Interstitial Pneumonia Related to Undifferentiated Connective Tissue Disease Pathologic Pattern and Prognosis

Transcription

1 [ Original Research Diffuse Lung Disease ] Interstitial Pneumonia Related to Undifferentiated Connective Tissue Disease Pathologic Pattern and Prognosis Ho-Cheol Kim, MD ; Wonjun Ji, MD ; Mi Young Kim, MD, PhD ; Thomas V. Colby, MD, PhD ; Se Jin Jang, MD, PhD ; Chang-Keun Lee, MD, PhD ; Seung Bong Han, PhD ; and Dong Soon Kim, MD, PhD BACKGROUND: Undifferentiated connective tissue disease (UCTD) involves conditions characterized by both having symptoms of connective tissue disease (CTD) and autoantibodies but not fulfilling the criteria of a specific CTD. The frequency or prognosis of the usual interstitial pneumonia (UIP) pattern in UCTD is unknown, which may be confused with idiopathic pulmonary fibrosis (IPF). This study aimed to investigate the frequency of the UIP pattern in interstitial pneumonia related to UCTD and compare its prognosis with that of IPF and UCTD-nonspecific interstitial pneumonia (UCTD-NSIP). METHODS: The medical records of 788 patients presumptively diagnosed with idiopathic interstitial pneumonia at Asan Medical Center from January 2005 to December 2012 were retrospectively reviewed. UCTD was diagnosed according to the criteria by Corte and colleagues, and the prognoses were compared between UCTD-UIP and UCTD-NSIP and between UCTD-UIP and IPF. RESULTS: Among 105 patients with UCTD (13.3% of total subjects), 44 had a UIP pattern (by surgical lung biopsy: 24; by high-resolution CT scan: 20), 29 had a nonspecific interstitial pneumonia pattern (by surgical lung biopsy), and nine had an organizing pneumonia pattern (by biopsy). The overall survival of the UCTD-UIP group was shorter than that of the UCTD- NSIP group ( P 5.021) but significantly better than that of the IPF group ( P 5.042). CONCLUSIONS: A UIP pattern, which seems to be frequent in UCTD, showed a poorer prognosis than that of UCTD-NSIP. However, the prognosis of UCTD-UIP was significantly better than that of IPF, highlighting the importance of searching for underlying UCTD in suspected IPF cases. CHEST 2015; 147 ( 1 ): Manuscript received February 2, 2014; revision accepted August 10, 2014; originally published Online First September 11, ABBREVIATIONS: ANA 5 antinuclear antibody; CTD 5 connective tissue disease; Dlco 5 diffusing capacity of the lung for carbon monoxide; ENA 5 extractable nuclear antigen; HRCT 5 high-resolution CT; IIP 5 idiopathic interstitial pneumonia; ILD 5 interstitial lung disease; IPF 5 idiopathic pulmonary fibrosis; NSIP 5 nonspecific interstitial pneumonia; OP 5 organizing pneumonia; RF 5 rheumatoid factor; UCTD 5 undifferentiated connective tissue disease; UCTD-IP 5 interstitial pneumonia related to undifferentiated connective tissue disease; UIP 5 usual interstitial pneumonia AFFILIATIONS: From the Department of Pulmonary and Critical Care Medicine (Drs H.-C. Kim, Ji, and D. S. Kim) and Department of Radiology (Dr M. Y. Kim), Asan Medical Center, College of Medicine, University of Ulsan, Seoul, South Korea; Department of Laboratory Medicine and Pathology (Dr Colby), Mayo Clinic, Scottsdale, AZ; and the Department of Pathology (Dr Jang), Division of Rheumatology, Department of Internal Medicine (Dr Lee), and Department of Clinical Epidemiology and Biostatistics (Dr Han), Asan Medical Center, College of Medicine, University of Ulsan, Seoul, South Korea. Drs H.-C. Kim and Ji contributed equally as the co-first authors. FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study. CORRESPONDENCE TO: Dong Soon Kim, MD, PhD, Department of Pulmonary and Critical Care Medicine, Asan Medical Center, College of Medicine, University of Ulsan, 88 Olympic-ro 43-gil, Songpa-gu, Seoul, South Korea ; dskim615@gmail.com 2015 AMERICAN COLLEGE OF CHEST PHYSICIANS. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: /chest journal.publications.chestnet.org 165

