| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Kobe University Graduate School of Medicine, 7-5-2
Kusunoki-cho, Chuo-ku, Kobe, Hyogo 650-0017, Japan.
2 Philips Medical Systems, Tokyo, Japan.
Received May 3, 2007;
accepted after revision September 9, 2007.
Supported by grants-in-aid for scientific research from the Japanese
Ministry of Education, Culture, Sports, Science, and Technology (JSTS, KAKEN;
No. 18591346); the Knowledge Cluster Initiative of the Ministry of Education,
Culture, Sports, Science, and Technology, Japan; the Smoking Research
Foundation; and Philips Medical Systems.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Ten nonsmoking and 61 consecutive smoking-related COPD subjects underwent dynamic oxygen-enhanced MRI, CT, and pulmonary function tests. COPD subjects were classified into four clinical stages on the basis of the ATS–ERS guidelines. Wash-in time and relative enhancement ratio maps were generated by pixel-by-pixel analyses. Mean wash-in time and relative enhancement ratio were determined as averages of region of interest (ROI) measurements. CT-based functional lung volumes were measured on quantitative CT using the density-masked CT technique. For comparison of assessment capability for smoking-related functional loss, the three parameters were correlated with the percentage predicted forced expiratory volume in 1 second (%FEV1) and the percentage predicted diffusing capacity of the lung for carbon monoxide corrected for alveolar volume (%DLCO/VA). To determine the clinical stage classification capability, these parameters were statistically compared for nonsmoking subjects and all clinical stages of smoking-related COPD subjects.
RESULTS. Correlation between mean wash-in time and %FEV1 (r = –0.74, p < 0.0001) and between mean relative enhancement ratio and %DLCO /VA (r = 0.66, p < 0.0001) was better than that between CT-based functional lung volume and either %FEV1 (r = 0.61, p < 0.0001) or %DLCO/VA (r = 0.56, p < 0.0001). Mean wash-in time showed a significant difference between nonsmoking and smoking-related COPD subjects at all clinical stages (p < 0.05).
CONCLUSION. Dynamic oxygen-enhanced MRI has potential for pulmonary functional loss assessment and clinical stage classification of smoking-related COPD as does quantitative CT.
Keywords: chronic obstructive pulmonary disease (COPD) • lung • MRI • oxygen • smoking • ventilation
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Smoking-related COPD is assessed by using whole-lung pulmonary function tests and radiologic examinations and is characterized by increased airway obstruction and diffusional abnormalities detected in routine clinical practice [8]. Staging of COPD uses the guidelines of the American Thoracic Society–European Respiratory Society (ATS–ERS), the Global Initiative for Chronic Obstructive Lung Disease (GOLD), or the National Institute for Health and Clinical Excellence (NICE) [1, 8–10]. For radiologic examinations to assess smoking-related COPD, chest radiography, CT, and nuclear medicine ventilation–perfusion (V/Q) studies have also been adapted to evaluate morphologic changes or regional pulmonary functional changes. Currently, CT is most widely used in this setting, and in clinical and academic practice, several commercially available systems or large numbers of proprietary software and visual scoring systems, such as the national emphysema treatment trial (NETT) score, have been adapted for CT-based assessment of pulmonary emphysema [11–16].
Recently, it has been suggested that hyperpolarized noble gas MRI using 3He and oxygen-enhanced MRI may be useful for assessment of regional morphologic and functional changes in COPD and other pulmonary diseases [17–25]. Oxygen-enhanced MRI was first mentioned as a potential alternative approach in 1996 [20]. Since then, several investigators have reported that regional ventilation and alveolocapillary gas transfer of molecular oxygen could be detected with this method [17–25]. Moreover, a few investigators found that oxygen-enhanced MRI parameters correlated significantly with forced expiratory volume in 1 second (FEV1) and the diffusing capacity of the lung for carbon monoxide (DLCO) [22–25]. Oxygen-enhanced MRI may therefore also be capable of assessing alveolo-capillary gas transfer and the degree of airway obstruction. However, to our knowledge, the literature shows no publications dealing with direct comparison of the capability of quantitatively assessed CT and oxygen-enhanced MRI for smoking-related functional loss assessment and clinical stage classification of smoking-related COPD subjects.
