[
Original Research Chest Infections
]
Pulmonary Changes of Pleural TB Up-to-Date CT Imaging Jeong Min Ko, MD; Hyun Jin Park, MD; and Chi Hong Kim, MD
The objective of this study was to evaluate pulmonary abnormalities of pleural TB by CT scanning and to determine CT scan findings for the development of the paradoxical response (PR).
BACKGROUND:
CT scans were performed for 349 patients with pleural TB (between 2008 and 2013). We excluded 34 patients with coexisting pulmonary disease (n 5 13) or a totally collapsed lung (n 5 21). We analyzed CT scans focusing on pulmonary abnormalities such as the presence of consolidation, cavitation, interlobular septal thickening, and micronodules and their distribution. In addition, we recorded the development of PR during follow-up and statistically analyzed differences in clinical and CT scan findings between patients with and without PR.
METHODS:
A total of 270 of 315 patients (86%) had pulmonary abnormalities. Common CT scan findings were micronodules (n 5 209 [77%]), interlobular septal thickening (n 5 202 [75%]), and consolidation (n 5 120 [44%]). Cavitation was seen in 49 patients (18%). Among 209 with micronodules, the nodules were in the subpleural region (n 5 146 [70%]), peribronchovascular interstitium (n 5 113 [54%]), and centrilobular region (n 5 64 [31%]). PR occurred in 81 patients (26%), and patients with PR tended to be young, male, and without underlying disease (P , .05 by t test, Pearson x2 test). Subpleural micronodules were more common in patients with PR than in those without PR (Pearson x2, P 5 .025).
RESULTS:
Pulmonary abnormalities are very common in pleural TB. The most common CT scan findings were micronodules in the subpleural and peribronchovascular interstitium and interlobular septal thickening, suggesting the lymphatic spread of TB. In addition, PR is not rare in patients with pleural TB, especially in young, previously healthy, male patients who show subpleural nodules on initial CT scans. CHEST 2014; 146(6):1604-1611
CONCLUSIONS:
Manuscript received January 23, 2014; revision accepted June 2, 2014; originally published Online First June 19, 2014. ABBREVIATIONS: ADA 5 adenosine deaminase; AFB 5 acid-fast bacilli; LAP 5 lymphadenopathy; MDCT 5 multidetector CT; PCR 5 polymerase chain reaction; PR 5 paradoxical response AFFILIATIONS: From the Department of Radiology (Drs Ko and Park), and the Department of Internal Medicine (Dr Kim), St. Vincent’s Hospital, The Catholic University of Korea, Suwon-si, South Korea. FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study.
1604 Original Research
CORRESPONDENCE TO:
Hyun Jin Park, MD, Department of Radiology, St. Vincent’s Hospital, The Catholic University of Korea, 93 Ji-dong, Paldal-gu, Suwon-si, Kyeonggi-do, 403-720, South Korea; e-mail: [email protected] © 2014 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: 10.1378/chest.14-0196
[
146#6 CHEST DECEMBER 2014
]
Pleural TB develops from pulmonary lesions that form a small subpleural TB focus. Breakdown of these foci release their contents into the pleural space, which is followed by acute inflammation and exudation caused by a delayed hypersensitivity reaction. Operative specimens reveal that subpleural TB lesions are observed frequently in patients with pleural TB.1-3 However, the subpleural TB focus is seldom visualized on imaging, even though it plays a major role in the development of pleural TB. Most previous studies on CT scan findings of pulmonary changes of pleural TB have focused on bronchocentric lesions such as centrilobular nodules, a tree-in-bud appearance, and consolidation, because these are the most characteristic findings of active pulmonary TB.4,5 However, there are no reports describing subpleural nodules on CT scans as active TB lesions in pleural TB. Moreover, the subpleural nodules have additional clinical significance because they are thought to be responsible for the development of paradoxical response (PR).
Pleural TB can manifest as a primary disease from recent infection or as a reactivated disease from remote infection. TB pleurisy due to reactivation is more likely to be associated with radiographic parenchymal infiltrations. However, it is difficult to determine with absolute certainty whether individual patients have TB caused by recent or remote infection, because definitive evidence of the exact time of TB infection is usually not available.5 In addition, classification of pleural TB into primary and reactivation forms has no clinical implication.6 Therefore, we analyzed all the parenchymal changes of pleural TB with recent CT scanners regardless of the form, after discarding preconceptions about CT scan findings of active pulmonary TB. In addition, combined pulmonary changes on CT scan can be helpful in the differentiation of pleural TB from other pleural disease. We also examined the prevalence of PR and compared the characteristic clinical and CT scan findings of patients who showed PR during follow-up with those of patients who did not develop PR.
Materials and Methods
CT Scanning
Patients
CT scans were obtained with two CT scanners, LightSpeed VCT (General Electric Medical Systems) and Somatom Definition Flash (Siemens Heathcare), with or without IV administration of contrast medium (100 mL at rate of 2.0-2.5 mL/s). Attending physicians requested a CT scan with contrast enhancement if chest radiographs led them to suspect a malignancy or TB lymphadenitis. Scanning parameters included a 130-mA tube current, 120-kV tube voltage, 128 3 0.6-mm collimation, and pitch of 1.2. All images were reconstructed into axial images with a 5-mm slice thickness at 5-mm intervals and coronal images with a 3-mm slice thickness.
We selected patients with pleural TB who underwent CT scans at the time of diagnosis between May 2008 and September 2013. We excluded patients with other coexisting pulmonary disease or a large amount of pleural effusion resulting in total collapse of the lung. We retrospectively reviewed clinical features and pleural effusion parameters, including differential cell count and levels of lactate dehydrogenase and adenosine deaminase (ADA). We analyzed the CT scan findings of all patients and checked for the development of PR during follow-up. Patients were divided into two groups according to the development of PR, and clinical and CT scan findings were compared between the PR and non-PR groups. Diagnosis of Pleural TB and PR The diagnosis of pleural TB was based on a lymphocyte-prominent (. 50%) effusion that met Light’s criteria for exudative effusions, with demonstration of an elevated ADA level of . 40 IU/L, positive results for acid-fast bacilli (AFB) staining, growth of Mycobacterium tuberculosis from pleural fluid or sputum, positive results for TB in a polymerase chain reaction (PCR) assay, the presence of classic caseating granulomatous inflammation in the pleural tissue, or a successful response to anti-TB therapy.5,7 We defined PR as a radiographic worsening of pleural TB in patients who initially improved with anti-TB treatment; in these patients, the onset of PR must have occurred after the initiation of treatment, without evidence of positive AFB results in a sputum study. We determined that radiographic findings were worsened if pleural effusions increased or new pulmonary lesions developed. The time to development of PR was defined as the interval between the initiation of anti-TB treatment and the onset of PR, as defined previously. Patients who developed PR within 14 days after initiation of anti-TB treatment were excluded from the PR group.7,8
journal.publications.chestnet.org
CT Scan Analysis All CT scans and medical records were reviewed retrospectively by two chest radiologists (J. M. K. and H. J. P.) with 4 and 11 years of experience, respectively. Interpretation was based on a consensus between the two radiologists. CT scans obtained to assess pleural effusion were analyzed with specific attention to pulmonary findings, such as micronodules, consolidation, and cavitation. A micronodule was defined as a small rounded opacity with a diameter , 7 mm and was classified by distribution as centrilobular, perilymphatic (subpleural or peribronchovascular), or random.9 The presence of interlobular septal thickening and clustered or coalescent micronodules showing a cluster or galaxy sign was included in our analysis.10 We checked for the presence of lymphadenopathy (LAP) and extrathoracic TB involvement. Findings of old scar lesions such as bronchiectasis, fibrotic linear opacities, and ciccatricial emphysema were excluded from our CT scan analysis. Statistical Analysis We compared clinical and CT scan findings between the PR group and the non-PR group. All tests of significance were two-sided. Univariate comparisons between the PR and the non-PR group were performed using the Pearson x2 test or Fisher exact test for categorical variables and the Student t test for continuous variables. A P value , .05 was considered statistically significant.
