Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Seo, J. B.
Right arrow Articles by Chung, J. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Seo, J. B.
Right arrow Articles by Chung, J. H.
(Radiology. 1999;212:88-96.)
© RSNA, 1999

Experimental Studies

Shark Liver Oil–induced Lipoid Pneumonia in Pigs: Correlation of Thin-Section CT and Histopathologic Findings1

Joon Beom Seo, MD 2, Jung-Gi Im, MD, Woo Sun Kim, MD, Chang Kyu Seong, MD, Jae Woo Song, MD and Jin Haeng Chung, MD 3

1 From the Department of Radiology and Institute of Radiation Medicine (J.B.S., J.G.I., W.S.K., C.K.S.), and Department of Pathology (J.H.C.), Seoul National University College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul 110-744, South Korea; and Department of Radiology, Seoul City Boramae Hospital, Seoul (J.W.S.). Received May 27, 1998; revision requested July 16; final revision received November 2; accepted February 12, 1999. Address reprint requests to J.G.I. (e-mail: imjg@radcom.snu.ac.kr).


   Abstract
TOP
 Abstract
Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To evaluate sequential changes in thin-section computed tomographic (CT) findings after inducement of lipoid pneumonia and provide the histopathologic bases of these findings.

MATERIALS AND METHODS: Shark liver oil was administered to 12 sites in seven pigs. Thin-section CT scans were obtained within 1 hour and at 1 week, 2 weeks, 4 weeks, 8 weeks, 12 weeks, and 16 weeks after oil administration. Scans were assessed for opacity, distribution, location at the lobular level, extent, and volume of the lesions. The CT number in consolidation areas also was measured. Findings at CT were correlated with those in the histopathologic specimens.

RESULTS: Diffuse ground-glass opacity was noted on all immediately obtained scans. The opacity of the lesions was highest at 1 week; then it decreased gradually to an area of ground-glass opacity. The extent and volume of the lesions decreased at follow-up CT. Histopathologically, the lesions showed a lobular distribution sharply demarcated from the normal lungs. The lobules of decreased volume showed residual thickening of the alveolar walls with bronchiolectasis and mild collagen deposition of the interlobular septa. Pathologic examination of the low-attenuating consolidation area at CT revealed evidence of partial aeration.

CONCLUSION: Thin-section CT findings of lipoid pneumonia include ground-glass opacity and airspace consolidation, followed by complete or incomplete resolution with volume loss and septal thickening. Low-attenuating consolidation at CT does not always indicate the presence of fat.

Index terms: Computed tomography (CT), experimental, 60.12111, 60.12118 • Computed tomography (CT), thin-section, 60.12118 • Lung, CT, 60.12111, 60.12118 • Lung, diseases, 60.214, 60.253 • Pneumonitis, aspiration, 60.214


   Introduction
TOP
 Abstract
 Introduction
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Exogenous lipoid pneumonia is an uncommon condition that results from aspiration or inhalation of fatlike material. After Laughlen (1) in 1925 described exogenous lipoid pneumonia when he reported discovering oil droplets in the lungs during the autopsies of three children and one adult who had received mineral oil nose drops or oral laxatives, many clinical and experimental studies were conducted until the 1950s (2–9). Lipoid pneumonia has been encountered in patients who used mineral oils such as liquid paraffin, which is used in laxatives and nose drops (6,9–11); animal oils such as cod liver oil, which is commonly given to children (2,5); and vegetable oils such as sesame oil, which is used in medical suspensions (12,13). Several cases of lipoid pneumonia have originated from the use of traditional folk remedies (14–19). Currently, several cases of lipoid pneumonia caused by various materials are reported in the literature annually. Squalene is a lipid derived from shark liver oil. In some Asian countries, squalene is taken as a traditional folk remedy. Similar to the symptoms in most cases of oil aspiration, symptoms of shark liver oil–induced lipoid pneumonia are either absent or nonspecific, and the radiologic findings can simulate those of other diseases (16,19).

Lipoid pneumonia is difficult to diagnose because it mimics various diseases. In chest radiographic studies (8,14,20–25), various findings, from diffuse bilateral infiltration to solitary lung nodules that mimic neoplasm, have been described. Efforts were made to classify these radiographic findings and understand those that manifest as sequential changes after aspiration (20,21,23,24), but these efforts were limited owing to the lack of pathologic correlation. Depiction of low-attenuating consolidation at CT has been suggested as a diagnostic clue in lipoid pneumonia (10,15,16,26,27).

Although lipoid pneumonia is a well-recognized entity, our literature search disclosed that reports on the radiologic-pathologic correlation of this disease are limited (20,22,28). The purpose of this study was to evaluate the sequential changes in thin-section CT findings after inducement of lipoid pneumonia with shark liver oil and provide the histopathologic bases of these findings by correlating them with histopathologic findings.


   MATERIALS AND METHODS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Selection of Animals and Lipid Material
This study was approved by the animal research committee at our institution. Pigs were chosen for this study because their secondary lobules are well developed, similar to those in humans, and the size of their lungs is suitable for both handling and imaging (29). Seven 10–12-week-old Yorkshire pigs (15–20 kg) were used. Shark liver oil (Shark King Capsule; Banner Pharmacaps, Elizabeth, NJ) was selected as the lipid material because of our knowledge of its components, its convenience of preparation, and the recent increase in the number of clinical cases associated with use of this material in Korea. The capsule consists of 99.26% squalene and 0.74% vitamin E. The oil material in the capsule was aspirated with needle puncture, and 10 mL of the substance was prepared for each site.

Animal Preparation
The pigs were not allowed to eat for 6 hours before anesthesia was induced, and they were injected intramuscularly with 0.5 mg of atropine (Daehan Atropine; Daehan Pharmaceutical, Seoul, Korea) just before anesthesia. Anesthesia was induced with an intramuscular injection of 10 mg/kg ketamine hydrochloride (Ketara; Yuhan Yanhang, Seoul, Korea) and 4 mg/kg xylazine hydrochloride (Rompun; Bayer Korea, Seoul, Korea), and it was maintained with an intravenous injection of 5–10 mg/kg thiopental sodium (Pentothal; Choong Wae Pharmacy, Seoul, Korea). The animals were intubated with a 6.5-mm inside-diameter endotracheal tube after being paralyzed with an intravenous injection of 1 mg/kg succinylcholine chloride (Quelin; Abbott Laboratories, Chicago, Ill).

Administration of Shark Liver Oil
With fluoroscopic guidance, we introduced a 5.5-F balloon catheter (Selection Multi-Catheter; Clinical Supply, Hashima-gun, Japan) at the level of the lower lobar bronchus by using a guide wire with the pig in a semierect prone position. The balloon was inflated to prevent spillover of oil into the trachea or contralateral lung. Five milliliters of the prepared oil was injected slowly by hand. The animals were hyperventilated with a ventilator bag (Ambu bag; Ambu, Linthicum, Md) after the balloon was deflated to promote migration of the oil into the distal airspace. After keeping the animals in a semierect prone position for 15–20 minutes, we repeated the above procedure. Thus, a total dose of 10 mL was injected per site. In the first two pigs, oil was administered to the unilateral lung to compare it with the contralateral lung. The oil was administered bilaterally in the remaining five pigs.

