| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thoracic Imaging |
1 From the Department of Radiology (Y.O., T.H., M.N., H.W., K.S.), Division of Cardiovascular, Thoracic, and Pediatric Surgery (M.Y.), and Department of Internal Medicine, Division of Cardiovascular and Respiratory Medicine (M.S., Y.N.), Kobe University Graduate School of Medicine, 7-5-2 Kusunoki-cho, Chuo-ku, Kobe, Hyogo 650-0017, Japan; Department of Radiology, Beth Israel Deaconess Medical Center, Boston, Mass (H.H.); Department of Radiology, Kobe Ekisaikai Hospital, Kobe, Japan (D.T.); and Philips Medical Systems, Tokyo, Japan (M.V.C.). Supported in part by Grants-in-Aid for Scientific Research from the Japanese Ministry of Education, Culture, Sports, Science, and Technology (JSTS.KAKENHI [no. 14770454]) and the Smoking Research Foundation. Received January 5, 2004; revision requested March 3; final revision received July 19; accepted September 29. Address correspondence to Y.O. (e-mail: yosirad{at}kobe-u.ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Study received institutional review board approval, and informed patient consent was obtained. Thirty consecutive patients (16 men and 14 women, aged 44–81 years; mean age, 65 years) considered candidates for lung resection underwent oxygen-enhanced MR imaging, CT, perfusion scintigraphy, and measurement of forced expiratory volume in 1 second (FEV1). A respiratory-synchronized inversion-recovery half-Fourier single-shot turbo spin-echo MR sequence was used for data acquisition. Correlation of postsurgical lung function (postsurgical FEV1) as determined with oxygen-enhanced MR imaging (FEV1MR), quantitative assessment with CT (FEV1Quant), qualitative assessment with CT (FEV1Qual), and perfusion scintigraphy (FEV1PS) was conducted with actual postsurgical FEV1, and the limits of agreement of each were determined with Bland-Altman analysis.
RESULTS: Correlation between postsurgical FEV1MR and actual postsurgical FEV1 values was excellent (r2 = 0.81, P < .001); it was better than that of FEV1Qual (r2 = 0.76) and FEV1PS (r2 = 0.77) and similar to that of FEV1Quant (r2 = 0.81) values. The limits of agreement of FEV1MR were between –9.9% and 10.9%.
CONCLUSION: Oxygen-enhanced MR imaging can be used to predict posturgical lung function in patients with lung cancer, similar to quantitative CT.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In current medical practice, ventilation and perfusion lung scanning, combined with spirometry, is the most widely used radiologic examination for the evaluation of patients whose pulmonary function may not be adequate to tolerate resection on the basis of spirometry results alone (4). However, perfusion scintigraphy is often performed in patients with low predicted postsurgical forced expiratory volume in 1 second (FEV1). The reported correlation coefficients of predicted and actual postsurgical lung function with this method vary between 0.51 and 0.92 (4–11).
As an alternative approach for the evaluation of surgical risk in patients with lung cancer, quantitative or qualitative evaluation with computed tomography (CT) on the basis of lung attenuation or anatomy has been reported (2,11,12). Although the correlation between assessment of postsurgical lung function with quantitative CT and actual postsurgical lung function has been excellent, qualitative assessment (ie, simple calculation based on the number of segments to be resected) is used by many clinicians.
Oxygen-enhanced magnetic resonance (MR) imaging offers an alternative approach for imaging of pulmonary ventilation (13). Several investigators reported that oxygen-enhanced MR imaging could demonstrate regional ventilation and reflected diffusion of oxygen (13–22). In contrast to standard lung function tests that demonstrate global pulmonary ventilation, this MR imaging method provides information on regional delivery of oxygen (13–22).
We hypothesized that oxygen-enhanced MR imaging may have potential for prediction of postsurgical lung function. The purpose of our study was to prospectively determine if lung function as assessed with oxygen-enhanced MR imaging correlates with postsurgical lung function in patients with lung cancer, as compared with quantitative and qualitative CT and scintigraphy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Patients
Thirty consecutive patients with lung cancer that were considered candidates for lung resection (16 men and 14 women, aged 44–81 years; mean age, 65 years) underwent presurgical contrast material–enhanced multi–detector row CT, oxygen-enhanced MR imaging, perfusion scintigraphy, and pre- and postsurgical pulmonary function testing. All presurgical radiologic examinations were performed in variable order less than 1 week before or after MR examinations (mean, 4.3 days; range, 1–6 days). Twenty-four of 30 patients had adenocarcinoma, three had squamous cell carcinoma, two had large cell carcinoma, and one had small cell carcinoma. Final diagnoses of all patients were confirmed by means of pathologic diagnosis of the resected specimen. Our institutional review board approved this study, and written informed consent was obtained from each patient prior to joining the study.