We hypothesized that dynamic oxygen-enhanced MRI may correlate with not only alveolocapillary gas transfer but also airway obstruction in smoking-related COPD patients and may be used for assessment of regional functional loss and clinical stage in the same manner as quantitative CT on the basis of the density-masked CT technique. The purpose of the study reported here was to prospectively compare the capability of dynamic oxygen-enhanced MRI and quantitative CT for smoking-related pulmonary functional loss assessment and clinical stage classification of smoking-related COPD.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Ten nonsmoking subjects who were suspected of having an abnormal chest shadow on chest radiography but who were diagnosed as having no cardiopulmonary disease on the basis of MDCT and pulmonary function tests (eight men, two women; age range, 27–88 years; mean age, 54 years) prospectively underwent dynamic oxygen-enhanced MRI. Sixty-one consecutive smoking-related COPD subjects (48 men, 13 women; age range, 36–87 years; mean age, 58 years) prospectively underwent routine examinations including MDCT and pulmonary function tests and dynamic oxygen-enhanced MRI. All MDCT examinations and pulmonary function tests were performed in random order and 1 week or less before or after MR examinations (mean, 3.7 days; range, 0–6 days).
The history of long-term cigarette smoking of all subjects was assessed with the Brinkman index (cigarette consumption per day x years) [26], and the diagnosis of smoking-related COPD was based on physiologic findings and the results of pulmonary function tests and chest thin-section MDCT examinations and made according to American Thoracic Society criteria [27]. The subjects were also classified into the following four COPD stages using ATS–ERS guidelines: "At Risk for COPD," "Mild COPD," "Moderate COPD," "Severe or Very severe COPD" [9]. All pulmonary function tests were performed by three pulmonary physicians with 11, 14, and 23 years of experience, respectively, and diagnosis of smoking-related COPD and clinical stage classification of smoking-related COPD were decided by consensus by the same three pulmonary physicians. Detailed characteristics of nonsmoking subjects and smoking-related COPD subjects at all stages are shown in Table 1.
|
Routine Examination for Smoking-Related COPD
CT examination—All examinations were performed on an MDCT
scanner (Somatom Plus 4 Volume Zoom, Siemens Medical Solutions). The scans
were obtained from the lung apex to the diaphragm with the following
parameters: 4 x 1.0 mm collimation, 6:1 pitch, 300–350 field of
view, 512 x 512 matrix, 140 kV, 110 effective mAs, and reconstructed as
contiguous slices of 2.0-mm thickness by means of a standard algorithm. Before
CT examination, patients practiced their breathing to produce full and
sustained inspiration, and CT was performed during breath-holding at the end
of full inspiration.
Image analysis of quantitative CT—For a quantitative estimate of CT-based functional lung volume, we used the assessment method previously described [25, 28, 29]. After applying dual thresholds of –500 and –950 H to each subject, the trachea; main bronchus; right upper, middle, and lower bronchi; and left upper and lower bronchi were manually excluded from every slice. Total functional lung volume was then calculated by multiplying the area of each functionally relevant lung tissue by the slice thickness. The area of associated emphysema was excluded by applying the lower threshold value (–950 H), and areas of air space loss, such as those resulting from fibrosis and atelectasis due to previous tuberculosis, were also satisfactorily excluded by means of visual inspection of the functional lung volume map [25, 28, 29]. A chest radiologist with 15 years of experience performed all quantitative assessments of CT-based functional lung volume using commercially available software (Pulmo, Siemens Medical Solutions).
