Even though we have made much progress over the last decade, much still needs to be done. I am personally very pleased with the excellent management and authority of my coauthor, Joe Frank, in assuming the Acting Directorship of our DRRP. You will be hearing more from him in the immediate future about this research program, and he will be accessible to all of you, both individually and at our national meetings. We must continue to recruit representatives to fill responsible positions at the NIH, including Director for the DRRP at the National Cancer Institute. Further funding for equipment and personnel within the Diagnostic Radiology Research Laboratory also will be necessary. We are anticipating that the annual laboratory budget will increase further, but to provide additional resources, the Conjoint Committee plans to take an active role in encouraging foundations, corporations, and member societies to assist further with this effort. Most of the other disciplines within medicine--in fact, many individuals in your own medical school within other departments--have spent time at the NIH. It is this high-quality experience, in an environment that can provide both the research background and the cultivated personal relationships and interactions, that will foster a long-term research career. Clearly, the complementary role of both the intramural and extramural programs will allow diagnostic radiology to achieve excellence and true respect in the total research community of the Academy. Many people have been responsible for our success, and I have already mentioned a few.(ABSTRACT TRUNCATED AT 250 WORDS)

To test the potential of 1.5 Tesla magnetic resonance imaging (MRI) for assessing the protein concentration of pleural effusions, five pleural fluid analogs (saline + 0, 2, 4, 6, 8 g albumin/100 mL) and, for comparison, four saline dilutions of whole blood were evaluated in vitro. The relaxation rates (1/T1, 1/T2) of albumin solutions were determined by 1.5 T spectroscopy (MRS) and correlated with albumin concentration (1/T1:slope 0.02, r + 0.89, P less than .05; 1/T2: slope 0.16, r = 0.997, P less than .001). MRI studies of these solutions showed no significant correlation with 1/T1, but 1/T2 showed a positive correlation with albumin concentration (r = 0.98, P less than .01). Both MRI relaxation rates were significantly correlated with blood concentration, and slopes were greater than for albumin solutions. These preliminary studies, demonstrating differences in correlation between relaxation rates and the concentration of albumin and blood, suggest that MRI has the potential for differentiating pleural effusions of different chemical composition.

We have described an unusual variant of bronchopleural fistula whose ball valve characteristics led to a chronic tension pneumothorax without fluid collection in a postpneumonectomy space.

A retrospective study of 48 patients with non-Hodgkin lymphoma (NHL) who underwent 54 computed tomographic (CT) examinations of the chest evaluated the role of chest CT in the management of this disease. Of 18 cases in which chest radiographs were not indicative of NHL, CT scans showed abnormalities consistent with NHL in five (28%). Of 11 cases where radiographs were questionable, CT confirmed NHL in five (45%) and excluded it in six (55%). Of 25 cases where radiographs were consistent with NHL, CT confirmed the findings in 23 (92%) and added information in all 25. Chest CT affected management in eight of 25 treated and five of 19 untreated patients. It appears to be useful in untreated patients with stage I or II NHL but no definite radiographic abnormalities, or with abnormal radiographs but no extrathoracic spread, and in treated patients with questionable radiographs. CT is not helpful in untreated patients with stage III or IV NHL or treated patients with normal radiographs.

Oleic acid infusion in dogs produces a patchy, predominantly peripheral lesion on CT scans. This study correlates the pattern of oleic acid injury with the distribution of infused oleic acid and pulmonary blood flow. Radiolabeled oleic acid (I-125, 0.05 ml/kg) and radiolabeled 15-micron microspheres (Co-57) were infused into the right atria of 11 dogs. Oleic acid was given after the microspheres in six dogs and before microspheres in five dogs. Ten minutes after infusion, the lungs were removed. Four transverse slices (0.5 cm thick) of the lower lobes were taken from each dog and cubed. Samples were grouped into three regions of the transverse slice: outer, middle, and inner concentric rings. In both groups, I-125 (oleic acid) activity was greater in the outer than the middle and inner concentric layers (P less than 0.001). When Cobalt-57 microspheres were given before oleic acid, Cobalt-57 activity was marginally lower in the outer layer compared with the middle and inner layers. However, when oleic acid was given first, microsphere activity in the outer layer was significantly lower (P less than 0.001) than the middle layer. Thus, oleic acid was preferentially distributed to the peripheral regions of the lung, similar to the regions of injury on CT. This distribution did not correspond to the pattern of pulmonary blood flow as indicated by the microspheres. Immediately after oleic acid infusion, pulmonary blood flow to the periphery was reduced, reflecting a response to the predominantly peripheral injury by oleic acid.