Oxygen-enhanced MR Imaging Technique
Oxygen-enhanced MR images were acquired by using inhaled oxygen as a contrast agent at T1-weighted MR imaging. T1-weighted images were collected continuously by using a respiratory-synchronized half-Fourier centrically reordered inversion-recovery single-shot turbo spin-echo pulse sequence by using a 1.5-T imager (Gyroscan Intera T-15; Philips Medical Systems, Best, the Netherlands). For a 256 x 256 matrix, 132 phase-encoding steps were acquired, including four steps for phase correction. The interecho spacing was 4.0 msec. The effective echo time was 4 msec. The repetition time varied between 3200 and 5000 msec and was dependent on the respiratory cycle. Respiratory triggering was performed during the end of inhalation. The section thickness was 10 mm. The field of view was 450 x 450 mm. The inversion time was 900 msec. The number of signals acquired was three. Oxygen-enhanced MR images from three coronal sections for each patient were acquired for all studies.
Each patient inhaled room air first, followed by 100% oxygen (15 L/min) by using a nonrebreathing ventilation mask. To image each coronal section for oxygen-enhanced MR imaging, 10 dynamic series were acquired during each alternating period. Each oxygen-enhanced MR image was obtained 60 seconds after oxygen inhalation.
Oxygen-enhanced MR Image Analysis
Oxygen-enhanced MR image data were analyzed by using proprietary software (PRIDE; Philips Medical Systems, Best, the Netherlands) developed by IDL (AdamNet, Tokyo, Japan) and a personal computer (FMV-900; Fujitsu, Tokyo, Japan). The signal intensity for each voxel was measured for the series of images. To visualize the relative enhancement map of oxygen-enhanced MR imaging, oxygen-enhanced MR images were expressed as the percentage change between the oxygen-enhanced and baseline images after applying dual thresholds that were able to most suitably allow extraction of lung parenchyma in each patient at visual assessment by a chest radiologist with 10 years of experience (Y.O.). Each pixel on the percentage change map was calculated computationally as follows:
 |
To determine the mean oxygen enhancement in each lobe, the mean relative enhancement ratio (MRER) of each lobe was calculated as the mean relative enhancement ratio of each lobe on all sections. The relative enhancement ratio of each lobe on each section was determined from regions of interest (range, 300–20 000 mm2) placed in the right upper, right middle, right lower, left upper, and left lower lobe on each section by one chest radiologist (Y.O.). The postsurgical FEV1 (percentage predicted) with oxygen-enhanced MR imaging (FEV1MR) was calculated as follows:
 |
CT Examination
All examinations were performed by using a multi–detector row CT scanner (Somatom Plus 4 Volume Zoom; Siemens Medical Systems, Forchheim, Germany). The scans were performed from the lung apex to the diaphragm (four detectors with 1-mm detector collimation, 6:1 pitch, 300–350 field of view, 512 x 512 matrix, 140 kV, 110 effective mAs) and reconstructed with 7.5-mm section thickness. Before CT examination, patients practiced their breathing to produce full and consistent inspiration. CT was then performed during breath holding at the end of full inspiration. Contrast medium (Iopamiron 300; Shering Japan, Osaka, Japan) was administered intravenously via an antecubital vein at 2–3 mL/sec with a power injector (Auto Enhance-50; Nemoto, Tokyo, Japan) with an empiric scan delay of 20 seconds to delineate the boundaries between tumor and mediastinal structures.
CT Image Analysis
Quantitative estimation of postsurgical lung function.—For quantitative estimation of postsurgical lung function from the functional lung volume, we applied the assessment method described previously by Wu et al (11,12). After applying dual thresholds of –500 and –910 HU, total and regional functional lung volumes of the lung or lobe to be resected were calculated by multiplying the area of each functionally relevant lung tissue by section thickness. The area of associated emphysema was excluded by the lower threshold value (–910 HU), and the areas of tumor-related air space loss, such as the tumor itself and postobstructive atelectasis or areas of non–tumor-related air-space loss, such as fibrosis and atelectasis due to previous tuberculosis, were also excluded satisfactorily at visual inspection of the functional lung volume map. A chest radiologist with 10 years of experience (Y.O.) performed all quantitative assessments of functional lung volume by using commercially available software (Pulmo; Siemens Medical Systems).
