Cryptogenic Organizing Pneumonia: Serial High-Resolution CT Findings in 22 Patients

Cardiopulmonary Imaging Original Research Lee et al. High-Resolution CT of Cryptogenic Organizing Pneumonia Cardiopulmonary Imaging Original Research Ju Won Lee 1 Kyung Soo Lee 1 Ho Yun Lee 1 Man Pyo Chung 2 Chin A Yi 1 Tae Sung Kim 1 Myung Jin Chung 1 Lee JW, Lee KS, Lee HY, et al. Keywords: cryptogenic organizing pneumonia, lung CT, lung interstitial disease DOI:10.2214/AJR.09.3940 Received November 9, 2009; accepted after revision February 28, 2010. 1 Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-Dong, Kangnam-Ku, Seoul 135-710, Republic of Korea. Address correspondence to K. S. Lee (kyungs.lee@samsung.com). 2 Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea. AJR 2010; 195:916 922 0361 803X/10/1954 916 American Roentgen Ray Society Cryptogenic Organizing Pneumonia: Serial High-Resolution CT Findings in 22 Patients OBJECTIVE. We conducted a review of serial high-resolution CT (HRCT) findings of cryptogenic organizing pneumonia (COP). MATERIALS AND METHODS. Over the course of 14 years, we saw 32 patients with biopsy-confirmed COP. Serial HRCT scans were available for only 22 patients (seven men and 15 women; mean age, 52 years; median follow-up period, 8 months; range, 5 135 months). Serial CT scans were evaluated by two chest radiologists who reached a conclusion by consensus. Overall changes in disease extent were classified as cured, improved (i.e., 10% decrease in extent), not changed, or progressed (i.e., 10% increase in extent). When there were remaining abnormalities, the final follow-up CT images were analyzed to express observers ideas regarding what type of interstitial lung disease the images most likely suggested. RESULTS. The two most common patterns of lung abnormality on initial scans were ground-glass opacification (86% of patients [19/22]) and consolidation (77% of patients [17/22]), distributed along the bronchovascular bundles or subpleural lungs in 13 patients (59%). In six patients (27%), the disease disappeared completely; in 15 patients (68%), the disease was decreased in extent; and in one patient (5%), no change in extent was detected on follow-up CT. When lesions remained, the final follow-up CT findings were reminiscent of fibrotic nonspecific interstitial pneumonia in 10 of 16 patients (63%). CONCLUSION. Although COP is a disease with a generally good prognosis, most patients (73%) with COP have some remaining disease seen on follow-up CT scans, and, in such cases, the lesions generally resemble a fibrotic nonspecific interstitial pneumonia pattern. C ryptogenic organizing pneumonia (COP) is characterized histopathologically by plugs of granulation tissue lying within small airways, alveolar ducts, and alveoli and by chronic inflammatory cell infiltration in alveolar walls [1]. Patients with COP generally present with subacute illness, including shortness of breath, fever, malaise, and weight loss [1, 2]. Imaging findings of COP at presentation have been well characterized. Common chest radiographic findings include bilateral patchy areas of consolidation showing subpleural and lower lung zone predominance. High-resolution CT (HRCT) findings consist of consolidative areas or nodules distributed along the bronchovascular bundles or along the subpleural lungs [3 6]. Patients with COP manifest good prognoses with corticosteroid therapy [2]. However, despite this therapy, cases of relapse or progression have been reported [7, 8]. In addi- tion, in rare cases, a disease initially identified as COP progresses to respiratory failure and death [9]. Thus, more studies as to the exact clinical outcome of COP, particularly dealing with large number of patients, are needed. Moreover, to the best of our knowledge, the follow-up HRCT findings of the disease in a large patient cohort have yet to be reported. Thus, the principal objective of our study was to report serial HRCT findings of COP in a relatively large patient population and to look for clinical, pulmonary function test (PFT), or HRCT findings that might facilitate the prediction of patient prognoses. Materials and Methods Patient Enrollment and Demographics Our institutional review board approved this retrospective study with a waiver of patient informed consent. We reviewed all surgical biopsy files recorded from January 1995 to May 2008 in a single tertiary hospital and identified 72 patients who had a histopathologic diagnosis of organizing 916 AJR:195, October 2010