2 Undifferentiated connective tissue disease (UCTD) is a clinical entity as described in the rheumatology community characterized by features that are suggestive of connective tissue disease (CTD) but not fulfilling the classification criteria for a specific CTD. 1,2 Kinder et al 3 reported that most patients with a nonspecific interstitial pneumonia (NSIP) pattern on surgical lung biopsy met their case definition of UCTD. Later, several authors also reported the common occurrence of UCTD in surgical lung biopsy-proven NSIP, 4-6 suggesting that most idiopathic NSIP may actually be the lung manifestation of UCTD. 3 However, the diagnostic criteria for UCTD differ, and Kinder et al 3 used broad criteria, including nonspecific features like elevated erythrocyte sedimentation rate or gastroesophageal reflux, that might result in an overestimation of UCTD. Furthermore, although the NSIP pattern was first described and seems to be the predominant pattern in UCTD, no study, to our knowledge, has investigated the relative frequency of NSIP or usual interstitial pneumonia (UIP) patterns in UCTD. Vij et al 7 reported that the UIP pattern was the most common pattern in autoimmune-featured interstitial pneumonia, suggesting that the UIP pattern may be more frequent than previously thought. In other aspects, the prognosis of idiopathic pulmonary fibrosis (IPF) is much poorer than that of a UIP pattern related to classic CTD, with the possible exception of rheumatoid arthritis Moreover, autoantibodies like antinuclear antibody (ANA) and rheumatoid factor (RF) are frequently found in patients with IPF, 11 suggesting that UCTD may also be present but undiagnosed in these patients. Consequently, it is clinically important to know the frequency of UCTD in patients with presumptive diagnosis of IPF and the difference in prognosis between UCTD-UIP and IPF. This study aimed to investigate the relative frequency of UIP and NSIP patterns in patients with UCTD-related interstitial pneumonia (UCTD-IP) and their prognostic differences, and to compare the prognoses for UCTD-UIP and IPF. Materials and Methods Study Population A total of 788 patients seen at Asan Medical Center due to presumptive idiopathic interstitial pneumonia (IIP) from January 2005 to December 2012 were included in this study. Patients with classic CTDs or an exposure history in relation to the possible causes of interstitial lung disease (ILD), such as drugs or environmental agents, were excluded. Data were retrospectively collected from medical records. A thorough systematic history, including any rheumatologic symptoms and signs, with serologic testing for CTD was obtained from all subjects at the time of initial diagnosis and intermittently during follow-up. Many of the patients analyzed here had been included in our previous studies. 12,13 Because this study was a retrospective review of medical records, written informed consent was waived. The study protocol was approved by the institutional review board of Asan Medical Center (approval number ). Autoantibody Test All patients underwent serologic tests for autoantibodies at the time of diagnosis. ANA was tested in the serum with a commercially available prestandardized kit (ANA/HEp-2 Test System; ZEUS Scientific, Inc). Any sample that had a positive result above 1:160 titer was considered positive. Extractable nuclear antigens (ENAs) were tested with an ENA Combi ELISA kit (BL Diagnostika). A signal-to-cut-off ratio. 1.0 was considered positive. ENAs testing included anti-ro (SS-A), anti-la (SS-B), anti-scl-70, anti-ribonucleoprotein, and Jo-1 antibodies. RF was measured with a commercially available kit (RapiTexRF; Dade Behring Inc/ Siemens Medical Solutions) that uses slide latex agglutination for qualitative measurements. Any sample that had a positive result at above a 1:160 International Unit was considered positive. Histopathology Surgical lung biopsy was performed in 363 patients and transbronchial lung biopsy in 11 patients. The biopsy specimen slides of patients with UCTD were rereviewed independently by two pathologists (T. V. C., S. J. J.) who were blind to the clinical findings. Final consensus histopathologic diagnosis was made according to the American Thoracic Society/European Respiratory Society international consensus classification of IIPs. 14 High-Resolution CT Scan High-resolution CT (HRCT) scan was performed at baseline in all patients and followed up annually or at a time of acute change in the patient s condition. The HRCT scan images were reviewed in a blind manner by the radiologist (M. Y. K.) with. 10 years experience in ILD. The HRCT scan pattern was categorized as the typical UIP pattern or as a non-uip pattern. 