Oleic acid infusion, as a model of fat embolism, produces a predominantly peripheral lesion in the dog lung. The lung injury corresponds to the peripheral distribution of labeled oleic acid. The basis for this distribution of oleic acid is not known. Our hypothesis for this nonuniform distribution is that particle diameter plays a role in the subsequent distribution of infused oleic acid and the resulting lung injury. We injected 15-mu microspheres 85Sr and then 137-mu microspheres (141Ce) into the right atria of seven dogs, which were killed and the lungs removed. Analysis of the distribution of the two different diameter microspheres within axial slices from the left caudal lobe of each dog revealed a peripheral distribution of the larger diameter microspheres not seen with the smaller microspheres.

The results concerning the activity or inactivity of the alveolitis determined with our proposed scoring system were comparable to those determined with gallium scanning in 68.2% of the patients with biopsy-proven sarcoidosis in our study. Certainly, further evaluation of a larger series of patients over a longer period of time will be required to more adequately evaluate this classification method. We also believe this methodology is reproducible and can be utilized by experienced interpreters of chest radiographs. If we closely reevaluate the 32% of the cases in which the radiographic results disagreed with the gallium assessment of activity, two points worthy of mention become evident. In one group the gallium indices were borderline for activity whereas in the other group the radiographic criteria were most likely indicative of irreversible parenchymal changes. When the radiographs for the latter group were reevaluated for signs of fibrosis such as volume loss, cystic changes, and alteration in vascular patterns, it was possible to delineate a significant false-positive group of patients. We will be reporting the long-term results of our observations over the next several years as well as adding more groups of patients to our initial study analysis. At that time, correlation with all other measurable parameters in determining high-intensity alveolitis versus low-intensity alveolitis will be compared to these new chest radiographic profiles.

Hepatocellular neoplasms, including neoplastic nodules (NN), are the most commonly induced tumors resulting from chemical carcinogen evaluation. Our objective was to image neoplastic nodules using computed tomography. In a preliminary study using rats with diethylnitrosamine (DEN) induced tumors, lesions smaller than 1.5 cm were difficult to identify by CT. Since NN do not take up excess iron whereas normal liver does accumulate iron, we studied iron as a CT contrast material. Hemochromatosis was induced in 15 control rats by subcutaneous injections of iron dextran. A linear increase in iron-loading dose produces a linear CT liver enhancement (r = 0.97): 68, 80, 84, 94, and 104 HU at 0, one, two, four, and six weeks, respectively. No loss of enhancement was noted ten weeks later. Rat hepatic tissue was chemically analyzed after a similar iron-loading regimen. The iron concentration (microgram/g hepatic tissue) progressively increased during the first four weeks of loading and remained stable for three weeks following iron-loading. Four animals that had been given DEN and iron were examined by CT scanning to detect small NN. Iron-enhanced CT allowed the visualization of small lesions (less than 5 mm). Histopathologic sections confirmed a homogeneous pattern of iron uptake in normal liver with a deficiency of iron in NN. We conclude that CT scanning following iron-loading is a noninvasive method to detect small nodules and may provide a method to study the progression or regression of small liver nodules in rodents.

Fifty patients with clinically suspected lesions of the chest apex underwent computed tomography (CT) and plain film (PF) examinations in an attempt to define the underlying etiology as benign or malignant. CT was slightly more sensitive in diagnosing malignancy (90%) compared with PF (77%), with a higher predictive value of a negative test (85% vs. 70%), although the differences were not statistically significant. The specificity and predictive value of a positive test were similar for CT and PF. CT's advantage was better elucidation of the soft tissues of the chest apex. CT provided additional information beyond PF in 17/20 patients with benign conditions and in 29/30 patients with cancer. In 21 of these 29 cancer patients, CT findings significantly altered subsequent management.

The varied computed tomographic (CT) appearance of obstructive lobar collapse is described in 25 cases. CT is helpful in understanding the morphology and mechanisms of lobar collapse and in diagnosing atypical cases. The final shape of the collapsed lobe is dependent on the size and location of the obstructing tumor and on the amount of retained lung fluid. Focal bulging of the fissure (S-sign) is the most helpful sign in identifying the obstructing tumor. Differential enhancement could not separate the tumor from collapsed lung in six of eight (75%) of the cases studied with intravenous contrast, a finding at variance with previous reports. The limitations of CT in evaluating mediastinal or pleural invasion in the presence of lobar collapse are discussed.

To clarify the role of standard chest radiography in prostatic adenocarcinoma, the pulmonary manifestations of 198 patients with Stage D disease were evaluated. All patients were treated with chemotherapeutic protocols allowing for adequate clinical and radiographic correlation. Retrospective interpretation of serial chest radiographs revealed that 35% of our patients had visible intrathoracic abnormalities; however, only 24% of the patients had abnormalities attributable to intrathoracic metastases. Twenty-two percent of patients had pleural effusions, 16% reticular opacities, 3.5% reticulonodular opacities, 8% isolated or discrete pulmonary nodules, and 4.5% adenopathy. Etiologies of these opacities included metastatic disease in 93.5% of those with adenopathy and nodular or reticulonodular opacities, but 39% of pleural effusions and 52% of reticular opacities were best attributed to concomitant processes. Four patients had intrathoracic metastases without bone metastases. Standard chest radiography is a valuable screening procedure that should be correlated with clinical data to differentiate metastases from concomitant processes.