High-Resolution CT of Cryptogenic Organizing Pneumonia pneumonia. Among the 72 patients, 16 patients had underlying collagen vascular disease, 14 patients had concurrent pulmonary infection, and 10 had drug-associated lung diseases. We excluded these 40 patients from this study. Therefore, the remaining 32 patients had a histopathologic diagnosis of organizing pneumonia of unknown cause (i.e., COP). Among these 32 patients, 10 had received no follow-up CT or follow-up examination only within a 1-month interval. None of these 10 patients died of COP. For the remaining 22 patients, follow-up CT studies were obtained at least once, and the shortest follow-up period was 2 months. Thus, a total of 22 patients with COP (seven men and 15 women; mean [± SD] age, 52 ± 11.2 years; range, 28 70 years) were ultimately included in this study. The serial HRCT scans were obtained over a mean follow-up period of 18 ± 29.6 months (median, 7 months; range, 2 135 months). All patients underwent surgical lung biopsies (video-assisted thoracoscopic surgery biopsy, n = 21; open lung biopsy, n = 1). Surgical lung biopsy specimens were acquired from three lobes in two patients, from two lobes in 11 patients, and from one lobe in nine patients. PFT Initial PFT data were obtained for 19 patients. Forced spirometry and single-breath carbon monoxide diffusing capacity of the lung (DL CO ) were obtained with pulmonary function units (V max 22, SensorMedics Corporation). Forced vital capacity (FVC) and DL CO were expressed as a percentage of the predicted value based on height, age, sex, and ethnic origin. Additionally, arterial oxygen tension at room air (PaO 2 ) was measured at the time of biopsy or when PFT results were obtained. Serial trends in PFT results and PaO 2 values at follow-up studies were defined as improvement (i.e., an increase of > 15% in FVC or DL CO or an increase of > 15 mm Hg in PaO 2 ), decline (i.e., a decrease of > 15% in FVC or DL CO or a decrease of > 15 mm Hg in PaO 2 ), or stability (i.e., a change of < 15% in FVC or DL CO or a change of < 15 mm Hg in PaO 2 ). Image Acquisition Twenty-two patients underwent a total of 80 HRCT studies (mean, 3.6 ± 1.8 CT examinations per patient; range, 2 9 CT examinations). For the total of 44 CT studies performed for 22 patients as the initial and final follow-up scans, various helical CT scanners from different vendor companies with various numbers of detectors were used for CT image acquisition. A single-detector scanner was used for 17 studies, a 4-MDCT scanner was used for six studies, an 8-MDCT scanner was used for six studies, a 16-MDCT scanner was used for nine studies, a 40-MDCT scanner was used for two studies, and a 64-MDCT scanner was used for four studies. None of the patients received IV injections of contrast medium for the CT study. The scanning parameters were 120 kvp and 90 170 ma. With MDCT scanners, helical CT scans (beam width of 10 20 mm and beam pitch of 1.375 1.5) were obtained throughout the thorax, and the scan data were reconstructed with 1.0 1.25-mm section thickness and at 10-mm intervals, covering from the lung apices to the bottom of the lungs. With a single-detector scanner, 1.0-mm high-resolution lung-window CT images were obtained throughout the thorax at 10-mm TABLE 1: Clinical and Physiologic Analysis Results in 22 Patients With Biopsy-Proved Cryptogenic Organizing Pneumonia Clinical Result Initial Follow-Up Men, no. (%) of patients 7 (32) NA Smokers, no. (%) of patients 5 (23) NA Age at diagnosis (y) 52 ± 11.2 NA Duration from onset of symptom to treatment (mo) 4 NA Physiologic Forced vital capacity (% of predicted value) a 70 ± 21.6 a 85 ± 22.4 a Single-breath carbon monoxide diffusing capacity of the lung 75 ± 18.2 b 86.3 ± 3.9 c (% of predicted value) b Arterial oxygen tension at room air (mm Hg) d 80 ± 16.5 d 93 ± 9.4 e Note Except where noted, data are mean ± SD. NA = not applicable. a Data were available for 19 patients. b Data were available for 13 patients. c Data were available for 11 patients. d Data were available for 15 patients. e Data were available for 9 patients. intervals. In both single-detector CT and MDCT, data were reconstructed by using a bone algorithm. Image data were displayed directly on the monitors (four monitors with 1,536 2,048 image matrices, 8-bit viewable gray-scale, and 60-foot-lambert luminescence) of a PACS (PathSpeed or Centricity 