Pathologic quantification of connective tissue disease-associated versus idiopathic usual interstitial pneumonia.
The histologic differences between CTD-UIP and I-UIP have not been clearly defined. Published scoring systems for assessing interstitial lung disease have been qualitative and, therefore, subject to observer interpretation. Nonetheless, differences between CTD-UIP and I-UIP histology have been reported using qualitative assessments. For example, the number and extent of fibroblastic foci (FFs) may be lower in CTD-UIP than it is in I-UIP10-12, and signs of inflammation (such as lymphoid follicles with germinal centers) may be greater in CTD-UIP than they are in I-UIP. (13) Furthermore, nonspecific interstitial pneumonia (NSIP) may be more prevalent in CTD-associated interstitial lung disease than it is in idiopathic interstitial lung disease. (14,15) The histologic coexistence of the NSIP pattern (NSIP-P) and the UIP pattern is not well-established, although some studies have documented coexisting UIP and NSIP patterns (so-called discordant UIP) as well as potential progression from NSIP to combined NSIP-UIP. (16-18)
There is currently no comprehensive, quantitative, histologic scoring system to evaluate various histologic criteria that may help to distinguish CTD-UIP from I-UIP. Such a system would be subject to less interobserver interpretative variability than would qualitative assessment. Also, such a system could address the validity of published findings using qualitative systems. The purpose of our study was to develop a comprehensive, quantitative, histologic scoring system for assessment of UIP, to determine the distinguishing histologic features of CTD-UIP and I-UIP.
MATERIALS AND METHODS
This research was conducted with approval by the institutional review board of the University of Chicago Medical Center (Chicago, Illinois).
Using an existing clinicopathologic database of patients with interstitial lung disease, a search was performed to identify adults with a histologic diagnosis of UIP. Patients were excluded if a coexisting pulmonary diagnosis was present, such as cancer (primary lung or metastatic to lung), hypersensitivity pneumonitis, drug-induced pulmonary fibrosis, or acute lung injury. Thirty-five patients were included in the study. Patients additionally bearing a clinical diagnosis of connective tissue disease were noted, including rheumatoid arthritis (RA), systemic lupus erythematosus, Sjogren syndrome, polymyositis, dermatomyositis, scleroderma, mixed connective tissue disease, and undifferentiated connective tissue disease. Patients were categorized as CTD-UIP if they had a known CTD (n = 17; 49%), and as I-UIP if no CTD was present (n = 18; 51%). Within the CTD-UIP group, those with RA were subgrouped as RA-CTD (3 of 17; 18%), and those with other CTDs were subgrouped as Other-CTD (14 of 17; 82%). Diagnostic slides and blocks from all 35 patients identified by the database were obtained from the surgical pathology files. Specimens included lung wedge biopsies (n = 27; 77%) and explanted lungs (n = 8; 23%). All slides were reviewed, and the diagnosis of UIP was confirmed using the following histologic criteria: patchy involvement of tissue including areas of spared and involved parenchyma, where the "involved" parenchyma showed evidence of temporal heterogeneity (including older, dense fibrosis and younger FF) and spatial heterogeneity (including predilection for basilar and subpleural areas).
The biopsy specimens were included in the quantitative analysis (n = 27; I-UIP, 14 [52%]; CTD-UIP, 13 [48%], including 3 RA-CTD and 10 Other-CTD). To maintain temporal consistency, the explants were excluded from the quantitative analysis (ie, if biopsy represents "time 0," the explants represent varying time points after diagnosis and were, therefore, excluded). The number of slides per biopsy was variable ([greater than or equal to] 2). Therefore, to standardize analysis across all patients and to create a manageable workload, 2 slides per patient were chosen for analysis. Because the right lung was biopsied more frequently than the left lung was and because the lower lobes are involved by UIP more frequently and earlier than were the upper lobes, slides from the right, lower lobe were chosen. In the absence of a right lower lobe specimen, alternate specimens were chosen with the following priority: (1) left, lower lobe; (2) right, upper lobe; (3) left, upper lobe; (4) lingula; (5) right lung; (6) left lung. The presence of the UIP pattern on each of the 2 slides was verified before scanning.