From total functional lung volume (TFLV) and regional functional lung volume data, postsurgical FEV1 (percentage predicted), which was evaluated with quantitative assessment of CT findings (FEV1Quant), was calculated by using the following formula:
 |
The regional functional lung volume of the resected lung or lobe was determined as the sum of regional functional lung volumes calculated from regions of interest (range, 300–20 000 mm2) placed in the resected lobe or lung on each section by one chest radiologist (Y.O.).
Qualitative estimation of postsurgical lung function.—For qualitative estimation of postsurgical lung function, postsurgical FEV1 was obtained by using presurgical pulmonary function testing data and information on the number of bronchopulmonary segments removed according to the previously reported method frequently used in many surgical institutions (2,23–25). To determine the number of bronchopulmonary segments removed, all CT studies were interpreted by two chest radiologists who had 8 and 10 years of experience (T.H. and Y.O.), and final assessments were made by means of consensus. The postsurgical FEV1 value (percentage predicted) according to qualitative assessment of CT findings (FEV1Qual) was then estimated with the following formula:
 |
Perfusion Scintigraphic Technique
Standard perfusion scintigraphy was performed after intravenous administration of 185 MBq of technetium 99m (99mTc) macroaggregated albumin. Images were acquired with a gamma camera with a large field of view (e-CAM; Siemens Medical Systems) equipped with a medium-energy all-purpose collimator, according to the method described by Markos et al (8). The matrix size was 256 x 256, and the energy windows of 99mTc were 140 keV ± 14.
Perfusion Scintigraphic Image Analysis
All perfusion studies were analyzed by using the application software of the gamma camera (e-soft, version 2; Siemens Medical Systems). On both the anterior and posterior images, rectangular regions of interest (range, 7130–16 580 pixels) that were equal in size were drawn over the whole lung by one chest radiologist (Y.O.). Both lungs were divided into five regions of interest in the right upper, right middle, right lower, left upper, and left lower lobes in all patients. Then, perfusion of each lobe (QP) was evaluated as follows:
 |
From the QP findings, postsurgical FEV1 (percentage predicted) with perfusion scintigraphy (FEV1PS) was calculated as follows:
 |
Physiologic Index and Outcome Measures
Pulmonary function testing was performed according to American Thoracic Society standards by using an automatic spirometer (System 9; Minato Ikagaku, Osaka, Japan). All patients underwent pre- and postsurgical pulmonary function testing. All presurgical pulmonary function spirometric tests were performed within 2 weeks before MR examination (mean, 6.4 days). All postsurgical pulmonary function spirometric tests were performed within 6–24 weeks of surgery (mean, 10 weeks). Pulmonary function testing was performed according to American Thoracic Society standards (26,27).
Monitoring of Adverse Events
Adverse events related to 100% oxygen inhalation, such as dyspnea, chest pain, headache, dizziness, nausea, and vomiting, were monitored by two radiologists (Y.O., T.H.). Adverse events were evaluated as mild, moderate, or severe by using the following definitions: mild, the event was easily tolerated; moderate, the event interfered with normal activity; and severe, the event was incapacitating (led to inability to perform usual activity or work). Patients were monitored for the occurrence of adverse events from the start of 100% oxygen inhalation until completion of the examination.
Statistical Analysis
To determine the capacity of oxygen-enhanced MR imaging for estimation of postsurgical lung function, the correlation and the limits of agreement between estimated and actual postsurgical FEV1 (percentage predicted) were evaluated statistically. The limits of agreement between estimated and actual postsurgical FEV1 (percentage predicted) were analyzed with Bland-Altman analysis.
A P value of less than .05 was considered to indicate a statistically significant difference in all statistical analyses. The basic theory and application of the limits of agreement have been documented in the literature (28).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Comparison between estimated and actual postsurgical FEV1 values (percentage predicted) and the correlation and limits of agreement of FEV1MR were better than those of FEV1Qual and FEV1PS and similar to those of FEV1Quant. In addition, postsurgical FEV1 values (percentage predicted) estimated with oxygen-enhanced MR imaging had a tendency toward underestimation similar to that with quantitative CT, when compared with actual postsurgical FEV1 values (percentage predicted). The limits of agreement of oxygen-enhanced MR imaging are considered small enough for clinical purposes. Oxygen-enhanced MR imaging can be used for estimation of postsurgical lung function.