2.0; GE Healthcare Integrated Imaging Solutions). Both mediastinal (window width, 400 HU; window level, 20 HU) and lung (window width, 1,500 HU; window level, 700 HU) window images were available for analysis on the monitors. Analyses of Thin-Section CT Findings Patterns (consolidation, ground-glass opacification [GGO], nodule, reticulation, and honeycombing), extent (to the nearest 5%), and distribution of lung abnormalities at the initial and final followup CT scan were evaluated by two chest radiologists with 5 and 20 years of experience in thoracic CT interpretation, respectively, who reached their conclusions by consensus. Overall disease extent changes were evaluated by comparing the initial and the final follow-up CT scans [10], and then the changes were divided into the following categories: completely resolved, improved (i.e., 10% decrease in total extent), not changed, and progressed (i.e., 10% increase in extent). In cases of complete disappearance, the follow-up interval was defined as period between the initial CT and the first follow-up CT that showed completely resolved parenchymal lesions. After the estimation of lung lesion extent, the final follow-up CT images were given again to the same two observers, who were asked to express their ideas independently as to what disease the given images most suggested (i.e., COP, nonspecific interstitial pneumonia [NSIP], or usual interstitial pneumonia [UIP]), according to previously published CT findings of idiopathic interstitial pneumonias [2]. When there was disagreement regarding which residual pattern the interstitial lung disease on follow-up CT scans suggested, the final decision was arrived at by consensus. The presence of pleural effusion, pleural thickening, pericardial effusion, and lymphadenopathy were also recorded. Clinical Outcome of Patients We reviewed the patients medical records. Relevant data included clinical symptoms at initial presentation, time interval between onset of symptom and treatment, treatment regimens, and relapse and survival of the patients. Relapses of COP were defined as the appearance of new characteristic opacities on chest imaging, with compatible clinical features, after some period of stable or improved state. Patients survival was identified from medical records or by contacting the patient s family when necessary. AJR:195, October 2010 917

Lee et al. Statistical Analysis Statistical analyses were conducted by using SPSS software (version 12.0, SPSS, Inc.). The relationship between the changes in serial CT findings (patients with completely disappeared disease and with remaining disease) and the initial PFT or CT finding results was correlated. Univariate and multivariate analyses were conducted to seek clinical information or initial CT findings that would help to predict remaining disease on follow-up CT. Continuous data were compared between the complete resolution group and the remaining disease group by using the Mann-Whitney U test, and categorical data were compared by using Fisher s exact test. In all statistical analyses, p values less than 0.05 were considered statistically significant. Results The demographic and clinical findings of 22 patients with biopsy-confirmed COP are summarized in Table 1. All 22 patients presented with dyspnea, cough, or fever. Seventeen patients were nonsmokers, three were ex-smokers (mean duration, 19 pack-years), and two were smokers (mean duration, 33 pack-years). PFT Results PFT results are also summarized in Table 1. Among 19 patients for whom FVC results (percentage of predicted value) was available, an interval increase was noted in 11 patients (58%), and stability was noted in eight patients (42%) on follow-up examinations. None of the patients exhibited an interval decrease in FVC. Eleven patients had both TABLE 2: Initial and Follow-Up CT Patterns, Distribution, and Extent of Lung Involvement in 22 Patients With Cryptogenic Organizing Pneumonia Pattern Parameter No. (%) of Patients initial and follow-up DL CO examinations. Six patients (55%) were deemed stable, five patients (45%) had an interval increase, and none exhibited an interval decrease on follow-up DL CO examinations. Initial PaO 2 results were available for 15 patients. The mean PaO 2 was 80 ± 16.5 mm Hg. Nine patients had both initial and follow-up PaO 2. Three patients (33%) were stable, six patients (67%) showed an interval increase, and none exhibited an interval decrease