0
Original Research: Disorders of the Pleura |

A Pilot Study of Autofluorescence in the Diagnosis of Pleural DiseaseAutofluorescence Detection During Thoracoscopy FREE TO VIEW

Feng Wang, MMSc; Zhen Wang, MD; Zhaohui Tong, MD; Lili Xu, MMSc; Xiaojuan Wang, MMSc; Yanbing Wu, MD
Author and Funding Information

From the Department of Respiratory and Critical Care Medicine, Beijing Institute of Respiratory Medicine and Beijing Chaoyang Hospital, Capital Medical University, Beijing, China.

CORRESPONDENCE TO: Zhaohui Tong, MD, Department of Respiratory and Critical Care Medicine, Beijing Chaoyang Hospital, No. 8 Gong Ti Nan Lu, Chaoyang District, Beijing, China 100020; e-mail: tongzhaohuicy@sina.com


Dr F. Wang is currently at the Department of Respiratory Medicine, Fuxing Hospital, Capital Medical University, Beijing, China.

Drs F. Wang and Z. Wang contributed equally to this manuscript.

FUNDING/SUPPORT: This study was supported by Beijing Municipal Science & Technology Commission [No. Z131107002213107].

Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details.


Chest. 2015;147(5):1395-1400. doi:10.1378/chest.14-1351
Text Size: A A A
Published online

BACKGROUND:  Conventional medical thoracoscopy (MT), routinely performed in patients with pleural disease, does not always lead to a conclusive diagnosis. The endoscopic appearance of pleural diseases under white light could be misleading. Autofluorescence has been shown to be an interesting and effective diagnostic tool. The objective of this study was to evaluate the diagnostic value of autofluorescence imaging during MT.

METHODS:  Patients with undiagnosed pleural effusion admitted to our clinical center between August 2013 and February 2014 were enrolled. MT was performed first with white light and then by autofluorescence. Endoscopic results of different diseases were recorded, and biopsy specimens were obtained for pathologic analysis. We calculated the diagnostic sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the two methods by comparing them with the pathologic results.

RESULTS:  Thirty-seven eligible patients were studied, including 21 with malignancy, nine with tuberculous pleurisy, three with infective pleurisy, and four with no diagnosed condition. Autofluorescence revealed additional malignant lesions, which were missed under white light in five patients. The diagnostic sensitivity and NPV of autofluorescence were 100% (95% CI, 98.5%-100%) and 100% (95% CI, 93.9%-100%), respectively. Autofluorescence was superior to white light, with a sensitivity of 92.8% (95% CI, 89.3%-95.3%) and NPV of 76.8% (95% CI, 67.0%-84.4%). For the specificity and PPV, no significant difference was found.

CONCLUSIONS:  The advantage of autofluorescence is its high sensitivity and NPV. It is useful to detect microlesions and delineate the pathologic margins. Autofluorescence can benefit patients with its better visualization.

Figures in this Article

Pleural effusion is a common syndrome where establishing the diagnosis may be challenging. Medical thoracoscopy (MT), also known as local anesthesia thoracoscopy, is a minimally invasive endoscopic procedure used in the diagnosis of pleural disease. Its diagnostic value has been proven in many studies.1 However, conventional MT occasionally fails to reveal an abnormal pathologic result because the endoscopic findings from white light can be nonspecific for some pleural diseases. A previous study reported that conventional thoracoscopy is unable to find macroscopic abnormalities in 8% to 10% of malignant cases with pleural effusion for which the cytologic examination of hydrothorax was a positive solution.2 Therefore, a more effective diagnostic tool is needed for clinical practice.

For many years, fluorescence has been extensively studied with the aim of enhancing the optical contrast of malignant tissues against surrounding tissues. Now, fluorescence-based photodiagnosis has been a popular and widely applied technology. In this study, we chose an autofluorescence imaging (AFI) system (Olympus) as a simple, economic, and safe procedure without involving photosensitizer administration. Autofluorescence has already been demonstrated to be an interesting and effective tool in the diagnosis of malignancy during bronchoscopy, but it has not been demonstrated as effective during MT. This study focused on the endoscopic findings of various pleural diseases by AFI during MT with an aim to evaluate diagnostic value.