14 Because the typical UIP pattern in HRCT scans was recently shown to be highly specific for the pathologic UIP pattern in patients with rheumatoid arthritis similar to IPF, we used this HRCT scan pattern for the diagnosis of UCTD-UIP. 15 Diagnostic Criteria for UCTD and Follow-up Course The narrower diagnostic criteria for UCTD recommended by Corte et al 4 were used, that is, the presence of one or more of the specific autoantibodies combined with more than one symptom or sign of CTD (including Raynaud s phenomenon, sicca symptoms, arthralgia, morning stiffness, or proximal muscle weakness). 4,16 All final diagnoses were made via the multidisciplinary approach that included experienced pathologists, clinicians, and radiologists. The diagnosis of IPF was reconfirmed according to the new American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Latin American Thoracic Society guidelines. 17 The clinical follow-up courses and survival statuses of the patients until July 2013 were obtained from medical records, National Health Insurance of Korea records, and/or telephone interviews. Treatment responses were categorized as improvement, stable, or deterioration according to the change in the pulmonary function test. 17,18 Improvement/deterioration was defined as a change of 10% in the FVC and/or 15% in the diffusing capacity of the lung for carbon monoxide (D lco ). 18 Statistical Analysis The comparisons of baseline characteristics and treatments were made using a Student t test for continuous variables and a x 2 or Fisher exact test for categorical data. All P values were two-tailed, with statistical significance set at P,.05. Kaplan-Meier survival analysis was used to evaluate the differences in survival rates and risk factors for mortality were analyzed with Cox proportional hazards models. Statistical analyses were performed with SPSS, version 18.0 for Windows (IBM). 166 Original Research [ 147 # 1 CHEST JANUARY 2015 ]

3 Results Patient Characteristics and the Frequency of Pathologic Patterns in UCTD Among the 788 patients with presumptive IIP, 105 (13.3%) met the UCTD criteria. Patients with UCTD were younger (59.0 years vs 65.1 years) and more likely to be female (62.4% vs 21.8%) than patients with non-uctd-iip. The most common rheumatologic symptom was arthralgia (59.0%), and the most common laboratory finding was positive ANA (51.4%) ( Table 1 ). The k value for histologic classification between two pathologists was Among the 105 patients with UCTD, 44 had the UIP pattern (24 by surgical lung biopsy and 20 by the typical UIP pattern on HRCT scan), and 29 patients had the NSIP pattern (all diagnosed by surgical lung biopsy). An organizing pneumonia (OP) pattern was found in nine patients (by transbronchial lung biopsy in four patients, by surgical lung biopsy in five patients), and the pathologic type of the remaining 23 patients was uncertain because no surgical lung biopsy was performed. The UCTD-NSIP group (52.5 years of age) was younger than the UCTD-UIP group (62.8 years of age). BAL and baseline pulmonary function test results were not significantly different between the two groups ( Table 2 ). Among 683 patients with proven IIP, 499 (73.1%) were diagnosed as IPF, 81 (11.9%) as idiopathic NSIP (biopsy-proven), and 59 (8.6%) a cryptogenic OP (biopsy-proven; seven cases by transbroncheal lung biopsy). The pattern of the remaining patients was uncertain because no surgical lung biopsy was performed. Among the patients with the UIP pattern, the