SEARCH

SEARCH BY CITATION

Keywords:

  • chronic obstructive pulmonary disease ;
  • CT quantization ;
  • MR perfusion imaging ;
  • phenotype ;
  • pulmonary function test

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Introduction

Computed tomography (CT) and magnetic resonance imaging (MRI) can provide detailed anatomic structures and quantitative function information for chronic obstructive pulmonary disease (COPD).

Objectives

To prospectively clarify characteristics of pulmonary function test (PFT), CT volume parameters and magnetic resonance (MR) perfusion imaging in COPD patients with different high-resolution computed tomography (HRCT) phenotypes.

Methods

Sixty-two patients performed PFT, CT and MR perfusion imaging. COPD was classified into three phenotypes according to HRCT quantitative findings: A, E and M phenotype. Total lung volume (TLV), total emphysema volume (TEV) and emphysema index (EI) were quantitated by HRCT. In cases of perfusion defects (PDs), the shape and size were evaluated. The contrast between the normal lung and PDs was quantified by calculating their signal intensity ratio (RSI = SIPD/SInormal). The correlation was performed between PFT, CT and MR perfusion.

Results

There were 42 A phenotype, 9 E phenotype and 11 M phenotype. There was significant difference in forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) between A and M phenotype (P < 0.05). TEV and EI of A phenotype (0.4 ± 0.4 L and 8.0% ± 4.3%) were lower than those of E (1.0 ± 0.3 L and 18.6% ± 3.2%) or M phenotype (0.9 ± 0.2 L and 17.5% ± 1.7%). MR perfusion images showed circumscribed or diffuse patchy PDs. RSI of A phenotype was higher than that of E phenotype (20.3% ± 8.5% vs 11.8% ± 5.4%; P = 0.006). TEV and EI were moderate negatively correlated with diffusion function parameters. RSI was strongly correlated with FEV1% (A) and FEV1/FVC (M). FEV1/FVC was strongly correlated with TEV or EI (E).

Conclusion

There were different features and correlations between PFT, CT volume and MR perfusion in different phenotype, indicating each phenotype may have novel imaging method guiding clinical management.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Chronic obstructive pulmonary disease (COPD), a common preventable and treatable disease, is characterized by persistent airflow limitation that is usually progressive and associated with an enhanced chronic inflammatory response in the airways and the lung to noxious particles or gases. Exacerbations and comorbidities contribute to the overall severity in individual patients [1]. Airflow limitation is mainly caused by small airway remodeling and/or emphysema, and these pathologic changes directly determine the clinical management strategy [2, 3]. Therefore, differentiation of the two types of pathologic change is clinically important. Pulmonary function test (PFT), as an established clinical tool, is difficult to determine these pathologic changes, and cannot evaluate the regional function and morphological abnormalities. In contrast, imaging examinations have shown great advantages on the assessment of regional morphologic and function changes in COPD patients.

High-resolution computed tomography (HRCT) can provide detailed anatomic information and morphologic changes such as emphysema, bronchial wall thickening and air trapping. Thus, HRCT has been used to differentiate between airway-predominant and emphysema-predominant COPD [4]. Several studies have evaluated the relationship between HRCT phenotypes and responses to therapeutic intervention, or between HRCT phenotypes and clinical characteristics [3, 5]. In addition to detailed anatomic structures, computed tomography (CT) can make quantitative analysis on lung volume, airway and lung density. Previous studies have demonstrated that these quantitative measures correlate well with pathology [6]. Quantitative assessment of emphysema is most often based on determining the percentage of lung voxels below a specific threshold, such as –950 HU. Both automatic and semiautomatic quantitative evaluations have been applied in recent studies [2, 7-9].

In addition to morphological changes, ventilation and perfusion in COPD patients may also be impaired because of airway obstruction, parenchymal destruction and hypoxic vasoconstriction. Although magnetic resonance (MR) has some limits in lung morphological assessment because of intrinsic low spin density and heterogeneous susceptibility, it plays an important role in imaging of lung function including perfusion and ventilation. MR perfusion imaging has been widely used in the assessment of focal pulmonary parenchyma perfusion in various patients with pulmonary embolism, lung cancer and emphysema [10, 11].