Patients

Eligible patients presented with an undiagnosed exudative pleural effusion after at least one thoracocentesis and were admitted to Beijing Chaoyang Hospital between August 2013 and February 2014. Patient age ranged from 18 to 90 years. Patients with multiple pleural adhesions, transudatory pleural effusions, respiratory failure, coagulation disorders, anesthetic allergy, pregnancy, or any other MT contraindications were excluded. All patients received thoracic ultrasonography, chest CT scans, ECGs, and routine pleural effusion tests (eg, cell analysis, biochemical examination). This study was approved by the Institutional Review Board of Beijing Chaoyang Hospital (project approval number 12-ke-88). All patients signed information consent forms.

Equipment

The AFI videobronchoscope BF-F260 (the AFI system) was used as the fluorescence method during MT, whereas the flexirigid medical thoracoscope Olympus LTF-240 was used as a means of conventional white light mode control. The AFI system consists of three parts: an autofluorescence endoscope (BF-F260), a videoprocessor unit (EVIS LUCERA SPECTRUM [CV-260SL]; Olympus), and a xenon light source. An integrated filter enabled the selection of white light and excitation from the xenon light. The excitation light is blue (395-445 nm) and induces the autofluorescence (490-700 nm). There are two types of reflection lights: green (550 nm) and red (610 nm). The excitation light irradiates the examining tissues, and the resulting light signals are processed by the red-green-blue sequential videoscope system. Autofluorescence displays in green, and the green reflection light displays in red and the red reflection light in blue. In the autofluorescence examination, abnormal tissues are defined as an area that displays a different color from the surrounding tissues. In theory, normal mucosa appears green, the color of the autofluorescence light. Inflamed tissue appears blue because it contains a high concentration of hemoglobin that absorbs the green and red light signals. Malignant tissues appear red because the abnormally thickened epithelium absorbs most of the green autofluorescence signals3 (Table 1).

Table Graphic Jump Location
TABLE 1 ]  Clinical Judgment of Endoscopic Appearance

For AFI, the specimens without color change were considered negative. For WLT, only the normal-appearing pleura were considered negative. AFI = autofluorescence imaging; WLT = white light thoracoscopy.

Thoracoscopic Procedure

MT was performed in an endoscopy suite. For minimizing the visual bias, each procedure was completed by two pulmonologists working together. The patients were positioned on the nonaffected side, and a sterile field was prepared. The patients were given local anesthesia, staying conscious during the procedure. An entrance trocar was inserted into the thoracic cavity for inspection. After evacuating pleural effusions, the pleural cavity was first thoroughly inspected by conventional thoracoscope (white light thoracoscopy [WLT]) followed by AFI. Biopsy specimens of abnormal tissues identified by either method were examined. At the same time, random biopsy specimens were obtained in circumstances where no pathologic changes were detected. The specimens were stored in formalin and processed for pathologic analysis. The port incision was closed after the procedure, and a chest drainage tube was inserted for fluid evacuation and optimal expansion of the lung.

Outcome Evaluation

The specimens were categorized as AFI positive or negative and WLT positive or negative. The pathologist was blinded to the categorization and assessed each specimen separately. The pathologic reports were classified as malignancy, TB, infection, nonspecific inflammation, and normal. In this study, true positive was defined as abnormal endoscopic findings along with pathologic examination identifying as malignancy, TB, or infection. False positive was defined as abnormal endoscopic findings with pathologic examination indicating normal pleural or nonspecific inflamed tissues. On the other hand, a normal thoracoscopic result with an abnormal pathologic result was considered false negative. Normal endoscopic findings with benign biopsy specimens were defined as true negative.

The McNemar test was used to evaluate the differences between AFI and WLT, and consistency between the two methods was examined by κ test. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of both AFI and WLT were calculated. All probability values were evaluated by assuming a CI level of 95%. Statistical analyses were performed with SPSS for Windows, version 13.0 software (IBM Corporation).

Thirty-seven patients aged 22 to 85 years with undiagnosed exudative pleural effusions were enrolled, including 17 men and 20 women. Twenty patients had right-sided effusions, 16 had left-sided effusions, and one had bilateral effusions. Thirty-three patients received diagnoses, including 21 with malignancy, nine with tuberculous pleurisy, and three with infective pleurisy. Four patients did not receive a diagnosis because their pleural biopsy specimens indicated nonspecific inflammatory changes (Table 2).