Quantitative Digital Scoring System for UIP
All 54 slides were digitally scanned using ACIS II (Version 188.8.131.52; formerly, ChromaVision Medical Systems, Inc, San Juan Capistrano, California; now Clarient, Inc, Aliso Viejo, California) at X40 magnification and were analyzed using the Review software (Version 184.108.40.206; Clarient). Two reviewers (N.A.C. and A.N.H.), masked to the presence or absence of CTD, outlined the fragments of tissue on each slide, automatically generated the fragment area (in pixels), and summed the total tissue area per slide. For each slide, all FFs, lymphoid aggregates (LAs), and LA with germinal centers were outlined, and the area (in pixels) was automatically calculated. All pixel areas were converted to micrometers squared with the following conversion: 1 pixel = 1 [[micro]m.sup.2] For each slide, the number of FF was identified (FF count), and the sum of the areas of FF was calculated (FF area). For each slide, the FF count per total area was calculated, as well as the FF area per total area. The same was done for LAs, with and without germinal centers (Figure 1, A through D). To assess interobserver variability, 6 slides from 3 patients with I-UIP were chosen for analysis by a third reviewer (I.O.G.) following the same procedure as described above.
Qualitative Histologic Assessment for NSIP-P
Both the biopsy and explant specimens were included in the qualitative analysis (n = 27 [77%] and 8 [23%], respectively). Slides (the 2 chosen slides from each biopsy case, and all slides from the explant cases) were examined microscopically by 2 reviewers (N.A.C. and A.N.H.), who were masked to the presence or absence of CTD. The presence or absence of the NSIP-P away from the areas of UIP fibrosis was noted using the following histologic criteria: (1) the area was considered away" from UIP fibrosis if it was at least one X40 field (5 mm) away from the edge of dense fibrosis; (2) the cellular NSIP-P consisted of widened alveolar septae, occupied by interstitial inflammatory infiltrates, spanning at least one X40 field (5 mm); and (3) the fibrotic NSIP-P consisted of alveolar septae widened by dense fibrosis with few to no inflammatory infiltrates, spanning at least one X40 field (5 mm). If the entire portion of lung parenchyma on a given slide was affected by UIP, the slide was deemed inadequate to evaluate for NSIP-P.
Statistical analysis was performed with Stata 10.0 (StataCorp LP; College Station, Texas). The means of the 2 slides for each patient were analyzed for each measure described above. Patients were assigned to the following groups as previously described: I-UIP and CTD-UIP (containing both RA-CTD and Other-CTD). For each measure, the median and interquartile range (Q1, Q3) were determined for each group. Wilcoxon rank sum test and Fisher exact test were used to test for differences between patient groups for each measure. Additionally, the Wilcoxon signed-rank test was used to compare the judgments of the initial 2 reviewers to those of the third reviewer. Two-sided P-values less than .05 were considered statistically significant.
Quantitative Digital Scoring System for UIP
Findings included lower FF counts and lower FF areas in patients with CTD-UIP (0.0884/[[micro]m.sup.2] and 0.0016 [[micro]m.sup.2])as compared with patients with I-UIP (0.1134/[micro][m.sup.2] and 0.0033 [micro][m.sup.2])(P = .26 and P = .08, respectively). The LA counts and LA areas were greatest for patients with RA-CTD (0.0992/ [micro][m.sup.2] and 0.0040 [micro][m.sup.2])(P = .61 and P = .10, respectively) and were similar among the remaining patients in the Other CTD group (0.0728/[micro][m.sup.2] and 0.0022 [micro][m.sup.2]) and among patients with I-UIP (0.0852/[micro][m.sup.2] and 0.0024 [micro][m.sup.2])(P = .95 and P = .73, respectively), as well as among the CTD-UIP patient group as a whole (0.0751/[micro][m.sup.2] [micro][m.sup.2]and 0.0024 [micro][m.sup.2])(P = .88 and P = .36, respectively). These trends did not reach statistical significance (P values for all groups derived in comparison to I-UIP) (Table; Figure 2, A through D). The presence of LAs with germinal centers was rare and insufficient for further analysis.