Therefore, the combination of CT and magnetic resonance imaging (MRI) can provide detailed anatomic structures and quantitative function information as well. The characteristics of MR perfusion in COPD patients with different HRCT phenotypes are seldom reported, although they can help understand the pulmonary blood change in different phenotypes. The aim of this prospective study was to clarify the characteristics of PFT, CT volume analysis and MR perfusion imaging in COPD patients with different phenotypes.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Patients and PFT

Included in this prospective study were 62 COPD patients (53 male and 9 female) who admitted in the same Department of Respiration Medicine in one hospital and ranged in age from 44 to 79 years with a mean of 65.7 ± 9.3 years. The patient inclusion criteria were based on the American Thoracic Society criteria, which define COPD as airflow obstruction with a forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) <70% and no significant response to a bronchodilator test (after bronchodilator use: FEV1 increase <200 mL and/or reduction of airway resistance <15%) [12]. The mean body mass index (BMI) of all the patients was 22.7 ± 3.4 kg/m2. Of the 62 patients, 51 patients had a smoking history; the median and interquartile range of smoking index was 33 pack years and 20 pack years, respectively. The clinical symptoms were recorded according to the patients' complain, mainly including cough, sputum and wheezing. Five patients were hospitalized after they finished the study because of acute exacerbation, and received inhaled glucocorticoids /long-acting β-agonists treatment. The study was accepted by the local ethics committee, and prospective written informed consent for study participation was obtained from all subjects after the study purpose was fully explained.

PFT parameters including FEV1, FVC (forced vital capacity, L), FEV1/FVC ratio, FEV1%, maximum expiratory flow at 25% of the FVC (MEF 25) and diffusion capacity of the lung for carbon monoxide (DLCO) were measured using a spirometer (Jaeger/Toennies, Hoechberg, Germany). The subsequent CT scanning and MRI were performed within 24 h.

CT scan protocol and image analysis

Non-enhanced CT scan was acquired using a 256-MDCT (Philips Brilliance 256-slice; Philips Healthcare, Cleveland, OH, US) at the end of inspiratory breath-hold. Breath-hold training was carried out before each exam. The scanning parameters were as follows: thickness 1 mm, collimation 128 × 0.625 mm, rotation time 0.5 s, pitch 0.915 and velocity of 146.4 mm/s, high-resolution algorithms 120 kVp/110 mAs.

All images were evaluated on a post-processing workstation (Extended Brilliance Workspace TM, Philips Healthcare, Cleveland, OH, US). The morphological abnormalities, emphysema and bronchial wall thickness, were quantitatively assessed by a thoracic CT radiologist with more than 10-year experience, who was blind to the identities and clinical data of the subjects. Emphysema quantitative analysis was as follows: the total lung volume (TLV) was calculated automatically. Emphysema was defined as area with attenuation of −950 HU or less [13]. Semi-automatic analysis was applied into the total emphysema volume (TEV) evaluation. First, the window width and window level were set as 50 HU and –975 HU, respectively. The whole volume with attenuation of –950 HU or less was calculated automatically. Then, the emphysema volume was acquired by removing the air in the trachea, main bronchi and esophagus manually. The TEV/TLV ratio represented the emphysema index (EI = TEV/TLV). EI = 0, 0 < EI < 15%, EI ≥ 15% was defined as no emphysema, mild emphysema and apparent emphysema, respectively. Bronchial wall thickening was defined as the ratio of any bronchial wall to adjacent pulmonary artery diameter no less than 30% (B/PA ≥ 30%) [14].

According to the dominancy of emphysema and the presence of bronchial wall thickening, COPD was classified into three phenotypes [3]: A phenotype: no or mild emphysema (0 ≤ EI < 15%) with or without bronchial wall thickening; E phenotype: apparent emphysema (EI ≥ 15%) without bronchial wall thickening (B/PA < 30%); M phenotype: apparent emphysema (EI ≥ 15%) with bronchial wall thickening (B/PA ≥ 30%).

MR perfusion imaging and image analysis

MR perfusion imaging was performed on all the patients at the end of inspiratory breath-hold using a clinical 1.5T whole-body MR system (HDMR, GE Healthcare, Milwaukee, WI, USA) with a maximum gradient strength of 40 mT/m and a slew rate of 150 T/m/s. A time-resolution three-dimensional gradient-echo pulse sequence [a GE product LAVA (liver acquisition with volume acceleration)] with parallel acquisition was used for MR perfusion imaging. The following imaging parameters were used: TR/TE: 3.2 ms/1.5 ms; flip angle: 120; field of view: 35 cm × 35 cm; matrix: 256 × 160; bandwidth: 62.5 KHz; number of slices: 28 and zero padded to 56 slices; thickness: 6 mm; NEX, 0.75; ASSET factor: 2; phases: 6; and scan time: 4 s per phase. All injections were performed with an automatic power injector (Medrad, Warrendale, PA, USA). Thirty milliliter gadolinium-diethylene triamine pentaacetic acid was injected, followed by 20 mL saline flush into the antecubital vein at a rate of 3 mL/s [10].