Table Graphic Jump Location
TABLE 2 ]  Diagnoses in Enrolled Patients

Data are presented as No. (%).

A total of 491 biopsy specimens from 126 pleural sites (parietal and diaphragmatic pleura) were examined for pathologic changes. There were 380 specimens with AFI-positive and WLT-positive results. Pathologic examination revealed malignancy in 181 specimens, TB in 84 specimens, infection in 32 specimens, and nonspecific inflammatory change in 83 specimens. AFI-negative and WLT-negative results were found in 62 specimens, and the pathologic analysis showed normal pleura. A discrepancy existed in 37 specimens with AFI-positive but WLT-negative results, with 23 of these being malignant and 14 inflammations. Additionally, in 12 specimens, WLT showed porcelain-like plaques, but AFI indicated no fluorescence (ie, WLT positive with AFI-negative result), which proved to be nonspecific fibrous thickening (Table 3, e-Table 1).

Table Graphic Jump Location
TABLE 3 ]  Pathology and Endoscopic Appearance of Pleural Biopsy Specimens (n = 491)

− = AFI- or WLT-negative result; + = AFI- or WLT-positive result. See Table 1 legend for expansion of other abbreviations.

There was a statistical differentiation between AFI and WLT (P = .00). For pleural lesion detection, the rate of AFI-positive results was higher than that for WLT. Furthermore, the consistency of the two methods was statistically significant (κ = 0.66, P = .00). The sensitivity, specificity, PPV, and NPV of the two methods are summarized in Table 4. Although this result did not mean that AFI had achieved overall improvement in the diagnosis, we found the sensitivity and NPV of AFI to be extremely high; that AFI could detect all pleural diseases studied, including micrometastases (Ф < 3 mm); and that WLT missed additional pathologic changes (Fig 1). AFI was also useful in delineating the tumor margins because of its better visualization property. (Typical cases are shown in e-Figs 1-3 and Video 1.)

Table Graphic Jump Location
TABLE 4 ]  Diagnostic Probability Values of AFI and WLT

The difference in sensitivity and NPV between the two imaging methods was statistically significant by 95% CI comparison. NPV = negative predictive value; PPV = positive predictive value. See Table 1 legend for expansion of other abbreviations.

Figure Jump LinkFigure 1 –  A, Parietal pleura were generally normal apart from mild hyperemia during white light thoracoscopy. B, The same parietal pleura as in A are seen during the autofluorescence imaging examination. Some parts of the pleura, which were not visible by white light thoracoscopy, were found to present as purple-red, and a small cell lung cancer metastasis was confirmed by the color-changed area on biopsy specimen.Grahic Jump Location

Running Time: 10:41

Complications

No severe complications occurred in any patients studied. Minor postoperative complications included localized pain in two of the 37 patients (5.4%) and subcutaneous emphysema in another two patients (5.4%). No complication from the autofluorescence occurred. Furthermore, 5 to 10 min were required for additional fluorescence examination. This method proved to be as safe as the conventional thoracoscopy.

Application of fluorescence in the diagnosis of pleural diseases was first reported in 2002 by Prosst et al,4 who developed an animal pleural carcinosis model of human adenocarcinoma and measured photosensitizer (5-aminolevulinic acid [5-ALA]) accumulations in the tumor by applying indirect spectrometry; it was 11 times higher than the normal tissues. Fluorescence detected 30% more pleural malignant lesions than conventional thoracoscopy. Another group showed similar findings in a pig model,5 and another animal study by Ali et al6 found an interesting phenomenon of varying fluorescence diagnostic sensitivity among different tumor types after administration of 5-ALA.

