Published online Mar 07, 2014.
https://doi.org/10.3348/kjr.2014.15.2.286
Assessment of Perfusion Pattern and Extent of Perfusion Defect on Dual-Energy CT Angiography: Correlations between the Causes of Pulmonary Hypertension and Vascular Parameters
Abstract
Objective
To assess perfusion patterns on a dual-energy pulmonary CT angiography (DECTA) of pulmonary hypertension (PHT) with variable causes and to assess whether the extent of perfusion defect can be used in the severity assessment of PHT.
Materials and Methods
Between March 2007 and February 2011, DECTA scans of 62 consecutive patients (24 men, 38 women; mean age, 58.5 ± 17.3 [standard deviation] years; range, 19-87 years) with PHT were retrospectively included with following inclusion criteria; 1) absence of acute pulmonary thromboembolism, 2) maximal velocity of tricuspid regurgitation jet (TR Vmax) above 3 m/s on echocardiography performed within one week of the DECTA study. Perfusion patterns of iodine map were divided into normal (NL), diffuse heterogeneously decreased (DH), multifocal geographic and multiple peripheral wedging patterns. The extent of perfusion defects (PD), the diameter of main pulmonary artery (MPA) and the ratio of ascending aorta diameter/MPA (aortopulmonary ratio, APR) were measured. Pearson correlation analysis was performed between TR Vmax on echocardiography and CT imaging parameters.
Results
Common perfusion patterns of primary PHT were DH (n = 15) and NL (n = 12). The perfusion patterns of secondary PHT were variable. On the correlation analysis, in primary PHT, TR Vmax significantly correlated with PD, MPA and APR (r = 0.52, r = 0.40, r = -0.50, respectively, all p < 0.05). In secondary PHT, TR Vmax significantly correlated with PD and MPA (r = 0.38, r = 0.53, respectively, all p < 0.05).
Conclusion
Different perfusion patterns are observed on DECTA of PHT according to the causes. PD and MPA are significantly correlated with the TR Vmax.
INTRODUCTION
Pulmonary hypertension (PHT) is one of the most serious chronic disorders of the pulmonary circulation; it can cause right ventricular failure and even death. Because most patients tend to initially present with non-specific symptoms, they are usually diagnosed with PHT at a late stage, thus resulting in high mortality and morbidity. Therefore, it is important to discover potentially curable forms of PHT as early as possible. There have been several studies reported analyzing the perfusion patterns and determining the causes of PHT using pulmonary perfusion scintigraphy (1, 2). Abnormal radiotracer distribution in patients with PHT may be caused by some different mechanisms. One mechanism is arterial thromboembolic occlusion such as in chronic thromboembolic pulmonary hypertension (CTEPH). Another mechanism is parenchymal destruction such as in interstitial lung disease and emphysema or by distal arteriopathy such as in idiopathic PHT and other nonembolic forms (3, 4). A different imaging pattern is resulting from these mechanisms. However, the perfusion patterns of perfusion scans were only useful for the distinction between embolic and nonembolic causes and were not helpful for the distinction of histologic subtypes, primarily because of its limited spatial resolution. Although right-heart catheterization remains the gold standard for correct diagnosing of PHT (5), a superb noninvasive way to check the suspected PHT patients is the transthoracic Doppler echocardiography. Maximal velocity of tricuspid regurgitation jet (TR Vmax) has linear positive and negative correlations with systolic pulmonary arterial pressure and mean pulmonary arterial pressure measured by right-heart catheterization (6-10).
High-resolution computed tomography (CT) and CT angiography play a crucial role in the diagnostic work-up of PHT (11) as they show higher a sensitivity and specificity than lung perfusion scintigraphy for the evaluation of patients with PHT caused by various etiologies, particularly in patients with CTEPH (3, 12). Many attempts have been made to establish reliable and reproducible evaluation methods for the assessment of PHT severity using CT. Earlier results have shown that an increase in the diameter of the main pulmonary artery (PA) is a reliable indicator of PHT (13). The ratio of diameter of PA to the diameter of the ascending aorta (aortopulmonary ratio, APR) and the diameter of the segmental PA compared to that of the homologous segmental bronchus have also been investigated in order to predict the severity of PHT (14, 15). However, measuring the main PA diameter or the APR in order to predict the severity of PHT remains controversial. Recently, assessments of the lung-perfused blood volume (LPBV) or pulmonary perfusion have been made using contrast-enhanced dual-energy CT (DECT) in patients with acute pulmonary thromboembolism (PTE) (16-20). Although there were several studies about the evaluation of the diagnostic utility of LPBV for CTEPH (21-23), no studies were conducted yet to evaluate PHT of various causes other than CTEPH. The purpose of the present study was to assess the perfusion pattern seen on dual-energy pulmonary CT angiography (DECTA) of PHT with various causes and to investigate whether the extent of the perfusion defect can be used in the severity assessment of PHT.
MATERIALS AND METHODS
Patients
This retrospective study was approved by our Institutional Review Board and an informed consent was waived.
Between March 2007 and February 2011, 120 patients underwent DECTA with the clinical suspicion of PTE. The patients were included, whose TR Vmax was above 3 m/s on echocardiography. The echocardiography was performed within one week after the DECTA study. Of these patients, 58 patients with clinical diagnosis of acute PTE were excluded. Therefore, 62 patients (24 men, 38 women; mean age, 58.5 ± 17.3 [standard deviation, SD] years; range 19-87 years) comprised our study population. Right-heart catheterization was performed in 8 patients of total 62 patients.