All the MR raw data sets were sent to advantage workstation 4.3 (GE Healthcare) and all the post-processed images were interpreted by a thoracic MR radiologist with more than 10-year experience. Mask images were subtracted from each phase for a pure perfusion image. The homogeneity of perfusion images was visually assessed. In case of perfusion defects (PDs), the shape and size of these regions were evaluated according to the following standards. The shape was categorized into three types: wedge-shaped, circumscribed but not wedge-shaped and patchy. The size was also classified into three types: small, middle and large according to its maximum area on the coronal slice: small-sized ≤ 300 mm2, 300 mm2 < middle-sized ≤ 1000 mm2 and large-sized > 1000 mm2. In addition, the signal intensity of PD (SIPD) was measured. The contrast between normal lung and PD was quantitated by calculating their signal intensity ratio (RSI = SIPD/SInormal).

Statistical analysis

The statistical analysis of all data sets was performed with SPSS 18.0 software (SPSS Inc, Chicago, IL, USA). Comparisons between three phenotypes in PFT parameters, CT volume index and MR perfusion parameters were performed. In terms of normally distributed data, analysis of variance and Bonferroni method were used. Otherwise, statistics were analyzed using Kruskal–Wallis method and Mann–Whitney U-test. Spearman correlation analysis was used to evaluate the correlation between CT volume parameters and pulmonary diffusion function in the whole COPD patients. Spearman correlation analysis was also performed between PFT, CT volumetric parameters and MR perfusion parameters in different HRCT phenotype COPD patients.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The 62 COPD patients represented 42 cases of A phenotype, 9 cases of E phenotype and 11 cases of M phenotype. The symptoms of A and M phenotypes were relatively heavier than that of E phenotype. A phenotype mainly complained of little sputum and wheezing, and M phenotype complained of large sputum, wheezing and productive cough. The mean BMI of A phenotype patients was significantly higher than that of E phenotype patients (23.6 ± 3.5 kg/m2 vs 20.4 ± 3.3 kg/m2; P = 0.026). The mean BMI of M phenotype patients was 21.1 ± 2.8 kg/m2, which was not significantly different from that of A phenotype or E phenotype patients (Fig. 1). In terms of BMI less than 20 kg/m2, the proportion in A, E and M phenotypes was 19%, 44% and 45%, respectively. Smoking index was not significantly different between the three phenotypes (P > 0.05).

figure

Figure 1. Box plots of body mass index (BMI) (A) and forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) (B) in patients with different high-resolution computed tomography (HRCT) phenotypes.

Download figure to PowerPoint

Comparison of PFT between different COPD phenotypes

The PFT parameters are listed in Table 1. There was statistically significant difference in FEV1/FVC between A phenotype and M phenotype (P = 0.003) (Fig. 1). Significant difference was found in DLCO/VA between A phenotype and E phenotype (P = 0.004). DLCO single breath-hold and DLCO SB% of A phenotype was significant higher than that of M phenotype (P < 0.05). There was also significant difference between A and E phenotype in DLCO SB% (P = 0.001). No difference in FEV1, FEV1%, FVC and MEF 25 was observed between the three phenotypes.

Table 1. Clinical characteristic and pulmonary function test of three phenotypes of COPD patients
 A phenotypeE phenotypeM phenotype
  1. *Data are expressed in terms of median and interquartile range (in parentheses). Other values are given as mean ± SD.

  2. †,‡Significant statistical difference between two phenotypes (P < 0.05).

  3. COPD, chronic obstructive pulmonary disease; BMI, body mass index; PY, pack year; L, liter; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; FEV1/FVC, ratio of FEV1 to FVC; MEF 25, maximum expiratory flow 25% of FVC; DLCO SB, diffusion capacity of the lung for carbon monoxide within single breath-hold; DLCO/VA, diffusion constant.

