slowly and deeply breathing animals who inhaled 200-mmol-Gd/L aerosol, and the enhancement effect was significantly greater compared with that with the aerosol (P < 0.0001). The ventilation- and perfusion-weighted images clearly defined the regionally matched perfusion-ventilation deficits in all the AO models, and the regionally mismatched perfusion-ventilation in all the PE models., Conclusion: Gd-based aerosol can provide efficient lung enhancement in spontaneously and adequately breathing animals, using a relatively noninvasive aerosol delivery system. The combined use of Gd-based perfusion MR imaging may be acceptable for defining regionally impaired function associated with acute AO and PE.

Purpose: Gd-DTPA aerosol ventilation MR imaging was obtained using a modified aerosol delivery system with an aerosol reservoir to non-invasively assess regional lung ventilation in dogs., Material and Methods: Seven anesthetized, spontaneously breathing normal dogs inhaled 200 mmol Gd/l Gd-DTPA aerosol produced by an ultrasonic nebulizer, using an open-circuit aerosol delivery system with or without an aerosol reservoir. Fast gradient-echo MR images were sequentially acquired with an interval time of 1 min for 25 min before and after aerosol inhalation. The aerosol study was also performed using the aerosol delivery system with an aerosol reservoir in the same 7 dogs after airway obstruction with a balloon catheter, and in another 7 dogs after pulmonary arterial embolization with enbucrilate. An i.v. Gd-DTPA-enhanced dynamic MR study after i.v. bolus injection of a 0.1 mmol/kg dose of Gd-DTPA was combined to assess regional lung perfusion. Lung enhancement effect was evaluated by time-signal intensity curves and the subtracted ventilation- and perfusion-weighted images., Results: With or without the aerosol reservoir, the normal dog lungs were gradually and gravity-dependently enhanced with time after aerosol inhalation. The use of the aerosol reservoir, however, showed significantly greater lung enhancement without a significant increase in breathing rate and with minimal reduction in PaO2 of less than 5 mm Hg in these animals. The enhancement effect of i.v. injection of Gd-DTPA at pulmonary arterial perfusion phase was significantly greater compared to that of Gd-DTPA aerosol throughout the normal lungs, and the subtracted ventilation-weighted and perfusion-weighted images showed homogeneous but gravity-dependent aerosol deposition and perfusion. These images clearly defined the regionally matched perfusion-ventilation deficits in the lung regions distal to bronchial obstruction in all the airway obstruction dogs, and the regionally mismatched perfusion-ventilation in the embolized regions of all the pulmonary arterial embolization animals., Conclusion: Gd-based aerosol can non-invasively image regional lung ventilation in spontaneously breathing animals, using an adequate aerosol delivery system. The combined use of Gd-DTPA perfusion MR imaging may be acceptable for defining regionally impaired lung function associated with acute airway obstruction and pulmonary arterial embolization.

Rationale and Objectives: We conducted an animal study to describe and interpret the perfusion characteristics of oleic acid (OA)-injured lungs on gadopentetate dimeglumine (Gd-DTPA)-enhanced dynamic perfusion magnetic resonance (MR) imaging., Methods: Fourteen dogs received an intravenous OA infusion in the supine (n = 4), prone (n = 4), and right lateral decubitus (n = 6) positions, and 10 minutes later these animals in the same postures underwent the dynamic MR study. Regional Gd-DTPA kinetics was analyzed by the time-signal intensity (SI) curves and by qualitative functional map images of the mean transit time that was representative of the mean circulation time in the vascular bed and the average cumulative sum of the relative increases in SI representative of Gd-DTPA distribution volume during Gd-DTPA first pass. The results were compared with those in six control animals and in another six animals that underwent the MR study 3 minutes (n = 3) and 60 minutes (n = 3) after OA infusion. The MR findings were correlated with the distribution of lung damage and the infused OA particles as assessed by histology., Results: The dynamic MR study showed postural shifts on the gravity-dependent perfusion map of normal lungs. Contrast enhancement during Gd-DTPA first pass in the lung was lower and more heterogeneous in the OA-injured lung models than in controls but was followed by conversely greater and persistent enhancement during the Gd-DTPA redistribution phase. Regardless of the postures for OA infusion, these abnormalities were predominant in the dependent lungs and became more pronounced with time after OA infusion, where more prominent capillary obstruction with OA droplets and alveolar/interstitial edema were histologically observed. On the functional map images, greater mean transit time and the average cumulative sum of the relative increases in SI values were also predominantly distributed in the dependent lungs., Conclusions: Low and heterogeneous enhancement was observed during Gd-DTPA first pass but was followed by persistent enhancement during the Gd-DTPA redistribution phase, and predominant abnormalities in the dependent lungs may be characteristic features of the perfusion of OA-injured lungs. The histological correlations indicate that these abnormalities may reflect OA-induced pathophysiologies associated with capillary OA obstruction, increased vascular resistance, and capillary permeability/extravascular spaces and that lung damage may be gravity dependent.