Animal experiments have demonstrated that 5-ALA-induced fluorescence is more sensitive than conventional thoracoscopy in the diagnosis of intrathoracic malignancy. However, few studies have been done in human subjects due to potential complications caused by the photosensitizer. Baas et al7 selected 26 patients with undiagnosed pleural effusions or thickening in an advancing study. They examined 5-ALA-mediated fluorescence diagnosis during video-assisted thoracic surgery and found that fluorescence could enhance visualization of abnormal lesions, which led to more accurate staging for some of the patients with mesothelioma. However, no overall improvement was achieved by the fluorescence diagnostic method compared with the conventional method. In a similar study, Pikin et al8 reported a specificity, sensitivity, and diagnostic accuracy of 5-ALA-mediated fluorescence of 88.4%, 89.1%, and 88.9%, respectively. They concluded that the fluorescence diagnostic method improved the visualization of micromalignant lesions. The researchers also noticed a higher false-negative rate in fluorescence method. Besides the photosensitizer-mediated fluorescence method, Chrysanthidis and Janssen9 also studied autofluorescence diagnosis of pleural malignancy. Their study indicated easier discovery and more-precise mapping of the same malignant lesions by autofluorescence, reporting a sensitivity and specificity of 100% and 75%, respectively.

The highlight of the present study is the use of the autofluorescence technique in MT. Compared with photosensitizer-mediated fluorescence and thoracoscopic surgery, autofluorescence combined with MT is more convenient and economical for clinical practice with fewer complications. MT equipment is readily available to the pulmonologist, and it is possible to replace the thoracoscope with a bronchoscope. The results show the fluorescence technique to have a higher relative sensitivity but a comparatively lower specificity. We found that both the diagnostic sensitivity and the NPV of AFI are 100%. Based on these results, AFI seems to be superior to WLT (sensitivity, 92.8%; NPV, 76.8%). Specificity and PPV were not significant between the two methods. It is noteworthy that these data include the diagnosis of all types of pleural disease; in other words, positive pathologic results included malignancy, TB, and infection. If the positive pathologic findings were limited to malignancy only, the diagnostic specificity of AFI would change from 43.3% to 28.8%. However, the diagnostic specificity of WLT was 28.6%. Therefore, AFI showed no improvement in specificity in the diagnosis of malignancy. Early studies by a different group showed that 5-ALA-mediated fluorescence has a better diagnostic specificity in malignancy, which we did not prove in the present study. This might be explained by the characteristics of autofluorescence, which are a mixture of light reflections from different parts within tissues. In addition, fibrin, necrotic tissues, and pleural effusions could have interfered with macroscopic observation during the autofluorescence.

We are still unclear about the specificity of either photosensitizer-mediated fluorescence or autofluorescence in the diagnosis of various pleural diseases. Malignancy, TB, nonspecific inflammation, and even fat tissue showed a color change during the fluorescence examination. Biopsy specimens of the abnormal regions were AFI positive in most cases, but some only revealed nonspecific inflammation or fibrotic tissues. We noticed that malignancy usually displayed a bright purple-red color under AFI, whereas TB displayed light purple. This observation might be related to the degree of pathologic development. We also found that fat tissue could show a bright fuchsia color under AFI. Although AFI was giving a false-positive result, fat tissue could be easily identified by WLT, and this eliminated the need to examine a biopsy specimen. Some studies mentioned that fluorescence diagnostic methods could occasionally show false-negative results.4,79 This could be explained by the fact that the fluorescence can only examine superficial tissues and cannot detect deeper lesions. In the present study, 23 WLT false-negative results were found, but not one AFI false-negative result was identified. This means that WLT inspection might lead to more false-negative results than AFI.

Although we had carefully conducted thoracic examination by AFI and WLT in this study, a diagnosis still could not be reached in four patients. A possible explanation is that the pleural area was much larger than the airway tracts, and the autofluorescence light was insufficient to examine the whole cavity. Additionally, exhaustive examination of the pleural cavity was not possible with spontaneous breathing when the lung was not sufficiently collapsed under local anesthesia, and some microlesions might have been missed. We are still following up with these patients.