Number42911
BMI (kg/m2)23.6 ± 3.520.4 ± 3.321.1 ± 2.8
Smoking index (PY)*30 (36.5)40 (24.5)27.5 (22.1)
FEV1/FVC (%)64.5 ± 13.260.5 ± 13.951.9 ± 11.6
FEV1 (L)1.6 ± 0.81.4 ± 0.91.4 ± 0.9
FEV1% (%)59.3 ± 21.746.2 ± 24.447.5 ± 20.2
FVC (L)2.5 ± 0.92.4 ± 1.12.6 ± 1.1
MEF 25 (L/s)0.45 ± 0.270.44 ± 0.350.35 ± 0.26
DLCO SB6.1 ± 1.544.1 ± 1.514.2 ± 1.52
DLCO SB% (%)75.9 ± 14.5†‡50.3 ± 22.049.7 ± 12.3
DLCO/VA1.2 ± 0.350.65 ± 0.390.93 ± 0.23

Comparison of CT volume analysis between different COPD phenotypes

The CT volumetric parameters are listed in Table 2. The TEV and EI of A phenotype (0.4 ± 0.4 L, 8.0% ± 4.3%) were significant lower than those of E phenotype (1.0 ± 0.3 L, 18.6% ± 3.2%; P = 0.001, P = 0.001) or M phenotype (0.9 ± 0.2 L, 17.5% ± 1.7%; P = 0.001, P = 0.002). No difference in TEV or EI was found between E phenotype and M phenotype (P > 0.05). TLV in M phenotype was higher than that in A phenotype (5.3 ± 1.9 L vs 4.3 ± 1.3 L; P = 0.027), but was not different from that in E phenotype (5.1 ± 1.5 L, P > 0.05). A phenotype was not different from E phenotype in TLV (P > 0.05).

Table 2. CT volume parameters of three phenotypes of COPD patients
 A phenotypeE phenotypeM phenotype
  1. *,†,‡,§,¶Significant statistical difference between two phenotypes: *t = −2.283, P = 0.027; †t = −3.611, P = 0.001; ‡t = −3.369, P = 0.001; §t = −3.63, P = 0.001; ¶t = −3.273, P = 0.002.

  2. CT, computed tomography; TLV, total lung volume; TEV, total emphysema volume; EI, emphysema index; L, liter.

TLV (L)4.3 ± 1.3*5.1 ± 1.55.3 ± 1.9*
TEV (L)0.4 ± 0.4†‡1.0 ± 0.30.9 ± 0.2
EI (%)8.0 ± 4.3§¶18.6 ± 3.2§17.5 ± 1.7

Comparison of MR perfusion imaging between different COPD phenotypes

The MR perfusion images of all COPD patients appeared as circumscribed but not wedge-shaped or diffuse patchy PDs (Figs 2-4). No wedge-shaped PDs occurred in any patient. Each phenotype had its relatively specific features: A phenotype mainly showed small-sized PDs, E phenotype presented large-sized PDs, and M phenotype manifested as middle-sized PDs. The MR perfusion parameters are listed in Table 3. RSI of A phenotype was significantly higher than that of E phenotype (20.3% ± 8.5% vs 11.8% ± 5.4%; P = 0.006). No difference was observed in SIPD between different COPD phenotypes in our series (P > 0.05).

figure

Figure 2. A 58-year-old man with Global Initiative for Obstructive Lung Disease (GOLD) IV chronic obstructive pulmonary disease (COPD) [forced expiratory volume in 1 s (FEV1)% = 29.5%]. (A) No emphysema occurs on the high-resolution computed tomography (HRCT) image (A phenotype). (B) Magnetic resonance (MR) perfusion imaging shows heterogeneous perfusion, small circumscribed but not wedge-shaped perfusion defects occur in the both lower lobe.

Download figure to PowerPoint

figure

Figure 3. A 57-year-old woman with Global Initiative for Obstructive Lung Disease (GOLD) II chronic obstructive pulmonary disease (COPD) [forced expiratory volume in 1 s (FEV1)% = 54%]. (A) Apparent emphysema occurs on the high-resolution computed tomography (HRCT) image (E phenotype). (B) Magnetic resonance (MR) perfusion imaging shows large circumscribed but not wedge-shaped perfusion defects in the right upper lobe.

Download figure to PowerPoint

figure

Figure 4. A 58-year-old man with Global Initiative for Obstructive Lung Disease (GOLD) IV chronic obstructive pulmonary disease (COPD) [forced expiratory volume in 1 s (FEV1)% = 12.5%]. (A) Apparent emphysema with bronchial wall thickening occurs on the high-resolution computed tomography (HRCT) image (M phenotype). (B) Magnetic resonance (MR) perfusion imaging shows diffuse patchy perfusion defects in the both lungs.

Download figure to PowerPoint

Table 3. MRI perfusion values of three phenotypes of COPD
 A phenotypeE phenotypeM phenotype
  1. Values are means ± SD.

