Comparison of Fibrotic Sarcoidosis with Usual Interstitial Pneumonia

Charikleia Matsouri*, Katerina Malagari, Efthimia Alexopoulou, Alexios Kelekis, Maria Pomoni, Anastasia Pomoni, George D. Papadimitriou, Dimitrios A. Kelekis

  Open Access OPEN ACCESS  Peer Reviewed PEER-REVIEWED

Comparison of Fibrotic Sarcoidosis with Usual Interstitial Pneumonia

Charikleia Matsouri*1, Katerina Malagari1, Efthimia Alexopoulou1, Alexios Kelekis1, Maria Pomoni1, Anastasia Pomoni1, George D. Papadimitriou1, Dimitrios A. Kelekis1

1Imaging and Research Unit of University of Athens Medical School, Evgenidion Hospital, Athens, Greece

Abstract

Objectives: This retrospective study examines the differences of the fibrotic stage of sarcoidosis with idiopathic interstitial fibrosis using High-resolution Computed Tomography (HRCT), imaging features and quantification of the severity and extent of the disease. Methods: 81 patients with biopsy diagnosed pulmonary sarcoidosis and 81 patients with usual interstitial pneumonia (UIP) were included. All underwent HRCT scan. Findings were evaluated per lung lobe. The severity of disease was quantified per feature and overall with the Gay - Watters score. Results: Respectively, the two groups recorded: 52 (64.20%) vs 60 (74.07%) presented a reticular pattern, 62 (76.54%) vs 74 (91.36%) thickened septal lines, 24 (29.63%) vs 56 (69.14%) honeycomb pattern, 52 (64.20%) vs 43 (53.09%) ground glass attenuation, 7 (8.64%) vs 27 (33.33%) ground glass with traction bronchiectasis, 47 (59,26%) vs 25 (30.86%) micronodular pattern, 48 (60,49%) vs 1 (1.23%) hilar / mediastinal lymph node enlargement. The micronodular pattern showed a statistically significant association with sarcoidosis (p = 0.03), whereas ground glass opacities associated with bronchiectasis was statistically significant for UIP (p = 0.0021). Honeycomb was statistically significant for UIP (P = 0.0037). Lymph node enlargement was associated with sarcoidosis (p = 0.0001). Thickened septal lines seemed to be more frequent with UIP compared to sarcoidosis (p = 0.05). Overall, the severity of parenchymal involvement for sarcoidosis and UIP was 26 ± 19 / 43 ± 18 respectively (p < 0.001). Conclusions: The micronodular pattern highly indicates sarcoidosis, whereas honeycombing and ground glass with traction bronchiectasis are associated with UIP. These features can assist on differentiation, when disease presents itself with fibrotic HRCT characteristics at diagnosis.

At a glance: Figures

Cite this article:

  • Matsouri*, Charikleia, et al. "Comparison of Fibrotic Sarcoidosis with Usual Interstitial Pneumonia." American Journal of Medical Sciences and Medicine 1.3 (2013): 38-44.
  • Matsouri*, C. , Malagari, K. , Alexopoulou, E. , Kelekis, A. , Pomoni, M. , Pomoni, A. , Papadimitriou, G. D. , & Kelekis, D. A. (2013). Comparison of Fibrotic Sarcoidosis with Usual Interstitial Pneumonia. American Journal of Medical Sciences and Medicine, 1(3), 38-44.
  • Matsouri*, Charikleia, Katerina Malagari, Efthimia Alexopoulou, Alexios Kelekis, Maria Pomoni, Anastasia Pomoni, George D. Papadimitriou, and Dimitrios A. Kelekis. "Comparison of Fibrotic Sarcoidosis with Usual Interstitial Pneumonia." American Journal of Medical Sciences and Medicine 1, no. 3 (2013): 38-44.