1605
Results Patients
This study was approved by the Institutional Review Board of St. Vincent’s Hospital, which waived informed consent (VC13RISI0218). A total of 349 patients who had received a diagnosis of pleural TB underwent CT scans at the time of diagnosis. Among them, 34 patients were excluded because of total lung collapse (n 5 21) or coexisting pulmonary diseases such as reexpansion edema (n 5 5), pneumonia (n 5 3), malignancy (n 5 2), pneumoconiosis (n 5 2), and pulmonary edema (n 5 1). The mean age of the remaining 315 patients was 47 years, and 201 patients were men. Fifty-seven patients had chronic illness, and common comorbidities were diabetes (n 5 27), chronic liver disease (n 5 12), collagen vascular disease (n 5 5), chronic renal failure (n 5 4), and chronic heart failure (n 5 4). None of the patients had HIV infection. Diagnosis
Diagnosis of pleural TB was confirmed by positive microbiology results (AFB, n 5 31; culture, n 5 59; TB in PCR, n 5 38), increased ADA level (mean, 84 IU/L) in 294 patients, lymphodominant exudates (mean percent of lymphocyte, 77%) in 289 patients, and histopathology in 27 patients. CT Scan Findings
Among 315 patients, 270 (86%) had pulmonary abnormalities on CT scans. Common CT scan findings were micronodules (n 5 209 [77%]) and interlobular septal thickening (n 5 202 [75%]). Among 209 patients with micronodules, the nodules were present along the subpleural region in 146 patients (70%) and the peribronchovascular bundles in 113 (54%). Micronodules were present in the subpleural and/or perilymphatic region in 184 patients (88%) (Figs 1, 2). Centrilobular distribution was present in 64 patients (31%), and miliary nodules with random distribution were observed in one patient. Forty-four patients (21%) had both centrilobular and perilymphatic nodules together. Determination of micronodule distribution was not possible in 12 patients. In eight of 12 patients with indeterminate nodules, perilymphatic micronodules coexisted in different areas. Micronodules were clustered in 26 patients (8%), showing a galaxy or cluster sign (Fig 3). Consolidation was seen in 120 patients (44%), and cavitation was associated with the consolidation in 49 patients (18%). Intrathoracic LAP 1606 Original Research
Figure 1 – A 43-y-old man with right pleural TB, diagnosed by increased adenosine deaminase level (100 IU/L) within lymphodominant (98%) pleural effusion. Microbiology study was negative. A and B, Axial CT scans with a lung window setting at the level of the manubrium show multiple welldefined micronodules in the subpleural region of the right upper lobe (arrows), accompanying smooth interlobular septal thickening (empty arrows).
was present in 46 patients (15%), and extrathoracic TB involvement was observed in 21 patients (7%) (Tables 1-3).
[
146#6 CHEST DECEMBER 2014
]
Comparison of Characteristics Between PR and Non-PR Groups
PR occurred in 81 patients (26%) between 16 and 233 days (mean, 76 days) after initiation of anti-TB treatment. Compared with the non-PR group, the PR group was younger (mean, 37 years; P , .000), had a lower prevalence of comorbidities (11%, P 5 .012), and had a male predominance (74%, P 5 .020) (Table 4). There was no difference in CT scan findings between the two groups, except for the presence of subpleural micronodules: The PR group had a higher prevalence of subpleural nodules than did the non-PR group (56%, P 5 .025) (Table 4) (Fig 4).
Discussion Pleural TB still remains the most common cause of pleural effusion in areas with a high prevalence of TB and has become more frequent after the emergence of HIV and the widespread use of immunosuppressive drugs.1,3-5 Patients invariably have a small subpleural nidus of TB that shows fibrous and granulomatous inflammation.11 However, coexisting pulmonary changes that are suggestive of active TB have a relatively low prevalence on imaging.5,12,13 A key point of the pathogenesis of TB is that TB progresses in a bronchocentric manner. Therefore, radiologic findings have been interpreted from the same point of view. The CT scan findings of active pulmonary TB include centrilobular nodules and linear branching opacity, consolidation, and cavitation. In addition to bronchogenic dissemination, LAP or miliary nodules can be seen in patients with lymphatic and hematogeneous spread.14,15
Figure 2 – A 50-y-old man with bilateral pleural TB. Thoracocentesis revealed lymphodominant exudate (100%) with an increased adenosine deaminase level (145 IU/L). TB in a polymerase chain reaction was positive in the pleural fluid. Sputum study was negative. A, Axial CT scan at the level of the sternal notch shows smooth interlobular septal thickening along the anterior aspect of both upper lobes (arrows). Multiple micronodules are seen in the left upper lobe, located in the subpleural region and thickened interlobular septa (empty arrow). B, Axial CT scan at the level of the aortic arch shows multiple micronodules in the subpleural region of the left upper lobe. Some discrete micronodules, , 3 mm in size, are also seen along the broncho-
journal.publications.chestnet.org
Most of the patients (86%) had pulmonary abnormalities other than TB sequelae on their CT scans. The most common findings were interlobular septal thickening and micronodules. Eighty-eight percent of the micronodules were distributed along subpleural regions and peribronchovascular bundles, almost twice the number of centrilobular nodules. Both interlobular septal thickening and the presence of subpleural and peribronchovascular micronodules strongly indicate that the lesions involved mainly lymphatics. However, to date, TB has not been considered a perilymphatic disease. In fact, such CT scan findings, typically, negate TB in differential diagnosis. vascular bundles in the left upper lobe (arrowheads). Nearby airways are patent, representing peribronchovascular lesions, not intrabronchiolar lesions.
1607
TABLE 1
] Pulmonary Abnormal CT Scan Findings in 270 Patients With Pleural TB
CT Scan Findings
No. Patients
%
Micronodules
209
77
Interlobular septal thickening
202
75
26
8
120
44
Cavitation
49
18
Intrathoracic LAP
46
15
Extrathoracic involvement
21
7
Galaxy/clustered sign Consolidation
LAP 5 lymphadenopathy.
Interlobular septal thickening is known to be relatively common in miliary TB.15,16 Im et al17 reported that interlobular septal thickening was transiently observed on the CT scans of 34% of patients with active pulmonary TB. They suggested that the septal thickening may have been caused by a local increase in lymphatic flow from the exudative lesions in a pattern of lobular consolidation and not by caseous material in the septa, because the findings disappeared completely without any residual changes after TB treatment.16-18 However, we disagree with their assessment for several reasons. First, interlobular septal thickening was very common in patients (77%), whereas the prevalence of consolidation was relatively low (44%). Secondary changes of nearby consolidation are not sufficient to explain the higher prevalence of associated interlobular septal thickening. Second, we do not agree with the notion that lesions that disappear without any residue after TB treatment would not be granulomatous lesions. Third, one TB study involving CT scan and pathologic correlation showed that granulomas with caseous necrosis were present in the interlobular septa in a pathologic specimen of a patient with miliary TB.15 Therefore, interlobular septal thickening in the TABLE 2
] Distribution of Micronodules in 209 Patients
Distribution
No. Patients
Figure 3 – A 26-y-old man with right pleural TB, diagnosed by increased adenosine deaminase level (69 IU/L) in lymphodominant exudate (67%). Sputum acid-fast bacilli and TB in a polymerase chain reaction were all negative. A, Axial CT scan shows a coalescence of numerous discrete micronodules in the peribronchovascular bundles and subpleural region, resulting in a sarcoid cluster sign, in the right upper lobe. The micronodules are different from typical centrilobular nodules, which are more evenly spaced, larger, and less defined than those of this patient. B, Another CT scan at the carina level shows nodular thickening of bronchovascular bundles in the anterior segment of the right upper lobe. There is no evidence of any filling lesion within bronchioles.