  2. *Significant difference between two phenotypes (t = 2.855, P = 0.006).

  3. COPD, chronic obstructive pulmonary disease; MRI, magnetic resonance imaging; RSI, the signal intensity ratio of the perfusion defects and normal lung perfusion; SIPD, signal intensity of perfusion defects.

RSI (%)20.3 ± 8.5*11.8 ± 5.4*15.5 ± 5.5
SIPD13.1 ± 9.98.3 ± 5.09.2 ± 4.8

Correlation between pulmonary diffusion function and CT volume parameters

Table 4 showed that TEV and EI were moderate negatively correlated with pulmonary diffusion function parameters (DLCO SB, DLCO SB% or DLCO/VA) (r range −0.451 to −0.705, P < 0.05). DLCO SB% was also moderate negatively correlated with TLV (r = −0.507, P < 0.01).

Table 4. Correlation between CT volumetric parameters and pulmonary diffusion function in patients with COPD
CT volumetric parametersDLCO SBDLCO SB%DLCO/VA
  1. Data were expressed in correlation coefficient r (statistical significance, P), r (P).

  2. **Correlation is significant at the 0.01 level (two-tailed); *Correlation is significant at the 0.05 level (two-tailed).

  3. COPD, chronic obstructive pulmonary disease; CT, computed tomography; TLV, total lung volume; TEV, total emphysema volume; EI, emphysema index; DLCO SB, diffusion capacity of the lung for carbon monoxide within single breath-hold; DLCO/VA, diffusion constant.

TLV−0.216 (0.270)−0.507** (0.007)−0.290 (0.127)
TEV−0.502** (0.007)−0.670** (0.000)−0.451* (0.014)
EI−0.579** (0.001)−0.705** (0.000)−0.485** (0.008)

Correlation between PFT, CT volumetric parameters and MR perfusion parameters in different HRCT phenotype COPD patients

For A phenotype, moderate negative correlations were found between PFT parameters (FEV1/FVC, FEV1, FEV1%) and TEV or EI (r range −0.341 to −0.497, P < 0.05); weak to moderate positive correlations were found between all the PFT parameters (FEV1/FVC, FEV1, FEV1%, FVC and MEF 25) and MR perfusion parameters (SIPD and RSI) (r range 0.354–0.589, P < 0.05). The correlations between FEV1% and RSI (r = 0.589, P < 0.001), between FEV1 and RSI (r = 0.586, P < 0.001) were stronger (Table 5).

Table 5. Correlations between PFT, CT volumetric parameters and MR perfusion parameters in different HRCT phenotypej COPD patients
PFTHRCT phenotypeTLVTEVEISIPDRSI
  1. Data were expressed in correlation coefficient r (statistical significance, P), r (P).

  2. **Correlation is significant at the 0.01 level (two-tailed); *Correlation is significant at the 0.05 level (two-tailed).

  3. COPD, chronic obstructive pulmonary disease; CT, computed tomography; TLV, total lung volume; TEV, total emphysema volume; EI, emphysema index; SIPD, signal intensity of perfusion defects; RSI, the signal intensity ratio of the perfusion defects and normal lung perfusion; FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; FEV1/FVC, ratio of FEV1 to FVC; MEF 25, maximum expiratory flow 25%.

FEV1/FVCA−0.028 (0.862)−0.470** (0.002)−0.497** (0.001)0.439** (0.004)0.581** (0.000)
E−0.753* (0.019)−0.854** (0.003)−0.854** (0.003)0.335 (0.379)0.452 (0.222)
M−0.255 (0.450)−0.518 (0.102)−0.536 (0.089)0.622* (0.031)0.909** (0.000)
FEV1A0.213 (0.182)−0.374* (0.016)−0.447** (0.003)0.456** (0.002)0.586** (0.000)
E−0.636 (0.066)−0.837** (0.005)−0.837** (0.005)0.787* (0.012)0.845** (0.004)
M−0.209 (0.537)−0.427 (0.190)−0.491 (0.125)0.634* (0.027)0.718** (0.009)
FEV1%A0.143 (0.373)−0.341* (0.029)−0.393* (0.011)0.546** (0.000)0.589** (0.000)
E−0.600 (0.088)−0.767* (0.016)−0.767* (0.016)0.767* (0.016)0.783* (0.013)
M−0.345 (0.298)−0.582 (0.06)−0.545 (0.083)0.629* (0.028)0.650* (0.022)
FVCA0.313* (0.046)−0.278 (0.078)−0.366* (0.019)0.396** (0.010)0.531** (0.000)
E−0.450 (0.224)−0.667* (0.050)−0.667* (0.050)0.850** (0.004)0.850** (0.004)
M−0.264 (0.433)−0.536 (0.089)−0.564 (0.071)0.573 (0.051)0.636* (0.026)
MEF 25A0.268 (0.099)−0.220 (0.179)−0.303 (0.061)0.354* (0.025)0.439** (0.005)
E−0.503 (0.204)−0.683 (0.062)−0.683 (0.062)0.491 (0.217)0.587 (0.126)
M−0.273 (0.417)−0.627* (0.039)−0.627* (0.039)0.357 (0.255)0.322 (0.308)

