| Japanese Journal of Clinical Oncology | Pages |
Lipoid Pneumonia in Lung Cancer: Radiographic and Pathological Features
Introduction
Materials and Methods
Results
Discussion
References
Lipoid Pneumonia in Lung Cancer: Radiographic and Pathological Features
Methods: To elucidate the features of obstructive pneumonia, we radiolopathologically studied 147 patients with lung cancer that had been resected.
Results: Gross inspection of the resected materials revealed evidence of endogenous lipoid pneumonia in 33 of the 147 patients with radiography that corresponded to obstructive pneumonia. We classified the 33 cases into three types as follows: (1) type I lipoid pneumonia, localized to the lung parenchyma distal to an airway obstructed by a tumor (23 cases); (2) type II lipoid pneumonia, features of type I lipoid pneumonia and consecutively spreading to the adjacent segment whose airway was not affected (five cases); (3) type III lipoid pneumonia, features of type II lipoid pneumonia and spreading to the isolated segments (five cases). Lipoid pneumonia was found in 16 of 89 (18%) adenocarcinoma cases and in 17 of 55 (31%) squamous cell carcinoma cases. In type I lipoid pneumonia, squamous cell carcinoma cases were predominant over adenocarcinoma cases (14 vs nine cases), but in type III lipoid pneumonia, adenocarcinoma cases predominated (four vs one case). Further, in cases of type III lipoid pneumonia, radiographs frequently revealed that lung cancers were cavitated.
Conclusion: Lipoid pneumonia in lung cancer may be associated with factors that play a larger role than the cancer alone. It can be speculated that transbronchial dissemination of breakdown products of adenocarcinoma cells, including mucin, may contribute to the spread of the non-obstructive component of lipoid pneumonia, because the local physical effect of obstructed bronchus does not affect the non-obstructive component.
Introduction
In lung cancer, especially in squamous cell carcinoma of the lung, airways are frequently obstructed by tumor invasion. Radiographically, this condition was recognized as atelectasis and opacification of lung parenchyma distal to an obstructed airway (1,2).
In 1949, McDonald et al. (3) first described the latter finding as `obstructive pneumonitis', which is characterized by the presence of bronchial obstruction and the accumulation of lipid-filled macrophages. Although it had previously been believed that the obstructive pneumonitis was caused by infection mainly in the obstructed bronchus (3,4), in fact the condition is due to a non-infectious process, such as physical or chemical effects of airway blockage (1,2,5). Thus, we now understand that obstructive pneumonitis (obstructive pneumonia) on radiographs is synonymous with golden pneumonia in gross pathology and with endogenous lipoid pneumonia, which is characterized by numerous foamy macrophages in the air spaces with some degree of interstitial mononuclear infiltration (6).
In spite of the terminology `obstructive', the obstructive effect sometimes accounts for a much larger infiltrate on the usual chest radiograph than is caused by the tumor alone (6,7), but there have been few radiolopathological studies of this condition. We therefore assessed the radiolopathological details of this controversial pneumonia.
Materials and Methods
Between 1989 and 1993, 147 consecutive patients underwent lobectomy or pneumonectomy for lung cancer in Tokyo National Chest Hospital. Based on gross pathological inspection of the resected lungs, 44 of the 147 cases were found to have golden pneumonia. Radiographic review of the 44 cases revealed opacification of lung parenchyma in 38 cases and lobar atelectasis in six cases. Five of the 38 cases also had abscess formation that represented evidence of active bacterial infection and we deleted them from this study. We therefore studied 33 of 147 cases (22%) radiolopathologically.
In each of these 33 cases, lung specimens were obtained, fixed in 10% formalin and cut sagittally in 1 cm slices. Subsequently, we examined the relationship between the tumor and pneumonia by gross inspection of the lobar, segmental and subsegmental bronchi. Next, we investigated histopathological details of the tumor and pneumonia and also evaluated the relationship of these pathological findings with radiographic findings, including those by both plain and computed tomography (CT) chest radiographs of all patients, which were taken within 3 weeks before surgery. Further, we also reviewed medical records before surgery, which included results of bacteriological tests.
Statistical analysis was evaluated by the chi-squared test and P values of < 0.05 were considered to be significant.