When postsurgical lung function was predicted with oxygen-enhanced MR imaging, regional and total functional lung volumes were assessed on the basis of regional oxygen enhancement. One hundred percent molecular oxygen has a weak paramagnetic property and modulates the MR imaging signal intensity of blood and fluid, reflecting that of oxygen ventilation and diffusion (17,18,21,29–32). Therefore, oxygen enhancement surrounding lung cancer and the area of obstructive pneumonia demonstrated decreased oxygen enhancement (18). In addition, for patients with clinical or subclinical pulmonary emphysema, the areas of the lung affected by emphysema demonstrated decreased regional respiration and inhomogeneous and varying degrees of decreased oxygen enhancement (21). As a result of the maintained ventilation and perfusion relationship in patients with lung cancer, regional oxygen enhancement is considered to be the result of regional respiration (ie, ventilation, perfusion, diffusion of oxygen, and neural regulation affected by these other three parameters). Therefore, functional lung volume evaluated with oxygen-enhanced MR imaging is considered regional respiratory volume and demonstrates higher correlation coefficient and smaller limits of agreement with estimated and actual postsurgical FEV1 values compared with other radiologic methods, except quantitative CT.
At quantitative CT, estimated postsurgical FEV1 (percentage predicted) values were calculated from regional and total functional lung volumes, which omitted the nonfunctional lung volume due to pulmonary emphysema, tumor, atelectasis, and fibrosis from regional and/or whole lung volume based on attenuation in the lung. From our results and those of others (11,12), this method may be the most accurate for estimation of postsurgical lung function.
In contrast to quantitative CT, qualitative assessment of CT scans based on bronchopulmonary segments requires a simple calculation method. Many surgical institutions claim that this simple calculation provides satisfactory estimation of surgical risk. This is certainly true for low-risk candidates with good presurgical pulmonary function test results. However, the correlation coefficient and the limits of agreement of qualitative CT were lower and larger, respectively, than those of other methods in the present study. Moreover, the tendency of underestimation of postsurgical FEV1 values in this simple calculation was more than that of oxygen-enhanced MR imaging and quantitative CT because there is no evaluation of associated pulmonary emphysema and functional loss due to obstructive atelectasis or of the nonfunctional lung around the lung mass.
Perfusion scintigraphy has been the reference standard for estimation of postsurgical lung function after lung resection in patients with lung cancer. As a result of the low spatial resolution and underestimation of the associated functional loss, perfusion scintigraphy tends to lead to overestimation of postsurgical lung function and demonstrates lower correlation coefficient and larger limits of agreement with regard to actual postsurgical lung functions (5–11). These facts are compatible with our results. Perfusion scintigraphy may be replaced by quantitative CT or oxygen-enhanced MR imaging in the near future because of the better correlations and smaller limits of agreement of quantitative CT and oxygen-enhanced MR imaging with regard to actual postsurgical FEV1 values (percentage predicted).
There were some limitations in this study. First, in patients with severe emphysema, lung resection can improve postsurgical pulmonary function, similar to that seen in lung volume reduction surgery. The beneficial effects on lung elastic recoil and chest wall mechanics make it difficult to predict postsurgical lung function (33,34). Therefore, it is necessary for patients with lung cancer and severe emphysema to have their cardiopulmonary reserve evaluated with exercise lung function tests before lung resection, even though oxygen-enhanced MR imaging or other radiologic methods may lead to more accurate estimation of postsurgical lung function. Second, the administration of oxygen in patients with pulmonary disease may alter or modify existing pulmonary pathophysiology. For example, increased oxygenation in the airway may reverse existing hypoxic vasoconstriction (35). Thus, oxygen-enhanced MR imaging could demonstrate falsely increased oxygen delivery in a previously hypoxic segment or lobe, correlating with FEV1 (percentage predicted) and not clinical outcome.