For E phenotype, moderate to strong negative correlation was found between PFT parameters (FEV1/FVC, FEV1, FEV1% and FVC) and TEV or EI (r range −0.667 to −0.854, P < 0.05). Moderate to strong positive correlation was found between PFT parameters (FEV1, FEV1% and FVC) and MR perfusion parameters (SIPD and RSI) (r range 0.767–0.850, P < 0.05). The correlations between FEV1/FVC and TEV or EI (r = –0.854, P < 0.001), between FVC and SIPD or RSI (r = 0.850, P < 0.001) were stronger (Table 5).

For M phenotype, MEF 25 was moderate negatively correlated with TEV or EI (r = −0.627, P < 0.05). FEV1/FVC, FEV1 and FEV1% were moderate to strong positively correlated with SIPD or RSI (r range 0.622–0.909, P < 0.05). FVC was moderate positively correlated with RSI (r = 0.636, P < 0.001). The correlations between FEV1/FVC and RSI (r = 0.909, P < 0.001) was strongest (Table 5).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Airflow limitation in COPD is a complex phenomenon caused by small airway remodeling and emphysema. Small airway remodeling is characterized by epithelial cell hyperplasia, smooth muscle hypertrophy and mucous metaplasia. Increased mucus secretion with resultant luminal occlusion, epithelial layer thickening and increased mucus secretion altering airway surface tension and a resultant predisposition to expiratory collapse lead to airflow obstruction [15, 16]. Emphysema results in airflow limitation because of the decrease in elastic recoil, which causes small airway collapse during expiration. The FEV1/FVC of A phenotype was higher than that of M phenotype (P < 0.05), similar to previous study [3].

CT-based quantification delivers local information, which is essential in the individual planning of surgical, interventional or systemic treatment for COPD [7]. It has been reported that the emphysema volume quantified by CT may predict the severity of COPD. The visual assessment and software-based automatic volume analysis have a close correlation in terms of the emphysema volume measurement [17]. In present study, a semi-automatic method, setting a threshold value of –950 HU, was applied into emphysema volume quantitative assessment. Previous studies [3, 5] have visually evaluated HRCT phenotypes of COPD; quantitative assessment makes the classification more accurate recently. However, there is not a definitive quantitative threshold of apparent emphysema until now. Kitaguchi Y et al. used the ratio of low attenuation area to total lung area of the same slice to estimate the emphysema extent, and the ratio ranging from 0% to 25% of the single slice was classified into mild emphysema [5]. The present study used the volume ratio to classify the emphysema extent, and EI was 15% as the threshold of apparent emphysema. According to Table 2, TEV and EI of A phenotype were significant lower than those of E or M phenotype. This finding is consistent with the classification criteria of COPD according to the CT findings, the structural changes of A phenotype at the site of the small airway and that of E phenotype at the pulmonary parenchyma destruction. Previous studies [5] also showed milder lung hyperinflation occurred in A phenotype than other phenotypes. TLV in M phenotype was higher than that in A phenotype. This finding is also associated with the pathological change of COPD. M phenotype is the combination of small airway remodeling and emphysema, but it is difficult to estimate the relative contribution of them to airflow limitation. The correlation analysis showed that the DLCO was moderate negatively correlated with TEV and EI, which was in accordance with that the measurement of the DLCO allowed the detection of emphysema. Table 1 showed DLCO SB% and DLCO/VA in E phenotype were significant lower than that in A phenotype, which was accordance with the previous study [3].

