| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Experimental Studies |
1 From the First Dept of Internal Medicine (K. Ichikado, M.S., Y.G., K. Iyonaga, M.A.) and Dept of Radiology (T.Y., S.T., M.T.), Kumamoto Univ School of Medicine, 1-1-1 Honjo, Kumamoto, 860-0811, Japan; Depts of Radiology (T.J., O.H.) and Internal Medicine (Y.S.), Osaka Univ Medical School, Japan; Dept of Radiology, Kyoto Univ Hosp Shogoin, Japan (H.I.); Dept of Radiology, Ehime Univ Medical School, Japan (J.I.); and Dept of Radiology, Vancouver Hosp and Health Sciences Centre and Univ of British Columbia, Canada (N.L.M.). From the 1997 RSNA scientific assembly. Received Jul 16, 1999; revision requested Aug 26; revision received Oct 14; accepted Oct 26. Address correspondence to K. Ichikado (e-mail: ichikado@kaiju.medic.kumamoto-u.ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Eighteen juvenile pigs were exposed to more than 80% oxygen—for 24, 48, 72, 96, or 120 hours—or room air in sealed cages. Their removed lungs were inflated with air infused through the trachea and examined with thin-section CT. Two independent observers, without knowledge of the exposure times, compared 63 areas selected on the CT scans with the corresponding pathologic and histologic findings, which were evaluated independently by two pathologists.
RESULTS: CT findings correlated well with histologic findings (
= 0.86, P < .001), which corresponded to the pathologic phases of DAD. All areas of normal CT attenuation, eight of nine spared regions within areas of opacity, and two of 15 areas of ground-glass opacity corresponded to the early exudative pathologic phase of DAD. All areas that showed traction bronchiolectasis at CT corresponded to the early proliferative pathologic phase. There was good observer agreement regarding the interpretation of CT findings (
statistic, >0.60) and histologic results (
0.70).
CONCLUSION: Thin-section CT findings reflect the pathologic phases of DAD, although the early exudative phase cannot be specifically depicted by thin-section CT. Traction bronchiolectasis on a CT scan suggests progression to the proliferative phase.
Index terms: Animals • Computed tomography (CT), thin-section, 60.12111, 60.12118 • Lung, CT, 60.12111, 60.12118 • Lung, diseases, 60.26, 60.4133, 60.916 • Respiratory distress syndrome, adult (ARDS), 60.4133
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
The DAD that characterizes ARDS can be categorized into three pathologic phases: acute exudative, subacute proliferative, and chronic fibrotic (9–12). Lamy et al (13) reported that patients who demonstrated the acute exudative phase have a better prognosis than do those who show more advanced disease. Meduri et al (14,15) suggested that corticosteroids may be useful during the proliferative phase and bring about marked physiologic improvement or a lower prevalence of infectious complications, especially during the early proliferative phase.
Ichikado et al (7) found that certain findings on thin-section CT scans correlated with the pathologic phases of DAD in 14 patients with acute interstitial pneumonia, which is the idiopathic form of ARDS. In that study, the presence of traction bronchiectasis in areas that showed increased attenuation on CT scans was associated with the late proliferative or fibrotic phase of DAD. However, the authors did not clarify whether the early lesions of DAD can be detected at thin-section CT or which CT findings reflect progression to the proliferative phase, which may be a critical point in predicting the reversibility of the lesions. Our purpose was to conduct a more detailed study of the correlation between thin-section CT findings and the pathologic phases of DAD in an experimental porcine model.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
Hyperoxic Exposure
All of the pigs were housed in individual cages (Natsume Seisakusho, Tokyo, Japan) that were made of stainless steel and equipped with a duct for the outflow of gas, including carbon dioxide, and a Plexiglas window for observation of the animal. To produce high concentrations of oxygen within each cage, the cages were initially saturated with 100% oxygen at a flow of 10 L/min for 45 minutes. The concentration of oxygen was then maintained at 80%–85% with an inflow of 3 L/min, except during the daily cleaning of the cage, which took 10 minutes. The oxygen levels were measured daily by using an oxygen analyzer (Natsume Seisakusho). The temperature in the cage was kept between 22°C and 24°C. The cage was spacious enough to allow the animal to exercise, and food and water were available in the cage. To prevent the animals from developing pneumonia, ampicillin in a dry syrup form (25 mg/kg) was administered daily in the food mixture. Fifteen pigs were exposed to concentrations of oxygen of 80%–85%, and three animals each were exposed for 24, 48, 72, 96, and 120 hours. Three pigs that were exposed to room air in their separate cages for more than 72 hours served as control animals.