There were some technical limitations in this study. First, the difference in color changes during AFI for pleural disease was based on subjective physician decision (ie, visual observation rather than quantitative calculation). It was difficult to tell subtle fluorescent changes in some lesions, and we could not quantitatively evaluate the discrepancies between different pleural diseases. Second, we used the autofluorescence bronchoscope instead of thoracoscope during the fluorescence inspection. Bronchoscopes are considered more flexible, lacking the mechanical strength needed for accurately obtaining pleural specimens and, thus, making it more difficult to reach the same diagnostic accuracy compared with a rigid thoracoscope.10 To minimize the deviation, we used a flexirigid thoracoscope to obtain pleural biopsy specimens after using the autofluorescence bronchoscope to locate the lesion. Third, many biopsy specimens turned out to be nonspecific inflammations partly because of the small size or insufficient depth of the sampled tissue. Therefore, further improvement in obtaining thoracoscopic biopsy specimens is required to minimize the rate of missed diagnoses. For example, a closer approach by autofluorescence could help to detect subtle changes in a lesion. Obtaining biopsy specimens with sufficient depth and sampling the same pleural site repeatedly could also minimize false-negative results.

To our knowledge, we are the first group to report on the application of the autofluorescence diagnostic method during MT. The advantages of autofluorescence lie in its extremely high sensitivity and NPV. Autofluorescence combined with white light is helpful in identifying microlesions and delineating pathologic margins. AFI also contributes to the precise staging of advanced lung cancer with pleural effusions. For patients with AFI-negative results, physicians should be careful when evaluating causes for pleural effusions. These patients would benefit from this endoscopic imaging method.

Author contributions: F. W. and Z. T. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data collection and analysis. F. W. and Z. W. contributed to the data collection and analysis; F. W., Z. W., L. X., X. W., and Y. W. contributed to the MT procedure; F. W. contributed to the writing of the manuscript; Z. W. and Z. T. contributed to the review and editing of the manuscript; and L. X., X. W., and Y. W. contributed to the final approval of the manuscript.

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.

Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript.

Other contributions: The authors thank Mulan Jin, MD, and her team of pathologists for assistance in this study. The authors also thank Hong-jie Li and Zhan Lu for assistance during the MT procedure.

Additional information: The e-Figures, e-Table, and Video can be found in the Supplemental Materials and Multimedia sections of the online article.

5-ALA

5-aminolevulinic acid

AFI

autofluorescence imaging

MT

medical thoracoscopy

NPV

negative predictive value

PPV

positive predictive value

WLT

white light thoracoscopy

Blanc FX, Atassi K, Bignon J, Housset B. Diagnostic value of medical thoracoscopy in pleural disease: a 6-year retrospective study. Chest. 2002;121(5):1677-1683. [CrossRef] [PubMed]
 
Chakrabarti B, Ryland I, Sheard J, Warburton CJ, Earis JE. The role of Abrams percutaneous pleural biopsy in the investigation of exudative pleural effusions. Chest. 2006;129(6):1549-1555. [CrossRef] [PubMed]
 
Chiyo M, Shibuya K, Hoshino H, et al. Effective detection of bronchial preinvasive lesions by a new autofluorescence imaging bronchovideoscope system. Lung Cancer. 2005;48(3):307-313. [CrossRef] [PubMed]
 
Prosst RL, Winkler S, Boehm E, Gahlen J. Thoracoscopic fluorescence diagnosis (TFD) of pleural malignancies: experimental studies. Thorax. 2002;57(12):1005-1009. [CrossRef] [PubMed]
 
Vandermeulen L, Makris D, Mordon S, et al. Thoracoscopic findings and pharmacokinetics of inhaled fluorescein in a pig model. Respiration. 2010;80(3):228-235. [CrossRef] [PubMed]
 
Ali AH, Takizawa H, Kondo K, et al. 5-Aminolevulinic acid-induced fluorescence diagnosis of pleural malignant tumor. Lung Cancer. 2011;74(1):48-54. [CrossRef] [PubMed]
 
Baas P, Triesscheijn M, Burgers S, van Pel R, Stewart F, Aalders M. Fluorescence detection of pleural malignancies using 5-aminolaevulinic acid. Chest. 2006;129(3):718-724. [CrossRef] [PubMed]
 
Pikin O, Filonenko E, Mironenko D, Vursol D, Amiraliev A. Fluorescence thoracoscopy in the detection of pleural malignancy. Eur J Cardiothorac Surg. 2012;41(3):649-652. [CrossRef] [PubMed]
 