CT Scanning
Contiguous thick-section (ie, 5-mm collimation, 5-mm intervals, standard algorithm, through the thorax) and thin-section CT scans were obtained immediately (ie, within 1 hour) and at 1 week after the administration of oil in 12 sites in seven pigs, at 2 weeks in 10 sites in six pigs, at 4 weeks in eight sites in five pigs, at 8 weeks in six sites in four pigs, at 12 weeks in four sites in three pigs, and at 16 weeks in two sites in two pigs. The CT scans were obtained with a CT HiSpeed Advantage scanning system (GE Medical Systems, Milwaukee, Wis). Initially, thick-section CT scans were obtained for overall assessment of the lung lesions and selection of the appropriate level for thin-section CT. Thin-section CT scans were obtained at contiguous 30-mm lengths in a spiral mode with 1-mm collimation and 1-mm table speed and reconstructed at 1-mm intervals by using a high-spatial-frequency algorithm. The scanning parameters were 120 kVp, 230 mA, a 1-second rotation time, a 15–17-cm field of view, and a 512 x 512 matrix. At follow-up CT, the level of the thin-section CT scan was adjusted to that in the initial study.

All pigs were examined in the prone position. Both mediastinal window (width, 400 HU; level, 10 HU) and lung window (width, 1,500 HU; level, -700 HU) scans were obtained. All scans were obtained during suspended respiration at resting end expiration (functional residual capacity), which was induced by muscular paralysis with an intravenous injection of 1 mg/kg succinylcholine chloride, after hyperventilation with a ventilator bag, because suspended respiration at functional residual capacity was easy to maintain in animals with muscular paralysis. The animals were bred during follow-up periods. No complications such as infection or aspiration occurred. None of the pigs involved in this study died unexpectedly.

Lung Fixation, Radiography of Specimens, and Histologic Section Preparation
After all scheduled follow-up CT scans were obtained, the pigs were sacrificed by intravenously administering an excessive dose of thiopenthal sodium. The lungs were removed from the thorax, with care taken not to tear the visceral pleura. This procedure was performed within 2 hours after the final CT scans were obtained. Lung fixation and inflation was performed by using the method of Markarian and Dailey (30). A total of 12 sites of lipoid pneumonia were prepared: 1 week after oil administration, two sites from one pig were prepared; at 2 weeks, two sites from one pig; at 4 weeks, two sites from one pig; at 8 weeks, two sites from one pig; at 12 weeks, two sites from one pig; and at 16 weeks, two sites from two pigs.

Following fixation, the specimens were cut into 10-mm slices that corresponded to the section at in vivo thin-section CT, and radiographs of each specimen slice were obtained. A soft x-ray radiographic unit (Faxitron 43805N; Hewlett-Packard, Sunnyvale, Calif) and X-Omat V film (Kodak, Rochester, NY) were used to perform radiography of the specimens. Subsequently, thin 2–3-mm sections were obtained from the thick 10-mm slices, and radiographs of the specimens were obtained with technical parameters of 25 kVp, 3 mA, and 3–4-minute exposure. Six-micron-thick, 12 x 8-cm giant microscopic section slides from the selected 2–3-mm-thick sections were prepared with hematoxylin-eosin stain. Additional giant section slides from 12-week follow-up specimens were prepared with Masson trichrome stain to evaluate the deposition of collagen fibers.

Analyses of Thin-Section CT Scans
The CT scans were reviewed by three radiologists (J.G.I., J.W.S., J.B.S.). The lung window and mediastinal window images were analyzed together. Conclusions were made in consensus. The CT scans were assessed specifically for the intensity, distribution, and extent of opacification. The opacification was classified as ground-glass opacity or airspace consolidation. The distribution of the opacification was classified as either dependent or nondependent. Because the CT scans were obtained with the patient in the prone position, the anterior portion of the lung corresponded to the dependent area, and the posterior portion, to the nondependent area. The location of the lesion at the lobular level also was assessed. The opacity, extent of opacification, and volume of the affected area on follow-up CT scans were evaluated and compared with the findings in the same region on initial CT scans.

To measure the CT number in the consolidated lung, the data were transferred to an Advantage Windows workstation (GE Medical Systems, Milwaukee, Wis). The scans were magnified four times. Measurements of the consolidation area were obtained on mediastinal window images. Care was taken to avoid the air-attenuation area within the region-of-interest circles. We could measure the attenuation on all CT scans obtained 1–8 weeks after oil administration and that in five sites on the scans obtained immediately after administration, where the consolidation areas were detected. We regarded -10 HU as the threshold value of low-attenuating consolidation that was sufficient for the detection of fat on CT scans. The attenuation values of subcutaneous fat, lobar arteries, and oil in the squalene capsule also were measured for comparison.

The findings on thin-section CT scans were correlated with those on radiographs of the specimens and on giant microscopic slide specimens by two radiologists (J.G.I., J.B.S.) and one pathologist (J.H.C.) in consensus.


   RESULTS
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Thin-Section CT Findings
Areas of diffuse centrilobular or panlobular ground-glass opacity were seen on the CT scans obtained immediately after the administration of oil (Fig 1b). An area of low-attenuating consolidation in the dependent (anterior) portion was noted in five of 12 sites. One week after oil administration, consolidation in the dependent area was seen in all sites (Fig 1c). The degree of opacity decreased on CT scans at 2–4 weeks after the oil administration (Fig 1d). The opacity changed mostly to ground-glass opacity at 12–16 weeks after the oil administration (Table 1), with panlobular distribution in the dependent area and centrilobular distribution in the nondependent area (Fig 1). The extent of the lesions decreased with time, especially in the nondependent area. Thickening of the interlobular septa in the areas of intermediate and ground-glass opacity was observed at the 2-, 4-, and 8-week follow-up CT examinations (Table 1; Figs 1d, 2a). Smooth thickening of the interlobular septa was observed in the cleared lung on CT scans obtained 4–16 weeks after oil administration (Figs 1d, 1e, 3a). The affected lung decreased in volume gradually, with pleural retraction and crowding of the bronchovascular bundle and bronchiolectasis (Fig 1d–f).



View larger version (116K):
[in this window]
[in a new window]
  		Figure 1a. Serial axial CT scans at the level of the lower lobe of the left lung in a pig in the prone position obtained (a) before and (b–e) after endobronchial administration of shark liver oil, and (f) radiograph of a sliced specimen obtained at the same level. (a) CT scan obtained with 5-mm collimation before the oil injection shows the lung is clear except for a nonspecific ill-defined opacity (arrow) in the nondependent area and a peribronchial infiltration (arrowhead) in the dependent area, into which oil was not injected. (b) CT scan obtained immediately after oil administration shows diffuse ground-glass opacity in the lung. The areas of opacification are distributed mainly in the dependent (anterior) portion. Note the segmental or lobular distribution of the lesions (open arrows). Areas of ill-defined centrilobular ground-glass opacity (solid arrows) are noted in the nondependent area. (c) CT scan obtained 1 week after oil administration shows airspace consolidation in the dependent area. Note also the areas of centrilobular lesions (arrowheads) in the nondependent area. (d) CT scan obtained 4 weeks after oil administration shows a decreased degree of opacity. The opacity obscures the small lobular vessels but not the larger pulmonary vessels (arrowheads). The fine linear structures in the area of opacification represent thickening of interlobular septa (thin black arrows). Note the decreased extent of centrilobular opacity with areas of residual peribronchiolar opacification (curved arrow). Smooth thickening of the interlobular septum (open arrow) between the normal lobules also is noted. Note the straightening of the interlobar fissure (solid white arrows) compared with the bulging appearance of the corresponding interlobar fissure in b and c; the change in the fissure from bulging to straight suggests volume loss in the affected lung. (e) CT scan obtained 12 weeks after oil administration shows further volume decrease in the affected lung. Note the crowding of bronchovascular bundles within the lesion and the retraction of the interlobar fissure (open arrows). The attenuation of the lesion has decreased. The bronchiole (black arrow) in the area of opacification is irregular and slightly dilated. Smooth thickening of an interlobular septum (white arrow) is still seen. (f) Contact radiograph of the resected and sliced lung at a level similar to that in e shows retraction of the lung surface (curved arrows), which represents a volume decrease caused by lipoid pneumonia. Compared with the size of the adjacent normal lobule (white arrowheads), that of the affected lobule (black arrowheads) is markedly smaller.