CT Examination
After hyperoxia or room air exposure, the pigs were anesthetized with intramuscular and intravenous ketamine and then sacrificed with intravenous pentobarbital sodium. The lungs were removed, inflated with air at a pressure of 20–25 cm H2O, and scanned by using thin-section CT. All scans were obtained by using a HiSpeed Advantage scanner (GE Medical Systems, Milwaukee, Wis). The scans were obtained through the entire lung (4-mm interval, 1-mm collimation, 15–18-cm field of view, 512 x 512 matrix, 120 kVp, 100 mAs) and reconstructed by using a high-spatial-frequency algorithm. The scans were viewed at window levels appropriate for pulmonary parenchyma (window width, 1,200 or 1,500 HU; window level, -700 or -550 HU).
CT Assessment
The thin-section CT scans were read by two independent chest radiologists (T.J., T.Y.) who were unaware of the durations of exposure or the pathologic data. The scans were assessed for the presence or absence of ground-glass opacity, airspace consolidation, septal thickening, and traction bronchiolectasis and traction bronchiectasis, and the overall extent of parenchymal abnormalities. Ground-glass opacity was defined as an area of hazy increased opacification without obscuration of the underlying vascular markings. Airspace consolidation was considered to be present when an area of increased attenuation obscured the vascular markings. When areas of ground-glass opacity became heterogeneous relative to areas of relatively high attenuation, "heterogeneous attenuation" was noted. When areas that seemed to show normal attenuation were observed focally and segmentally on the scans, they were noted as "spared areas" (7). Septal thickening was recognized as an abnormal thickening of interlobular septa.
The criteria for deciding whether the airways were bronchiectatic or bronchiolectatic were based on the results of a previous study of the morphology and CT anatomy of the porcine lung (16) and on the CT findings in the control animals in the present study. Because pulmonary arteries of approximately the same caliber as the bronchi collapsed during the lung removal process, the usual criteria for identifying bronchiectatic bronchi of a larger caliber than that of the accompanying arteries were not necessarily used. Therefore, an irregularly dilated bronchus within an area of parenchymal abnormality was recognized as traction bronchiectasis. The lobular architecture of porcine lung tissue is similar to that of human lung tissue (16). Traction bronchiolectasis was recognized as irregularly dilated bronchioles within areas of increased attenuation. A mosaic pattern of attenuation was considered to be present when lobular or multilobular areas of variable attenuation were seen at CT (17,18).
When a CT abnormality was present, the extent of involvement of the entire lung—that is, the percentage of lung involvement—was estimated visually. The percentages of lung involvement for each abnormality were averaged together for the three animals that were subjected to the same degree of hyperoxia, and the mean was recorded as absent (-), mild (+, 0%–50%), moderate (++, 50%–75%), or severe (+++, >75%) (Table 1). The areas of increased attenuation were also assessed for laterality of involvement, patchy or diffuse distribution, and upper, middle, or lower zonal predominance. The upper zone was defined as the area above the level of the carina; the middle zone, the area between the level of the carina and the branching of the right lower lobe bronchus; and the lower zone, the area below the branching of the right lower lobe bronchus. Each of these levels previously has been shown to correspond to approximately one-third of the lung volume in pigs (16).