In COPD patients, reduced ventilation because of airway obstruction and parenchyma destruction would result in hypoxic vasoconstriction and further lead to local pulmonary blood flow (PBF) reduction. The reduction of the pulmonary vascular bed and mechanical vascular compression by the hyperinflated lungs also contributes to the reduced perfusion. Therefore, accurate quantitative measurements of regional perfusion changes are important for improved understanding of lung pathophysiology in COPD. Our previous study has showed that CT and MR were more sensitive to the early abnormal changes of lung than PFT. Moreover, MR perfusion imaging showed a higher positive rate of abnormalities than CT images in the control [18]. Table 5 showed that the correlation coefficients between FEV1 (or FEV1%), and RSI were higher than that between FEV1 (or FEV1%) and CT volumetric parameters in three different phenotypes. The findings indicate that MR perfusion may have great potential to evaluate and monitor the blood flow change of COPD. All the COPD patients showed circumscribed but not wedge-shaped or diffuse patchy PDs. This shape of PDs was different from that of pulmonary embolism (wedge-shaped). Bauer et al. [19] revealed that a patchy PD pattern was most commonly associated with emphysematous or fibrotic changes followed by fluid collections in the interstitial or alveolar space; circumscribed but not wedge-shaped PD was associated with fluid collections in the interstitial, alveolar or pleural space followed by tumors and located bullae; and wedge-shaped PD was most commonly associated with occlusive pulmonary embolism in the supplying artery. Except for the special shape of PD in COPD patients, the size of PD was also somewhat different between the three phenotypes. The size of PDs in A, M and E phenotype was small-, middle- and large-sized, respectively. This may be explained by the pathological difference between different phenotypes. In COPD patients, the quantitative evaluation of three-dimensional perfusion showed diffusely decreased PBF and pulmonary blood volume (PBV), and the changes were heterogeneous [20]. Based on our previous study [10], RSI, instead of PBF and PBV, was used to demonstrate the perfusion abnormalities. Table 3 showed that RSI of A phenotype was higher than that of E phenotype, which indicated that the influence of emphysema on pulmonary perfusion may be more significant than that of small airway remodeling.

There were significant different features and correlation between PFT, CT volume analysis and MR perfusion in different phenotype, indicating each phenotype may have novel imaging method. According to Table 5, TEV, EI, SIPD and RSI were optimized parameters for A phenotype, TEV and EI for E phenotype, and RSI for M phenotype. This finding can guide the clinician to select the proper imaging method for follow up. Weight loss, muscle wasting and tissue depletion are commonly seen in COPD patients [3]. The mean BMI of A phenotype was significantly higher than that of E phenotype, and this finding is similar to previous studies [3, 5]. The mechanism underlying weight loss is not yet completely understood, but an imbalance in ongoing processes of protein degradation and replacement seems to be the reason [21].

There are several limitations in this study. First, the sample size was relatively small, especially for E and M phenotype. Second, CT volume quantitative analysis requires a semi-automatic method using manual extraction of the tracheobronchial tree and esophagus, but they could not be extracted thoroughly in some cases, leading to overestimation of emphysema. Finally, only inspiratory CT was performed without comparison with expiratory CT, which may affect the quantitative accuracy of air trapping.