Pulmonary Function Test
Pulmonary function testing was performed with an automatic spirometer
(System 9, Minato Ikagaku) and according to American Thoracic Society
standards [30,
31]. We measured forced vital
capacity (FVC), FEV1, lung volume, and DLCO, and
evaluated the ratio of FEV1 to FVC (FEV1/FVC%:
percentage predicted), the percentage predicted FEV1
(%FEV1), and the percentage predicted DLCO corrected for
alveolar volume (%DLCO/VA) according to the
ATS–ERS guide-lines
[32]. DLCO was
estimated for 10-second breath-holding.
Dynamic Oxygen-Enhanced MRI Technique
Dynamic oxygen-enhanced MR images were obtained with inhaled oxygen as a T1
contrast agent. T1-weighted images were continuously collected by means of a
respiratory synchronized centrically reordered HASTE pulse sequence and a
1.5-T scanner (Gyroscan Intera T-15, Philips Medical Systems). For a 256
x 256 matrix, 132 phase-encoding steps were acquired including four
steps for phase correction. The interecho spacing was 4.0 milliseconds, the
effective TE was 4 milliseconds, and the TR varied between 3,200 and 5,000
milliseconds depending on the respiratory cycle. A pneumatic belt was used for
respiratory triggering at the end of expiration. The section thickness was 10
mm, the field of view, 450 mm x 450 mm; and the inversion time (TI), 900
milliseconds. The in-plane phase-encoding direction was cranial to caudal. One
excitation was used to obtain three coronal sections for each oxygen-enhanced
MR examination. The sections were obtained 30 mm anterior to section two
(section one), at the center of the lung and including the carina and
bilateral main bronchus (section two), and 30 mm posterior to section two
(section three).
Patients and healthy volunteers inhaled room air first, followed by 100% oxygen (15 L/min) using a nonrebreathing ventilation mask. Administration of 100% oxygen in this manner has been shown to produce oxygen concentrations in the lung of between 60% and 80% [33]. The following paradigm for oxygen inhalation was used to obtain all coronal sections: Subjects first breathed 21% oxygen (room air) for 30 dynamic series, then 100% oxygen for 30 dynamic series, and finally room air for 30 dynamic series (90 dynamic series in total). Time intervals between first and second or between second and third dynamic oxygen-enhanced MR acquisitions were fixed at 5 minutes for each subject. The total time for each dynamic oxygen-enhanced MR examination was less than 30 minutes.
Image Analysis of Dynamic Oxygen-Enhanced MRI
Oxygen-enhanced MR image data were analyzed using the software (PRIDE,
Philips Medical Systems; developed as IDL by AdamNet Ltd.) and run on a Dell
Precision PC. The signal intensity–time course curve for each voxel on
oxygen-enhanced MR images was measured for a series of images and
computationally fitted as follows:
 | (1) |
 | (2) |
The regional distribution of wash-in time for the pixels was expressed as maps color coded from dark blue to red. The regional distribution of relative enhancement ratios for the pixels was expressed as maps color coded from dark blue to red. All oxygen-enhanced MR images were visualized as the wash-in time and as relative enhancement maps after the application of dual thresholds that were most suitable for extracting lung parenchyma according to visual assessment by a chest radiologist with 12 years of experience.
For correlation with the Brinkman index, FEV1/FVC %, %FEV1, and %DLCO/VA, and for determination of the capability for clinical stage assessment of smoking-related COPD on dynamic oxygen-enhanced MRI, mean wash-in time and mean relative enhancement ratio for each subject were determined as the averages of wash-in times and relative enhancement ratios measured from three regions of interest (ROIs) drawn over both lungs on three coronal sections (one ROI from each of the three coronal images). Areas of airspace loss, such as those resulting from fibrosis and atelectasis because of previous tuberculosis, were also satisfactorily excluded by means of visual inspection at ROI placement on dynamic oxygen-enhanced MRI.