Import into BibTeX Import into EndNote Import into RefMan Import into RefWorks

1. Introduction

Sarcoidosis, a multisystemic chronic inflammatory condition of unknown etiology, is characterized by noncaseous epithelioid cell granulomas and changes in tissue architecture. The mediastinal and hilar lymph nodes enlargement is the most common imaging feature in the chest, presented in approximately 80-90% of the patients [1]. The most common pulmonary symptoms are cough, dyspnea and chest pain. Other symptoms include fatigue, night sweats, weight loss, and erythema nodosum. 50% of the reported cases is asymptomatic, with abnormalities detected incidentally at chest radiography [2].

A diagnosis of sarcoidosis is established on the basis of compatible clinical-radiological findings and histological evidence of the presence of noncaseous epithelioid cell granulomas in one or more organs. High-resolution computed tomography (HRCT) provides detailed information regarding lung involvement and hence can be helpful for distinguishing active inflammation from irreversible fibrosis in selected patients with stage 2 or 3 sarcoidosis [3]. Nodules, ground-glass opacities, and alveolar opacities are suggestive of granulomatous inflammation that may be reversed with therapy, whereas honeycomb pattern, architectural distortion, volume loss, and traction bronchiectasis are indicative of irreversible fibrosis [4]. In fact, fibrotic interstitial disease is a common pattern at presentation. HRCT may also be useful for verifying specific diagnoses in patients with atypical clinical manifestations or unusual radiographical features [3, 5].

Usual interstitial pneumonitis, or idiopathic pulmonary fibrosis (UIP/IPF), is a common type of interstitial pneumonia with progressive diffuse parenchymal lung changes and unclear pathophysiology [6]. The median survival is poor, recording 2-3 years [7]. Restrictive diseases, such as UIP/IPF, result in decreased lung volumes and distortion of the normal anatomy. In diffuse lung diseases, the HRCT is used as a significant tool. Several studies have shown that HRCT of the chest with certain features can be used as a surrogate in order to recognize the pathological pattern of UIP with high degrees of confidence [8].

In this context, the present study investigates the differences between the fibrotic stage of sarcoidosis and fibrosis in UIP/IPF using High-resolution Computed Tomography by comparing individually the imaging features, the severity and the extent of the disease of each entity. This may be of assistance in imaging differentiation between the two entities, since both may feature fibrotic changes at HRCT at presentation [9] and represent a considerable proportion of interstitial lung disease.

2. Material and Methods

In this retrospective study, performed between January and August 2012, a population of 162 patients with biopsy documented lung disease (Imaging and Research Unit of University of Athens Medical School, Evgenidion Hospital) were classified in two groups. Group A included 81 patients with pulmonary sarcoidosis (51 women, 30 men, mean age 57.8 ± 12.3 years) and group B included 81 patients with usual interstitial pneumonia -UIP- (23 women, 58 men mean age 69.3 ± 9.8 years) respectively. Patient selection included consecutive patients presenting with pulmonary fibrosis in which lung biopsy documented either sarcoidosis or UIP.

High-resolution computed tomography scanning of the lung was evaluated; the images were obtained by using 1mm collimation every 10mm throughout the thorax. The scans were obtained having the patient in supine position at full inspiration and were reconstructed by using a high-spatial-frequency algorithm. All images were viewed at window settings optimized for assessment of lung parenchyma. HRCT features were evaluated per lung lobe (ground glass, reticular pattern, micronodular, thickened septal lines, ground glass with traction bronchiectasis and bronchiolectasis, honeycomb, hilar and mediastinal lymph node enlargement). Quantification of the severity and extent of each feature was also performed and compared. Gay and Watters technique was used and all HRCT findings scored on a scale 0-5 [10]. These scores were also summed into a total CT score for each patient and for each disease separately. Hence the severity and distribution of HRCT features was recorded regarding the 81 patients with sarcoidosis and the 81 patients with UIP.

Statistical analysis: Fisher’s exact test was used to evaluate differences between the proportions. A value of p< 0,05 was considered statistically significant.

Ethical considerations: This retrospective study complies with the Declaration of Helsinki.