Results
The ages of the 33 patients (28 men and 5 women) at presentation ranged from 39 to 82 years (mean 61 years). None had had prior chemotherapy or radiotherapy before surgery, nor had they any history of intake of oily substances (in Japan, oily substances are not used to treat sinus symptoms or constipation). Lobectomy was performed in 24 patients and pneumonectomy in nine. The pathological stage of lung cancers was Stage I in 13, Stage II in five and Stage III in 15 patients. Lipoid pneumonia was found in 16 of 89 (18%) adenocarcinoma cases, in 17 of 55 (31%) squamous cell carcinoma cases and in none of the three small cell carcinoma cases. The frequency of lipoid pneumonia was higher in squamous cell carcinoma than in adenocarcinoma cases, but the difference was not significant (P = 0.11). The pneumonia was then classified into three types as shown in Fig. 1: type I pneumonia was localized to the lung parenchyma distal to an airway obstructed by a tumor (pure obstructive pneumonia); type II pneumonia had the features of type I but with consecutive spreading to the adjacent segment whose airway was not affected; and in type III pneumonia, there was spreading to an isolated segment in addition to the features of type II.
Figure 1. Classification of endogenous lipoid pneumonia in lung cancer. Figure 2. Radiographic findings of type I lipoid pneumonia. A chest CT radiograph showing small, sparse opacification (arrowhead) around the tumor shadow.
Type I pneumonia (23 cases) was radiographically identified as a small, sparse opacification around the tumor shadow (Fig. 2), pathologically corresponding to air spaces filled with an accumulation of foamy or vacuolated macrophages (endogenous lipoid pneumonia) accompanied by eosinophilic proteinaceous materials within the air spaces (Fig. 3). The degree of lymphocytic infiltration to the alveolar wall and/or intra-alveolar organization varied among the cases. Type II pneumonia (five cases) was radiographically shown to have a reticulonodular shadow and pathologically found to be lipoid pneumonia, similar to type I pneumonia. Type III pneumonia (five cases) was radiographically indicated by a widespread reticulonodular shadow (Fig. 4) and with some pathological elements corresponding to type I lipoid pneumonia, such as the accumulation of foamy or vacuolated macrophages in the air spaces. The pneumonia in the isolated segments was not accompanied by eosinophilic proteinaceous materials within alveolar spaces (Fig. 5). The relationship between the type of lipoid pneumonia and the features of cancer is shown in Table 1. Three patients who had small tumors (< 30 mm in diameter) all had type I lipoid pneumonia, but in cases with larger tumors, there seemed to be no correlation between tumor size and type of lipoid pneumonia. In both histological types, type I pneumonia was the most common pattern. However, there were some differences in the distribution of the patterns. The majority of cases of squamous cell carcinoma manifested type I pneumonia (14/17), whereas about half of adenocarcinoma cases (7/16) manifested type II or III pneumonia. Evaluation of cancer cell differentiation revealed that the proportion of poorly differentiated carcinoma cases rose in type II and rose further in type III pneumonia. In relation to patterns of lipoid pneumonia, radiographic features of the lung cancers were also analyzed (Table 2). It was revealed that the lung cancers were frequently cavitated in cases accompanied by type II and III pneumonia and that, especially, type III pneumonia was frequently seen in cavitated, poorly differentiated adenocarcinoma cases. Table 1.
Table 2. 
Figure 3. Pathological findings of type I lipoid pneumonia. (A) Gross finding showing a tumor (squamous cell carcinoma) and golden pneumonia distal to obstructed airways by the tumor (obstructive pneumonia). (B) Histopathological finding showing accumulation of foamy macrophages in the air spaces (endogenous lipoid pneumonia). Note concomitance of eosinophilic materials within air spaces.
A
B
Figure 4. Radiographic findings of type III lipoid pneumonia. (A) Chest plain radiograph showing a widespread reticulonodular shadow and a cavitated tumor shadow (arrow). (B) Chest CT radiograph showing a cavitated tumor shadow in the right upper lobe and a reticulonodular shadow around the tumor shadow. (C) Chest CT radiograph also showing a reticulonodular shadow in the right middle lobe and lower lobe (S6).
A
B
C
Figure 5. Pathological findings of type III lipoid pneumonia in the same patient as in Fig. 4. (A) Gross finding showing a cavitated tumor (adenocarcinoma, arrow) accompanied by golden pneumonia in the right upper lobe. (B) Gross finding showing golden pneumonia in the right lower lobe (S6) (arrowhead). Note no tumorous lesion found in the S6. (C) Histopathological finding showing endogenous lipoid pneumonia without eosinophilic materials within the air spaces.