1608 Original Research
%
Perilymphatic distribution
184
88
Subpleural
146
70
Peribronchovascular
113
54
64
31
1
0
12
6
a
Centrilobular distribution Random distribution Indeterminate
aNodules were considered perilymphatic when they were in at least one of the subpleural region, peribronchovascular interstitium, or interlobular septa.
[
146#6 CHEST DECEMBER 2014
]
TABLE 3
] Comparison of Clinical Features Between PR and Non-PR PR (n 5 81 [26%])
Clinical Feature Age, mean, y
37
Male ratio
0.74
Non-PR (n 5 234 [74%]) 50 0.59
P Value .000a .012a
Presence of comorbidity, %
11
21
.020a
Lymphocyte, mean, %
79
77
.254
ADA level, mean, IU/L
92
85
.136
Positive AFB, %
12
16
.517
Positive PCR, %
13
12
.812
Positive culture, %
16
19
.594
ADA 5 adenosine deaminase; AFB 5 acid-fast bacilli; PCR 5 polymerase chain reaction; PR 5 paradoxical response. aStatistically significant (P , .05).
patients should be considered interstitial involvement of TB.
regions via the pulmonary lymphatics, TB may cause pleural TB.
In contrast to the interlobular septal thickening, micronodules in the subpleural region or peribronchovascular bundles associated with pleural or pulmonary TB are mentioned very rarely in previous reports. Traditionally, micronodules with such a distribution represent lymphatic spread of the disease. In the aforementioned study by Lee et al,15 there were some cases with necrotizing granulomas along the bronchovascular and subpleural interstitium in pathologic specimens. A high prevalence of subpleural micronodules together with interlobular septal thickening indicates that TB involves mainly the pulmonary lymphatics in many of the patients with pleural TB.9,19,20 After reaching the subpleural
Considering the high prevalence, we do not think that subpleural micronodules and interlobular septal thickening are new radiologic findings; instead, such lesions have previously been neglected or considered nonspecific. Moreover, nodules along the peribronchovascular interstitium could have been mistaken for centrilobular nodules and linear branching opacities. Actually, Poey et al21 reported that interstitial nodules and thickening were more common than centrilobular nodules in their TB study using highresolution CT scans. However, they considered interstitial lesions to be nonspecific or negligible findings in cases of active TB.
TABLE 4
] Comparison of CT Scan Findings Between PR and Non-PR
CT Scan Finding
PR (%)
Non-PR (%)
P Value
Pleural thickening
96
93
.183
Micronodules
74
64
.077
Subpleural nodules
56
42
.025a
Centrilobular nodules
22
19
.503
Peribronchovascular nodularity
38
35
.642
Septal thickening
65
64
.892
Galaxy
8
8
.970
Consolidation
36
38
.892
Cavitation
20
14
.234
Intrathoracic LAP
15
14
.734
5
7
.533
Extrathoracic TB
See Table 1 and 3 legends for expansion of abbreviations. Statistically significant (P , .05).
a
journal.publications.chestnet.org
Some studies have reported some cases in which active pulmonary TB shares certain CT scan signs with sarcoidosis. These are a sarcoid galaxy sign and a sarcoid cluster sign, which correspond to clusters of multiple micronodules with or without formation of a coalescent mass.6,10,22 Eight percent of patients had clustered nodules showing a galaxy or cluster sign. Heo et al22 mentioned that the nodules of the galaxy sign did not appear to coincide with any reported pattern of pulmonary TB such as bronchogenically spread nodules, but instead these may have a gross pathology similar to that of nodules of sarcoidosis. We also think that micronodules along the subpleural and peribronchovascular interstitium are probably caused by granulomas in the interstitium, as in sarcoidosis. In addition, considering that centrilobular nodules range from a few millimeters to approximately 1 cm in size, are often ill defined, and appear to be evenly spaced, the micronodules of these signs are most likely not centrilobular lesions, but instead are perilymphatic lesions.23 1609
PR was not a rare condition in our study, occurring in 26% of patients with pleural TB. This result is similar to that of Cheng et al.8 In their study, the risk factors for PR were younger age, a high serum albumin level, and a low proportion of lymphocyte and a high proportion of neutrophils in the pleural fluid. They did not include CT scan findings in their analysis. In these patients, there was no difference in pleural fluid parameters. However, PR was more common in younger, male, and previously healthy individuals without any comorbidities. The presence of subpleural nodules was the only CT scan finding that differed between the PR and nonPR groups. It is surprisingly consistent with the speculation by Choi et al.26 Considering the combined risk factors, PR seems to be more common in TB that involves subpleural lymphatics and is in a primary form.
Figure 4 – A 27-y-old man with pleural TB and paradoxical response. Thoracocentesis revealed lymphodominant exudate (90%) with increased adenosine deaminase level (78 IU/L). A, Axial CT scan of inferior pulmonary vein level shows fluid collection in left pleural space. Smooth interlobular septal thickening is seen in the lingular division of the left upper lobe (arrowheads). Also seen are several small nodules along the right interlobar fissures and the subpleural region of the right lower lobe (arrows). B, One-month follow-up axial CT scan after initiation of anti-TB chemotherapy shows complete regression of the left pleural effusion and the interlobular septal thickening of the left lingula. New subpleural consolidation is seen along the right minor fissure (empty arrow), despite the fact that the TB mainly involved the left pleura initially. Video-assisted thoracoscopic biopsy proved TB pathologically. However, microbiology results were all negative. The subpleural lesion regressed without any change of initial TB regimen later.
PR is the unusual expansion or new formation of a TB lesion during TB treatment, which is one phenomenon of the immune recontstitution inflammatory syndrome.8 PR has been reported extensively in patients who are HIV-positive who receive highly active antiretroviral therapy.24,25 However, it can also occur in patients who are HIV-negative, especially in extrapulmonary TB.7,26,27 Choi et al26 reported that the most common CT scan findings of PR in patients with pleural TB were peripheral pulmonary nodules or masses that mostly abut the normal or thickened pleura. Given that most patients with pleural TB had concomitant subpleural diseases and showed the characteristic subpleural location of PR, they speculated that subpleural TB lesions undergo transient worsening and become radiographically evident during treatment of TB pleurisy. 1610 Original Research
Pulmonary TB can be a lymphatic disease.28 An animal study revealed that Mycobacterium tuberculosis causes lymphatic vasculitis in the lungs, thus, providing a direct conduit from the lungs to the regional nodes. Moreover, the pulmonary lymphatics are themselves sites of infection and could be a site of latent infection.29 This lymphatic model can explain how TB reaches the subpleural lymphatics, which will penetrate the visceral pleura and cause pleural TB. Given that lymphatic spread is a major route of primary TB, it is easy to understand the high prevalence of pleural TB in a primary form. This could also explain the atypical manifestation of TB resembling sarcoidosis and the negative sputum results despite extensive pulmonary lesions with micronodules on CT scans. The major weak point of this study is that an open lung biopsy specimen was not obtained in any of these patients. Therefore, the CT scan findings cannot be correlated with the pathologic findings and our results are assumed. However, those findings have been recognized as representing perilymphatic disease for a long time and in many studies.19,20,30 Another weak point is that CT images of 3- to 5-mm thickness were used for the evaluation of micronodules and interlobular septal thickening. In our institute, contrast-enhanced multidetector CT (MDCT) scanning rather than high-resolution CT scanning is performed routinely when evaluating TB, because the main reason for the scan is to examine the presence of lymphadenitis or other associated findings or to rule out diseases other than pulmonary TB, such as lung cancer. Nonetheless, the resolution of MDCT scans has improved, which facilitates the evaluation of subtle abnormalities such as micronodules and septal lines, and, if needed, multiplanar reconstruction with thin-slice thickness is feasible by applying the
[
146#6 CHEST DECEMBER 2014
]
appropriate software. The fact that the results of our study can be applied to routine MDCT scan protocols may actually be advantageous.31 In addition, we retrospectively analyzed the patients who had been given a diagnosis of pleural TB, and, thus, the diagnostic performance or usefulness of CT scanning could not be evaluated. Therefore, further study is necessary to determine whether CT scan findings can differentiate pleural TB from other inflammatory and infectious pleural diseases such as empyema.