Table 1. Incidence and distribution of each HRCT featrue in the 81 patients with sarcoidosis

3. Results

Of the 81 patients with sarcoidosis, 52 (64.20%) presented a reticular pattern, 62 presented thickened septal lines (76.54%), 24 presented honeycomb pattern (29.63%), 52 ground glass attenuation (64.20%), 7 ground glass presented traction bronchiectasis and bronchiolectasis (8.64%), 47 micronodular pattern (59.26%) and 48 hilar and mediastinal lymph node enlargement (60.49%).

The distribution and severity findings were summarized in Table 1; the reticular pattern had a peripheral distribution in 13 patients (25.00%), the central peribronchial distribution in 40 patients (76.92%) and the involved upper lobes in 38 patients (73.08%), the middle lobe in 29 patients (55.77%) and the lower lobe in 41 patients (78.85%). The severity score of parenchymal reticular involvement was 30.22%.

62 patients (76.54%) recorded thickened septal lines and 52 (83.87%) with peripheral distribution, 28 (45.16%) with central peribronchial distribution. 35 (56.45%) showed distribution in the upper lobes, 34 (54.84%) in the middle lobe and 43 (69.35%) in the lower lobe. The severity score was 23.06%.

52 patients (64.20%) recorded ground glass opacities and 33 (63.46%) with peripheral distribution, 14 (26.92%) with central peribronchial distribution. 20 (38.46%) showed distribution in the upper lobes, 36 (69.23%) in the middle lobe and 26 (50%) in the lower lobe. The severity score was 17.04%.

Ground glass opacities associated with traction bronchiectasis and bronchiolectasis were observed in 7 patients (8.64%). A peripheral distribution was recorded in 2 (28.57%), central peribronchial distribution in 3 (42.86%), involved upper lobes in 1 (14.29%), middle lobe in 4 (57.14%) and lower lobe in 6 (85.71%). The severity score was 2.47%.

Micronodular opacities were found in 47 patients (59.26%). A peripheral distribution was noticed in 36 (76.60%), central peribronchial distribution in 15 (31.91%), upper lobes involvement in 24 (51.06%), middle lobe in 33 (70.21%) and lower lobe in 29 (61.70%). The severity score was 15.65%.

Honeycomb pattern was found in 24 patients (29.63%). The honeycomb pattern had a peripheral distribution in 24 (100%), central peribronchial distribution in 5 (20.83%) and upper lobes involvement in 14 (58.33%), middle lobe in 16 (66.67%) and lower lobe in 21 (87.50%). The severity score was 14.02%.

Mediastinal and hilar lymph node enlargement was found in 48 patients (60.49%).

Overall, the severity of parenchymal involvement was 26 ± 19. Figure 1 shows the distribution of the severity or parenchymal involvement in the patients of sarcoidosis and it clearly reveals that the severity is mild in these patients (Figure 1). The semiquantitative severity assessment showed mild involvement (score 0-49) in 73 patients (90.12%), and mild severity (score 50-99) in 8 patients (9.88%) respectively.

Figure 1. Severity distribution in 81 patients with sarcoidosis shoes minimal to moderate involvement
Figure 2. Severity distribution in 81 patients with UIP is shoes more advanced disease severity

Table 2. Incidence and distribution of each HRCT feature in the 81 patients with UIP

Of the 81 patients with usual interstitial pneumonia (UIP), 60 presented reticular pattern (74.07%), 74 (91.36%) presented thickened septal lines (91.36%), 56 presented honeycomb pattern (69.14%), 43 with ground glass attenuation (53.09%), 27 with ground glass with traction bronchiectasis and bronchiolectasis (33.33%), 25 with micronodular pattern (30.86%) and 1 with hilar and mediastinal lymph node enlargement (1.23%).

Regarding distribution and severity, findings are summarized in Table 2; the reticular pattern had a peripheral distribution in 60 patients (100%), central peribronchial distribution in 10 patients (16.67%) and involved upper lobes in 57 (95.00%), middle lobe in 59 (98.33%) and lower lobe in 55 (91.67%). The severity score of parenchymal reticular involvement was 49.38%.