prevalence of UCTD was 8.1%. Among the NSIP pattern, UCTD was found in 26.3%. Treatment and Response of UCTD-IP Treatment with steroids alone or with immunosuppressant was given to 77 patients (73.3%) according to the attending physician s judgment. The remaining patients were not treated either because the disease was mild, or due to patient refusal. There were no significant differences in the treatment regimen or response to therapy between the UCTD-UIP and UCTD-NSIP groups (Table 3 ). Comparison of the Prognosis Between UCTD-UIP and UCTD-NSIP The survival period of patients with UCTD-NSIP was significantly longer than those with UCTD-UIP TABLE 1 ] Baseline Characteristics of 105 Patients With UCTD Characteristics UCTD-IP Age, mean SD, y Sex, male 29 (27.6) Smoking (ever smoker) 30 (28.6) Symptoms Arthralgia 62 (59.0) Sicca symptom 56 (53.3) Raynaud phenomenon 32 (30.5) Morning stiffness 11 (10.5) Muscle weakness 1 (1.0) Autoantibody ANA (titer 160) 54 (51.4) SS-A (Ant-Ro) 40 (38.1) Anti-Jo-1 15 (14.3) Scl (11.4) RF (titer 160) 6 (5.7) Anti-RNP 5 (4.8) SS-B (Ant-La) 2 (1.9) Positive two antibodies 23 (21.9) Positive three antibodies 4 (3.8) HRCT scan pattern Typical UIP pattern 32 (30.5) Non-UIP pattern 73 (69.5) Surgical lung biopsy (No. 558) UIP 24 (41.4) NSIP 29 (50.0) OP 5 (8.6) Final pattern UIP 44 NSIP 29 OP 9 Uncertain 23 Values are reported as mean SD or as frequency (%). ANA 5 antinuclear antibody; HRCT 5 high-resolution CT; NSIP 5 nonspecific interstitial pneumonia; OP 5 organizing pneumonia; RF 5 rheumatoid factor; RNP 5 ribonucleoprotein; SS 5 Sjogren syndrome; UCTD 5 undifferentiated connective tissue disease; UCTD-IP 5 interstitial pneumonia related to undifferentiated connective tissue disease; UIP 5 usual interstitial pneumonia. ( P 5.021) ( Fig 1 ). The univariate Cox proportional hazards model showed that the UCTD-NSIP pattern was a significant independent prognostic factor (hazard ratio, 0.256; 95% CI, ; P 5.033). However, in the multivariate Cox proportional hazards model, a UCTD-NSIP pattern was not a significant independent prognostic factor ( Table 4 ). journal.publications.chestnet.org 167

4 TABLE 2 ] Comparison of the Baseline Features Between the UCTD-UIP and UCTD-NSIP Groups Features UIP NSIP P Value Patient number 44 (41.9) 29 (27.6) Age, y ,.005 Sex, male 15 (34.1) 6 (20.7).216 Smoking (ever smoker) 16 (36.4) 6 (20.7).153 BAL (UIP: No. 5 34; NSIP: No. 5 18) WBC count, / m L Neutrophil, % Lymphocyte, % Baseline PFT (UIP: No. 5 44; NSIP: No. 5 29) FVC, % FEV 1, % D LCO, % TLC, % MWT, distance Values are reported as mean SD or as frequency (%). 6MWT 5 6-min walk test; D LCO 5 diffusing capacity of the lung for carbon monoxide; PFT 5 pulmonary function test; TLC 5 total lung capacity. See Table 1 legend for expansion of other abbreviations. Because the pathologic pattern was not confirmed in 23 patients, the clinical features and survival curves of these patients were compared separately. As shown in Table 2, there was no difference in the clinical features between UCTD-UIP and UCTD-NSIP except age and the age of these 23 patients with uncertain pathologic pattern was years, similar to UCTD-UIP TABLE 3 ] Clinical Course of the Patients With UCTD-UIP and UCTD-NSIP Characteristics UCTD-UIP UCTD-NSIP P Value Patient number Treatment regimen.695 Steroid only 17 (38.6) 13 (44.8) Steroid 1 immunosuppressant 14 (31.8) 10 (34.5) None 13 (29.5) 6 (20.7) Treatment response (UCTD-UIP: No. 5 31; UCTD-NSIP: No. 5 23) Improvement 5 (16.1) 3 (13.0) Stable 19 (61.3) 17 (73.9) Deterioration 7 (22.6) 3 (13.0).619 Median follow-up, mo Survival (%) y y Values are reported as frequency (%). See Table 1 legend for expansion of abbreviations. group ( years) ( e-table 1 ). And the survival of these patients was also similar to that of the UCTD-UIP group ( e-fig 1 ). Comparison of the Survival Between IPF and UCTD-UIP The baseline characteristics of the IPF and UCTD-UIP groups were summarized in Table 5. Kaplan-Meier survival analysis showed significantly better prognosis of the UCTD-UIP group (n 5 44) than the IPF group ( P 5.042) (Fig 2 ). The univariate Cox proportional Figure 1 Comparison of the survival curves of the patients with UCTD-UIP and UCTD-NSIP. NSIP 5 nonspecific interstitial pneumonia; UCTD 5 undifferentiated connective tissue disease; UIP 5 usual interstitial pneumonia. 168 Original Research [ 147 # 1 CHEST JANUARY 2015 ]