|
Histopathologic Evaluation
For histopathologic evaluation, each left lung was inflated, fixed with 10% formalin that was injected through the left main bronchus, and immersed in the fixative for 2 days. The formalin-fixed lungs were cut into 4-mm sections in the same plane as the CT scans. Specimens from the 63 areas were obtained for histopathologic analysis and comparison with the CT scans. The cut surfaces of the specimens were compared with the CT scans, with special attention paid to such anatomic landmarks as the bronchial bifurcations and the pulmonary vasculature. The lung area specimens were mapped out on the basis of CT findings. To quantitatively evaluate the lesions induced by hyperoxia, each specimen area was evaluated histologically by two independent pathologists without knowledge of the duration of hyperoxic exposure or the CT findings. The following 13 histologic features that are characteristic of DAD were observed: epithelial destruction, capillary congestion, interstitial edema, intraalveolar edema, hemorrhage, mononuclear infiltrates, polymorphonuclear infiltrates, septal thickening, hyaline membrane formation, microatelectasis, type II pneumocyte hyperplasia, fibroblast proliferation, and interstitial collagenous deposition. The severity of these findings was graded on a four-point scale as follows: 0, absent; 1, mild; 2, moderate; 3, severe. An overall histologic score (mean ± SEM) was determined for each lung area on the basis of the total score in each respective area.
Each area was classified into a pathologic phase of DAD according to the following criteria: The early exudative phase was characterized by epithelial destruction, capillary congestion, and interstitial and intraalveolar edema. The late exudative phase was characterized by hyaline membrane formation and microatelectasis in addition to the criteria for the early exudative phase. The early proliferative phase was characterized by proliferation of type II pneumocytes, fibroblast proliferation, and loose intraalveolar fibrosis. The late proliferative phase was characterized by type II pneumocyte hyperplasia, with the proliferation of fibroblasts not only within the interstitium but also within the airspaces. The fibrotic phase was characterized by extensive interstitial fibrosis caused by the proliferation of numerous fibroblasts and the deposition of collagen (9,10,19). Correlations between the overall histologic scores and the pathologic phases of DAD were determined statistically.
Pathologic Basis of CT Findings
To evaluate the pathologic basis of CT findings, the excised right lungs were inflated and fixed by using the Heitzman method (20). Briefly, the lungs were distended through the main bronchus with fixative fluid that contained polyethylene glycol 400, 95% ethyl alcohol, 40% formalin, and plain water in proportions of 10:5:2:3. The specimens were immersed in the fixative for 2 days, and then the fixed lungs were air dried. After fixation, first-screening thin-section CT was performed by using the same scanner as that used to initially image the excised and inflated lungs and under the same conditions as those described earlier. Selected areas of these lungs were cut into 1-cm-thick transverse sections, and serial thin-section CT scans of these sections were obtained at 1-mm thickness.
A microslicer (DTK-3000W; Dosaka EM, Kyoto, Japan) was used to cut selected areas smaller than 7 x 5 cm into serial 1-mm-thick sections that corresponded to the CT scans of the specimen sections. A radiograph of each section was obtained in contact with a fine-grain film (Softex, fine grain; Fuji Medical, Tokyo, Japan) at 10 kVp, 1,260 mAs, and a 30-cm tube-film distance. When histologic confirmation was needed, serial 20-µm-thick microscopic sections were obtained. The thin-section CT scans were compared directly with the stereomicroscopic views, contact radiographs, and histologic sections of the same areas.
Data Analysis
Interobserver variability for the presence of parenchymal abnormalities, the extent and distribution of the lesions on the CT scans, and the scoring of the severity and pathologic phases was quantified by using the coefficient of agreement. To quantify the relationship between the histologic scores and the pathologic phases, the Mann-Whitney test was used. The correlations between the histologic scores and the CT findings were quantified by using the Spearman correlation coefficient.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
statistic for normal attenuation was 0.92; for spared area, 0.92; for ground-glass opacity, 0.82; for airspace consolidation, 0.75; for heterogeneous attenuation, 0.74; for septal thickening, 0.82; and for traction bronchiolectasis, 0.70. The statistic for the extent of each lung abnormality also was good (>0.60).