Third, physiologic index and outcome measurement after surgery ranged between 6 and 24 weeks and were not assessed at a specific time. This fact affected our actual postsurgical lung function and our statistical comparison between predicted and actual postsurgical lung function. Fourth, in the present study, we assessed the regional and total functional lung volumes from relative enhancement ratios acquired in a limited number of sections because of the limited MR examination time in a clinical situation. To estimate precisely the regional and total lung functional volume from the oxygen-enhanced MR images, however, it would be better for us to obtain images of the entire lung at oxygen-enhanced MR imaging. Therefore, it is necessary for us to develop new oxygen-enhanced MR sequences for assessment of regional oxygen enhancement in the entire lung and to demonstrate the relationship between regional oxygen enhancement and functional lung volume.
Fifth, during quantitative assessment of functional lung on chest CT images of the entire lung, we adapted 7.5-mm images because of the limited computer performance. However, it would be better for us to use thinner sections for assessment of pulmonary emphysema. This fact is the potential limitation for presurgical assessment of regional and total functional lung volume and estimation of postsurgical lung function by using quantitative CT. Therefore, in future study, we would adapt thin-section CT for quantitative assessment of functional lung after installation of a computer with a central processing unit that has more advanced capabilities.
Sixth, the population of the present study was small, and the comparison of each case of estimated postsurgical lung function with long-term outcome after lung resection could not be evaluated. Furthermore, although estimated postsurgical FEV1 values (percentage predicted) have been reported as the most sensitive parameter for lung resection, we did not compare the oxygen-enhanced MR parameter with other pulmonary functional parameters. Therefore, larger prospective studies are warranted to demonstrate the real implications for determination of postsurgical lung function and outcomes in patient care with each radiologic examination.
In conclusion, oxygen-enhanced MR imaging and quantitative CT correlate with postsurgical lung function in patients with lung cancer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: FEV1 = forced expiratory volume in 1 second • FEV1MR = FEV1 at oxygen-enhanced MR imaging • FEV1Qual = FEV1 at qualitative CT • FEV1Quant = FEV1 at quantitative CT • FEV1PS = FEV1 at perfusion scintigraphyabbx
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, Y.O.; study concepts and design, Y.O.; literature research, Y.O.; clinical studies, Y.O., T.H., M.N., D.T., H.W., M.Y., M.S., Y.N., K.S.; data acquisition, Y.O., T.H., M.N., M.V.C.; data analysis/interpretation, Y.O., M.V.C.; statistical analysis, Y.O.; manuscript preparation, Y.O.; manuscript definition of intellectual content and editing, Y.O., H.H.; manuscript revision/review and final version approval, Y.O., H.H., K.S.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. K. Bernstein and S. Deshpande Preoperative Evaluation for Thoracic Surgery Seminars in Cardiothoracic and Vascular Anesthesia, June 1, 2008; 12(2): 109 - 121. [Abstract] [PDF]
|
|
|
|
 |
|
 Y. Ohno, T. Iwasawa, J. B. Seo, H. Koyama, H. Takahashi, Y.-M. Oh, Y. Nishimura, and K. Sugimura Oxygen-enhanced Magnetic Resonance Imaging versus Computed Tomography: Multicenter Study for Clinical Stage Classification of Smoking-related Chronic Obstructive Pulmonary Disease Am. J. Respir. Crit. Care Med., May 15, 2008; 177(10): 1095 - 1102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Ohno, H. Koyama, M. Nogami, D. Takenaka, S. Matsumoto, M. Obara, and K. Sugimura Dynamic Oxygen-Enhanced MRI Versus Quantitative CT: Pulmonary Functional Loss Assessment and Clinical Stage Classification of Smoking-Related COPD Am. J. Roentgenol., February 1, 2008; 190(2): W93 - W99. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Colice, S. Shafazand, J. P. Griffin, R. Keenan, and C. T. Bolliger Physiologic Evaluation of the Patient With Lung Cancer Being Considered for Resectional Surgery: ACCP Evidenced-Based Clinical Practice Guidelines (2nd Edition) Chest, September 1, 2007; 132(3_suppl): 161S - 177S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Ohno, H. Koyama, M. Nogami, D. Takenaka, S. Matsumoto, M. Yoshimura, Y. Kotani, and K. Sugimura Postoperative Lung Function in Lung Cancer Patients: Comparative Analysis of Predictive Capability of MRI, CT, and SPECT Am. J. Roentgenol., August 1, 2007; 189(2): 400 - 408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Hopkins, D. L. Levin, K. Emami, S. Kadlecek, J. Yu, M. Ishii, and R. R. Rizi Advances in magnetic resonance imaging of lung physiology J Appl Physiol, March 1, 2007; 102(3): 1244 - 1254. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