In conclusion, PFT, CT volume parameters and MR perfusion imaging have some characteristic features in different HRCT phenotypes of COPD. A phenotype has lower TEV and EI, higher perfusion and BMI in comparison with E phenotype. This finding may help gain further insights into pathophysiological changes of each phenotype of COPD.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors would like to thank the Youth Fund of the National Natural Science Foundation of China (81000602), the Natural Science Foundation of Shanghai (10ZR1438900) and the National Natural Science Foundation of China (81171333 and 30970800) for the financial support.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease (Updated 2013). Available at: http://www.goldcopd.org/guidelines-global-strategy-for-diagnosis-management.html (accessed January 2013).
  • 2
    Matsuoka S, Yamashiro T, Washko GR, Kurihara Y, Nakajima Y, Hatabu H. Quantitative CT assessment of chronic obstructive pulmonary disease. Radiographics. 2010;30: 5566.
  • 3
    Fujimoto K, Kitaguchi Y, Kubo K, Honda T. Clinical analysis of chronic obstructive pulmonary disease phenotypes classified using high-resolution computed tomography. Respirology. 2006;11: 731740.
  • 4
    Makita H, Nasuhara Y, Nagai K, Ito Y, Hasegawa M, Betsuyaku T, Onodera Y, Hizawa N, Nishimura M. Characterization of phenotypes based on severity of emphysema in chronic obstructive pulmonary disease. Thorax. 2007;62: 932937.
  • 5
    Kitaguchi Y, Fujimoto K, Kubo K, Honda T. Characteristics of COPD phenotypes classified according to the findings of HRCT. Respir Med. 2006;100(10): 17421752.
  • 6
    Xu Y, Sonka M, McLennan G, Guo J, Hoffman EA. MDCT-based 3-D texture classification of emphysema and early smoking related lung pathologies. IEEE Trans Med Imaging. 2006;25: 464475.
  • 7
    Heussel CP, Herth FJ, Kappes J, Hantusch R, Hartlieb S, Weinheimer O, Kauczor HU, Eberhardt R. Fully automatic quantitative assessment of emphysema in computed tomography: comparison with pulmonary function testing and normal values. Eur Radiol. 2009;19: 23912402.
  • 8
    Ley-Zaporozhan J, Ley S, Weinheimer O, Iliyushenko S, Erdugan S, Eberhardt R, Fuxa A, Mews J, Kauczor HU. Quantitative analysis of emphysema in 3D using MDCT: influence of different reconstruction algorithms. Eur J Radiol. 2008;65: 228234.
  • 9
    Daimon T, Fujimoto K, Tanaka K, et al. Volume of pulmonary lobes and segments in chronic obstructive pulmonary diseases calculated using newly developed three-dimensional software. Jpn J Radiol. 2009;27: 115122.
  • 10
    Fan L, Liu SY, Sun F, Xiao XS. Assessment of pulmonary parenchyma perfusion with FAIR in comparison with DCE-MRI-Initial results. Eur J Radiol. 2009;70(1): 4148.
  • 11
    Ley-Zaporozhan J, Ley S, Eberhardt R, Weinheimer O, Fink C, Puderbach M, Eichinger M, Herth F, Kauczor HU. Assessment of the relationship between lung parenchymal destruction and impaired pulmonary perfusion on a lobar level in patients with emphysema. Eur J Radiol. 2007;63: 7683.
  • 12
    Pauls S, Gulkin D, Feuerlein S, Muche R, Kruger S, Schmidt SA, Dharaiya E, Brambs HJ, Hetzel M. Assessment of COPD severity by computed tomography: correlation with lung functional testing. Clin Imaging. 2010;34: 172178.
  • 13
    Matsuoka S, Kurihara Y, Yagihashi K, Nakajima Y. Quantitative assessment of peripheral airway obstruction on paired expiratory/inspiratory thin section computed tomography in chronic obstructive pulmonary disease with emphysema. J Comput Assist Tomogr. 2007;31: 384389.
  • 14
    Roberts HR, Wells AU, Milne DG, Rubens MB, Kolbe J, Cole PJ, Hansell DM. Airflow obstruction in bronchiectasis correlation between computed tomography features and pulmonary function tests. Thorax. 2000;55: 198204.
  • 15
    James AL, Wenzel S. Clinical relevance of airway remodeling in airway diseases. Eur Respir J. 2007;30: 134155.
  • 16
    Hogg JC, Chu F, Utokaparch S, et al. The nature of small-airway obstruction in chronic obstructive pulmonary disease. N Engl J Med. 2004;350: 26452653.
  • 17
    Desai SR, Hansell DM, Walker A, MacDonald SL, Chabat F, Wells AU. Quantification of emphysema: a composite physiologic index derived from CT estimation of disease extent. Eur Radiol. 2007;17: 911918.
  • 18
    Fan L, Xia Y, Guan Y, Yu H, Zhang TF, Liu SY, Li B. Capability of differentiating smokers with normal pulmonary function from COPD patients: a comparison of CT pulmonary volume analysis and MR perfusion imaging. Eur Radiol. 2013;23(5): 12341241.
  • 19
    Bauer RW, Kerl JM, Weber E, Weisser P, Korkusuz H, Lehnert T, Jacobi V, Vogl TJ. Lung perfusion analysis with dual energy CT in patients with suspected pulmonary embolism – influence of window settings on the diagnosis of underlying pathologies of perfusion defects. Eur J Radiol. 2011;80(3): e476482.
  • 20
    Ohno Y, Hatabu H, Murase K, Higashino T, Kawamitsu H, Watanabe H, Takenaka D, Fujii M, Sugimura K. Quantitative assessment of regional pulmonary perfusion in the entire lung using three dimensional ultra fast dynamic contrast-enhanced magnetic resonance imaging: preliminary experience in 40 subjects. J Magn Reson Imaging. 2004;20: 353365.
  • 21
    Wouters EFM, Creutzberg EC, Schols AMWJ. Systemic effects in COPD. Chest. 2002;121: 127130.