 


View larger version (120K):
[in this window]
[in a new window]
  		Figure 1b. Serial axial CT scans at the level of the lower lobe of the left lung in a pig in the prone position obtained (a) before and (b–e) after endobronchial administration of shark liver oil, and (f) radiograph of a sliced specimen obtained at the same level. (a) CT scan obtained with 5-mm collimation before the oil injection shows the lung is clear except for a nonspecific ill-defined opacity (arrow) in the nondependent area and a peribronchial infiltration (arrowhead) in the dependent area, into which oil was not injected. (b) CT scan obtained immediately after oil administration shows diffuse ground-glass opacity in the lung. The areas of opacification are distributed mainly in the dependent (anterior) portion. Note the segmental or lobular distribution of the lesions (open arrows). Areas of ill-defined centrilobular ground-glass opacity (solid arrows) are noted in the nondependent area. (c) CT scan obtained 1 week after oil administration shows airspace consolidation in the dependent area. Note also the areas of centrilobular lesions (arrowheads) in the nondependent area. (d) CT scan obtained 4 weeks after oil administration shows a decreased degree of opacity. The opacity obscures the small lobular vessels but not the larger pulmonary vessels (arrowheads). The fine linear structures in the area of opacification represent thickening of interlobular septa (thin black arrows). Note the decreased extent of centrilobular opacity with areas of residual peribronchiolar opacification (curved arrow). Smooth thickening of the interlobular septum (open arrow) between the normal lobules also is noted. Note the straightening of the interlobar fissure (solid white arrows) compared with the bulging appearance of the corresponding interlobar fissure in b and c; the change in the fissure from bulging to straight suggests volume loss in the affected lung. (e) CT scan obtained 12 weeks after oil administration shows further volume decrease in the affected lung. Note the crowding of bronchovascular bundles within the lesion and the retraction of the interlobar fissure (open arrows). The attenuation of the lesion has decreased. The bronchiole (black arrow) in the area of opacification is irregular and slightly dilated. Smooth thickening of an interlobular septum (white arrow) is still seen. (f) Contact radiograph of the resected and sliced lung at a level similar to that in e shows retraction of the lung surface (curved arrows), which represents a volume decrease caused by lipoid pneumonia. Compared with the size of the adjacent normal lobule (white arrowheads), that of the affected lobule (black arrowheads) is markedly smaller.

 


View larger version (108K):
[in this window]
[in a new window]
  		Figure 1c. Serial axial CT scans at the level of the lower lobe of the left lung in a pig in the prone position obtained (a) before and (b–e) after endobronchial administration of shark liver oil, and (f) radiograph of a sliced specimen obtained at the same level. (a) CT scan obtained with 5-mm collimation before the oil injection shows the lung is clear except for a nonspecific ill-defined opacity (arrow) in the nondependent area and a peribronchial infiltration (arrowhead) in the dependent area, into which oil was not injected. (b) CT scan obtained immediately after oil administration shows diffuse ground-glass opacity in the lung. The areas of opacification are distributed mainly in the dependent (anterior) portion. Note the segmental or lobular distribution of the lesions (open arrows). Areas of ill-defined centrilobular ground-glass opacity (solid arrows) are noted in the nondependent area. (c) CT scan obtained 1 week after oil administration shows airspace consolidation in the dependent area. Note also the areas of centrilobular lesions (arrowheads) in the nondependent area. (d) CT scan obtained 4 weeks after oil administration shows a decreased degree of opacity. The opacity obscures the small lobular vessels but not the larger pulmonary vessels (arrowheads). The fine linear structures in the area of opacification represent thickening of interlobular septa (thin black arrows). Note the decreased extent of centrilobular opacity with areas of residual peribronchiolar opacification (curved arrow). Smooth thickening of the interlobular septum (open arrow) between the normal lobules also is noted. Note the straightening of the interlobar fissure (solid white arrows) compared with the bulging appearance of the corresponding interlobar fissure in b and c; the change in the fissure from bulging to straight suggests volume loss in the affected lung. (e) CT scan obtained 12 weeks after oil administration shows further volume decrease in the affected lung. Note the crowding of bronchovascular bundles within the lesion and the retraction of the interlobar fissure (open arrows). The attenuation of the lesion has decreased. The bronchiole (black arrow) in the area of opacification is irregular and slightly dilated. Smooth thickening of an interlobular septum (white arrow) is still seen. (f) Contact radiograph of the resected and sliced lung at a level similar to that in e shows retraction of the lung surface (curved arrows), which represents a volume decrease caused by lipoid pneumonia. Compared with the size of the adjacent normal lobule (white arrowheads), that of the affected lobule (black arrowheads) is markedly smaller.

 


View larger version (115K):
[in this window]
[in a new window]
  		Figure 1d. Serial axial CT scans at the level of the lower lobe of the left lung in a pig in the prone position obtained (a) before and (b–e) after endobronchial administration of shark liver oil, and (f) radiograph of a sliced specimen obtained at the same level. (a) CT scan obtained with 5-mm collimation before the oil injection shows the lung is clear except for a nonspecific ill-defined opacity (arrow) in the nondependent area and a peribronchial infiltration (arrowhead) in the dependent area, into which oil was not injected. (b) CT scan obtained immediately after oil administration shows diffuse ground-glass opacity in the lung. The areas of opacification are distributed mainly in the dependent (anterior) portion. Note the segmental or lobular distribution of the lesions (open arrows). Areas of ill-defined centrilobular ground-glass opacity (solid arrows) are noted in the nondependent area. (c) CT scan obtained 1 week after oil administration shows airspace consolidation in the dependent area. Note also the areas of centrilobular lesions (arrowheads) in the nondependent area. (d) CT scan obtained 4 weeks after oil administration shows a decreased degree of opacity. The opacity obscures the small lobular vessels but not the larger pulmonary vessels (arrowheads). The fine linear structures in the area of opacification represent thickening of interlobular septa (thin black arrows). Note the decreased extent of centrilobular opacity with areas of residual peribronchiolar opacification (curved arrow). Smooth thickening of the interlobular septum (open arrow) between the normal lobules also is noted. Note the straightening of the interlobar fissure (solid white arrows) compared with the bulging appearance of the corresponding interlobar fissure in b and c; the change in the fissure from bulging to straight suggests volume loss in the affected lung. (e) CT scan obtained 12 weeks after oil administration shows further volume decrease in the affected lung. Note the crowding of bronchovascular bundles within the lesion and the retraction of the interlobar fissure (open arrows). The attenuation of the lesion has decreased. The bronchiole (black arrow) in the area of opacification is irregular and slightly dilated. Smooth thickening of an interlobular septum (white arrow) is still seen. (f) Contact radiograph of the resected and sliced lung at a level similar to that in e shows retraction of the lung surface (curved arrows), which represents a volume decrease caused by lipoid pneumonia. Compared with the size of the adjacent normal lobule (white arrowheads), that of the affected lobule (black arrowheads) is markedly smaller.