Unlabelled: Suga K, Mikawa M, Ogasawara N, et al. Potential of Gd-DTPA-mannan liposome particles as a pulmonary perfusion MRI contrast agent: An initial animal study. Invest Radiol 2001;36:136-145., Rationale and Objectives: A paramagnetic, particle-type MR contrast agent, (Gd-diethylenetriamine pentaacetic acid [DTPA]-mannan-cholesterol)-coated liposomes, was designed to localize in the lung by the mechanism of capillary blockade, and the potential of this agent for pulmonary perfusion MRI was experimentally investigated., Methods: Before and up to 60 minutes after slow injection of this contrast agent, MR images were sequentially acquired at 10-second intervals along the same transaxial plane of the lung by using a gradient-echo pulse sequence with a short echo time of 1.2 ms on a 1.5-T MR scanner. After the minimal dose for obtaining a sufficient lung enhancement effect was determined in five rabbits, the time course of the enhancement effect was evaluated in six dogs by arterial blood gas analysis. The efficacy of MRI for detecting perfusion defects was evaluated in seven other dogs with pulmonary embolism., Results: Normal lungs were dose-dependently enhanced by this agent, and with a 2.0 mL/kg dose, dependent lungs were enhanced by more than 201%, with an average half-life of the enhancement effect of 35.7 +/- 5.3 minutes. With less than this dose (1.0-1.5 mL/kg), all of the embolized lung portions were clearly identified as perfusion defects. The prolonged enhancement effect allowed the acquisition of subsequent multisectional lung images, thus facilitating the assessment of anatomic location and extent of the perfusion defects. The reduction of PaO2 in room air after injection was within 5 mm Hg in both normal and embolized animals., Conclusions: These initial, experimental results show that paramagnetically labeled liposome particles may be a successful MR contrast agent for pulmonary perfusion imaging.

Study Objectives: Dynamic MRI and (133)Xe single-photon emission CT (SPECT) were used to directly evaluate the interaction of regional respiratory mechanics and lung ventilatory function in pulmonary emphysema., Methods: Respiratory diaphragmatic and chest wall (D/CW) motions were analyzed by sequential MRI of fast-gradient echo pulse sequences during two to three respiratory cycles in 28 patients with pulmonary emphysema, including 9 patients undergoing lung volume reduction surgery (LVRS). The extent of air trapping in the regional lung was quantified by the (133)Xe retention index (RI) on three-dimensional (133)Xe SPECT displays., Results: By contrast to healthy subjects (n = 6) with regular, synchronous D/CW motions, pulmonary emphysema patients showed reduced, irregular, or asynchronous motions in the hemithorax or location with greater (133)Xe retention, with significant decreases in the maximal amplitude of D/CW motions (MADM and MACWM; p < 0.0001 and p < 0.05, respectively). The removal of (133)Xe retention sites by LVRS effectively and regionally improved D/CW motions in nine patients, with significant increases in MADM and MACWM (p < 0.01 and p < 0.001, respectively). In a total of 40 studies of the 28 patients including post-LVRS studies, normalized MADM and MACWM correlated with percent predicted FEV(1) (r = 0.881, p < 0.0001; and r = 0.906, p < 0.0001, respectively), and also with (133)Xe RI in each hemithorax (r = -0.871, p < 0 0.0001; and r = -0.901, p < 0 0.0001, respectively.), Conclusions: This direct comparison of regional respiratory mechanics with lung ventilation demonstrated a close interaction between these impairments in pulmonary emphysema. The present techniques provide additional sensitivity for evaluating pathophysiologic compromises in pulmonary emphysema, and may also be useful for selecting resection targets for LVRS and for monitoring the effects.