Thickened septal lines were seen in 74 patients (91.36%), showing peripheral distribution in 74 (100%), central peribronchial distribution in 60 (81.08%) and upper lobes involvement in 71 (95.95%), middle lobe in 74 (100%) and lower lobe in 73 (98.65%). The severity score was 49.53%.

Ground glass opacities were found in 43 patients (53.09%), had a peripheral distribution in 43 (100%), central peribronchial distribution in 7 (16.28%) and upper lobes involvement in 25 (58.14%), middle lobe in 36 (83.72%) and lower lobe in 34 (79.07%). The severity score was 20.74%.

Ground glass opacities associated with traction bronchiectasis and bronchiolectasis were observed in 27 patients (33.33%), showing a peripheral distribution in 7 (25.93%), a central peribronchial distribution in 22 (81.48%) and an upper lobes involvement in 9 (33.33%), a middle lobe in 24 (88.89%) and a lower lobe in 23 (85.19%). The severity score was 10.42%.

Micronodular opacities were found in 25 patients (30.86%), having a peripheral distribution in 25 (100%), central peribronchial distribution in 6 (24.00%) and upper lobes involvement in 13 (52.00%), middle lobe in 21 (84.00%) and lower lobe in 24 (96.00%). The severity score was 11.46%.

Honeycomb pattern was found in 56 patients (69.14%), presenting peripheral distribution in 56 (100%), central peribronchial distribution in 6 (10.71%) and involved upper lobes in 36 (64.29%), middle lobe in 47 (83.93%) and lower lobe in 56 (100%). The severity score was 31.46%.

Mediastinal and hilar lymph node enlargement was found in 1 patient (1.23%).

Overall, the severity of parenchymal involvement was 43±18. Figure 2 shows the distribution of the severity or parenchymal involvement in the patients with UIP and it clearly reveals that the severity is more advanced in these patients (Figure 2). The semiquantitative severity assessment showed mild involvement (score 0-49) in 52 patients (64.20%), and mild severity (score 50-99) in 29 patients (35.8%) respectively.

Table 3. Comparative frequency of HRCT features in SARC and UIP and statistical significance

Table 4. Distribution of each feature per lobe and in the axial plane

Table 3 summarizes the statistical differences in the frequency of each particular imaging finding (Table 3). It shows that the micronodular pattern is highly associated with sarcoidosis at a statistical significant level (p = 0.03), whereas ground glass opacities associated with bronchiectasis and bronchiolectasis are statistically significant for UIP (p=0.0021). In addition honeycomb is statistically significant for UIP (P=0.0037). Moreover, lymph node enlargement is associated with sarcoidosis (p=0.0001). Table 4 illustrates the differences in frequency between the various imaging features between sarcoidosis and UIP along with statistical significance for each feature. Thickened septal lines seem to be more frequent to UIP compared to sarcoidosis (p = 0.05) (Table 4). Figure 3 illustrates the differences in the severity of involvement between the two disease entities (Figure 3) showing that a more severe disease is present in UIP.

Figure 3. Severity distribution for both disease entities shows more severe involvement in UIP

4. Discussion

End-stage pulmonary sarcoidosis is typically characterized by severe interstitial fibrosis. The fibrotic stage of sarcoidosis and idiopathic interstitial fibrosis concerns the 60% of fibrosis [11].

In this study, we examine the differences between pulmonary sarcoidosis and UIP with high-resolution computed tomography (HRCT) regarding imaging features and quantification of the severity and extent of the disease. The study compared the HRCT findings per lobe separately.