Pathologic Findings
The pathologic findings induced by hyperoxia in this model ranged from features characteristic of the early exudative phase of DAD to those characteristic of the early proliferative phase. No bacterial or fungal infection was found in any animal at histologic examination. The histologic scores corresponded to the pathologic phases. There was a significant difference in the histologic scores between the specimens showing the early exudative phase and those showing the late exudative phase and between the specimens showing the late exudative phase and those showing the early proliferative phase (P < .001). The pathologic phases, with the number of areas evaluated, and the corresponding overall histologic scores were as follows: early exudative phase (21 areas), 5.3 ± 0.5; late exudative phase (21 areas), 16.2 ± 0.9; and early proliferative phase (18 areas), 23.1 ± 0.9. In contrast, the mean histologic score of the areas in the control animals was 1.0 ± 0.0. The total of 21 areas that corresponded to the early exudative phase of DAD were observed in the animals subjected to hyperoxia for 24, 48, 72, or 96 hours. The 21 areas that corresponded to the late exudative phase were seen in the animals subjected to hyperoxia for 72 or 96 hours. The 18 areas that corresponded to the early proliferative phase were seen in the animals exposed for 96 and 120 hours.
There was a good agreement between observers in the evaluation of the histologic results; the statistics were as follows: epithelial destruction, 0.90; capillary congestion, 0.90; interstitial edema, 0.90; alveolar edema, 0.95; hemorrhage, 0.89; mononuclear infiltrate, 0.73; polymorphonuclear infiltrate, 0.63; septal thickening, 0.83; hyaline membrane formation, 0.62; microatelectasis, 0.90; type II pneumocyte hyperplasia, 0.86; and fibroblastic proliferation, 0.94.
CT Findings and Pathologic Phases
The CT findings, histologic scores, and pathologic phases are shown in Table 2. The CT findings correlated with the histologic scores ( = 0.86, P <.001) (Fig 5), although there was some overlap between the CT findings and the histologic scores. The three areas of normal attenuation on the CT scans obtained in the three control animals were confirmed histologically to be normal. On the CT scans obtained in the hyperoxic pigs, 11 areas that had an appearance similar to the areas of normal attenuation seen on the images obtained in the control animals were histologically determined to show the early exudative phase of DAD (Fig 1). Nine spared areas that appeared within areas of increased attenuation showed the early exudative (eight areas) or late exudative (one area) phase at histologic analysis. At histologic analysis, 15 areas of ground-glass opacity associated with septal thickening showed the early exudative phase (two areas), late exudative phase (10 areas) (Fig 2), or early proliferative (three areas) phase. The three areas of early proliferative disease were noted as type II pneumocyte hyperplasia and loose fibroblastic proliferation.
|
|
The mosaic pattern of attenuation seen on the CT scans in all the animals exposed to a high concentration of oxygen for 72 or more hours consisted of spared areas and areas of increased attenuation such as ground-glass opacity and/or heterogeneous attenuation. The relationships between the areas showing the mosaic pattern and the pathologic phases were as follows: The nine spared areas seen at CT corresponded to the early exudative (eight areas) or late exudative (one area) phase histologically. The seven areas of ground-glass opacity seen at CT corresponded to the early exudative (two areas), late exudative (four areas), or early proliferative (one area) phase histologically. The heterogeneous attenuation seen in five areas at CT corresponded to the early proliferative phase (five areas) histologically. The mosaic pattern of attenuation seen on the CT scans was attributed to the histologic heterogeneity of the lesions between lobules (Figs 2, 3).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
In the current study, we found a close correlation between thin-section CT findings and the pathologic phases of DAD, from the late exudative phase to the early proliferative phase. Furthermore, we observed a close correlation between normal CT findings and the early exudative phase of DAD. The problem is that CT did not enable us to differentiate normal from early abnormal disease, which is the most treatable phase of DAD.
Three (20%) of the 15 areas of ground-glass opacity with septal thickening, six (38%) of the 16 areas of heterogeneous attenuation, and all the areas of increased attenuation with dilated bronchioles within secondary lobules on the CT scans corresponded histologically to the early proliferative phase of DAD. Accordingly, the presence of dilated bronchioles within areas of increased attenuation was the most reliable sign of the transition from the exudative phase of DAD to the early proliferative phase. The dilation of bronchioles apparently resulted from interstitial fibroblastic proliferation and microatelectasis in the stage of early proliferative change rather than from an intrinsic bronchiolar abnormality. Therefore, the finding of dilated bronchioles was regarded as traction bronchiolectasis. These findings complement those of Ichikado et al (7), who analyzed the findings in 14 patients with acute interstitial pneumonia and found that the patients who had areas of increased attenuation with traction bronchiectasis at CT had histologic findings of the late proliferative or fibrotic phase of DAD.