Acknowledgments Author contributions: H. J. P. is the guarantor of the content of this manuscript, including the data and analysis. J. M. K., H. J. P., and C. H. K. contributed to the conception and design, acquisition of data, analysis, and interpretation of data; drafting of the submitted article or critical revision of it for important intellectual content; and final approval of the version to be published. 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.
9.
10.
11.
12.
References 1. Berger HW, Mejia E. Tuberculous pleurisy. Chest. 1973;63(1):88-92. 2. Stead WW, Eichenholz A, Stauss HK. Operative and pathologic findings in twenty-four patients with syndrome of idiopathic pleurisy with effusion, presumably tuberculous. Am Rev Tuberc. 1955;71(4):473-502. 3. Seibert AF, Haynes J Jr, Middleton R, Bass JB Jr. Tuberculous pleural effusion. Twenty-year experience. Chest. 1991;99(4):883-886. 4. Seiscento M, Vargas FS, Bombarda S, et al. Pulmonary involvement in pleural tuberculosis: how often does it mean disease activity? Respir Med. 2011;105(7): 1079-1083. 5. Kim HJ, Lee HJ, Kwon S-Y, et al. The prevalence of pulmonary parenchymal tuberculosis in patients with tuberculous pleuritis. Chest. 2006;129(5):1253-1258. 6. Liam C-K, Lim KH, Wong C-M. Tuberculous pleurisy as a manifestation of primary and reactivation disease in a region with a high prevalence of tuberculosis. Int J Tuberc Lung Dis. 1999;3(9):816-822. 7. Jeon K, Choi WI, An JS, et al. Paradoxical response in HIV-negative patients with pleural tuberculosis: a retrospective multicentre study. Int J Tuberc Lung Dis. 2012;16(6):846-851. 8. Cheng VC, Yam WC, Woo PC, et al. Risk factors for development of paradoxical response during antituberculosis therapy
journal.publications.chestnet.org
13.
14.
15.
16.
17.
18.
19.
20.
Conclusions In conclusion, interlobular septal thickening and micronodules in the subpleural and peribronchovascular regions are the most common CT scan findings of pleural TB. Radiologically, lymphatic spread of TB is more common than bronchogenic spread in the lungs of patients with pleural TB. In addition, the presence of subpleural nodules is a risk factor for later development of PR, especially in young male patients without underlying disease.
in HIV-negative patients. Eur J Clin Microbiol Infect Dis. 2003;22(10):597-602. Austin JH, Müller NL, Friedman PJ, et al. Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society. Radiology. 1996;200(2):327-331. Marchiori E, Zanetti G, Barreto MM, de Andrade FT, Rodrigues RS. Atypical distribution of small nodules on high resolution CT studies: patterns and differentials. Respir Med. 2011;105(9):1263-1267. Valdés L, Alvarez D, San José E, et al. Tuberculous pleurisy: a study of 254 patients. Arch Intern Med. 1998; 158(18):2017-2021. Yilmaz MU, Kumcuoglu Z, Utkaner G, Yalniz O, Erkmen G. Computed tomography findings of tuberculous pleurisy. Int J Tuberc Lung Dis. 1998;2(2):164-167. Hulnick DH, Naidich DP, McCauley DI. Pleural tuberculosis evaluated by computed tomography. Radiology. 1983;149(3):759-765. Im J-G, Itoh H, Lee KS, Han MC. CT-pathology correlation of pulmonary tuberculosis. Crit Rev Diagn Imaging. 1995;36(3):227-285. Lee JY, Lee KS, Jung K-J, et al. Pulmonary tuberculosis: CT and pathologic correlation. J Comput Assist Tomogr. 2000;24(5): 691-698. McGuinness G, Naidich DP, Jagirdar J, Leitman B, McCauley DI. High resolution CT findings in miliary lung disease. J Comput Assist Tomogr. 1992;16(3): 384-390. Im JG, Itoh H, Shim Y-S, et al. Pulmonary tuberculosis: CT findings—early active disease and sequential change with antituberculous therapy. Radiology. 1993;186(3):653-660. Voloudaki AE, Tritou IN, Magkanas EG, Chalkiadakis GE, Siafakas NM, Gourtsoyiannis NC. HRCT in miliary lung disease. Acta Radiol. 1999;40(4): 451-456. Raman SP, Pipavath SN, Raghu G, Schmidt RA, Godwin JD. Imaging of thoracic lymphatic diseases. AJR Am J Roentgenol. 2009;193(6):1504-1513. Remy-Jardin M, Beuscart R, Sault MC, Marquette CH, Remy J. Subpleural micronodules in diffuse infiltrative lung
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
diseases: evaluation with thin-section CT scans. Radiology. 1990;177(1): 133-139. Poey C, Verhaegen F, Giron J, Lavayssiere J, Fajadet P, Duparc B. High resolution chest CT in tuberculosis: evolutive patterns and signs of activity. J Comput Assist Tomogr. 1997;21(4):601-607. Heo J-N, Choi YW, Jeon SC, Park CK. Pulmonary tuberculosis: another disease showing clusters of small nodules. AJR Am J Roentgenol. 2005;184(2):639-642. Murata K, Itoh H, Todo G, et al. Centrilobular lesions of the lung: demonstration by high-resolution CT and pathologic correlation. Radiology. 1986;161(3):641-645. Jung JW, Shin JW, Kim JY, et al. Risk factors for development of paradoxical response during anti-tuberculosis treatment in HIV-negative patients with pleural tuberculosis. Tohoku J Exp Med. 2011;223(3):199-204. Cheng VC, Ho PL, Lee RA, et al. Clinical spectrum of paradoxical deterioration during antituberculosis therapy in non-HIV-infected patients. Eur J Clin Microbiol Infect Dis. 2002;21(11):803-809. Choi YW, Jeon SC, Seo HS, et al. Tuberculous pleural effusion: new pulmonary lesions during treatment. Radiology. 2002;224(2):493-502. Cheng SL, Wang HC, Yang PC. Paradoxical response during antituberculosis treatment in HIV-negative patients with pulmonary tuberculosis. Int J Tuberc Lung Dis. 2007;11(12): 1290-1295. Behr MA, Waters WR. Is tuberculosis a lymphatic disease with a pulmonary portal? Lancet Infect Dis. 2014;14(3):250-255. Basaraba RJ, Smith EE, Shanley CA, Orme IM. Pulmonary lymphatics are primary sites of Mycobacterium tuberculosis infection in guinea pigs infected by aerosol. Infect Immun. 2006;74(9):5397-5401. Raoof S, Amchentsev A, Vlahos I, Goud A, Naidich DP. Pictorial essay: multinodular disease: a high-resolution CT scan diagnostic algorithm. Chest. 2006;129(3):805-815. Ko JM, Kim KJ, Park SH, Park HJ. Bronchiectasis in active tuberculosis. Acta Radiol. 2013;54(4):412-417.