Regarding interobserver variability, there were no statistically significant differences between pathologists involving any of the assessed parameters (FF count, FF area, LA count, and LA area; P = .59, P = .11, P = .11, and P = .11, respectively).
[FIGURE 1 OMITTED]
Qualitative Histologic Assessment for NSIP-P
Analysis of biopsies demonstrated that the prevalence of the NSIP-P was higher in patients with CTD-UIP compared with patients with I-UIP (P = .005). The subgroups RA-CTD and Other-CTD also demonstrated higher rates of NSIP-P compared with I-UIP (P = .03 and P = .04, respectively) (Table; Figure 2, E).
Analysis of explanted lungs showed that 1 of 4 patients with I-UIP (25%) had cellular NSIP-P in all lobes and 1 (25%) had patchy fibrotic NSIP-P. Three of 4 patients with CTD-UIP (75%) had cellular NSIP-P in all lobes. Two of these patients (2 of 3; 67%) had scleroderma and one (1 of 3; 33%) had undifferentiated connective tissue disease. In both groups, patients without multilobar NSIP-P demonstrated widened alveolar walls only in the areas adjacent to fibrosis, which was not seen in areas distant from fibrosis (Figure 3, A through F).
We report the development of a comprehensive, quantitative, histologic scoring system to evaluate histologic criteria thought to be important in distinguishing CTD-UIP from I-UIP. We found that although the patients with CTD-UIP demonstrated a trend toward slightly fewer, smaller FFs, as compared with those with I-UIP, the differences were not statistically significant. Our quantitative findings, therefore, do not completely validate previously published qualitative studies demonstrating fewer FFs in patients with connective tissue disease. (10-12) The presence of fewer FFs is thought to be associated with increased survival in patients with UIP. (19)
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
We also found through our quantitative assessment that the number and size of LAs were similar between patients with CTD-UIP and those with I-UIP. A similar study suggested that germinal centers are the best distinguishing feature between CTD-UIP and I-UIP. (13) Although we considered LAs with germinal centers separately, the prevalence of germinal centers was not high enough in our samples to make adequate comparisons. Interestingly, there was a trend toward more, larger LAs in the RA subset in our study. This finding might help identify patients with RA who present initially with pulmonary symptoms, which has important prognostic and therapeutic implications.
The most striking findings from our study involved the prevalence of NSIP pattern. There was a significantly higher prevalence of NSIP-P in all CTD-UIP cases (both RA-CTD and Other-CTD) compared with I-UIP cases. In explanted lung specimens, the pattern was most often cellular NSIP and, when identified, was present in all lobes. Therefore, when tissue from all lobes is available for analysis, the identification of multilobar NSIP-P in the setting of UIP may suggest an underlying connective tissue disease. It is important to make the diagnosis of NSIP-P in areas away from the UIP fibrosis because widened alveolar walls (mimicking NSIP) were present in areas adjacent to fibrosis in both CTD-UIP and I-UIP tissues. Previous studies have suggested that the NSIP-P is more common in patients with CTD-associated interstitial lung disease. (14,15) Also, a recent, longitudinal study (18) has shown that some patients with progressive NSIP develop subpleural honeycombing and histologic features suggestive of UIP, providing support for the possible coexistence of these 2 histologically distinct patterns. Our study was not designed to assess the overall prevalence of NSIP in patients with CTD but, rather, to evaluate the presence of the NSIP-P in the setting of UIP. Our finding may help shed light on the pathogenesis and progression of CTD-associated interstitial lung disease.