In a CT-pathologic correlation study involving 26 patients with chronic diffuse lung disease, Remy-Jardin et al (22) found traction bronchiolectasis or bronchiectasis within areas of ground-glass opacity to be an indirect sign of established lung fibrosis. It has been reported that the extent of traction bronchiectasis may reflect the severity of fibrosis in idiopathic pulmonary fibrosis, with the presence of proximal segmental or subsegmental bronchial dilatation being a sign of more advanced fibrosis (23). Similarly, in a study by Howling et al (24) involving 16 patients with ARDS, the dilatation of airways within areas of ground-glass opacity at CT tended to persist at follow-up and were accompanied by the CT features of supervening lung fibrosis.
The histologic findings observed in the early proliferative phase of DAD are related to repair processes of lung injury (12,19). Corticosteroids can accelerate the repair reaction and have been reported to be most effective in the early proliferative phase of DAD before the development of acellular fibrosis with deranged alveolar architecture, which is seen in the late proliferative and fibrotic phases (14,15,21). The previous findings of Ichikado et al (7) and the findings in the current study confirm that dilated bronchioles and bronchi within areas of increased attenuation on CT scans reflect the transition from the exudative to the fibroproliferative phase of DAD. In the current study, the appearance of traction bronchiolectasis at CT was the earliest and most reliable sign of fibroproliferative changes of DAD, and this stage may have greater potential for response to corticosteroids.
The pathologic phases of DAD are known to vary from area to area in the same patient (9,10). In our study, histologic differences in the lesions from area to area in terms of severity and temporality resulted in a mosaic pattern of attenuation on the thin-section CT scans. The mosaic pattern of attenuation at CT has been attributed to three conditions: small airway disease, pulmonary vascular disease, and infiltrative lung disease (17,18). In small airway disease, areas of hypoattenuation are caused by air trapping. In pulmonary vascular disease, these areas are related to decreased perfusion. However, in infiltrative disease, the mosaic pattern reflects the presence of areas with parenchymal abnormalities and areas of spared lung (17,18). Because our experimental model of DAD-related ARDS was induced by the spontaneous inhalation of a high concentration of oxygen and there were no histologic abnormalities in the bronchi and bronchioles, the differences in the severity and temporality of the lesions may have been attributed to regional physiologic differences in ventilation and/or perfusion.
Histologic features of the early exudative phase of DAD were seen in all 11 areas of normal attenuation, in eight of nine spared areas within the areas of increased attenuation, and in two of 15 areas of ground-glass opacity on the CT scans. However, the early exudative changes were not specifically detected on the thin-section CT scans, and, thus, CT may not depict the onset of lung injury to the early exudative phase of DAD. Fracica et al (5) investigated the pathologic and physiologic features of DAD-related ARDS in baboons. According to their report, no substantial changes in cardiopulmonary physiologic parameters were produced by exposing the animals to more than 60 hours of excess oxygen. Therefore, it would be difficult to diagnose the early exudative changes of DAD-related ARDS by using radiologic and physiologic examination methods. Other methods for the detection of early DAD-related ARDS lesions are needed.
Our study had several limitations. We selected pigs as the experimental animal because the porcine lung is of an appropriate size for radiologic study and has a lobular architecture that is similar to that of the human lung, except for the well-developed interlobular septa and the relatively thick pleura in the porcine lung (16). The CT finding of thickened interlobular septa is not common to the DAD observed in human lungs (7,8,24). This difference could be related to physiologic and/or structural differences between human and porcine lungs. The postmortem thin-section CT scans of isolated lung specimens provide greater detail than do in vivo scans. Furthermore, these images have no artifacts, which can limit the evaluation of findings in living patients.