1611
Original Research Chest Infections
]
Pulmonary Changes of Pleural TB Up-to-Date CT Imaging Jeong Min Ko, MD; Hyun Jin Park, MD; and Chi Hong Kim, MD
The objective of this study was to evaluate pulmonary abnormalities of pleural TB by CT scanning and to determine CT scan findings for the development of the paradoxical response (PR).
BACKGROUND:
CT scans were performed for 349 patients with pleural TB (between 2008 and 2013). We excluded 34 patients with coexisting pulmonary disease (n 5 13) or a totally collapsed lung (n 5 21). We analyzed CT scans focusing on pulmonary abnormalities such as the presence of consolidation, cavitation, interlobular septal thickening, and micronodules and their distribution. In addition, we recorded the development of PR during follow-up and statistically analyzed differences in clinical and CT scan findings between patients with and without PR.
METHODS:
A total of 270 of 315 patients (86%) had pulmonary abnormalities. Common CT scan findings were micronodules (n 5 209 [77%]), interlobular septal thickening (n 5 202 [75%]), and consolidation (n 5 120 [44%]). Cavitation was seen in 49 patients (18%). Among 209 with micronodules, the nodules were in the subpleural region (n 5 146 [70%]), peribronchovascular interstitium (n 5 113 [54%]), and centrilobular region (n 5 64 [31%]). PR occurred in 81 patients (26%), and patients with PR tended to be young, male, and without underlying disease (P , .05 by t test, Pearson x2 test). Subpleural micronodules were more common in patients with PR than in those without PR (Pearson x2, P 5 .025).
RESULTS:
Pulmonary abnormalities are very common in pleural TB. The most common CT scan findings were micronodules in the subpleural and peribronchovascular interstitium and interlobular septal thickening, suggesting the lymphatic spread of TB. In addition, PR is not rare in patients with pleural TB, especially in young, previously healthy, male patients who show subpleural nodules on initial CT scans. CHEST 2014; 146(6):1604-1611
CONCLUSIONS:
Manuscript received January 23, 2014; revision accepted June 2, 2014; originally published Online First June 19, 2014. ABBREVIATIONS: ADA 5 adenosine deaminase; AFB 5 acid-fast bacilli; LAP 5 lymphadenopathy; MDCT 5 multidetector CT; PCR 5 polymerase chain reaction; PR 5 paradoxical response AFFILIATIONS: From the Department of Radiology (Drs Ko and Park), and the Department of Internal Medicine (Dr Kim), St. Vincent’s Hospital, The Catholic University of Korea, Suwon-si, South Korea. FUNDING/SUPPORT: The authors have reported to CHEST that no funding was received for this study.
1604 Original Research
CORRESPONDENCE TO:
Hyun Jin Park, MD, Department of Radiology, St. Vincent’s Hospital, The Catholic University of Korea, 93 Ji-dong, Paldal-gu, Suwon-si, Kyeonggi-do, 403-720, South Korea; e-mail: [email protected] © 2014 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: 10.1378/chest.14-0196
[
146#6 CHEST DECEMBER 2014
]
Pleural TB develops from pulmonary lesions that form a small subpleural TB focus. Breakdown of these foci release their contents into the pleural space, which is followed by acute inflammation and exudation caused by a delayed hypersensitivity reaction. Operative specimens reveal that subpleural TB lesions are observed frequently in patients with pleural TB.1-3 However, the subpleural TB focus is seldom visualized on imaging, even though it plays a major role in the development of pleural TB. Most previous studies on CT scan findings of pulmonary changes of pleural TB have focused on bronchocentric lesions such as centrilobular nodules, a tree-in-bud appearance, and consolidation, because these are the most characteristic findings of active pulmonary TB.4,5 However, there are no reports describing subpleural nodules on CT scans as active TB lesions in pleural TB. Moreover, the subpleural nodules have additional clinical significance because they are thought to be responsible for the development of paradoxical response (PR).
Pleural TB can manifest as a primary disease from recent infection or as a reactivated disease from remote infection. TB pleurisy due to reactivation is more likely to be associated with radiographic parenchymal infiltrations. However, it is difficult to determine with absolute certainty whether individual patients have TB caused by recent or remote infection, because definitive evidence of the exact time of TB infection is usually not available.5 In addition, classification of pleural TB into primary and reactivation forms has no clinical implication.6 Therefore, we analyzed all the parenchymal changes of pleural TB with recent CT scanners regardless of the form, after discarding preconceptions about CT scan findings of active pulmonary TB. In addition, combined pulmonary changes on CT scan can be helpful in the differentiation of pleural TB from other pleural disease. We also examined the prevalence of PR and compared the characteristic clinical and CT scan findings of patients who showed PR during follow-up with those of patients who did not develop PR.
Materials and Methods
CT Scanning
Patients
CT scans were obtained with two CT scanners, LightSpeed VCT (General Electric Medical Systems) and Somatom Definition Flash (Siemens Heathcare), with or without IV administration of contrast medium (100 mL at rate of 2.0-2.5 mL/s). Attending physicians requested a CT scan with contrast enhancement if chest radiographs led them to suspect a malignancy or TB lymphadenitis. Scanning parameters included a 130-mA tube current, 120-kV tube voltage, 128 3 0.6-mm collimation, and pitch of 1.2. All images were reconstructed into axial images with a 5-mm slice thickness at 5-mm intervals and coronal images with a 3-mm slice thickness.
We selected patients with pleural TB who underwent CT scans at the time of diagnosis between May 2008 and September 2013. We excluded patients with other coexisting pulmonary disease or a large amount of pleural effusion resulting in total collapse of the lung. We retrospectively reviewed clinical features and pleural effusion parameters, including differential cell count and levels of lactate dehydrogenase and adenosine deaminase (ADA). We analyzed the CT scan findings of all patients and checked for the development of PR during follow-up. Patients were divided into two groups according to the development of PR, and clinical and CT scan findings were compared between the PR and non-PR groups. Diagnosis of Pleural TB and PR The diagnosis of pleural TB was based on a lymphocyte-prominent (. 50%) effusion that met Light’s criteria for exudative effusions, with demonstration of an elevated ADA level of . 40 IU/L, positive results for acid-fast bacilli (AFB) staining, growth of Mycobacterium tuberculosis from pleural fluid or sputum, positive results for TB in a polymerase chain reaction (PCR) assay, the presence of classic caseating granulomatous inflammation in the pleural tissue, or a successful response to anti-TB therapy.5,7 We defined PR as a radiographic worsening of pleural TB in patients who initially improved with anti-TB treatment; in these patients, the onset of PR must have occurred after the initiation of treatment, without evidence of positive AFB results in a sputum study. We determined that radiographic findings were worsened if pleural effusions increased or new pulmonary lesions developed. The time to development of PR was defined as the interval between the initiation of anti-TB treatment and the onset of PR, as defined previously. Patients who developed PR within 14 days after initiation of anti-TB treatment were excluded from the PR group.7,8
journal.publications.chestnet.org
CT Scan Analysis All CT scans and medical records were reviewed retrospectively by two chest radiologists (J. M. K. and H. J. P.) with 4 and 11 years of experience, respectively. Interpretation was based on a consensus between the two radiologists. CT scans obtained to assess pleural effusion were analyzed with specific attention to pulmonary findings, such as micronodules, consolidation, and cavitation. A micronodule was defined as a small rounded opacity with a diameter , 7 mm and was classified by distribution as centrilobular, perilymphatic (subpleural or peribronchovascular), or random.9 The presence of interlobular septal thickening and clustered or coalescent micronodules showing a cluster or galaxy sign was included in our analysis.10 We checked for the presence of lymphadenopathy (LAP) and extrathoracic TB involvement. Findings of old scar lesions such as bronchiectasis, fibrotic linear opacities, and ciccatricial emphysema were excluded from our CT scan analysis. Statistical Analysis We compared clinical and CT scan findings between the PR group and the non-PR group. All tests of significance were two-sided. Univariate comparisons between the PR and the non-PR group were performed using the Pearson x2 test or Fisher exact test for categorical variables and the Student t test for continuous variables. A P value , .05 was considered statistically significant.