According to our study, micronodules were 59.26% in sarcoidosis and 30.86% in UIP (p=0.0335) respectively. We found that the distribution in sarcoidosis was greater in middle zone with the micronodules located bilaterally. Nunes H et al refer that nodules are the hallmark of pulmonary sarcoidosis, seen in 80% to 100% of all patients at HRCT but less frequently in stage IV [12]. In our study, nodules were small, and had a pattern of irregular and poorly circumscribed margins. A perilymphatic distribution of micronodular lesions is the most common parenchymal disease pattern seen in patients with pulmonary sarcoidosis. Peribronchial distribution was seen in 31.91% with an upper lobe predominance in 51.06%, showing a slight differentiation from the commonly described patterns. A fact that may indicate, that when sarcoidosis is represented at first diagnosis with parenchymal involvement, the distribution is less typical. HRCT literature shows small nodules (2-4 mm in diameter), usually with a bilateral and symmetrical distribution, predominantly but not invariably in the upper and middle zones [13]. The severity score of micronodules was low in our study, reaching a 15.65%. A percentage, that is in concordance with the literature [3]. The micronodules were found most often in the subpleural peribronchovascular interstitium and less often in the septal lines.

Hunninghake et al [14], in their study concluded that the lower lung honeycombing and the upper lung irregular lines were the only independent predictors and, by using only these two factors, a diagnosis of UIP could be established with a sensitivity of 74%, a specificity of 81%, and a positive predictive value of 85%. Other studies have consistently found that honeycombing is the best discriminator between UIP and other conditions [15]. We found honeycomb pattern in 69.14% of patients with UIP and 29.63% in patients with pulmonary sarcoidosis (p=0.0037). The distribution was mostly peripheral. Abehsera et al showed that the fibrotic changes of pulmonary sarcoidosis, particularly bronchial distortion and honeycombing, were in the upper and middle zones in most patients. In our study, patients with fibrotic sarcoidosis were biopsy-documented with non caseating granulomas and abscence of fibroblastic foci characteristic of sarcoidosis and UIP respectively, excluding therefore the possibility of coexistence of these entities. In addition, on follow up of the patients the diagnosis did not change based on clinical observations or treatment response. Bronchial distortion was also mainly central, and honeycombing was peripheral.

We observed fibrosis in upper, middle and lower lobes. Honeycomb on CT is an important prognostic determinant in patients with UIP. Gotway et al [16] declare that Flaherty et al [17, 18] refer to a definite (confident) HRCT diagnosis of UIP on the basis of basilar honeycombing which portends a worse survival for individuals without honeycombing on the HRCT than for those with a histopathological diagnosis of UIP.

In our study, ground gland opacities associated with traction bronchiectasis and bronchiolectasisis presented a statistically significant correlation to UIP (p = 0.0021) and the distribution was mostly central. In the study by Baughman et al [19] traction bronchiectasis tend to be an upper lobe predominant process and can have relative sparing of the lower lobes in pulmonary sarcoidosis, whereas we recorded a lower lobe predominant process. When ground glass opacities are associated with reticular lines or traction bronchiectasis, they usually indicate histologic fibrosis [20]. Abeshera et al study showed that thickened septal lines were obviously fibrotic due to their irregularities, angulations and association to some signs of fissural and bronchial distortion [21]. Previous studies have shown that they were irreversible [22]. In our study, thickened septal lines seem to be more frequent to UIP compared to sarcoidosis (p = 0.05). Fibrotic lesions in UIP are often located in the periphery of the secondary pulmonary lobules, adjacent to septal lines. Lynch A D et al refer that scans considered consistent with IPF, were significantly more likely to show honeycombing, traction bronchiectasis and bronchiolectasis, and lower lobe volume loss, with less likelihood to show ground glass attenuation, decreased attenuation, mosaic attenuation, and centrilobular nodules compared to scans which were not consistent with IPF [23].

We found 60.49% enlargement of lymph nodes in sarcoidosis and 1.23% in UIP at a statistically significant level (p = 0.0001). In sarcoidosis, the enlargement lymph nodes were hilar and mediastinal, bilateral and symmetric. This is the most common pattern in sarcoidosis. Baughman et al supported that adenopathy can help in distinguishing sarcoidosis from other interstitial lung diseases [19]. The presence of adenopathy alone or along with parenchymal disease has been found in more than half of sarcoidosis patients at the time of diagnosis. CT scan is far more sensitive for detecting adenopathy. Various interstitial lung diseases can cause some adenopathy, but the enlargement is usually present only to a mild extent [19].