 


View larger version (113K):
[in this window]
[in a new window]
  		Figure 1e. Serial axial CT scans at the level of the lower lobe of the left lung in a pig in the prone position obtained (a) before and (b–e) after endobronchial administration of shark liver oil, and (f) radiograph of a sliced specimen obtained at the same level. (a) CT scan obtained with 5-mm collimation before the oil injection shows the lung is clear except for a nonspecific ill-defined opacity (arrow) in the nondependent area and a peribronchial infiltration (arrowhead) in the dependent area, into which oil was not injected. (b) CT scan obtained immediately after oil administration shows diffuse ground-glass opacity in the lung. The areas of opacification are distributed mainly in the dependent (anterior) portion. Note the segmental or lobular distribution of the lesions (open arrows). Areas of ill-defined centrilobular ground-glass opacity (solid arrows) are noted in the nondependent area. (c) CT scan obtained 1 week after oil administration shows airspace consolidation in the dependent area. Note also the areas of centrilobular lesions (arrowheads) in the nondependent area. (d) CT scan obtained 4 weeks after oil administration shows a decreased degree of opacity. The opacity obscures the small lobular vessels but not the larger pulmonary vessels (arrowheads). The fine linear structures in the area of opacification represent thickening of interlobular septa (thin black arrows). Note the decreased extent of centrilobular opacity with areas of residual peribronchiolar opacification (curved arrow). Smooth thickening of the interlobular septum (open arrow) between the normal lobules also is noted. Note the straightening of the interlobar fissure (solid white arrows) compared with the bulging appearance of the corresponding interlobar fissure in b and c; the change in the fissure from bulging to straight suggests volume loss in the affected lung. (e) CT scan obtained 12 weeks after oil administration shows further volume decrease in the affected lung. Note the crowding of bronchovascular bundles within the lesion and the retraction of the interlobar fissure (open arrows). The attenuation of the lesion has decreased. The bronchiole (black arrow) in the area of opacification is irregular and slightly dilated. Smooth thickening of an interlobular septum (white arrow) is still seen. (f) Contact radiograph of the resected and sliced lung at a level similar to that in e shows retraction of the lung surface (curved arrows), which represents a volume decrease caused by lipoid pneumonia. Compared with the size of the adjacent normal lobule (white arrowheads), that of the affected lobule (black arrowheads) is markedly smaller.

 


View larger version (128K):
[in this window]
[in a new window]
  		Figure 1f. Serial axial CT scans at the level of the lower lobe of the left lung in a pig in the prone position obtained (a) before and (b–e) after endobronchial administration of shark liver oil, and (f) radiograph of a sliced specimen obtained at the same level. (a) CT scan obtained with 5-mm collimation before the oil injection shows the lung is clear except for a nonspecific ill-defined opacity (arrow) in the nondependent area and a peribronchial infiltration (arrowhead) in the dependent area, into which oil was not injected. (b) CT scan obtained immediately after oil administration shows diffuse ground-glass opacity in the lung. The areas of opacification are distributed mainly in the dependent (anterior) portion. Note the segmental or lobular distribution of the lesions (open arrows). Areas of ill-defined centrilobular ground-glass opacity (solid arrows) are noted in the nondependent area. (c) CT scan obtained 1 week after oil administration shows airspace consolidation in the dependent area. Note also the areas of centrilobular lesions (arrowheads) in the nondependent area. (d) CT scan obtained 4 weeks after oil administration shows a decreased degree of opacity. The opacity obscures the small lobular vessels but not the larger pulmonary vessels (arrowheads). The fine linear structures in the area of opacification represent thickening of interlobular septa (thin black arrows). Note the decreased extent of centrilobular opacity with areas of residual peribronchiolar opacification (curved arrow). Smooth thickening of the interlobular septum (open arrow) between the normal lobules also is noted. Note the straightening of the interlobar fissure (solid white arrows) compared with the bulging appearance of the corresponding interlobar fissure in b and c; the change in the fissure from bulging to straight suggests volume loss in the affected lung. (e) CT scan obtained 12 weeks after oil administration shows further volume decrease in the affected lung. Note the crowding of bronchovascular bundles within the lesion and the retraction of the interlobar fissure (open arrows). The attenuation of the lesion has decreased. The bronchiole (black arrow) in the area of opacification is irregular and slightly dilated. Smooth thickening of an interlobular septum (white arrow) is still seen. (f) Contact radiograph of the resected and sliced lung at a level similar to that in e shows retraction of the lung surface (curved arrows), which represents a volume decrease caused by lipoid pneumonia. Compared with the size of the adjacent normal lobule (white arrowheads), that of the affected lobule (black arrowheads) is markedly smaller.

 

View this table:
[in this window]
[in a new window]
  		TABLE 1. Changes in Thin-Section CT Findings after Endobronchial Administration of Shark Liver Oil
 


View larger version (117K):
[in this window]
[in a new window]
  		Figure 2a. Histopathologic bases of thickening of interlobular septa at CT in the early phase (2–8 weeks after oil administration) of lipoid pneumonia in a pig. (a) Axial CT scan at the level of the lower lobe of the left lung obtained 2 weeks after oil administration shows airspace consolidation. Note the areas of faint linear opacification (arrows) in the consolidated lung. Ill-defined centrilobular opacity (arrowhead) is seen in the nondependent area. (b) Corresponding photomicrograph demonstrates thickened interlobular septa (arrows) with cellular infiltrates of lymphocytes, histiocytes, and young fibroblasts, and telangiectatic change in the capillaries. Note the alveolar wall thickening and the inflammatory exudate filling the alveolar spaces (arrowheads). (Hematoxylin-eosin stain; original magnification, x10.)

 


View larger version (182K):
[in this window]
[in a new window]
  		Figure 2b. Histopathologic bases of thickening of interlobular septa at CT in the early phase (2–8 weeks after oil administration) of lipoid pneumonia in a pig. (a) Axial CT scan at the level of the lower lobe of the left lung obtained 2 weeks after oil administration shows airspace consolidation. Note the areas of faint linear opacification (arrows) in the consolidated lung. Ill-defined centrilobular opacity (arrowhead) is seen in the nondependent area. (b) Corresponding photomicrograph demonstrates thickened interlobular septa (arrows) with cellular infiltrates of lymphocytes, histiocytes, and young fibroblasts, and telangiectatic change in the capillaries. Note the alveolar wall thickening and the inflammatory exudate filling the alveolar spaces (arrowheads). (Hematoxylin-eosin stain; original magnification, x10.)