Chrysanthidis MG, Janssen JP. Autofluorescence videothoracoscopy in exudative pleural effusions: preliminary results. Eur Respir J. 2005;26(6):989-992. [CrossRef] [PubMed]
 
Tassi G, Marchetti G. Minithoracoscopy: a less invasive approach to thoracoscopy. Chest. 2003;124(5):1975-1977. [CrossRef] [PubMed]
 

Figures

Figure Jump LinkFigure 1 –  A, Parietal pleura were generally normal apart from mild hyperemia during white light thoracoscopy. B, The same parietal pleura as in A are seen during the autofluorescence imaging examination. Some parts of the pleura, which were not visible by white light thoracoscopy, were found to present as purple-red, and a small cell lung cancer metastasis was confirmed by the color-changed area on biopsy specimen.Grahic Jump Location

Tables

Table Graphic Jump Location
TABLE 1 ]  Clinical Judgment of Endoscopic Appearance

For AFI, the specimens without color change were considered negative. For WLT, only the normal-appearing pleura were considered negative. AFI = autofluorescence imaging; WLT = white light thoracoscopy.

Table Graphic Jump Location
TABLE 2 ]  Diagnoses in Enrolled Patients

Data are presented as No. (%).

Table Graphic Jump Location
TABLE 3 ]  Pathology and Endoscopic Appearance of Pleural Biopsy Specimens (n = 491)

− = AFI- or WLT-negative result; + = AFI- or WLT-positive result. See Table 1 legend for expansion of other abbreviations.

Table Graphic Jump Location
TABLE 4 ]  Diagnostic Probability Values of AFI and WLT

The difference in sensitivity and NPV between the two imaging methods was statistically significant by 95% CI comparison. NPV = negative predictive value; PPV = positive predictive value. See Table 1 legend for expansion of other abbreviations.

Running Time: 10:41

References

Blanc FX, Atassi K, Bignon J, Housset B. Diagnostic value of medical thoracoscopy in pleural disease: a 6-year retrospective study. Chest. 2002;121(5):1677-1683. [CrossRef] [PubMed]
 
Chakrabarti B, Ryland I, Sheard J, Warburton CJ, Earis JE. The role of Abrams percutaneous pleural biopsy in the investigation of exudative pleural effusions. Chest. 2006;129(6):1549-1555. [CrossRef] [PubMed]
 
Chiyo M, Shibuya K, Hoshino H, et al. Effective detection of bronchial preinvasive lesions by a new autofluorescence imaging bronchovideoscope system. Lung Cancer. 2005;48(3):307-313. [CrossRef] [PubMed]
 
Prosst RL, Winkler S, Boehm E, Gahlen J. Thoracoscopic fluorescence diagnosis (TFD) of pleural malignancies: experimental studies. Thorax. 2002;57(12):1005-1009. [CrossRef] [PubMed]
 
Vandermeulen L, Makris D, Mordon S, et al. Thoracoscopic findings and pharmacokinetics of inhaled fluorescein in a pig model. Respiration. 2010;80(3):228-235. [CrossRef] [PubMed]
 
Ali AH, Takizawa H, Kondo K, et al. 5-Aminolevulinic acid-induced fluorescence diagnosis of pleural malignant tumor. Lung Cancer. 2011;74(1):48-54. [CrossRef] [PubMed]
 
Baas P, Triesscheijn M, Burgers S, van Pel R, Stewart F, Aalders M. Fluorescence detection of pleural malignancies using 5-aminolaevulinic acid. Chest. 2006;129(3):718-724. [CrossRef] [PubMed]
 
Pikin O, Filonenko E, Mironenko D, Vursol D, Amiraliev A. Fluorescence thoracoscopy in the detection of pleural malignancy. Eur J Cardiothorac Surg. 2012;41(3):649-652. [CrossRef] [PubMed]
 
Chrysanthidis MG, Janssen JP. Autofluorescence videothoracoscopy in exudative pleural effusions: preliminary results. Eur Respir J. 2005;26(6):989-992. [CrossRef] [PubMed]
 
Tassi G, Marchetti G. Minithoracoscopy: a less invasive approach to thoracoscopy. Chest. 2003;124(5):1975-1977. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).
Supporting Data

Online Supplement

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543