Gay and Watters’ HRCT scoring system uses a scale of 0-5 for ground glass opacity and linear opacity and estimates the severity and the extent of each disease [10]. In our study, we estimated all HRCT findings, per each lobe separately and assessed the severity and extent of pulmonary sarcoidosis and UIP and then we calculated the total severity for each patient. UIP in an adequate population of patients presents moderate severity of disease, while in pulmonary sarcoidosis the majority of patients exhibit mild severity. Shigemitsou et al [11] observed that when patients with sarcoidosis showed rapid progression of the disease, they were ultimately led to lung transplantation. Our work revealed that differences in the severity of involvement between the two disease entities and indicated that more severe degree of disease is present in UIP. The severity of each disease is related to the patient’s prognosis. Study limitations include the retrospective nature of the study, and the relatively small number of patients.

In conclusion, the major discriminators between the two entities at presentation is the micronodular pattern favoring sarcoidosis, the honeycomb pattern indicating UIP, the ground glass opacities with traction bronchiectasis favoring UIP and the lymph node enlargement which most frequently found in sarcoidosis. Severity scores are more advanced in UIP.

Financial Disclosure and Conflicts of Interest

All authors declare that there is no conflict of interest.

References

[1]  Rockoff S D, Rohatgi K. Prashant. Unusual manifestations of thoracic sarcoidosis. A J R 1985; 144: 513-528.
In article      
 
[2]  Lynch JP 3rd, White ES. Pulmonary sarcoidosis. Eur Respir Monograph 2005, 32: 105-129.
In article      CrossRef
 
[3]  Criado E, Sanchez M., Ramirez J, Arguis P, de Caralt MT, Perea JR, Xaubet A. Pulmonary Sarcoidosis: Typical and atypical manifestations at High-Resolution CT with pathologic correlation. RadioGraphics 2010; 30: 1567-1586.
In article      CrossRefPubMed
 
[4]  Baughman RP, Winget DB, Bowen EH, Lower EE. Predicting respiratory failure in sarcoidosis patients. Sarcoidosis Vasc Diffuse Lung Dis 1997; 14(2): 154-158.
In article      PubMed
 
[5]  Hennebicque AS, Nunes H, Brillet PY, Moulahi H, Valeyre D, Brauner MW. CT findings in severe thoracic sarcoidosis. Eur Radiol 2005; 15: 23-30.
In article      CrossRefPubMed
 
[6]  Schmidt SL, Sundaram B, Flaherty KR. Diagnosing fibrotic lung disease: when is high-resolution computed tomography sufficient to make a diagnosis of idiopathic pulmonary fibrosis. Respirology 2009; 14(7): 934-939.
In article      CrossRefPubMed
 
[7]  American Thoracic Society/European Respiratory Society International multidisciplinary consensus classification of idiopathic interstitial pneumonias. Am J Respir Crit Care Med 2002; 165: 277-304.
In article      CrossRefPubMed
 
[8]  Sverzellatti N, De Fillipo M, Bartalena T, Piciucchi S, Zompatori M. High-resolution computed tomography in the diagnosis and follow-up of idiopathic pulmonary fibrosis. Chest Radiology Radiol Med 2010; 115: 526-538.
In article      CrossRefPubMed
 
[9]  Silva CIS, Muller LN, Hansell MD, Lee SHK, Nicholson GA, Wells UA. Nonspecific interstitial pneumonia and idiopathic pulmonary fibrosis: Change in pattern and distribution of disease over time. Radiology 2008 247 (1).
In article      CrossRefPubMed
 
[10]  Gay SE, Kazerooni EA, Toews GB, Lynch JP 3rd, Gross BH, Cascade PN, Spizarny DL, Flint A, Schork MA, Whyte RI, Popovich J, Hyzy R, Martinez FJ. Idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 1998; 157: 1063-1072.
In article      CrossRefPubMed
 