5 TABLE 4 ] Predicting Factors for Mortality in Patients With UCTD-IP Assessed by Cox Proportional Hazards Model Parameters Hazard Ratio 95% CI P Value Univariate analysis NSIP diagnosis Sex, female Age Ever smokers BAL WBC count, / m L BAL neutrophil, % BAL lymphocyte, % FVC, % FEV 1, % D LCO, % TLC, % MWT, distance Treatment with steroid or immunosuppressant Multivariate analysis a NSIP diagnosis Age Sex, female FVC See Table 1 and 2 legends for expansion of abbreviations. a For the multivariate analysis, a total of 73 patients (UCTD-UIP 44 patients, UCTD-NSIP 29 patients) were evaluated. The covariates for the full model were diagnosis, age, sex, and FVC. Among the covariates significant in the univariate analysis, TLC was not included in the full model because of high correlation with FVC. hazards model revealed that UCTD-UIP was a significant independent prognostic factor for a reduced mortality of 43% compared with the IPF group (hazard ratio, 0.576; 95% CI, ; P 5.045). On multivariate analysis, UCTD-UIP pattern was a significant independent prognostic factor (hazard ratio, 0.565; 95% CI, ; P 5.030) ( Table 6 ). Interestingly, our data suggested that the treatment with steroid or immunosuppressant had a negative effect (worse prognosis) in the patients with IPF, however, it should be interpreted cautiously because the sicker, progressing patients are more likely to receive such therapy. Comparison of the Survival Between Idiopathic NSIP and UCTD-NSIP We compared UCTD-NSIP with idiopathic NSIP ( e-table 2 ). Kaplan-Meier survival analysis showed no significant differences in survival between the two groups (e-fig 2 ). Follow-up and Development of Definite CTD The median follow-up duration of patients with UCTD was 32.3 months (interquartile range, months). During follow-up, three patients (2.9%) developed definite CTD. Clinical characteristics of these three patients are presented in e-table 3. Discussion Our current study showed that the UIP pattern was as common as the NSIP pattern in patients with UCTD, and the prognosis of UCTD-UIP was better than that of IPF, although there was no difference in survival between UCTD-NSIP and idiopathic NSIP. Our data support the importance of differentiation between IPF and UCTD-UIP, although UCTD may not be easily recognized and may be misdiagnosed as IPF. Although the presence of underlying CTD is crucial both for prognosis and treatment in the patients with interstitial pneumonia, there is a subset of patients with symptoms, signs, and serologic tests suggestive of an underlying autoimmune disease but they do not fulfill all of the diagnostic criteria for a specific CTD. These patients have been designated as having UCTD. However, a conceptual discrepancy exists for UCTD between rheumatologists and pulmonologists. 19,20 Furthermore, the diagnostic criteria for UCTD are highly variable. Therefore, several different terminologies have been proposed to categorize these patients, such as lungdominant CTD by Fischer et al 19 and autoimmunefeatured ILD by Vij et al. 7 However, the criteria of autoimmune-featured ILD included several nonspecific symptoms like weight loss, gastroesophageal reflux, or dysphagia and those for lung-dominant CTD included histopathologic features of surgical lung biopsy specimen. Therefore, we decided to investigate UCTD using the criteria accepted by both rheumatologists and pulmonologists. In contrast to other UCTD studies, 2,5 we have reviewed all of the patients with presumptive IIP, including those with undetermined histopathologic patterns due to the lack of surgical lung biopsy. Although this study was a retrospective analysis, we prospectively collected all data on clinical features and the presence of blood autoantibodies from the beginning. We found that 13.3% of the patients with presumptive IIP had UCTD, which is lower than the 21% reported by Corte et al 4 who only investigated the patients on whom surgical lung biopsy was performed, resulting in higher prevalence of the NSIP pattern (45%) compared with other series (in our cohort, 11.9%). Because UCTD is more common in the NSIP pattern, consequently they might have found higher frequency of UCTD among the journal.publications.chestnet.org 169