1605
Results Patients
This study was approved by the Institutional Review Board of St. Vincent’s Hospital, which waived informed consent (VC13RISI0218). A total of 349 patients who had received a diagnosis of pleural TB underwent CT scans at the time of diagnosis. Among them, 34 patients were excluded because of total lung collapse (n 5 21) or coexisting pulmonary diseases such as reexpansion edema (n 5 5), pneumonia (n 5 3), malignancy (n 5 2), pneumoconiosis (n 5 2), and pulmonary edema (n 5 1). The mean age of the remaining 315 patients was 47 years, and 201 patients were men. Fifty-seven patients had chronic illness, and common comorbidities were diabetes (n 5 27), chronic liver disease (n 5 12), collagen vascular disease (n 5 5), chronic renal failure (n 5 4), and chronic heart failure (n 5 4). None of the patients had HIV infection. Diagnosis
Diagnosis of pleural TB was confirmed by positive microbiology results (AFB, n 5 31; culture, n 5 59; TB in PCR, n 5 38), increased ADA level (mean, 84 IU/L) in 294 patients, lymphodominant exudates (mean percent of lymphocyte, 77%) in 289 patients, and histopathology in 27 patients. CT Scan Findings
Among 315 patients, 270 (86%) had pulmonary abnormalities on CT scans. Common CT scan findings were micronodules (n 5 209 [77%]) and interlobular septal thickening (n 5 202 [75%]). Among 209 patients with micronodules, the nodules were present along the subpleural region in 146 patients (70%) and the peribronchovascular bundles in 113 (54%). Micronodules were present in the subpleural and/or perilymphatic region in 184 patients (88%) (Figs 1, 2). Centrilobular distribution was present in 64 patients (31%), and miliary nodules with random distribution were observed in one patient. Forty-four patients (21%) had both centrilobular and perilymphatic nodules together. Determination of micronodule distribution was not possible in 12 patients. In eight of 12 patients with indeterminate nodules, perilymphatic micronodules coexisted in different areas. Micronodules were clustered in 26 patients (8%), showing a galaxy or cluster sign (Fig 3). Consolidation was seen in 120 patients (44%), and cavitation was associated with the consolidation in 49 patients (18%). Intrathoracic LAP 1606 Original Research
Figure 1 – A 43-y-old man with right pleural TB, diagnosed by increased adenosine deaminase level (100 IU/L) within lymphodominant (98%) pleural effusion. Microbiology study was negative. A and B, Axial CT scans with a lung window setting at the level of the manubrium show multiple welldefined micronodules in the subpleural region of the right upper lobe (arrows), accompanying smooth interlobular septal thickening (empty arrows).
was present in 46 patients (15%), and extrathoracic TB involvement was observed in 21 patients (7%) (Tables 1-3).
[
146#6 CHEST DECEMBER 2014
]
Comparison of Characteristics Between PR and Non-PR Groups
PR occurred in 81 patients (26%) between 16 and 233 days (mean, 76 days) after initiation of anti-TB treatment. Compared with the non-PR group, the PR group was younger (mean, 37 years; P , .000), had a lower prevalence of comorbidities (11%, P 5 .012), and had a male predominance (74%, P 5 .020) (Table 4). There was no difference in CT scan findings between the two groups, except for the presence of subpleural micronodules: The PR group had a higher prevalence of subpleural nodules than did the non-PR group (56%, P 5 .025) (Table 4) (Fig 4).
Discussion Pleural TB still remains the most common cause of pleural effusion in areas with a high prevalence of TB and has become more frequent after the emergence of HIV and the widespread use of immunosuppressive drugs.1,3-5 Patients invariably have a small subpleural nidus of TB that shows fibrous and granulomatous inflammation.11 However, coexisting pulmonary changes that are suggestive of active TB have a relatively low prevalence on imaging.5,12,13 A key point of the pathogenesis of TB is that TB progresses in a bronchocentric manner. Therefore, radiologic findings have been interpreted from the same point of view. The CT scan findings of active pulmonary TB include centrilobular nodules and linear branching opacity, consolidation, and cavitation. In addition to bronchogenic dissemination, LAP or miliary nodules can be seen in patients with lymphatic and hematogeneous spread.14,15
Figure 2 – A 50-y-old man with bilateral pleural TB. Thoracocentesis revealed lymphodominant exudate (100%) with an increased adenosine deaminase level (145 IU/L). TB in a polymerase chain reaction was positive in the pleural fluid. Sputum study was negative. A, Axial CT scan at the level of the sternal notch shows smooth interlobular septal thickening along the anterior aspect of both upper lobes (arrows). Multiple micronodules are seen in the left upper lobe, located in the subpleural region and thickened interlobular septa (empty arrow). B, Axial CT scan at the level of the aortic arch shows multiple micronodules in the subpleural region of the left upper lobe. Some discrete micronodules, , 3 mm in size, are also seen along the broncho-
journal.publications.chestnet.org
Most of the patients (86%) had pulmonary abnormalities other than TB sequelae on their CT scans. The most common findings were interlobular septal thickening and micronodules. Eighty-eight percent of the micronodules were distributed along subpleural regions and peribronchovascular bundles, almost twice the number of centrilobular nodules. Both interlobular septal thickening and the presence of subpleural and peribronchovascular micronodules strongly indicate that the lesions involved mainly lymphatics. However, to date, TB has not been considered a perilymphatic disease. In fact, such CT scan findings, typically, negate TB in differential diagnosis. vascular bundles in the left upper lobe (arrowheads). Nearby airways are patent, representing peribronchovascular lesions, not intrabronchiolar lesions.
1607
TABLE 1
] Pulmonary Abnormal CT Scan Findings in 270 Patients With Pleural TB
CT Scan Findings
No. Patients
%
Micronodules
209
77
Interlobular septal thickening
202
75
26
8
120
44
Cavitation
49
18
Intrathoracic LAP
46
15
Extrathoracic involvement
21
7
Galaxy/clustered sign Consolidation
LAP 5 lymphadenopathy.
Interlobular septal thickening is known to be relatively common in miliary TB.15,16 Im et al17 reported that interlobular septal thickening was transiently observed on the CT scans of 34% of patients with active pulmonary TB. They suggested that the septal thickening may have been caused by a local increase in lymphatic flow from the exudative lesions in a pattern of lobular consolidation and not by caseous material in the septa, because the findings disappeared completely without any residual changes after TB treatment.16-18 However, we disagree with their assessment for several reasons. First, interlobular septal thickening was very common in patients (77%), whereas the prevalence of consolidation was relatively low (44%). Secondary changes of nearby consolidation are not sufficient to explain the higher prevalence of associated interlobular septal thickening. Second, we do not agree with the notion that lesions that disappear without any residue after TB treatment would not be granulomatous lesions. Third, one TB study involving CT scan and pathologic correlation showed that granulomas with caseous necrosis were present in the interlobular septa in a pathologic specimen of a patient with miliary TB.15 Therefore, interlobular septal thickening in the TABLE 2
] Distribution of Micronodules in 209 Patients
Distribution
No. Patients
Figure 3 – A 26-y-old man with right pleural TB, diagnosed by increased adenosine deaminase level (69 IU/L) in lymphodominant exudate (67%). Sputum acid-fast bacilli and TB in a polymerase chain reaction were all negative. A, Axial CT scan shows a coalescence of numerous discrete micronodules in the peribronchovascular bundles and subpleural region, resulting in a sarcoid cluster sign, in the right upper lobe. The micronodules are different from typical centrilobular nodules, which are more evenly spaced, larger, and less defined than those of this patient. B, Another CT scan at the carina level shows nodular thickening of bronchovascular bundles in the anterior segment of the right upper lobe. There is no evidence of any filling lesion within bronchioles.