 


View larger version (126K):
[in this window]
[in a new window]
  		Figure 3a. Histopathologic bases of interlobular septa seen on CT scans in the later phase (4–16 weeks) of lipoid pneumonia in a pig. (a) Axial CT scan at the level of the lower lobe of the right lung obtained 8 weeks after endobronchial administration of shark liver oil shows smooth thickening of the interlobular septa (boxed area and arrows) in the aerated portion of the lung. Also note the lobular airspace consolidation with volume loss. (b) Photomicrograph of a resected specimen from the boxed area in a demonstrates widening of the interlobular septum with ectatic lymphatic channels and capillaries (arrows); there are residual cellular infiltrates in the septum. (Hematoxylin-eosin stain; original magnification, x40.) The residual thickening of the alveolar walls explains the ground-glass opacity seen at CT (boxed area in a). Photomicrograph of the specimen prepared with Masson stain (not shown) showed wavy, green-stained linear structures in the septa, which confirmed collagen fibers.

 


View larger version (161K):
[in this window]
[in a new window]
  		Figure 3b. Histopathologic bases of interlobular septa seen on CT scans in the later phase (4–16 weeks) of lipoid pneumonia in a pig. (a) Axial CT scan at the level of the lower lobe of the right lung obtained 8 weeks after endobronchial administration of shark liver oil shows smooth thickening of the interlobular septa (boxed area and arrows) in the aerated portion of the lung. Also note the lobular airspace consolidation with volume loss. (b) Photomicrograph of a resected specimen from the boxed area in a demonstrates widening of the interlobular septum with ectatic lymphatic channels and capillaries (arrows); there are residual cellular infiltrates in the septum. (Hematoxylin-eosin stain; original magnification, x40.) The residual thickening of the alveolar walls explains the ground-glass opacity seen at CT (boxed area in a). Photomicrograph of the specimen prepared with Masson stain (not shown) showed wavy, green-stained linear structures in the septa, which confirmed collagen fibers.

 
The measured CT attenuation value of the oil in the capsule ranged from -210 to -200 HU, whereas that of the subcutaneous fat and lobar arteries ranged from -120 to -40 HU and from 0 to 40 HU, respectively. The measured CT attenuation value of the consolidated lung ranged from -150 to 50 HU (Fig 4a). In all sites, within the airspace consolidation areas, an area of low attenuation below -10 HU was identified. The attenuation value of the low-attenuating portion of the consolidated lung measured immediately after oil administration was markedly low, with a range of -150 to -120 HU. The attenuation of the consolidation was highest on the CT scans obtained 1 week after oil administration and reflective of the extensive inflammatory process involving the alveolar spaces and alveolar wall.



View larger version (141K):
[in this window]
[in a new window]
  		Figure 4a. (a, b) Axial CT scans and (c) photomicrograph of shark liver oil–induced lipoid pneumonia in the earlier phase obtained in a pig 1 week after oil administration. (a) The measured CT numbers of the inferior vena cava (circle 1), consolidated lung (circle 2), and subcutaneous fat (circle 3) are 26 HU, -17 HU, and -65 HU, respectively. (b) The attenuation of the consolidated lung in the dependent area is low (arrow), similar to that of the subcutaneous fat. The measured CT number is -85 HU. (c) Photomicrograph of the area in b shows intraalveolar lipid droplets (arrows) surrounded by numerous macrophages and polymorphonuclear lymphocytes filling most of the alveolar spaces. In this case, one can regard the low attenuation of the consolidation in b as fat without a CT number measurement. Note the widening of the alveolar walls (arrowheads), with infiltration by inflammatory cells. This finding explains the nature of the consolidation in a and b. (Hematoxylin-eosin stain; original magnification, x400.)

 


View larger version (135K):
[in this window]
[in a new window]
  		Figure 4b. (a, b) Axial CT scans and (c) photomicrograph of shark liver oil–induced lipoid pneumonia in the earlier phase obtained in a pig 1 week after oil administration. (a) The measured CT numbers of the inferior vena cava (circle 1), consolidated lung (circle 2), and subcutaneous fat (circle 3) are 26 HU, -17 HU, and -65 HU, respectively. (b) The attenuation of the consolidated lung in the dependent area is low (arrow), similar to that of the subcutaneous fat. The measured CT number is -85 HU. (c) Photomicrograph of the area in b shows intraalveolar lipid droplets (arrows) surrounded by numerous macrophages and polymorphonuclear lymphocytes filling most of the alveolar spaces. In this case, one can regard the low attenuation of the consolidation in b as fat without a CT number measurement. Note the widening of the alveolar walls (arrowheads), with infiltration by inflammatory cells. This finding explains the nature of the consolidation in a and b. (Hematoxylin-eosin stain; original magnification, x400.)

 


View larger version (174K):
[in this window]
[in a new window]
  		Figure 4c. (a, b) Axial CT scans and (c) photomicrograph of shark liver oil–induced lipoid pneumonia in the earlier phase obtained in a pig 1 week after oil administration. (a) The measured CT numbers of the inferior vena cava (circle 1), consolidated lung (circle 2), and subcutaneous fat (circle 3) are 26 HU, -17 HU, and -65 HU, respectively. (b) The attenuation of the consolidated lung in the dependent area is low (arrow), similar to that of the subcutaneous fat. The measured CT number is -85 HU. (c) Photomicrograph of the area in b shows intraalveolar lipid droplets (arrows) surrounded by numerous macrophages and polymorphonuclear lymphocytes filling most of the alveolar spaces. In this case, one can regard the low attenuation of the consolidation in b as fat without a CT number measurement. Note the widening of the alveolar walls (arrowheads), with infiltration by inflammatory cells. This finding explains the nature of the consolidation in a and b. (Hematoxylin-eosin stain; original magnification, x400.)

 
Correlation of Thin-Section CT with Radiography of Pathologic Specimens and Histopathology
The histopathologic findings of airspace consolidation that were visible at 1- and 2-week follow-up CT were exudate of lipid droplets, lipid-laden macrophages, and polymorphonuclear leukocytes in the alveolar space, along with thickening of the alveolar walls with cellular infiltrates (Table 2, Fig 4c). Areas of ground-glass opacity on CT scans were seen as areas of fine reticular opacity on radiographs of the specimens, which corresponded to the "washout" of alveolar space with a variable degree of residual intraalveolar exudate and the thickened alveolar wall with residual cellular infiltrates and activation of pneumonocytes (Fig 2b). Thickening of the alveolar wall decreased in severity over time (Fig 3b), which explains the gradual decrease in the intensity of the opacification on sequential CT scans.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Radiologic-Histopathologic Correlation of Shark Liver Oil–induced Lipoid Pneumonia
 
The well-demarcated lobular opacity at CT corresponded to a lobular distribution of cellular infiltrates, whereas the centrilobular opacity in the nondependent area at CT and radiography of the specimens corresponded to peribronchiolar infiltration by inflammatory cells (Fig 5).



View larger version (118K):
[in this window]
[in a new window]
  		Figure 5a. Lobular distribution of shark liver oil–induced lipoid pneumonia in a pig. (a) Axial CT scan of the right lung obtained 1 week after administration of shark liver oil demonstrates airspace consolidation (arrows) with lobular or segmental distribution in the dependent area and areas of centrilobular opacity (arrowheads) in the nondependent area. (b) Close-up view of the corresponding area on the contact radiograph of a specimen from the lung shows a lobule (arrow) with centrilobular peribronchiolar airspace consolidation. Also note the consolidation of the surrounding lung. (c) Photomicrograph of the area in b shows centrilobular airspace consolidation (arrow) with cellular infiltrates adjacent to the bronchiole. The adjacent cellular infiltrates occupying whole lobules correspond to the lobular or segmental consolidation in a and b. (Hematoxylin-eosin stain; original magnification, x4.)