6 TABLE 5 ] Comparison Between IPF and UCTD-UIP Features IPF UCTD-UIP P Value Patient number Age, y Sex, male 390 (78.2) 15 (34.1),.005 Smoking (ever smoker) 355 (71.1) 16 (36.4),.005 Surgical lung biopsy 204 (40.9) 24 (54.5).080 BAL (IPF: No ; UCTD-UIP: No. 5 34) WBC count, / m L Neutrophil, % Lymphocyte, % Baseline PFT, frequency (%) (IPF: No ; UCTD-UIP: No. 5 44) FVC FEV D LCO TLC MWT, distance Median follow-up, mo Survival (%) y y Values are reported as mean SD or as frequency (%). IPF 5 idiopathic pulmonary fibrosis. See Table 1 and 2 legends for expansion of other abbreviations. total subjects. Vij et al 7 reported that the prevalence of autoimmune-featured ILD was 52% of presumable IIP; however, their criteria included several nonspecific features as mentioned previously. Figure 2 Kaplan-Meier survival curves of the patients with UCTD-UIP and IPF. IPF 5 idiopathic pulmonary fibrosis. See Figure 1 legend for expansion of other abbreviations. Although the frequency of UCTD was three times higher in the NSIP than in the UIP pattern, our current study showed that in patients with ILD and UCTD, the UIP pattern was as frequent as or more frequent than the NSIP pattern, which was consistent with the findings of Vij et al 7 (25 UIP pattern vs two NSIP pattern among 63 patients with autoimmune-featured ILD). We also confirmed that among the patients with UCTD, the overall prognosis was better for patients with the NSIP pattern than those with the UIP pattern, although not significant on multivariate analysis. Furthermore, Kaplan-Meier survival analysis showed a significant difference between UCTD-UIP and IPF; the presence of UCTD in the patients with the UIP pattern was a significant independent prognostic factor on multivariate analysis. Previously, Corte et al 4 reported that UCTD was not associated with a survival benefit, which may be due to smaller numbers (total 101 patients) of the subject compared with ours. Recently, Strand et al 21 also reported that there was no significant difference in survival between UCTD-UIP and IPF. However, the number of their subjects was small (UCTD-UIP: 19 patients). Furthermore, there was no mention about the diagnostic criteria of UCTD, which may affect the 170 Original Research [ 147 # 1 CHEST JANUARY 2015 ]

7 TABLE 6 ] Predicting Factors for Mortality in Patients With IPF and UCTD-UIP Assessed by Cox Proportional Hazards Model Parameters Hazard Ratio 95% CI P Value Univariate analysis UCTD-UIP diagnosis Sex, female Age Ever smokers BAL WBC count, / m L BAL neutrophil, % BAL lymphocyte, % FVC, % ,.001 FEV 1, % ,.001 D LCO, % ,.001 TLC, % ,.001 6MWT, distance ,.001 Treatment with steroid or immunosuppressant Multivariate analysis a UCTD-UIP diagnosis Age Sex, female FVC ,.001 Treatment with steroid or immunosuppressant See Table 1, 2, and 5 legends for expansion of abbreviations. a For the multivariate analysis, the covariates for the full model were diagnosis, age, sex, and FVC. Among the covariates significant in the univariate analysis, BAL neutrophil was not included in the multivariate analysis because of the high proportion of the missing data and FEV 1, and TLC due to high correlation with FVC. results. Our previous study showed that the mere presence of autoantibody had no influence on survival in the patients with IPF. 12 This study demonstrated, for the first time, to our knowledge, that the prognosis of UCTD-UIP was better than that of IPF, reenforcing the importance of diagnosing associated UCTD in