A
B
C
Lung cancer
Type of lipoid pneumonia
Type I
Type II
Type III
Tumor size (diameter)
< 30 mm
3
0
0
31-60 mm
15
4
4
>61 mm
5
1
1
Histological type
Adenocarcinoma
9
3
4
Squamous cell carcinoma
14
2
1
Cell differentiation
Poorly differentiated
6
2
3
Moderately differentiated
12
3
2
Well differentiated
5
0
0
Total
23
5
5
Histological type
Cavitated lung cancer (%)
Type I
Type II
Type III
Adenocarcinoma
0 (0)
1 (33)
3 (75)
Squamous cell carcinoma
2 (14)
1 (50)
1 (100)
Total
2 (9)
2 (40)
4 (80)
A review of clinical data revealed that pathogenic bacteria, such as Pseudomonas aeruginosa and Klebsiella pneumoniae, was detected in sputum culture of six patients (two patients in each of types I, II and III pneumonia) and intravenous antibiotics were administered to these patients before surgery. In these cases, histopathological observation revealed that neutrophilic infiltration was localized in the obstructive airways of major components of lipoid pneumonia with minimal parenchymal involvement and that neutrophilic infiltration in non-obstructive airways was not evident in adjacent components in type II pneumonia or in an isolated component in type III pneumonia.
Discussion
We have demonstrated that endogenous lipoid pneumonia in lung cancer often spreads distal to the non-obstructive bronchus and we classified the pneumonia into three types. Type I pneumonia was identical with pure `obstructive pneumonia', and the most interesting pattern of lipoid pneumonia in this study was type III lipoid pneumonia (the combination of obstructive and non-obstructive pneumonia).
The pathogenesis of lipoid pneumonia is complex and has been thought to be related to several mechanisms, such as retained epithelial secretion (2,6), cell breakdown (6), leakage from vessels (6), prolonged hypoxia (5) and local oxygen and carbon dioxide tension (8). In the present study, type I pneumonia may be associated with all or some of these mechanisms. Eosinophilic proteinaceous materials in the air spaces are also caused by similar mechanisms, such as a transudate from alveolar capillaries, retained surfactant and secretions of Clara cells (2). In type III lipoid pneumonia, however, mechanisms of lipoid pneumonia in the isolated segments (non-obstructive pneumonia) seem to be different from those of type I lipoid pneumonia, because physical effects of obstructive bronchus, such as leakage from vessels, hypoxia or local oxygen and carbon dioxide tensions, do not extend to the isolated segments whose airway was not obstructed. Lack of eosinophilic materials within air spaces in non-obstructive pneumonia also suggests that the pneumonia was not associated with local physical effects. Therefore, we consider that non-obstructive lipoid pneumonia may be associated with transbronchial dissemination of breakdown products of cancer cells, especially those of poorly differentiated adenocarcinoma cells and secretions including mucin, because type III pneumonia was frequently seen in cases with cavitated adenocarcinoma. Type II pneumonia may be influenced by such transbronchial dissemination in addition to the local effect seen in type I pneumonia. As for the contribution of infectious changes to endogenous lipoid pneumonia, infection is generally localized in airways because the surrounding lung is already consolidated by the non-infectious inflammatory changes, limiting the spread of bacteria (2). We also consider that infectious changes may not have been associated with progression of type II and III pneumonia in this study, because neutrophilic infiltration was localized only in obstructive airways of major components of lipoid pneumonia.
The cause of the differences in radiographic images between type I, II and III pneumonia is unclear, but it may be associated with differences of degree of pneumonia and of concomitant lesions, such as intraalveolar materials and/or organization.
References
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 19 Aug 1998
Copyright©Japanese Journal of Clinical Oncology, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
S. L. Betancourt, S. Martinez-Jimenez, S. E. Rossi, M. T. Truong, J. Carrillo, and J. J. Erasmus
Lipoid Pneumonia: Spectrum of Clinical and Radiologic Manifestations
Am. J. Roentgenol.,
January 1, 2010;
194(1):
103 - 109.
[Abstract]
[Full Text]
[PDF]
S. K. Mathai, A. N. Rubinowitz, R. J. Homer, F. Detterbeck, and E. L. Herzog
Of Lungs, Lipids, and Lollipops
Chest,
November 1, 2009;
136(5):
1420 - 1423.
[Full Text]
[PDF]
R. L. Hunter, M. R. Olsen, C. Jagannath, and J. K. Actor
Multiple Roles of Cord Factor in the Pathogenesis of Primary, Secondary, and Cavitary Tuberculosis, Including a Revised Description of the Pathology of Secondary Disease
Ann. Clin. Lab. Sci.,
January 1, 2006;
36(4):
371 - 386.
[Abstract]
[Full Text]
[PDF]
S. C. Gaerte, C. A. Meyer, H. T. Winer-Muram, R. D. Tarver, and D. J. Conces Jr
Fat-containing Lesions of the Chest
RadioGraphics,
October 1, 2002;
22(90001):
S61 - 78.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (7)
Request Permissions
Google Scholar
Articles by Tamura, A
Articles by Mori, M
Search for Related Content
PubMed
PubMed Citation
Articles by Tamura, A
Articles by Mori, M
Social Bookmarking
What's this?