OCALIZED lucent lung lesions depicted by the standard chest radiograph are relatively rare, except for those occurring in patients with generalized chronic obstructive pulmonary disease. The standard chest radiograph provides an inexpensive, sensitive means for determining the presence of most clinically significant focal, thinwalled, sharply demarcated areas of avascularity. Some of the limitations of the chest radiograph in this regard are: (a) it permits only a broad differential diagnosis; (b) it sometimes does not define the full extent of disease; (c) it sometimes does not allow localization to the precise anatomic compartment; and (d) it sometimes fails to demonstrate a small accompanying lesion adjacent to or remote from the lesion under investigation. Even with these radiographic limitations, other available data including clinical history, pulmonary function studies, and laboratory results significantly enhance the radiologist’s diagnostic accuracy. For those few but potentially life-threatening localized lucent lesions, more aggressive and risky procedures must be employed for diagnostic or therapeutic purposes, such as biopsy, chest tube insertion, or even thoracotomy. With the advent of CT, considerable improvement in the overall diagnostic accuracy of these localized lucent lung lesions has been obtained. By defining more precisely the location, margins, thickness, contents, and overall configuration, CT significantly narrows the differential diagnosis. If the CT characteristics of the lesion are nondiagnostic, interventional procedures such as bronchoscopy or percutaneous aspiration biopsy can more easily be obtained by CT guidance. The CT cross-sectional display and geometric resolution sometimes more clearly define previously recognized standard radiographic features of complicated versus uncomplicated and benign versus malignant lucent lung lesions. In addition, CT uniquely distinguishes pleural from subpleural lucent lesions. Probably the most important role CT serves is in demonstrating occult opaque or lucent lesions in proximity to the original lucency or in the contralateral lung. The presence or absence of abnormal pulmonary vascularity, adenopathy, calcification, or small pleural

The computed tomographic (CT) appearance of pulmonary edema induced by elevated left atrial (LA) pressure was examined. Dogs, in the prone position, were scanned during suspended ventilation at functional residual capacity. A surgically implanted LA balloon was inflated to elevate LA pressure for 30 to 140 minutes to a mean pressure of 29.8 mm Hg. Lung water, measured gravimetrically, averaged 14.7 ml/kg body weight compared with 5.7 ml/kg in nonedema control dogs. Lung density in dogs with edema was 69.5% higher than base-line density, while in the control group final lung density was only 4% higher than base line. Analysis of regional density indicated that there were greater increases in density in more central and dependent (ventral) zones of the lung and relatively smaller increases in nondependent (dorsal) peripheral zones. These results are in contrast to the previously reported pattern of density change seen with oleic acid injury in which density increases were primarily in peripheral zones of the lung.

Forty patients with coarctation of the aorta were compared with 80 normal subjects matched by sex and age. The aortic arch was invisible on the lateral radiograph in 84% of patients preoperatively and 88% postoperatively. Only 6 normal subjects (8%) had a similar finding. The authors conclude that an obscured aortic arch on the lateral view could be a clue to the presence of coarctation. Possible explanations for this phenomenon include (a) dilatation of the brachiocephalic arteries (particularly the left subclavian artery), obscuring the superior margin of the arch; (b) hypoplasia and anteromedial displacement of the distal portion of the arch; and (c) diffuse hypoplasia of the arch combined with focal coarctation.

Takasugi, J.; Godwin, J. D.; Halvorsen, R. E.; Williford, M. W.; Silverman, P. M.; Putman, C. E. Author Information

A noninvasive method for differentiating transudative and exudative pleural effusions could make thoracentesis unnecessary in many cases. Since computed tomography (CT) can be used to measure tissue densities, it has a potential use in assessing density of intrathoracic fluid. This study was performed to test this possibility. Plastic syringes containing water, saline, and three saline-albumin solutions with densities up to 1.021 g/cc were scanned individually in air, in a chest phantom, and in a dog thorax during apnea. We found significant positive correlations between specific gravity, and CT density; coefficients ranged from 0.85 to 0.99. Regression lines for the different test conditions were not statistically different. However, CT number variability, as indicated by the average standard deviation (SD), did differ for the test conditions: 1) Syringes alone, average SD was 3.8 HU; 2) chest phantom, average SD was 5.5 HU adjacent to rib and 5.8 HU anterior to vertebral column; 3) dog chest average SD was 19.2 HU with heart motion and 9.4 HU without heart motion. These results suggest that heart motion is a major factor in CT number variability in vivo, and that differentiation of transudates from exudates is not possible with current CT methods.