1608 Original Research
%
Perilymphatic distribution
184
88
Subpleural
146
70
Peribronchovascular
113
54
64
31
1
0
12
6
a
Centrilobular distribution Random distribution Indeterminate
aNodules were considered perilymphatic when they were in at least one of the subpleural region, peribronchovascular interstitium, or interlobular septa.
[
146#6 CHEST DECEMBER 2014
]
TABLE 3
] Comparison of Clinical Features Between PR and Non-PR PR (n 5 81 [26%])
Clinical Feature Age, mean, y
37
Male ratio
0.74
Non-PR (n 5 234 [74%]) 50 0.59
P Value .000a .012a
Presence of comorbidity, %
11
21
.020a
Lymphocyte, mean, %
79
77
.254
ADA level, mean, IU/L
92
85
.136
Positive AFB, %
12
16
.517
Positive PCR, %
13
12
.812
Positive culture, %
16
19
.594
ADA 5 adenosine deaminase; AFB 5 acid-fast bacilli; PCR 5 polymerase chain reaction; PR 5 paradoxical response. aStatistically significant (P , .05).
patients should be considered interstitial involvement of TB.
regions via the pulmonary lymphatics, TB may cause pleural TB.
In contrast to the interlobular septal thickening, micronodules in the subpleural region or peribronchovascular bundles associated with pleural or pulmonary TB are mentioned very rarely in previous reports. Traditionally, micronodules with such a distribution represent lymphatic spread of the disease. In the aforementioned study by Lee et al,15 there were some cases with necrotizing granulomas along the bronchovascular and subpleural interstitium in pathologic specimens. A high prevalence of subpleural micronodules together with interlobular septal thickening indicates that TB involves mainly the pulmonary lymphatics in many of the patients with pleural TB.9,19,20 After reaching the subpleural
Considering the high prevalence, we do not think that subpleural micronodules and interlobular septal thickening are new radiologic findings; instead, such lesions have previously been neglected or considered nonspecific. Moreover, nodules along the peribronchovascular interstitium could have been mistaken for centrilobular nodules and linear branching opacities. Actually, Poey et al21 reported that interstitial nodules and thickening were more common than centrilobular nodules in their TB study using highresolution CT scans. However, they considered interstitial lesions to be nonspecific or negligible findings in cases of active TB.
TABLE 4
] Comparison of CT Scan Findings Between PR and Non-PR
CT Scan Finding
PR (%)
Non-PR (%)
P Value
Pleural thickening
96
93
.183
Micronodules
74
64
.077
Subpleural nodules
56
42
.025a
Centrilobular nodules
22
19
.503
Peribronchovascular nodularity
38
35
.642
Septal thickening
65
64
.892
Galaxy
8
8
.970
Consolidation
36
38
.892
Cavitation
20
14
.234
Intrathoracic LAP
15
14
.734
5
7
.533
Extrathoracic TB
See Table 1 and 3 legends for expansion of abbreviations. Statistically significant (P , .05).
a
journal.publications.chestnet.org
Some studies have reported some cases in which active pulmonary TB shares certain CT scan signs with sarcoidosis. These are a sarcoid galaxy sign and a sarcoid cluster sign, which correspond to clusters of multiple micronodules with or without formation of a coalescent mass.6,10,22 Eight percent of patients had clustered nodules showing a galaxy or cluster sign. Heo et al22 mentioned that the nodules of the galaxy sign did not appear to coincide with any reported pattern of pulmonary TB such as bronchogenically spread nodules, but instead these may have a gross pathology similar to that of nodules of sarcoidosis. We also think that micronodules along the subpleural and peribronchovascular interstitium are probably caused by granulomas in the interstitium, as in sarcoidosis. In addition, considering that centrilobular nodules range from a few millimeters to approximately 1 cm in size, are often ill defined, and appear to be evenly spaced, the micronodules of these signs are most likely not centrilobular lesions, but instead are perilymphatic lesions.23 1609
PR was not a rare condition in our study, occurring in 26% of patients with pleural TB. This result is similar to that of Cheng et al.8 In their study, the risk factors for PR were younger age, a high serum albumin level, and a low proportion of lymphocyte and a high proportion of neutrophils in the pleural fluid. They did not include CT scan findings in their analysis. In these patients, there was no difference in pleural fluid parameters. However, PR was more common in younger, male, and previously healthy individuals without any comorbidities. The presence of subpleural nodules was the only CT scan finding that differed between the PR and nonPR groups. It is surprisingly consistent with the speculation by Choi et al.26 Considering the combined risk factors, PR seems to be more common in TB that involves subpleural lymphatics and is in a primary form.
Figure 4 – A 27-y-old man with pleural TB and paradoxical response. Thoracocentesis revealed lymphodominant exudate (90%) with increased adenosine deaminase level (78 IU/L). A, Axial CT scan of inferior pulmonary vein level shows fluid collection in left pleural space. Smooth interlobular septal thickening is seen in the lingular division of the left upper lobe (arrowheads). Also seen are several small nodules along the right interlobar fissures and the subpleural region of the right lower lobe (arrows). B, One-month follow-up axial CT scan after initiation of anti-TB chemotherapy shows complete regression of the left pleural effusion and the interlobular septal thickening of the left lingula. New subpleural consolidation is seen along the right minor fissure (empty arrow), despite the fact that the TB mainly involved the left pleura initially. Video-assisted thoracoscopic biopsy proved TB pathologically. However, microbiology results were all negative. The subpleural lesion regressed without any change of initial TB regimen later.
PR is the unusual expansion or new formation of a TB lesion during TB treatment, which is one phenomenon of the immune recontstitution inflammatory syndrome.8 PR has been reported extensively in patients who are HIV-positive who receive highly active antiretroviral therapy.24,25 However, it can also occur in patients who are HIV-negative, especially in extrapulmonary TB.7,26,27 Choi et al26 reported that the most common CT scan findings of PR in patients with pleural TB were peripheral pulmonary nodules or masses that mostly abut the normal or thickened pleura. Given that most patients with pleural TB had concomitant subpleural diseases and showed the characteristic subpleural location of PR, they speculated that subpleural TB lesions undergo transient worsening and become radiographically evident during treatment of TB pleurisy. 1610 Original Research
Pulmonary TB can be a lymphatic disease.28 An animal study revealed that Mycobacterium tuberculosis causes lymphatic vasculitis in the lungs, thus, providing a direct conduit from the lungs to the regional nodes. Moreover, the pulmonary lymphatics are themselves sites of infection and could be a site of latent infection.29 This lymphatic model can explain how TB reaches the subpleural lymphatics, which will penetrate the visceral pleura and cause pleural TB. Given that lymphatic spread is a major route of primary TB, it is easy to understand the high prevalence of pleural TB in a primary form. This could also explain the atypical manifestation of TB resembling sarcoidosis and the negative sputum results despite extensive pulmonary lesions with micronodules on CT scans. The major weak point of this study is that an open lung biopsy specimen was not obtained in any of these patients. Therefore, the CT scan findings cannot be correlated with the pathologic findings and our results are assumed. However, those findings have been recognized as representing perilymphatic disease for a long time and in many studies.19,20,30 Another weak point is that CT images of 3- to 5-mm thickness were used for the evaluation of micronodules and interlobular septal thickening. In our institute, contrast-enhanced multidetector CT (MDCT) scanning rather than high-resolution CT scanning is performed routinely when evaluating TB, because the main reason for the scan is to examine the presence of lymphadenitis or other associated findings or to rule out diseases other than pulmonary TB, such as lung cancer. Nonetheless, the resolution of MDCT scans has improved, which facilitates the evaluation of subtle abnormalities such as micronodules and septal lines, and, if needed, multiplanar reconstruction with thin-slice thickness is feasible by applying the
[
146#6 CHEST DECEMBER 2014
]
appropriate software. The fact that the results of our study can be applied to routine MDCT scan protocols may actually be advantageous.31 In addition, we retrospectively analyzed the patients who had been given a diagnosis of pleural TB, and, thus, the diagnostic performance or usefulness of CT scanning could not be evaluated. Therefore, further study is necessary to determine whether CT scan findings can differentiate pleural TB from other inflammatory and infectious pleural diseases such as empyema.