Monitoring Adverse Events of Dynamic Oxygen-Enhanced MRI
Adverse events related to 100% oxygen inhalation such as dyspnea, chest
pain, headache, dizziness, nausea, and vomiting were monitored. Such events
were evaluated as mild, moderate, or severe according to the following
criteria: mild, the event was easily tolerated; moderate, the event interfered
with normal activity; and severe, the event was incapacitating (causing
inability to perform everyday activities or work). Subjects were monitored for
the occurrence of adverse events from the start of 100% oxygen inhalation
until completion of the examination.
Statistical Analysis
To compare the efficacy of dynamic oxygen-enhanced MRI and quantitative CT
for smoking-related functional loss assessment, mean wash-in time, mean
relative-enhancement ratio, and CT-based functional lung volume were
correlated with Brinkman index, FEV1 / FVC%, %FEV1, and
%DLCO/VA.
To determine the capability of the two methods for clinical stage classification, mean wash-in time, mean relative-enhancement ratio, and CT-based functional lung volume of nonsmoking subjects were compared with those of subjects at all stages of COPD by using the analysis of variance followed by Tukey's honest significant difference (HSD) multiple comparison test. A p value less than 0.05 was considered significant for all statistical analyses.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
Mean wash-in time showed good correlation with the Brinkman index (r = 0.70, p < 0.0001), FEV1/FVC% (r = –0.74, p < 0.0001), and % FEV1 (r = –0.74, p < 0.0001) and had moderate correlation with %DLCO / VA (r = –0.57, p < 0.0001). Mean relative enhancement ratio showed moderate correlation with the Brinkman index (r = –0.55, p < 0.0001), FEV1/FVC% (r = 0.62, p < 0.0001), % FEV1 (r = 0.68, p < 0.0001), and %DLCO/VA (r = 0.66, p < 0.0001). In addition, CT-based functional lung volume showed moderate correlation with the Brinkman index (r = –0.57, p < 0.0001), FEV1/FVC% (r = 0.57, p < 0.0001), %FEV1 (r = 0.61, p < 0.0001), and %DLCO/VA (r = 0.56, p < 0.0001).
Comparison of Dynamic Oxygen-Enhanced MR Indexes and CT-Based Functional Lung Volume for Nonsmoking and Smoking-Related COPD Subjects at All Stages
The results of a statistical comparison of mean wash-in time, mean relative
enhancement ratio, and functional volume for nonsmoking subjects and
smoking-related COPD subjects at all stages are shown in
Table 3.
|
Mean wash-in time, mean relative enhancement ratio, and functional lung volume of nonsmoking subjects and smoking subjects assessed as "At Risk for COPD" and "Mild COPD" were significantly different from those of "Moderate COPD" and "Severe or Very Severe COPD" smoking subjects (p < 0.05). In addition, the mean wash-in time of nonsmoking subjects was significantly different from that of smoking subjects at "At Risk for COPD" and "Mild COPD" (p < 0.05). Finally, the mean wash-in time of smoking subjects assessed as "Moderate COPD" showed a significant difference from that of "Severe or Very Severe COPD" smoking subjects (p < 0.05).
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Cigarette smoking has a prominent effect on the structure and function of the bronchial mucous glands in the bronchi or large airways and also induces bronchiolar narrowing, inflammation, and fibrosis in small airways. Pathologic changes in the lungs of patients with smoking-related COPD lead to corresponding physiologic changes characteristic of the disease [1, 34, 35]. Expiratory airflow limitation is the hallmark physiologic change in COPD and is caused primarily by fixed airway obstruction and a consequent increase in airway resistance. These changes in the small airways are believed to be responsible for much of the airflow obstruction demonstrable in patients with smoking-related COPD, especially those classified as having "Mild COPD" [1, 34, 35]. In addition, cigarette smoke induces destruction of the connective tissue matrix of alveolar walls [1, 34, 35]. In advanced COPD, peripheral airways obstruction, parenchymal destruction, and pulmonary vascular abnormalities reduce the lungs' capacity for gas change, producing hypoxemia and, later, hypercapnia [35].