 


View larger version (140K):
[in this window]
[in a new window]
  		Figure 5b. Lobular distribution of shark liver oil–induced lipoid pneumonia in a pig. (a) Axial CT scan of the right lung obtained 1 week after administration of shark liver oil demonstrates airspace consolidation (arrows) with lobular or segmental distribution in the dependent area and areas of centrilobular opacity (arrowheads) in the nondependent area. (b) Close-up view of the corresponding area on the contact radiograph of a specimen from the lung shows a lobule (arrow) with centrilobular peribronchiolar airspace consolidation. Also note the consolidation of the surrounding lung. (c) Photomicrograph of the area in b shows centrilobular airspace consolidation (arrow) with cellular infiltrates adjacent to the bronchiole. The adjacent cellular infiltrates occupying whole lobules correspond to the lobular or segmental consolidation in a and b. (Hematoxylin-eosin stain; original magnification, x4.)

 


View larger version (163K):
[in this window]
[in a new window]
  		Figure 5c. Lobular distribution of shark liver oil–induced lipoid pneumonia in a pig. (a) Axial CT scan of the right lung obtained 1 week after administration of shark liver oil demonstrates airspace consolidation (arrows) with lobular or segmental distribution in the dependent area and areas of centrilobular opacity (arrowheads) in the nondependent area. (b) Close-up view of the corresponding area on the contact radiograph of a specimen from the lung shows a lobule (arrow) with centrilobular peribronchiolar airspace consolidation. Also note the consolidation of the surrounding lung. (c) Photomicrograph of the area in b shows centrilobular airspace consolidation (arrow) with cellular infiltrates adjacent to the bronchiole. The adjacent cellular infiltrates occupying whole lobules correspond to the lobular or segmental consolidation in a and b. (Hematoxylin-eosin stain; original magnification, x4.)

 
Thickened interlobular septa in the area of opacification corresponded to collagen fiber deposition with infiltration by lymphocytes, young fibroblasts, and histiocytes, with telangiectatic change in the capillaries of the septa (Fig 2b). The smooth thickening of the septa surrounding the normal lung during the late phase (4–16 weeks after oil administration) corresponded to dilated lymphatics and capillaries and residual cellular infiltrates (Table 2, Fig 3b). Histologic examination of specimens prepared with Masson trichrome stain revealed deposition of collagen fibers in the septa.

The decreased volume of the involved lung at CT corresponded to lobular volume loss with retraction of the overlying pleura and thickening of the interlobular septa at radiography of the specimens (Fig 6a). Histologic examination with Masson trichrome stain revealed evidence of fibrosis in the interlobular septa and pleura. There was no area of collagen fiber deposition in the alveolar wall. Areas of lobular collapse were seen on 8- and 12-week follow-up histologic specimens (Fig 6b).



View larger version (164K):
[in this window]
[in a new window]
  		Figure 6a. (a) Contact radiograph (axial plane) and (b) photomicrograph of a porcine lung specimen show the bases of volume decreases in shark liver oil–induced pneumonia at CT. (a) Close-up view on the contact radiograph of the specimen obtained 4 weeks after endobronchial administration of oil shows the decreased size of the pulmonary lobule (arrowheads) in the affected lung compared with the size of that in the adjacent normal lung (open arrows). Also note the retraction of the surface (curved arrow) of the affected lobule. The increased opacity was due to residual wall thickening of the alveoli at histologic examination (see Fig 3b). (b) Photomicrograph of a specimen from the pig, which was sacrificed 8 weeks after oil administration, shows a diamond-shaped area (arrow) demarcated sharply by interlobular septa. This area represents total collapse of a secondary pulmonary lobule. (Hematoxylin-eosin stain; original magnification, x10.)

 


View larger version (159K):
[in this window]
[in a new window]
  		Figure 6b. (a) Contact radiograph (axial plane) and (b) photomicrograph of a porcine lung specimen show the bases of volume decreases in shark liver oil–induced pneumonia at CT. (a) Close-up view on the contact radiograph of the specimen obtained 4 weeks after endobronchial administration of oil shows the decreased size of the pulmonary lobule (arrowheads) in the affected lung compared with the size of that in the adjacent normal lung (open arrows). Also note the retraction of the surface (curved arrow) of the affected lobule. The increased opacity was due to residual wall thickening of the alveoli at histologic examination (see Fig 3b). (b) Photomicrograph of a specimen from the pig, which was sacrificed 8 weeks after oil administration, shows a diamond-shaped area (arrow) demarcated sharply by interlobular septa. This area represents total collapse of a secondary pulmonary lobule. (Hematoxylin-eosin stain; original magnification, x10.)

 
The area of low attenuation on CT scans contained both fat and partly aerated alveoli at pathologic examination. However, the aeration was not seen at CT, even after magnification and modification of the window settings on the workstation.


   DISCUSSION
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In our study, the earliest CT finding of lipoid pneumonia was diffuse ground-glass opacity, which was prominent in the dependent area and had a centrilobular or panlobular distribution (Fig 1b). The centrilobular opacities were due to peribronchiolar airspace consolidation, which may correspond to the "acinar pattern" on radiographs previously described by Weill et al (22), who produced the "rosette and stippled" patterns of lipoid pneumonia by using intrabronchial instillation of mineral oil in anesthetized dogs and in clinical cases.

The airspace consolidation on CT scans at 1 week was due to the filling of the alveolar space with exudate and inflammatory cells, which is associated with thickening of the alveolar wall (Fig 4c). The degree of opacification decreased on 2- and 4-week follow-up studies and changed to ground-glass opacity. Histologically, a marked decrease in the extent of intraalveolar lipid droplets, lipid-laden macrophages, and inflammatory cells was noted. These changes were considered to be due to expectoration of lipid droplets and inflammatory cells and drainage through the lymphatic vessels.

The thickening of interlobular septa in the opacified area of the lungs in our study was mainly due to cellular infiltrates (Fig 2b). The smooth septal thickening surrounding the resolved lung on 4-week or later follow-up CT scans (Fig 3a) was due to ectatic changes in the lymphatics and capillaries, with residual cellular infiltrates and fibrosis (Fig 3b). We suggest that this finding may reflect lymphostasis induced by increased absorption of lipid droplets or lipid-laden macrophages, because the lymphatic channels are believed to be the route of clearance of exudates accumulated in the alveolar and interstitial spaces (2,3,6,7,31). Another possible theory is the proximal obstruction of lymphatics by large lipid droplets. Brody and Levin (28) reported interlobular septal thickening that manifested as Kerley B lines on the chest radiographs obtained in four patients with pathologically confirmed lipoid pneumonia.

The combination of ground-glass opacity and thickened interlobular septa can mimic alveolar proteinosis, which is a form of endogenous lipoid pneumonia (Fig 2a) (32). Recently, Lee et al (33) and Franquet et al (34) reported the crazy-paving pattern at CT in patients with exogenous lipoid pneumonia. In our experimental study, the partial filling of alveolar space with intraalveolar exudate, thickening of the alveolar wall with cellular infiltrates, and activation of pneumonocytes, combined with the thickening of interlobular septa with dilatation of the lymphatics, cellular infiltrates, and a varied degree of fibrosis were the histopathologic bases of the crazy-paving pattern at CT (Fig 2b).