patients with suspected IPF. Considering the failure of prior studies demonstrating the difference between IPF and UCTD (using broad criteria)-uip, 4,7,21 it is important to use strict diagnostic criteria for UCTD, both relevant symptoms and serology (higher titer). Because there was no significant difference in prognosis between UCTD-NSIP and idiopathic NSIP, we focused on comparing UCTD-UIP and IPF. Three of the patients developed definite CTD during follow-up as reported before showing the necessity of close monitoring regardless of the histologic pattern. There are several limitations to this study. First, our study was retrospective in design. However, the subjects were from prospectively recruited cohorts with little missing data. Another limitation was that the pathologic pattern of 30.5% of the patients with UCTD-IP was not confirmed. However, the only significant difference between UCTD-UIP and UCTD-NSIP was age ( Table 2 ) and the age of UCTD-undetermined pathology group was similar to the UCTD-UIP group so as to the survival curves. If all of these patients had had an NSIP pattern, the results would not have been different except that the prevalence of UCTD-NSIP would have increased to 49.5%. Because the old age (. 65 years, particularly among males) is strongly associated with UIP-pattern pathology in patient IIP but without the typical UIP pattern of HRCT scans (without honeycombing), 25 we reanalyzed the data as old age male patients with UCTD without biopsy (n 5 2) having UCTD-UIP. The results were the same (better survival of UCTD-UIP, hazard ratio, 0.529; 95% CI, ; P 5.026). Third, although we adopted narrow criteria of UCTD including a high titer of ANA (. 1:160), it may be somewhat liberal and ANA titer could change, if repeated. In the subjects, ANA titer was 1:1280 in six patients, 1:640 in nine, 1:320 in 23, and 1:160 in 16 patients. Repeated tests were performed in 28 patients; the results were journal.publications.chestnet.org 171

8 the same (. 1:160) in 25 patients. The titer was reduced in three patients, however, they were measured during the therapy with steroid and immunosuppressant. Furthermore, there is an innate difficulty in distinguishing arthralgias from osteoarthritis. However, most of the subjects, especially with UCTD were seen by a rheumatologist, and we think the possibility of misclassification was low. Finally, although this is the largest series of UCTD-IP cases reported to date, the numbers are still relatively small and from a single center. Thus, a multinational multiinstitutional prospective study with a larger number of patients is needed in the future. In conclusion, our current study findings confirm that a UIP pattern is frequently seen in patients with UCTD. Moreover, a significant proportion of patients with IPF have underlying UCTD and it is important to differentiate these patients from those with IPF because of the better prognosis of patients with UCTD-UIP. Acknowledgments Author contributions: D. S. K. has been identified as the guarantor of the paper, taking responsibility for the integrity of the work as a whole, from inception to published article. D. S. K. contributed to study conception and design; H.-C. K., W. J., M. Y. K., C.-K. L., and T. V. C. contributed to data acquisition; H.-C. K., W. J., and D. S. K. contributed to data analysis and interpretation; S. B. H. performed statistical analysis; H.-C. K., W. J., and D. S. K. contributed to the writing of the manuscript; and D. S. K, H.-C. K., W. J., M. Y. K., C.