There were several limitations to our study, one of which was the small sample size. Analyzing a larger number of both patients with CTD-UIP and patients with I-UIP may have revealed greater differences. Furthermore, we could not control for a possible surgical sampling bias in the biopsy specimens. Lower-lobe samples are usually most severely affected by fibrosis and, hence, are less likely to reveal an NSIP-P. Our finding of a significantly increased prevalence of multilobar NSIP-P, therefore, emphasizes the need for surgeons to sample all lobes in cases of interstitial lung disease. Finally, although we attempted to assess the validity of germinal centers as being a distinguishing feature between CTD-UIP and I-UIP, (13) our samples had too few germinal centers for sufficient analysis.
In summary, we found the coexistence of the UIP pattern and the NSIP-P in multiple lobes to be an important feature in distinguishing CTD-UIP from I-UIP, with CTD demonstrating an increased prevalence of multilobar, cellular NSIP-P in areas away from UIP fibrosis. Additionally, patients with CTD-UIP demonstrated smaller FFs and patients with RA and UIP demonstrated larger lymphoid aggregates than did patients with I-UIP, but those differences were not statistically significant. Overall, histologic assessment of UIP using our comprehensive, quantitative scoring system could not distinguish CTD-UIP from IUIP based on FFs and LAs and, therefore, does not support previously published, qualitative assessments that use similar histologic features to distinguish these 2 entities.
(1.) Lynch JP III, Saggar R, Weigt SS, Zisman DA, White ES. Usual interstitial pneumonia. Semin Respir Crit Care Med. 2006;27(6):634-651.
(2.) de Lauretis A, Veeraraghavan S, Renzoni E. Review series: aspects of interstitial lung disease: connective tissue disease-associated interstitial lung disease: how does it differ from IPF?: how should the clinical approach differ? Chron Respir Dis. 2011;8(1):53-82.
(3.) Rice AJ, Wells AU, Bouros D, et al. Terminal diffuse alveolar damage in relation to interstitial pneumonias: an autopsy study. Am J Clin Pathol. 2003; 119(5):709-714.
(4.) Churg A, Muller NL, Silva CI, Wright JL. Acute exacerbation (acute lung injury of unknown cause) in UIP and other forms of fibrotic interstitial pneumonias. Am J Surg Pathol. 2007;31(2):277-284.
(5.) Kondoh Y, Taniguchi H, Kawabata Y, Yokoi T, Suzuki K, Takagi K. Acute exacerbation in idiopathic pulmonary fibrosis: analysis of clinical and pathologic findings in three cases. Chest. 1993;103(6):1808-1812.
(6.) Akira M. Computed tomography and pathologic findings in fulminant forms of idiopathic interstitial pneumonia. J Thorac Imaging. 1999;14(2):76-84.
(7.) Churg A, Wright JL, Tazelaar HD. Acute exacerbations of fibrotic interstitial lung disease. Histopathology. 2011;58(4):525-530.
(8.) Su R, Bennett M, Jacobs S, et al. An analysis of connective tissue disease-associated interstitial lung disease at a US tertiary care center: better survival in patients with systemic sclerosis. J Rheumatol. 2011;38(4):693-701.
(9.) Park JH, Kim DS, Park IN, et al. Prognosis of fibrotic interstitial pneumonia: idiopathic versus collagen vascular disease-related subtypes. Am J Respir Crit Care Med. 2007;175(7):705-711.
(10.) Enomoto N, Suda T, Kato M, et al. Quantitative analysis of fibroblastic foci in usual interstitial pneumonia. Chest. 2006;130(1):22-29.
(11.) Flaherty KR, Colby TV, Travis WD, et al. Fibroblastic foci in usual interstitial pneumonia: idiopathic versus collagen vascular disease. Am J Respir Crit Care Med. 2003;167(10):1410-1415.
(12.) King TE Jr, Schwarz MI, Brown K, et al. Idiopathic pulmonary fibrosis: relationship between histopathologic features and mortality. Am J Respir Crit Care Med. 2001;164(6):1025-1032.
(13.) Song JW, Do KH, Kim MY, Jang SJ, Colby TV, Kim DS. Pathologic and radiologic differences between idiopathic and collagen vascular disease-related usual interstitial pneumonia. Chest. 2009;136(1):23-30.