The decreased volume of the affected lung on CT scans during the later stage (ie, 14–16 weeks after oil administration) was due to fibrosis of the interlobular septa and pleura, with residual thickening of the alveolar wall without fibrosis (Fig 3b). Another explanation for the volume decrease was lobular collapse, which was detected in 8- and 12-week follow-up specimens (Fig 6b).

In 1932, Pierson (20) performed a systematic evaluation of radiographic findings by correlating them with clinical data and performing a gross inspection of autopsy specimens. He reported that pathologic changes register on radiographs and that radiologic findings vary according to the different phases of the disease. In 1937, Ikeda (6) introduced an adult form of lipoid pneumonia that resembles chronic granulomatous pneumonitis or tumor, which is different from the acute bronchopneumonic form that occurs in children. He suggested that although the same material is aspirated, the difference in the amount of oil, the manner and duration of its administration, and the character of the local inflammatory reaction cause the discrepancy. Hampton et al (21) and Kennedy et al (24) used this concept in assessing radiologic findings.

Recently, Lee et al (33) analyzed thin-section CT findings of pathologically or clinically confirmed shark liver oil–induced lipoid pneumonia in 25 patients. Six of ten patients with a diffuse ground-glass opacity pattern at CT had a recent history of taking a large amount of squalene through the nose, whereas seven patients who had a consolidation pattern had a history of taking squalene for several months. All of the five patients who had an interstitial thickening pattern had a history of taking squalene for more than 1 year.

There are several reports on the usefulness of CT in diagnosing lipoid pneumonia by detecting areas of low attenuation that range from 140 HU to -17 HU in the consolidation or mass (10,15,16,26,27). There was wide variation in the CT numbers of the consolidated lungs in our study. The CT numbers ranged from -150 HU to 50 HU, which is a higher range of attenuation than that in the oil in the capsule (-210 to -200 HU). As Joshi and Cholankeril (27) stated earlier, and as we confirmed with histologic studies, the inflammation surrounding the oil material in the lung can account for a higher CT attenuation value. In our study, the attenuation of the consolidation was highest on CT scans obtained 1 week after oil administration, which reflected the extensive inflammatory process involving the alveolar spaces and alveolar wall. However, even on 1-week follow-up CT scans, an area of low attenuation below -10 HU was identified in the airspace consolidation in all sites.

Another point to keep in mind when regarding the low-attenuating consolidation as a sign of lipoid pneumonia is that partial volume averaging of the partly aerated alveoli can also appear as low-attenuating consolidation. To avoid this artifact, we magnified the images on the monitor four times and reviewed both the mediastinal window and lung window images when localizing the region-of-interest circles. However, pathologic review of the area of consolidation at CT where we measured the CT numbers showed evidence of partial aeration. We think that the measurement of attenuation for the diagnosis of lipoid pneumonia should be limited to the area of a masslike lesion in the chronic stage and to areas of compact consolidation in the earlier stage—that is, from immediately to about 1 week after aspiration of lipid materials, when most of the alveolar spaces are filled with lipid material and inflammatory exudate (Fig 4).

There have been several case reports of magnetic resonance (MR) imaging findings of lipoid pneumonia (10,35–37). In three reports (10,35,36) in which MR imaging findings of spin-echo sequences were described, both T1- and T2-weighted images displayed a high signal intensity that was equal to or slightly less than that of the subcutaneous fat. However, these signal intensity changes cannot always represent a fat component, because animal fat is more likely to produce necrotizing hemorrhagic pneumonia owing to its higher content of free fatty acid (3,5,7), and the signal intensity of blood can be high on both T1- and T2-weighted images. Recently, Cox et al (37) reported on the usefulness of chemical shift imaging in the diagnosis of lipoid pneumonia.

The degree and quality of tissue reaction to aspirated oil are quite variable and depend on the quality and frequency of aspiration, the chemical characters of the oil itself, and the effects of other substances that may be aspirated at the same time (3,5,7). In our study, the reaction of parenchyma after injection of shark liver oil was similar to those previously reported, but it was milder than the reaction after instillation of other animal fats. There was no area of alveolar hemorrhage or necrosis on histologic studies in the earlier stage, and mild septal fibrosis without distortion of the alveolar structures was noted in the later stage. Although squalene is extracted from shark liver oil, it is an intermediate form of biosynthetic cholesterol and different from common animal fat, which can be hydrolyzed into fatty acid by using lipase (38). A small amount of administered oil may be another explanation for the absence of a severe inflammatory reaction.

The method of lipid administration in our study was different from that in clinical settings; we made a single intrabronchial injection rather than repeated aspirations of small amounts. However, because the basic pathologic changes in the lung after oil administration in humans seem similar to those seen in our study, we think that the serial thin-section CT findings that correlated with the histopathologic findings in our study could be referred to when treating patients with squalene-induced pneumonia. Another shortcoming of our study is that the follow-up after administration of oil did not exceed 16 weeks. Thus, we can provide data on only the acute and subacute stages of lipoid pneumonia. Finally, all of the CT scans were obtained at functional residual capacity rather than at full inspiration. However, when the CT scans were compared with the corresponding contact radiographs of inflated and sliced specimens, there was no abnormal opacification that was thought to be due to the difference in the degree of inflation. We obtained CT scans before the oil injection in all of the pigs at functional residual capacity and could not find an area of substantially abnormal opacity that affected the CT findings of lipoid pneumonia (Fig 1a).

At first, we injected the oil endobronchially with the pig in a supine position by using a catheter without a balloon. The CT scans obtained immediately after the administration of oil showed dissemination of the oil into the upper and contralateral areas of the lung. We then placed the animal in a semierect prone position and used a balloon catheter to prevent regurgitation of the oil. We added a hyperventilation procedure with a ventilator bag just after the administration of oil to promote migration of the oil into the distal airspace. This modification allowed us to localize the pneumonic process to the unilateral lower lobe. We think localization of pneumonia is important for comparing the affected and normal lungs in the same animal and for decreasing the rate of mortality and morbidity. By using this model, it is feasible to produce pneumonia induced by various agents such as gastric juice in the pig; this may be useful for studying the pathophysiologic and radiologic changes of aspiration pneumonia.

In conclusion, the serial thin-section CT findings of exogenous lipoid pneumonia after a single endobronchial administration of oil in pigs were ground-glass opacity and airspace consolidation, followed by complete or incomplete resolution with lobular volume loss and septal thickening.Practical application: These results provide basic knowledge of serial changes in CT findings in cases of exogenous lipoid pneumonia and provide the histopathologic background of the CT findings. The results of our study also show that low-attenuating consolidation as a sign of lipoid pneumonia should be considered carefully, because partial volume averaging of partly aerated lung can cause a false-positive interpretation.


   Acknowledgments
 
We express our gratitude to Sung Hwan Hong, MD, Sun Won Park, MD, and Hyuk Jae Choi for their technical assistance with the animal experiments and management and to Eun Hoi Goo for his help in CT scanning. We also thank Jeong Wook Seo, MD, for advice in histopathologic interpretations.


   Footnotes
 
2 Current address: Department of Radiology, Gachon Medical College, Gil Medical Center, Inchon, Korea. Back

3 Current address: Department of Pathology, Korea Cancer Center Hospital, Seoul. Back

Author contributions: Guarantors of integrity of entire study, J.G.I., J.B.S.; study concepts, J.G.I., J.B.S., W.S.K.; study design, J.G.I., J.B.S.; definition of intellectual content, J.G.I., J.B.S., W.S.K.; literature research, J.B.S.; experimental studies, J.B.S., C.K.S.; data acquisition, J.B.S., C.K.S.; data analysis, J.G.I., J.B.S., J.W.S., J.H.C.; manuscript preparation and editing, J.B.S., J.G.I.; manuscript review, J.B.S., J.G.I., W.S.K., J.H.C.