Acknowledgments Author contributions: H. J. P. is the guarantor of the content of this manuscript, including the data and analysis. J. M. K., H. J. P., and C. H. K. contributed to the conception and design, acquisition of data, analysis, and interpretation of data; drafting of the submitted article or critical revision of it for important intellectual content; and final approval of the version to be published. 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.
9.
10.
11.
12.
References 1. Berger HW, Mejia E. Tuberculous pleurisy. Chest. 1973;63(1):88-92. 2. Stead WW, Eichenholz A, Stauss HK. Operative and pathologic findings in twenty-four patients with syndrome of idiopathic pleurisy with effusion, presumably tuberculous. Am Rev Tuberc. 1955;71(4):473-502. 3. Seibert AF, Haynes J Jr, Middleton R, Bass JB Jr. Tuberculous pleural effusion. Twenty-year experience. Chest. 1991;99(4):883-886. 4. Seiscento M, Vargas FS, Bombarda S, et al. Pulmonary involvement in pleural tuberculosis: how often does it mean disease activity? Respir Med. 2011;105(7): 1079-1083. 5. Kim HJ, Lee HJ, Kwon S-Y, et al. The prevalence of pulmonary parenchymal tuberculosis in patients with tuberculous pleuritis. Chest. 2006;129(5):1253-1258. 6. Liam C-K, Lim KH, Wong C-M. Tuberculous pleurisy as a manifestation of primary and reactivation disease in a region with a high prevalence of tuberculosis. Int J Tuberc Lung Dis. 1999;3(9):816-822. 7. Jeon K, Choi WI, An JS, et al. Paradoxical response in HIV-negative patients with pleural tuberculosis: a retrospective multicentre study. Int J Tuberc Lung Dis. 2012;16(6):846-851. 8. Cheng VC, Yam WC, Woo PC, et al. Risk factors for development of paradoxical response during antituberculosis therapy
journal.publications.chestnet.org
13.
14.
15.
16.
17.
18.
19.
20.
Conclusions In conclusion, interlobular septal thickening and micronodules in the subpleural and peribronchovascular regions are the most common CT scan findings of pleural TB. Radiologically, lymphatic spread of TB is more common than bronchogenic spread in the lungs of patients with pleural TB. In addition, the presence of subpleural nodules is a risk factor for later development of PR, especially in young male patients without underlying disease.
in HIV-negative patients. Eur J Clin Microbiol Infect Dis. 2003;22(10):597-602. Austin JH, Müller NL, Friedman PJ, et al. Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society. Radiology. 1996;200(2):327-331. Marchiori E, Zanetti G, Barreto MM, de Andrade FT, Rodrigues RS. Atypical distribution of small nodules on high resolution CT studies: patterns and differentials. Respir Med. 2011;105(9):1263-1267. Valdés L, Alvarez D, San José E, et al. Tuberculous pleurisy: a study of 254 patients. Arch Intern Med. 1998; 158(18):2017-2021. Yilmaz MU, Kumcuoglu Z, Utkaner G, Yalniz O, Erkmen G. Computed tomography findings of tuberculous pleurisy. Int J Tuberc Lung Dis. 1998;2(2):164-167. Hulnick DH, Naidich DP, McCauley DI. Pleural tuberculosis evaluated by computed tomography. Radiology. 1983;149(3):759-765. Im J-G, Itoh H, Lee KS, Han MC. CT-pathology correlation of pulmonary tuberculosis. Crit Rev Diagn Imaging. 1995;36(3):227-285. Lee JY, Lee KS, Jung K-J, et al. Pulmonary tuberculosis: CT and pathologic correlation. J Comput Assist Tomogr. 2000;24(5): 691-698. McGuinness G, Naidich DP, Jagirdar J, Leitman B, McCauley DI. High resolution CT findings in miliary lung disease. J Comput Assist Tomogr. 1992;16(3): 384-390. Im JG, Itoh H, Shim Y-S, et al. Pulmonary tuberculosis: CT findings—early active disease and sequential change with antituberculous therapy. Radiology. 1993;186(3):653-660. Voloudaki AE, Tritou IN, Magkanas EG, Chalkiadakis GE, Siafakas NM, Gourtsoyiannis NC. HRCT in miliary lung disease. Acta Radiol. 1999;40(4): 451-456. Raman SP, Pipavath SN, Raghu G, Schmidt RA, Godwin JD. Imaging of thoracic lymphatic diseases. AJR Am J Roentgenol. 2009;193(6):1504-1513. Remy-Jardin M, Beuscart R, Sault MC, Marquette CH, Remy J. Subpleural micronodules in diffuse infiltrative lung
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
diseases: evaluation with thin-section CT scans. Radiology. 1990;177(1): 133-139. Poey C, Verhaegen F, Giron J, Lavayssiere J, Fajadet P, Duparc B. High resolution chest CT in tuberculosis: evolutive patterns and signs of activity. J Comput Assist Tomogr. 1997;21(4):601-607. Heo J-N, Choi YW, Jeon SC, Park CK. Pulmonary tuberculosis: another disease showing clusters of small nodules. AJR Am J Roentgenol. 2005;184(2):639-642. Murata K, Itoh H, Todo G, et al. Centrilobular lesions of the lung: demonstration by high-resolution CT and pathologic correlation. Radiology. 1986;161(3):641-645. Jung JW, Shin JW, Kim JY, et al. Risk factors for development of paradoxical response during anti-tuberculosis treatment in HIV-negative patients with pleural tuberculosis. Tohoku J Exp Med. 2011;223(3):199-204. Cheng VC, Ho PL, Lee RA, et al. Clinical spectrum of paradoxical deterioration during antituberculosis therapy in non-HIV-infected patients. Eur J Clin Microbiol Infect Dis. 2002;21(11):803-809. Choi YW, Jeon SC, Seo HS, et al. Tuberculous pleural effusion: new pulmonary lesions during treatment. Radiology. 2002;224(2):493-502. Cheng SL, Wang HC, Yang PC. Paradoxical response during antituberculosis treatment in HIV-negative patients with pulmonary tuberculosis. Int J Tuberc Lung Dis. 2007;11(12): 1290-1295. Behr MA, Waters WR. Is tuberculosis a lymphatic disease with a pulmonary portal? Lancet Infect Dis. 2014;14(3):250-255. Basaraba RJ, Smith EE, Shanley CA, Orme IM. Pulmonary lymphatics are primary sites of Mycobacterium tuberculosis infection in guinea pigs infected by aerosol. Infect Immun. 2006;74(9):5397-5401. Raoof S, Amchentsev A, Vlahos I, Goud A, Naidich DP. Pictorial essay: multinodular disease: a high-resolution CT scan diagnostic algorithm. Chest. 2006;129(3):805-815. Ko JM, Kim KJ, Park SH, Park HJ. Bronchiectasis in active tuberculosis. Acta Radiol. 2013;54(4):412-417.
1611