The signal intensity–time course curve on dynamic oxygen-enhanced MRI is thought to be the result of regional respiration (i.e., ventilation, perfusion, alveolocapillary oxygen transfer [diffusion of oxygen], and neural regulation affected by the first three parameters). CT-based functional lung volume, on the other hand, has been used for quantitative assessment of destruction in lung parenchyma. Therefore, our results suggest dynamic oxygen-enhanced MRI may have potential for providing more accurate assessment of the influence of cigarette smoke than quantitatively assessed thin-section MDCT.
%FEV1 and FEV1/FVC% have been adapted for assessment of the degree of airway obstruction and of the clinical stage of COPD patients [36]. Our results indicate that mean wash-in time is likely to be more affected by ventilation than is the mean relative-enhancement ratio during dynamic oxygen-enhanced MRI. In addition, dynamic oxygen-enhanced MRI may provide a more sensitive assessment of the clinical stage of smoking-related COPD patients than quantitatively assessed thin-section MDCT.
Alveolocapillary gas transfer is normally defined as pulmonary diffusing
capacity, and in routine clinical practice, measured DLCO is
usually normalized by lung volume to yield %DLCO/VA. The
factors that determine the rate of diffusion of oxygen through the
alveolar–capillary barrier are specified in Fick's law for diffusion
[37]. Fick's law for diffusion
is shown in a simplified form as follows
[37]:
 | (3) |
Cigarette smoke induces destruction of lung parenchyma and reduces gas exchange due to reduction of the surface of the barrier available for diffusion. Our results therefore suggest that the mean relative enhancement ratio appears to be more affected than the mean wash-in time by alveolar-capillary oxygen transfer, even if measurements of %DLCO /VA and mean relative enhancement ratio are affected by ventilation. In addition, dynamic oxygen-enhanced MRI may produce a more sensitive assessment of the lung parenchymal destruction than quantitatively assessed thin-section MDCT.
Comparison of dynamic oxygen-enhanced MRI parameters and the CT-based functional lung volume of nonsmoking subjects and smoking-related COPD subjects at all stages showed that mean wash-in time clearly resulted in significant differences between nonsmoking subjects and smoking subjects at all clinical stages. In view of higher correlations among the mean wash-in time, Brinkman index, %FEV1, and FEV1/FVC%, previously described underlying physiopathology in cigarette smoking and smoking-related COPD, and the significant difference between nonsmoking subjects and smoking-related COPD subjects at all stages, our results suggest that mean wash-in time should be considered a more sensitive parameter than mean relative enhancement ratio or CT-based functional lung volume for clinical stage assessment in smoking-related COPD subjects.
There are some limitations in this study. First, the administration of oxygen to patients with pulmonary diseases may alter or modify existing pulmonary pathophysiology in both nonsmoking and smoking-related COPD subjects, especially advanced COPD patients. Second, we adopted the respiratory triggering technique to correct misregistration or change in the lung during respiration and did not use any other correction method for this purpose. This may cause problems, especially for assessment of regional oxygen enhancement near the diaphragm. For future studies, we will therefore adopt other correction methods, such as the navigator–echo technique with cardiac triggering or computational registration techniques for accurate assessment of regional oxygen enhancement near the diaphragm.
Third, we examined 10 nonsmoking subjects and 61 smoking-related COPD patients at various clinical stages. However, the actual number of smoking-related COPD subjects in each clinical stage group was relatively small and varied among groups. In addition, no justifications for the numbers of patients or control subjects were established for this study, resulting in a limited statistical significance of power analysis for comparison of clinical stage assessment of smoking-related COPD by using dynamic oxygen-enhanced MR indexes and CT-based functional lung volume. A large-scale prospective study with justification for the numbers of nonsmoking subjects and smoking-related COPD subjects in each clinical stage group is warranted to determine the potential of dynamic oxygen-enhanced MRI for clinical stage assessment of smoking-related COPD.