[11]  Shigemitsu H, Oblad JM, Sharma OP, Koss MN. Chronic interstitial pneumonitis in end-stage sarcoidosis. Eur Respir J. 2010; 35(3): 695-697.
In article      CrossRefPubMed
 
[12]  Nunes H, Brillet P, Valeyre D, Brauner M, Wells A. Imaging in sarcoidosis. Seminars in Respiratory and Critical Care Medicine 2007; 28(1).
In article      CrossRefPubMed
 
[13]  Koyama T, Ueda H, Togashi K, Umeoka S, Kataoka M, Nagai S. Radiologic manifestations of sarcoidosis in various organs. Radiographics 2004; 24(1): 87-104.
In article      CrossRefPubMed
 
[14]  Hunninghake GW, Lynch DA, Galvin JR, Gross BH, Müller N, Schwartz DA, King TE Jr, Lynch JP 3rd, Hegele R, Waldron J, Colby TV, Hogg JC. Radiologic findings are strongly associated with a pathologic diagnosis of usual interstitial pneumonia. Chest 2003; 124: 1215-1223.
In article      CrossRefPubMed
 
[15]  Sumikawa H, Johkoh T, Ichikado K, et al. Usual interstitial pneumonia and chronic idiopathic interstitial pneumonia: Analysis of CT appearance in 92 patients. Radiology 2006; 241(1): 258-266.
In article      CrossRefPubMed
 
[16]  Gotway MB, Freemer MM, King TE Jr. Challenges in pulmonary fibrosis 1: Use of High-resolution CT scanning of the lung for the evaluation of patients with idiopathic interstitial pneumonias. Thorax 2007; 62: 546-553.
In article      CrossRefPubMed
 
[17]  Flaherty KR, Thwaite EL, Kazerooni EA, Gross BH, Toews GB, Colby TV, Travis WD, Mumford JA, Murray S, Flint A, Lynch JP 3rd, Martinez FJ. Thorax. 2003; 58(2): 143-148.
In article      CrossRefPubMed
 
[18]  Flaherty KR, Toews GB, Travis WD, Colby TV, Kazerooni EA, Gross BH, Jain A, Strawderman RL 3rd, Paine R, Flint A, Lynch JP 3rd, Martinez FJ. Clinical significance of histological classification of idiopathic interstitial pneumonia. Eur Respir J. 2002; 19(2): 275-283.
In article      CrossRefPubMed
 
[19]  Baughman RP, Lower EE & Gibson K. Pulmonary manifastations of sarcoidosis. Presse Med 2012; 41: e289-e302.
In article      CrossRefPubMed
 
[20]  Misumi S and Lynch DA. Idiopathic pulmonary fibrosis/usual interstitial pneumonia: imaging diagnosis, spectrum of abnormalities, and temporal progression. Am Thorax Soc 2006; 3: 307-314.
In article      CrossRefPubMed
 
[21]  Abehsera M, Valeyre D, Grenier P, Jailet H, Battesti JP, Brauner MW. Sarcoidosis with pulmonary fibrosis: CT patterns and correlation with pulmonary function. A J R 2000; 174: 1751-1757.
In article      
 
[22]  Althoff-Souza C, Nestor L. Muller, Julia Flint, Joanne L. Wright, Andrew Churg. Idiopathic pulmonary fibrosis: Spectrum of High-Resolution CT findings. AJR 2005; 185: 1531-1539.
In article      CrossRefPubMed
 
[23]  Lynch DA, Godwin JD, Safrin S, Starko KM, Hormel P, Brown KK, Raghu G, King TE Jr, Bradford WZ, Schwartz DA, Richard Webb W. High-resolution computed tomography in idiopathic pulmonary fibrosis. Am J Respir Crit Care Med. 2005; 172(4): 488-493.
In article      CrossRefPubMed
 
comments powered by Disqus
  • CiteULikeCiteULike
  • MendeleyMendeley
  • StumbleUponStumbleUpon
  • Add to DeliciousDelicious
  • FacebookFacebook
  • TwitterTwitter
  • LinkedInLinkedIn