To evaluate impaired respiratory mechanics in pulmonary emphysema, dynamic breathing magnetic resonance imaging (BMRI) was acquired with fast-gradient echo pulse sequences at fixed thoracic planes over two to three slow, deep respiratory cycles in 6 controls and 28 patients with pulmonary emphysema including 9 patients undergoing lung volume reduction surgery (LVRS). Respiratory motions of the diaphragm and chest wall (D/CW) were assessed by a cine-loop view, a fusion display of maximal inspiratory and expiratory images, and the time-distance curves. By contrast with normal subjects with regular synchronous D/CW motions, the patients frequently showed reduced, irregular, or asynchronous motions, with significant decreases in the maximal amplitude of D/CW motions (MAD and MACW), and the length of apposition of the diaphragm (LAD) (P < 0.0001, P < 0.001, P < 0. 01, respectively). After LVRS, nine patients showed improvements in D/CW configuration and mobility, with significantly increased MAD, MACW, and LAD (P < 0.01, P < 0.0001, and P < 0.05, respectively). In 40 studies of 28 patients including the post-LVRS examinations, the normalized MAD and MACW significantly correlated with %FEV(1) (r = 0. 881 and r = 0.906; P < 0.0001, respectively). BMRI seems useful for noninvasively and directly assessing the impaired respiratory mechanics associated with abnormal ventilation in pulmonary emphysema, and also for monitoring the effects of LVRS. J. Magn. Reson. Imaging 1999;10:510-520., (Copyright 1999 Wiley-Liss, Inc.)

An initial attempt to superimpose in-111-labeled platelet SPECT over CT/MRI in two patients with intracardiac thrombus is presented. One patient was a 65-year-old woman with a history of mitral and tricuspid valve replacement. This patient had a large thrombus measuring 74 x 43 x 34 mm in the right atrium. Fusion imaging with CT showed increased in-111 activity on the surface of the thrombus, which was visualized as circular activity on a planar image. The other patient was a 72-year-old woman who had a thrombus measuring 17 x 14 x 12 mm at the cardiac apex. Planar imaging showed two discrete areas of abnormal in-111 activity. Superimposition of the SPECT image over MR revealed that one of these areas corresponded to the small thrombus, whereas the other related to a previous subendomyocardial biopsy. In both patients, the activity ratio of the abnormal in-111 accumulation versus the cardiac chamber on SPECT was greater than that on planar images. The correlation of in-111 platelet SPECT and CT/MRI was useful in accurately localizing the intracardiac thrombus.

MR imaging during arteriography was performed to evaluate the clinical utility to patients with hepatocellular carcinoma. After a conventional hepatic arteriography, the patients were transferred to the MR unit and MR arteriography was obtained with the gradient echo technique synchronously starting manual injection of Gd-DTPA diluted with normal saline into the hepatic artery using 5-F Cobra angiographic catheter. The tumors showed high signal intensity relative to surrounding normal parenchyma and a clear liver-to-lesion contrast was obtained. Especially in one case, a conventional hepatic arteriography did not show the recurrent mass, however it was demonstrated as a high signal intensity area on MR arteriography. Therefore, this new technique seemed to be useful for the detection of hepatocellular carcinoma.

To study the potential of MRI to demonstrate the architecture of the tracheobronchial wall, resected tracheobronchi of rabbits were examined with MRI. As a result, epithelium and cartilage were demonstrated as relatively low intensity bands on T1- and T2-weighted images (T1WI and T2WI). The subepithelium was of intermediate-high intensity on T1WI and of markedly high intensity on T2WI and Gd-DTPA enhanced T1WI. Peritracheobronchial fat was of markedly high intensity on T1WI and of relatively low intensity on T2WI. In conclusion, MRI was found to have the potential to demonstrate the architecture of the tracheobronchial wall as 4 layers on each sequence.