-K. L., T. V. C., and S. B. H. contributed to manuscript review and acceptance. Financial/nonfinancial disclosures: The authors have reported to CHEST that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Additional information: The e-figures and e-tables can be found in the Supplemental Materials section of the online article. References 1. Mosca M, Tani C, Neri C, Baldini C, Bombardieri S. Undifferentiated connective tissue diseases (UCTD). Autoimmun Rev ;6(1): Mosca M, Tani C, Carli L, Bombardieri S. Undifferentiated CTD: a wide spectrum of autoimmune diseases. Best Pract Res Clin Rheumatol ;26(1): Kinder BW, Collard HR, Koth L, et al. Idiopathic nonspecific interstitial pneumonia: lung manifestation of undifferentiated connective tissue disease? Am J Respir Crit Care Med ;176(7): Corte TJ, Copley SJ, Desai SR, et al. Significance of connective tissue disease features in idiopathic interstitial pneumonia. Eur Respir J ; 39 ( 3 ): Suda T, Kono M, Nakamura Y, et al. Distinct prognosis of idiopathic nonspecific interstitial pneumonia (NSIP) fulfilling criteria for undifferentiated connective tissue disease (UCTD). Respir Med ;104(10): Lunardi F, Balestro E, Nordio B, et al. Undifferentiated connective tissue disease presenting with prevalent interstitial lung disease: case report and review of literature. Diagn Pathol ;6: Vij R, Noth I, Strek ME. Autoimmunefeatured interstitial lung disease: a distinct entity. Chest ; 140 ( 5 ): Park JH, Kim DS, Park IN, et al. Prognosis of fibrotic interstitial pneumonia: idiopathic versus collagen vascular diseaserelated subtypes. Am J Respir Crit Care Med ;175(7): Kim EJ, Collard HR, King TE Jr. Rheumatoid arthritis-associated interstitial lung disease: the relevance of histopathologic and radiographic pattern. Chest ;136(5): Kim EJ, Elicker BM, Maldonado F, et al. Usual interstitial pneumonia in rheumatoid arthritis-associated interstitial lung disease. Eur Respir J ;35(6): Lee JS, Kim EJ, Lynch KL, et al. Prevalence and clinical significance of circulating autoantibodies in idiopathic pulmonary fibrosis. Respir Med ;107(2): Kang BH, Park JK, Roh JH, et al. Clinical significance of serum autoantibodies in idiopathic interstitial pneumonia. J Korean Med Sci ;28(5): Song JW, Do KH, Jang SJ, Colby TV, Han S, Kim DS. Blood biomarkers MMP-7 and SP-A: predictors of outcome in idiopathic pulmonary fibrosis. Chest ;143(5): Demedts M, Costabel U. ATS/ERS international multidisciplinary consensus classification of the idiopathic interstitial pneumonias. Eur Respir J ;19(5): Assayag D, Elicker BM, Urbania TH, et al. Rheumatoid arthritis-associated interstitial lung disease: radiologic identification of usual interstitial pneumonia pattern. Radiology ;270(2): Mosca M, Neri R, Bombardieri S. Undifferentiated connective tissue diseases (UCTD): a review of the literature and a proposal for preliminary classification criteria. Clin Exp Rheumatol ;17(5): Raghu G, Collard HR, Egan JJ, et al ; ATS/ ERS/JRS/ALAT Committee on Idiopathic Pulmonary Fibrosis. An official ATS/ ERS/JRS/ALAT statement: idiopathic pulmonary fibrosis: evidence-based guidelines for diagnosis and management. Am J Respir Crit Care Med ;183(6): Marie I, Josse S, Hatron PY, et al. Interstitial lung disease in anti-jo-1 patients with antisynthetase syndrome. Arthritis Care Res (Hoboken) ;65(5): Fischer A, West SG, Swigris JJ, Brown KK, du Bois RM. Connective tissue diseaseassociated interstitial lung disease: a call for clarification. Chest ; 138 ( 2 ): Vij R, Strek ME. Diagnosis and treatment of connective tissue disease-associated interstitial lung disease. Chest ; 143 (3): Strand MJ, Sprunger D, Cosgrove GP, et al. Pulmonary function and survival in idiopathic vs secondary usual interstitial pneumonia. Chest ; 146 ( 3 ): Bodolay E, Csiki Z, Szekanecz Z, et al. Five-year follow-up of 665 Hungarian patients with undifferentiated connective tissue disease (UCTD). Clin Exp Rheumatol ;21(3): Danieli MG, Fraticelli P, Salvi A, Gabrielli A, Danieli G. Undifferentiated connective tissue disease: natural history and evolution into definite CTD assessed in 84 patients initially diagnosed as early UCTD. Clin Rheumatol ;17(3): Guerrero LF, Rueda JC, Arciniegas R, Rueda JM. Undifferentiated connective tissue disease in a rheumatology center in Cali, Colombia: clinical features of 94 patients followed for a year. Rheumatol Int ;33(4): Fell CD, Martinez FJ, Liu LX, et al. Clinical predictors of a diagnosis of idiopathic pulmonary fibrosis. Am J Respir Crit Care Med ; 181 ( 8 ): Original Research [ 147 # 1 CHEST JANUARY 2015 ]