(14.) Tansey D, Wells AU, Colby TV, et al. Variations in histological patterns of interstitial pneumonia between connective tissue disorders and their relationship to prognosis. Histopathology. 2004;44(6):585-596.
(15.) Nakamura Y, Chida K, Suda T, et al. Nonspecific interstitial pneumonia in collagen vascular diseases: comparison of the clinical characteristics and prognostic significance with usual interstitial pneumonia. Sarcoidosis Vasc Diffuse Lung Dis. 2003;20(3):235-241.
(16.) Monaghan H, Wells AU, Colby TV, du Bois RM, Hansell DM, Nicholson AG. Prognostic implications of histologic patterns in multiple surgical lung biopsies from patients with idiopathic interstitial pneumonias. Chest. 2004; 125(2):522-526.
(17.) Flaherty KR, Travis WD, Colby TV, et al. Histopathologic variability in usual and nonspecific interstitial pneumonias. Am J Respir Crit Care Med. 2001; 164(9):1722-1727.
(18.) Schneider F, Hwang DM, Gibson K, Yousem SA. Nonspecific interstitial pneumonia: a study of 6 patients with progressive disease. Am J Surg Pathol. 2012;36(1):89-93.
(19.) Tiitto L, Bloigu R, Heiskanen U, Paakko P, Kinnula VL, Kaarteenaho-Wiik R. Relationship between histopathological features and the course of idiopathic pulmonary fibrosis/usual interstitial pneumonia. Thorax. 2006;61(12):1091 1095.
Nicole A. Cipriani, MD; Mary Strek, MD; Imre Noth, MD; Ilyssa O. Gordon, MD, PhD; Jeff Charbeneau, MS; Jerry A. Krishnan, MD, PhD; Thomas Krausz, MD; Aliya N. Husain, MD
Accepted for publication June 1, 2012.
From the Departments of Pathology (Drs Cipriani, Gordon, Krausz, and Husain), and Medicine, Division of Pulmonary and Critical Care Medicine (Drs Strek and Noth), the University of Chicago, Chicago, Illinois; and the Department of Pulmonary, Critical Care, and Sleep Medicine, University of Illinois Hospital & Health Sciences System, Chicago (Mr Charbeneau and Dr Krishnan).
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Nicole A. Cipriani, MD, Department of Pathology, the University of Chicago, 5841 S Maryland Ave, MC 6101, P-615, Chicago, IL 60637 (e-mail: Nicole.Cipriani@uchospitals.edu).
Median Value for Each Measure Analyzed in Each Group, Biopsy Specimens Measure I-UIP CTD-UIP RA-CTD Other- CTD FF count, 0.1134 0.0884 0.0846 0.0961 No./[micro][m.sup.2] FF area, 0.0033 0.0016 0.0014 0.0024 [micro][m.sup.2] LA count, 0.0852 0.0751 0.0992 0.0728 No./[micro][m.sup.2] LA area, 0.0024 0.0024 0.0040 0.0022 [micro][m.sup.2] NSIP prevalence, % 18.18 88.89 100 83.33 Abbreviations: CTD-UIP, connective tissue disease-associated usual interstitial pneumonia; FF, fibroblastic focus; I-UIP, idiopathic interstitial pneumonia; LA, lymphocyte aggregate; NSIP, nonspecific interstitial pneumonia; Other-CTD, non-rheumatoid arthritis subtypes of connective tissue disease-associated usual interstitial pneumonia; RA-CTD, rheumatoid arthritis subtype of connective tissue disease-associated usual interstitial pneumonia.
|Printer friendly Cite/link Email Feedback|
|Author:||Cipriani, Nicole A.; Strek, Mary; Noth, Imre; Gordon, Ilyssa O.; Charbeneau, Jeff; Krishnan, Jerry A|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Oct 1, 2012|
|Previous Article:||Tertium non datur? Legitimacy of a third diagnostic category in melanocytic lesions.|
|Next Article:||The separation of benign and malignant mesothelial proliferations.|