   References
TOP
 Abstract
 Introduction
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Laughlen GF. Studies on pneumonia following nasopharyngeal injections of oil. Am J Pathol 1925; 1:407-415.
  2. Pinkerton H. Oils and fats: their entrance into and fate in the lungs of infants and children—a clinical and pathologic report. Am J Dis Child 1927; 33:259-285.
  3. Pinkerton H. The reaction to oils and fats in the lung. Arch Pathol 1928; 5:380-401.
  4. Graef I. Pulmonary changes due to the aspiration of lipids and mineral oil. Am J Pathol 1935; 11:862-868.
  5. Ikeda K. Oil aspiration pneumonia (lipoid pneumonia): clinical, pathologic and experimental considerations. Am J Dis Child 1935; 49:985-1006.
  6. Ikeda K. Lipoid pneumonia of the adult type (paraffinoma of the lung): report of five cases. Arch Pathol 1937; 23:470-492.
  7. Paterson JLH. An experimental study of pneumonia following the aspiration of oily substances: lipoid cell pneumonia. J Pathol Bacteriol 1938; 46:151-164.
  8. Volk BW, Nathanson L, Losner S, Slade WR, Jacobi M. Incidence of lipoid pneumonia in a survey of 389 chronically ill patients. Am J Med 1951; 10:316-324.
  9. Jampolis RW, McDonald JR, Clagett OT. Mineral oil granuloma of the lung: an evaluation of methods for identification of mineral oil in tissue. Int Abs Surg 1953; 97:105-119.
  10. Bréchot JM, Buy JN, Laaban JP, Rochemaure J. Computed tomography and magnetic resonance findings in lipoid pneumonia. Thorax 1991; 46:738-739.[Abstract]
  11. Brown AC, Slocum PC, Puthoff SL, Wallace WE, Foresman BH. Exogenous lipoid pneumonia due to nasal application of petroleum jelly. Chest 1994; 105:968-969.[Abstract/Free Full Text]
  12. Brody H. Focal lipid granulomatosis of the lung following instillation of ionized poppyseed oil. Arch Pathol 1943; 35:744-749.
  13. Storrs RP, McDonald JR, Good CA. Lipoid granuloma of the lung following bronchography with iodized oil. J Thorac Surg 1949; 18:561-568.[Medline]
  14. Annobil SH, Ogunbiyi AO, Benjamin B. Chest radiographic findings in childhood lipoid pneumonia following aspiration of animal fat. Eur J Radiol 1993; 16:217-220.[Medline]
  15. de Oliveira GA, Del Caro SR, Bender Lamago CM, Mercon de Vargas PR, Vervloet VEC. Radiographic plain film and CT findings in lipoid pneumonia in infants following aspiration of mineral oil used in the treatment of partial small bowel obstruction by Ascaris lumbricoides. Pediatr Radiol 1985; 15:157-160.[Medline]
  16. Lee KS, Müller NL, Hale V, Newell Jr JD, Lynch KA, Im JG. Lipoid pneumonia: CT findings. J Comput Assist Tomogr 1995; 19:48-51.[Medline]
  17. Hugosson CO, Riff EJ, Moore CCM, Akhtar M, Tufenkeji HT. Lipoid pneumonia in infants: a radiographical-pathological study. Pediatr Radiol 1991; 21:193-197.[Medline]
  18. Balakrishman S. Lipoid pneumonia in infants and children in South India. Br Med J 1973; 4:329-331.
  19. Asnis DS, Saltzman HP, Melchert A. Shark oil pneumonia: an overlooked entity. Chest 1993; 103:976-977.[Abstract/Free Full Text]
  20. Pierson JW. Some unusual pneumonias associated with the aspiration of fats and oils in the lungs. AJR 1932; 27:572-579.
  21. Hampton SO, Bickham CE, Winship T. Lipoid pneumonia. AJR 1955; 73:938-949.
  22. Weill H, Ferrans VJ, Gay RM, Ziskind MM. Early lipoid pneumonia: roentgenologic, anatomic, and physiologic characteristics. Am J Med 1964; 36:370-376.[Medline]
  23. Genereux GP. Lipids in the lungs: radiologic-pathologic correlation. J Can Assoc Radiol 1970; 21:2-15.[Medline]
  24. Kennedy JD, Costello P, Balikian JP, Herman PG. Exogeneous lipoid pneumonia. AJR 1981; 136:1145-1149.[Abstract/Free Full Text]
  25. Wright BA, Jeffrey PH. Lipoid pneumonia. Semin Respir Infect 1990; 5:314-321.[Medline]
  26. Wheeler PS, Stitik FP, Hutchins GM, Klinefelter HF, Siegelman SS. Diagnosis of lipoid pneumonia by computed tomography. JAMA 1981; 245:65-66.[Abstract]
  27. Joshi RR, Cholankeril JV. Computed tomography in lipoid pneumonia. J Comput Assist Tomogr 1985; 6:211-213.
  28. Brody JS, Levin B. Interlobular septa thickening in lipid pneumonia. AJR 1962; 88:1061-1069.
  29. Todo G, Herman PG. High-resolution computed tomography of the pig lung. Invest Radiol 1986; 21:689-696.[Medline]
  30. Markarian B, Dailey ET. Preparation of inflated lung specimens. In: Heitzman ER, eds. The lung: radiologic-pathologic correlations. 2nd ed. St Louis, Mo: Mosby–Year Book, 1984; 4-12.
  31. French JE, Morris B. The uptake and storage of lipid particles in lymph-gland in the rat. J Pathol Bacteriol 1960; 79:11-19.[Medline]
  32. Godwin JD, Müller NL, Takasugi JE. Pulmonary alveolar proteinosis: CT findings. Radiology 1988; 169:609-613.[Abstract/Free Full Text]
  33. Lee JS, Im JG, Song KS, Seo JB, Lim TH. Exogenous lipoid pneumonia: high-resolution CT findings. Eur Radiol 1999; 9:287-291.[Medline]
  34. Franquet T, Giménez A, Bordes R, Rodríguez-Arias JM, Castella J. The crazy-paving pattern in exogenous lipoid pneumonia: CT-pathologic correlation. AJR 1998; 170:315-317.[Abstract/Free Full Text]
  35. Carrillon Y, Tixier E, Revel D, Cordier JF. MR diagnosis of lipoid pneumonia. J Comput Assist Tomogr 1988; 12:876-877.[Medline]
  36. Carette MF, Grivaux M, Monad B, Capeau F, Lebreton C, Bigot JM. MR findings in lipoid pneumonia. AJR 1989; 153:1097-1098.[Free Full Text]
  37. Cox JE, Choplin RH, Chiles C. Chemical-shift MRI of exogeneous lipoid pneumonia. J Comput Assist Tomogr 1996; 20:465-467.[Medline]
  38. Budavari S, ed. The Merck index 11th ed. Rahway, NJ: Merck, 1989; 1383.



This article has been cited by other articles:


Home page
 RadiologyHome page
W. R. Webb
Thin-Section CT of the Secondary Pulmonary L