on follow-up PaO 2 examinations. Serial Thin-Section CT Findings The initial and follow-up CT scan findings are summarized in Table 2. The two most Initial CT Average Extent (%) No. (%) of Patients Follow-Up CT Average Extent (%) Consolidation 17 (77) 27 1 (5) 20 Ground-glass opacification 19 (86) 33 16 (72) 21 Nodule 7 (32) 10 1 (5) 10 Reticulation 4 (18) 10 11 (50) 12 Honeycombing 0 0 2 (9) 8 Distribution Lower lung predominance 12 (55) NA 10 (45) NA Subpleural 9 (41) NA 8 (36) NA Peribronchovascular 5 (23) NA 4 (18) NA Note NA = not applicable. A common patterns of lung abnormality on the initial scans were GGO (observed in 19 patients [86%]) and consolidation (observed in 17 patients [77%]), and they were distributed along the bronchovascular bundles or subpleural lungs in 13 (59%) of 22 patients. All initial scans showed lung abnormality in the lower lobes, and lower lobe predominance was noted in 12 patients (55%). Pleural effusion and subcarinal lymphadenopathy were detected in one patient each. Pericardial effusion was not detected in any patient. On the follow-up CT scans, the disease disappeared completely in six of 22 patients (27%) (Fig. 1), 15 (68%) showed a decrease in disease severity (Figs. 2 and 3), one (5%) Fig. 1 43-year-old woman with cryptogenic organizing pneumonia showing complete resolution on follow-up CT scans. A, Transverse thin-section (1.0-mm section thickness) CT scan obtained at level of basal segmental bronchi shows bilateral patchy ground-glass opacification (GGO) lesions distributed along bronchovascular bundles in lower lung zones. Linear consolidation surrounds internal GGO, so-called reversed halo sign (arrows), in right lower lobe. There was no parenchymal opacity in upper and middle lung zones (not shown here). Total extent of parenchymal abnormalities on CT scans was 20% (consolidation, 10%; GGO, 10%). B, Nine-month follow-up CT scan obtained at similar level to panel A shows complete resolution of lung lesions without remaining disease. B 918 AJR:195, October 2010

High-Resolution CT of Cryptogenic Organizing Pneumonia showed no change in disease extent, and none of the patients exhibited interval progression. The mean follow-up period for the patients whose disease had completely disappeared was 6 ± 2.3 months (range, 2 9 months), and that for the patients with remaining disease was 23 ± 33.6 months (range, 5 135 months). The most common CT findings on follow-up CT scans were GGO (observed in 16 patients [73%]), followed by reticulation (observed in 11 patients [50%]). Among the 16 (73%) patients who had remaining disease on follow-up CT scans, 10 patients (63%) showed a pattern of lung abnormalities most similar to that of fibrotic NSIP (Fig. 2). The principal findings were GGO and reticulation involving predominantly subpleural and basal lungs. Traction bronchiectasis was noted in two of these 10 patients. The 10 patients exhibited neither consolidation nor honeycombing. In two patients (13%), CT findings composed by only GGO without any predominance in distribution, although nonspecific, reminded the observer of hypersensitivity pneumonitis. In another two patients (13%), honeycombing with lower lung zone predominance was noted; thus, the CT findings were most similar to the UIP pattern (Fig. 3). Traction bronchiectasis was also present in these two patients. One patient each (6%) showed a cellular NSIP pattern (subpleural and lower lung zone predominant GGO, with little reticulation and without traction bronchiectasis) and residual COP pattern on their followup CT scans, in the opinion of the reviewers. A Fig. 2 56-year-old woman with cryptogenic organizing pneumonia showing improvement but remaining disease. A, Transverse thin-section (1.0-mm section thickness) CT scan obtained at level of liver dome shows bilateral patchy areas of consolidation and nodules distributed along bronchovascular bundles in lower lung zones. Total extent of parenchymal abnormalities on CT scans was 50% (consolidation, 40%; nodule, 10%). B, Nine-month follow-up CT scan obtained at similar level to panel A shows decreased extent of parenchymal lesions. Total extent of remaining lung lesions was 30% (ground-glass opacification, 20%; reticulation, 10%). Note associated traction bronchiectasis (arrows) in both lower lobes. Remaining abnormalities are suggestive of fibrotic nonspecific interstitial pneumonia. In the last case, in which the residual disease was finally