Fourth, although dynamic oxygen-enhanced MRI was obtained at the end of expiration with respiratory gating, we obtained all CT images at the end of inspiration and did not use respiratory gating. Even with breathing practice, it is known that breath-holding volumes vary by up to 20%, and the level of inspiration affects the appropriate cutoff value on density-masked CT for quantitative assessment of smoking-related COPD and X-ray tube drift [38–41]. Therefore, each CT-based functional lung volume may be affected both by the level of inspiration of the subject and by the CT scanner.
Fifth, although our results showed that the dynamic oxygen-enhanced MRI indexes featured superior correlation with smoking-related pulmonary functional loss and classified clinical stages of smoking-related COPD subjects more accurately, CT is much more widely used in routine clinical practice. In addition, quantitative CT assessed by a density-masked CT technique could assess the degree of lung parenchymal destruction and not evaluate small airways disease, although our smoking-related COPD patients included emphysema and obstructive bronchiolitis. Moreover, dynamic oxygen-enhanced MR indexes in our study were compared only with CT-based functional lung volume assessed with commercially available software on the basis of density-masked CT and could not be compared with that assessed with other software that was based on other methodologies [12–16, 29]. Therefore, dynamic oxygen-enhanced MRI may be considered to play a complementary role in the assessment of smoking-related pulmonary functional loss and clinical stage classification in smoking-related COPD subjects with CT.
Sixth, we were unable to directly compare the capability of the two methods examined here with that of nuclear medicine studies and hyperpolarized noble gas MRI. Recently developed hyperpolarized noble gas MRI using 3He is a reportedly superb new method for pulmonary functional imaging. Although the molecular weight of hyperpolarized 3He is different from that of oxygen, this technique visualizes the gas itself, thus visualizing the airway and air spaces as static or dynamic images. This technique measures the size of the air space by using the diffusion MR technique and has a new capability for pulmonary functional imaging [17–19]. The only drawbacks of these hyperpolarized noble gas techniques are the need for laser equipment and specialized radiofrequency transmitter–receiver coils as well as the cost of the noble gas. In contrast, oxygen is safe, inexpensive, and readily available. Conventional clinical proton MR scanners can be used for oxygen-enhanced MRI without any modification. The underlying physiology for oxygen-enhanced MRI seems to be different from that for hyperpolarized noble gas and can provide information based on the regional ventilation, perfusion, and oxygen transfer between the alveoli and capillary bed. Functional data derived from hyperpolarized noble gas and oxygen-enhanced MRI are thus likely to be different and possibly complementary.
Seventh, although the capability of dynamic oxygen-enhanced MRI for pulmonary functional loss assessment and clinical stage classification was assessed in the present study, we could not evaluate the utility for management of COPD. Therefore, to determine the real significance of dynamic oxygen-enhanced MRI, we may have to plan further studies to directly compare the efficacy of dynamic oxygen-enhanced MRI, quantitative and qualitatively assessed thin-section CT based on other methodologies, nuclear medicine studies and hyperpolarized noble gas MRI, and utility for management of smoking-related COPD patients.
In conclusion, dynamic oxygen-enhanced MRI and quantitatively assessed CT using the density-masked CT technique have potential for pulmonary functional loss assessment and clinical stage classification of smoking-related COPD. Detailed correlation analysis of dynamic oxygen-enhanced MRI parameters and pulmonary function test showed that dynamic oxygen-enhanced MRI may well correlate with not only DLCO but also FEV1 in smoking-related COPD patients. Direct assessment of the signal intensity–time course curve obtained from dynamic oxygen-enhanced MR data yields maps that show correlation with not only the diffusing capacity of the lung but also with airflow limitation in the lung.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