The goal of our radiologic-histopathologic correlation study was to establish the relationship between CT findings and the pathologic phases of DAD. We believe that the use of the isolated lung was a reasonable and meaningful means to reach this goal. Finally, we evaluated only DAD induced by hyperoxia. The causes of DAD-related ARDS vary (9,10), so the present study model of hyperoxia-induced DAD alone may not be a sufficient mirror of DAD induced by other causative agents such as sepsis. Further investigation of thin-section CT in living patients with ARDS is necessary not only to determine the CT-histopathologic analysis correlations but also to assess the clinical and prognostic implications of thin-section CT findings. Ideally, such an investigation would be broadened to include the study of DAD-related ARDS in general, as seen in the more common cases of this disease, which are caused by sepsis or drugs.
In conclusion, our use of highly concentrated inspired oxygen in pigs produced progressive lung injury characterized by DAD, which is the pathologic hallmark of ARDS. The thin-section CT findings correlated well with the histologic characteristics and pathologic phases of DAD. The presence of traction bronchiolectasis within areas of increased attenuation at CT was the earliest, most reliable sign of the transition from the exudative phase to the proliferative phase, whereas the early exudative phase of DAD could not be detected by using thin-section CT.Practical applications: The results of the present study provide the histopathologic background of thin-section CT findings and show the usefulness and limitations of CT findings when applied to patients with early DAD-related ARDS. The presence of traction bronchiolectasis within areas of increased attenuation at CT as an initial sign of a proliferative or reparative process may be useful for predicting the response to corticosteroids. Even if the CT scan shows areas of normal attenuation, the presence of the early exudative phase is not excluded.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, M.S., M.A.; study concepts, K. Ichikado, T.J.; study design, K. Ichikado; definition of intellectual content, M.T., H.I., J.I., N.L.M.; literature research, K. Ichikado, M.S., Y.G.; experimental studies, K. Ichikado, Y.G., T.Y., O.H., Y.S.; data acquisition, K. Ichikado, Y.G.; data analysis, K. Iyonaga, T.J., O.H.; statistical analysis, S.T.; manuscript preparation, K. Ichikado; manuscript editing, N.L.M.; manuscript review, T.J., H.I., J.I., N.L.M.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Ichikado, M. Suga, H. Muranaka, Y. Gushima, H. Miyakawa, M. Tsubamoto, T. Johkoh, N. Hirata, T. Yoshinaga, Y. Kinoshita, et al. Prediction of Prognosis for Acute Respiratory Distress Syndrome with Thin-Section CT: Validation in 44 Cases Radiology, December 1, 2005; 238(1): 321 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Ichikado, M. Suga, N. L. Muller, H. Taniguchi, Y. Kondoh, M. Akira, T. Johkoh, N. Mihara, H. Nakamura, M. Takahashi, et al. Acute Interstitial Pneumonia: Comparison of High-Resolution Computed Tomography Findings between Survivors and Nonsurvivors Am. J. Respir. Crit. Care Med., June 1, 2002; 165(11): 1551 - 1556. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P. G. Padley, S. J. Jordan, P. Goldstraw, A. U. Wells, and D. M. Hansell Asymmetric ARDS Following Pulmonary Resection: CT Findings—Initial Observations Radiology, March 21, 2002; (2002) 2232010721. [Abstract] [Full Text]
|
|
|
|
 |
|
 Y. Gushima, K. Ichikado, M. Suga, T. Okamoto, K. Iyonaga, K. Sato, H. Miyakawa, and M. Ando Expression of matrix metalloproteinases in pigs with hyperoxia-induced acute lung injury Eur. Respir. J., November 1, 2001; 18(5): 827 - 837. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Ohno, H. Hatabu, D. Takenaka, S. Adachi, M. Van Cauteren, and K. Sugimura Oxygen-Enhanced MR Ventilation Imaging of the Lung: Preliminary Clinical Experience in 25 Subjects Am. J. Roentgenol., July 1, 2001; 177(1): 185 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P. G. Padley, S. J. Jordan, P. Goldstraw, A. U. Wells, and D. M. Hansell Asymmetric ARDS Following Pulmonary Resection: CT Findings—Initial Observations Radiology, May 1, 2002; 223(2): 468 - 473. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