classified as cellular NSIP, the two observers had different ideas about the remaining lesion pattern one classified it as continuing COP, and the other classified it as cellular NSIP pattern. Clinical Outcomes of Patients All 22 patients had begun a course of corticosteroid treatment a median of 3 months (range, 1 24 months) from symptom onset. Two of 22 patients (9%) had clinically recurring COP. Both of the patients survived, and their final follow-up periods were 147 and 72 months, respectively. In the former patient, follow-up CT scans obtained 135 months after the initial scan revealed several new subpleural nodular lesions in both upper lobes, in addition to the usual interstitial pneumonialike fibrosis in the middle and lower lung zones. The latter patient was deemed cured on the final available follow-up CT scans. Afterward, the patient had several episodes of relapse and exhibited an interval increase in the extent of bilateral parenchymal opacities, with lower lung zone predominance on the final follow-up chest radiographs. Six patients were lost to clinical long-term follow-up, and the remaining 16 patients were still surviving. None of the patients died of COP, or for any other reason. Interrelationship Between Imaging and Laboratory Findings Neither clinical information nor the pattern of parenchymal abnormalities on the initial CT scans differed significantly between the two groups (patients with complete disappearance of lung lesions on follow-up CT [n = 6] and with remaining lung abnormalities [n = 16]; p > 0.05). None of the variables on the initial CT findings helped to predict the follow-up CT finding results (Table 3). The initial FVC (percentage of predicted value) and initial DL CO (percentage of predicted value) results for patients with completely resolved disease were significantly higher than those of patients with remaining disease (p < 0.05) (Table 3). Discussion It has been shown that most patients who respond to corticosteroids show complete clearing or are left with small residual opacities [2]. If reticular opacities are the initial radiographic findings in patients with COP, the patient is less likely to respond to corticosteroids and may progress to lung fibrosis [11]. In our study, four patients exhibited reticulation on their initial CT images. These four patients had remaining diseases on their followup CT scans (mean follow-up period, 15 months; range, 5 55 months), three patients exhibited decreased extent of abnormalities, and one patient manifested unchanged disease extent. However, unlike a previous study [11], in the present study, the pattern of parenchymal abnormalities on the initial CT scans clearly did not constitute a prognosis-determining factor on a univariate or multivariate analysis. We cannot explain the difference in results between the previous study and ours. B AJR:195, October 2010 919

Lee et al. In the previous study [11], the results were based on chest radiographic finding analyses, not on CT examination results, as in ours. In addition, in both studies, the number of patients included was small. In our study, six patients (27%) had entered complete remission on follow-up CT examination. To the best of our knowledge, there has been no study thus far regarding the follow-up CT findings in COP. In a previous study [12], 24 of 50 patients (48%) manifested completely normal follow-up chest radiographs. In another study dealing with 12 immunocompromised patients with organizing pneumonia [13], five patients (42%) A C E Fig. 3 35-year-old man with cryptogenic organizing pneumonia showing decrease in total extent of disease (improvement) but remaining disease. A C, Transverse thin-section (1.0-mm section thickness) CT scans obtained at levels of trachea (A), superior segmental bronchi of lower lobes (B), and liver dome (C), respectively, show bilateral consolidation, ground-glass opacification (GGO), and nodules of lower lung zone and subpleural distribution. Total extent of parenchymal abnormalities on CT scans were 70% (consolidation, 40%; GGO, 20%; nodules, 10%). D F, Follow-up CT scans (135 months) obtained at similar levels to panels A, B, and C, respectively, show decreased extent of parenchymal lesions. Total extent of remaining lesions was 40% (GGO, 20%; reticulation, 10%; honeycombing [arrows], 10%). Remaining abnormalities show subpleural and lower lung zone predominance and simulate usual interstitial pneumonia pattern. B D F 920 AJR:195, October 2010

High-Resolution CT of Cryptogenic Organizing Pneumonia TABLE 3: Comparison of Clinical, Imaging, and Pulmonary Function Test Findings Between Patients With Complete Disappearance and Remaining Disease Clinical information Parameter exhibited complete resolution in follow-up radiographs obtained at a mean of 15 weeks. Of the 12 patients, four had available followup CT scans, and all four patients had remaining disease as shown by such findings as focal reticulation, nodules, consolidation, or GGO. There were two patients (9%) in our study whose follow-up clinical status showed some progression of disease after the stable period, which is suggestive of relapse. One patient showed no change in the extent of parenchymal opacity between the initial and final follow-up radiographs. The other patient had a slightly increased extent of parenchymal opacity between the initial and final follow-up radiographs. The proportion (9%) of relapsed COP cases in the current study is comparable to that reported in some previous studies [9, 14]. However, in the other studies, more than half of the patients with COP had relapsed disease [7, 15]. Largerscale and prospective studies will be required to search for relapse rates and to detect any correlation between relapse and remaining disease on follow-up CT scans. In 10 patients (63% of 16 patients who had remaining disease), the findings of follow-up CT scans reminded the observers of fibrotic Patients With Complete Disappearance (n = 6) NSIP. The principal findings indicated that the remaining diseases were GGO and reticulation, predominantly involving the subpleural or basal lungs. The findings revealed neither consolidations nor honeycombing. It is interesting to note that the remaining lung findings of COP, even after corticosteroid treatment, mimic those of fibrotic NSIP. To the best of our knowledge, this is the first report to specifically address the follow-up CT findings. It is important, however, to note that the follow-up assessments were based only on the experienced observers best estimate of the CT pattern. Radiologic-pathologic correlation would be necessary to confirm these opinions regarding treated patients with COP. It should also be elaborated that the imaging findings of NSIP are diverse [16, 17]. The patterns of lung abnormalities in NSIP range from consolidation or GGO to reticulation. Therefore, the similarities between the CT findings of residual COP and NSIP may not be unexpected news. Moreover, there is potential difficulty in separating COP and NSIP histopathologically [18]. Because half of NSIP biopsy specimens contain areas of organizing pneumonia component at the time of diagnosis, it may be possible Patients With Remaining Disease (n = 16) Men, no. (%) of patients 1 (17) 6 (38) 0.616 Smokers, no. (%) of patients 1 (17) 4 (25) 1.000 Age (y) 50 ± 12.0 53 ± 11.2 0.800 Duration from onset of symptom to treatment (mos) 5 ± 3.9 4 ± 5.6 0.331 CT findings Total extent (%) 45 ± 26.6 58 ± 20.4 0.189 Consolidation (%) 13 ± 12.1 24 ± 15.9 0.157 Ground-glass opacification (%) 28 ± 31.9 29 ± 23.9 0.822 Nodule (%) 3 ± 5.2 3 ± 4.8 0.964 Reticulation (%) 0 3 ± 4.5 0.205 Honeycombing (%) 0 0 1.000 Pulmonary function test findings Forced vital capacity (% of predicted value) 93 ± 7.2 58 ± 16.2 a 0.001 b Single-breath carbon monoxide diffusing capacity of the lung (% of predicted value) 86 ± 15.8 65 ± 14.2 c 0.038 b Arterial oxygen tension at room air (mm Hg) 88 ± 12.6 d 76 ± 17.0 e 0.133 Note Except where noted, data are mean ± SD. a Data were available for 13 patients. b Statistically significant. c Data were available for seven patients. d Data were available for four patients. e Data were available for 11 patients. that some of the cases of COP in our series might have been misdiagnosed and may, in fact, have represented NSIP. To our knowledge, there have been no studies thus far conducted in which a correlation was drawn between the physiologic data and the follow-up CT results. In one study [7], no significant difference was detected between nonrelapsing and relapsing groups in terms of PFT results. In our study, the initial FVC and initial DL co values of patients with completely resolved disease were significantly higher than those for patients with remaining disease. Additionally, one important finding on PFT results is that, although several patients had fibrotic pattern of NSIP on follow-up HRCT scans, there was no evidence of any crucial decline in pulmonary function. This study has some limitations. First, the study design was retrospective. Second, the number of patients included in the study was rather small. Only 22 of the original 32 patients with biopsy-proven COP were included in the study. The remaining 10 patients did not have follow-up CT examinations. This may have introduced a major bias into the investigation, because those patients who did not undergo follow-up CT examinations p AJR:195, October 2010 921

Lee et al. may have done extremely well clinically and may not have had residual disease. Third, the follow-up intervals varied, with the shortest follow-up period being only 2 months, and this may have affected the relapse data [7]. In conclusion, although COP is a disease generally associated with good prognoses, the majority of patients (73%) with COP manifest signs of remaining disease of interstitial lung disease pattern on follow-up CT scans. When lesions remain, they largely mimic the fibrotic NSIP pattern. Higher initial FVC and DL CO levels in patients with COP may predict complete resolution of parenchymal opacity on follow-up CT scans. Recurrence after corticosteroid treatment occurs in approximately 10% of patients. References 1. Davison AG, Heard BE, McAllister WA, Turner- Warwick ME. Cryptogenic organizing pneumonitis. Q J Med 1983; 52:382 394 2. American Thoracic Society/European Respiratory Society. International multidisciplinary consensus classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med 2002; 165:277 304 3. Muller NL, Guerry-Force ML, Staples CA, et al. Differential diagnosis of bronchiolitis obliterans with organizing pneumonia and usual interstitial pneumonia: clinical, functional, and radiologic findings. Radiology 1987; 162:151 156 4. Lee KS, Kullnig P, Hartman TE, Muller NL. Cryptogenic organizing pneumonia: CT findings in 43 patients. AJR 1994; 162:543 546 5. Muller NL, Staples CA, Miller RR. Bronchiolitis obliterans organizing pneumonia: CT features in 14 patients. AJR 1990; 154:983 987 6. Akira M, Yamamoto S, Sakatani M. Bronchiolitis obliterans organizing pneumonia manifesting as multiple large nodules or masses. AJR 1998; 170: 291 295 7. Lazor R, Vandevenne A, Pelletier A, Leclerc P, Court-Fortune I, Cordier JF. The Groupe d Etudes et de Recherche sur les Maladles Orphelines Pulmonaires (GERM O P). Cryptogenic organizing pneumonia: characteristics of relapses in a series of 48 patients. Am J Respir Crit Care Med 2000; 162:571 577 8. Cohen AJ, King TE Jr, Downey GP. Rapidly progressive bronchiolitis obliterans with organizing pneumonia. Am J Respir Crit Care Med 1994; 149:1670 1675 9. Izumi T, Kitaichi M, Nishimura K, Nagai S. Bronchiolitis obliterans organizing pneumonia: clinical features and differential diagnosis. Chest 1992; 102:715 719 10. Shin KM, Lee KS, Chung MP, et al. Prognostic determinants among clinical, thin-section CT, and histopathologic findings for fibrotic idiopathic interstitial pneumonias: tertiary hospital study. Radiology 2008; 249:328 337 11. Cordier JF, Loire R, Brune J. Idiopathic bronchiolitis obliterans organizing pneumonia: definition of characteristic clinical profiles in a series of 16 patients. Chest 1989; 96:999 1004 12. Epler GR, Colby TV, McLoud TC, Carrington CB, Gaensler EA. Bronchiolitis obliterans organizing pneumonia. N Engl J Med 1985; 312:152 158 13. Logan PM, Miller RR, Muller NL. Cryptogenic organizing pneumonia in the immunocompromised patient: radiologic findings and follow-up in 12 patients. Can Assoc Radiol J 1995; 46:272 279 14. Yamamoto M, Ina Y, Kitaichi M, Harasawa M, Tamura M. Clinical features of BOOP in Japan. Chest 1992; 102:21S 25S 15. Barroso E, Hernandez L, Gil J, Garcia R, Aranda I, Romero S. Idiopathic organizing pneumonia: a relapsing disease: 19 years of experience in a hospital setting. Respiration 2007; 74:624 631 16. Park JS, Lee KS, Kim JS, et al. Nonspecific interstitial pneumonia with fibrosis: radiographic and CT findings in seven patients. Radiology 1995; 195:645 648 17. Johkoh T, Muller NL, Colby TV, et al. Nonspecific interstitial pneumonia: correlation between thinsection CT findings and pathologic subgroups in 55 patients. Radiology 2002; 225:199 204 18. Travis WD, Hunninghake G, King TE, et al. Idiopathic nonspecific interstitial pneumonia: report of an American Thoracic Society project. Am J Respir Crit Care Med 2008; 177:1338 1347 922 AJR:195, October 2010