Scan and Contrast Injection Protocol
CT scans were performed using a DECT protocol recommended by the manufacturer of the two types of dual-source scanners (Somatom Definition [n = 56] and Somatome Definition Flash [n = 6] scanner, Siemens Healthcare, Forchheim, Germany). For example, the protocol for the Somatom Definition was as follows: 512 × 512 matrix; 14 × 1.2 mm collimation; 50 effective mAs at 140 kV and 210 effective mAs at 80 kV; 0.5 pitch and 0.33-second rotation time. Images were acquired in a single breath-hold in the craniocaudal direction from the lung apex to the costophrenic angles. A power injector was used to administer 100 mL of iodine contrast material (iopromide, Ultravist 370, Bayer Schering Pharma, Berlin, Germany) at a rate of 3.5 mL/s. A fixed scan delay of 20 seconds was used.
Perfusion Image Generation
Data from the 80 kV, 140 kV and weighted average images were transferred to a workstation (MultiModality Workplace, Siemens Healthcare, Forchheim, Germany). The weighted average images were automatically generated from a combination of the 140 and 80 kV data with a weighting factor of 1:4 (140:80 kV). These weighted average images were used as conventional pulmonary CT angiographic images. A color-coded iodine image based on the material decomposition theory was obtained using the LPBV application of the workstation software (Syngo Dual Energy). The application class was designed for iodine extraction and the material parameters for iodine extraction were as follows: -1000 Hounsfield units (HU) for air at 80 kV; -1000 HU for air at 140 kV; 60 HU for soft tissue at 80 kV; 54 HU for soft tissue at 140 kV, 2 for relative contrast enhancement; -960 HU for minimum value; -300 HU for maximum value and 4 for range. A color-coded image with a rainbow color setting was used as perfusion image where the red color indicates the highest iodine enhancement and the violet color the lowest enhancement. The window settings of the perfusion image were fixed to a level of 50 HU and a width of 100 HU. Only data displayed inside the dual-energy field of view (FOV) were used for the dual-energy evaluation.
Image Analysis
Based on the clinical features and image findings, the patients were classified as primary and secondary PHT. The secondary PHT was further subdivided into four groups according to the cause of PHT, i.e., cardiogenic, vascular, chronic obstructive pulmonary disease (COPD) and interstitial lung disease. Vascular causes included chronic thromboembolism, vascular occlusion due to fibrosing mediastinitis, and Takayasu's arteritis.
Two reviewers blinded to the conventional pulmonary CTA findings independently assessed the perfusion patterns and visually assessed the extent of perfusion defect (PD) of the iodine map. Normal perfusion was seen as red to yellow color, moderately decreased perfusion as green to blue, and no perfusion as dark blue to black on the color-coded perfusion images.
According to its morphology the perfusion pattern of the iodine map was categorized into four classes: normal (NL), no perfusion defect; diffuse heterogeneously decreased (DH), heterogeneous perfusion defect with intervening areas of preserved perfusion; multifocal geographic (MG), large areas of one color with variably scalloped borders sharply interfaced with another color; and multiple peripheral wedging patterns, a wedge-shaped sharply demarcated perfusion defect in the peripheral area (Fig. 1). Discordant results regarding the perfusion pattern were resolved in consensus during the reading session. The extent of the average perfusion defect was calculated on a 5% scale from three different coronal planes including the ascending aorta, tracheal carina and the vertebra level. The widest diameter perpendicular to the long axis of the main PA (MPA) and APR was measured with computer calipers at the level of the pulmonary arterial bifurcation.
Fig. 1
They are color-coded perfusion images and lung setting images in 4 patients diagnosed with PHT. A, B. 42-year-old woman diagnosed with primary PHT shows normal perfusion pattern with normal lung parenchyma. C, D. 65-year-old woman with primary PHT shows diffuse heterogenously decreased pattern with normal lung parenchyma. PHT = pulmonary hypertension They are color-coded perfusion images and lung setting images in 4 patients diagnosed with PHT. E, F. 87-year-old man with secondary PHT (COPD) shows multifocal geographic pattern with emphysema. G, H. 55-year-old woman with secondary PHT (CTEPH) shows multiple peripheral wedging pattern with mosaic attenuation. COPD = chronic obstructive pulmonary disease, CTEPH = chronic thromboembolic pulmonary hypertension, PHT = pulmonary hypertension
Examples showing four different types of perfusion patterns and parenchymal lung images in various causes of PHT.
Statistical Analysis
Statistical software packages (SPSS version 19, SPSS Inc., Chicago, IL, USA; MedCalc version 7.3, MedCalc Software, Mariakerke, Belgium) were used. The results were expressed as mean ± SD. Agreement between the two readers regarding the perfusion patterns was assessed using kappa statistics. Agreement between the same two readers regarding the extent of the perfusion defect was analyzed using the intraclass correlation coefficient (ICC). Pearson correlation and multiple linear regression analysis were evaluated between TR Vmax on echocardiography and the CT imaging parameters. In addition, Spearman correlation was evaluated between mean pulmonary arterial pressure by right-heart catheterization and PD, MPA and APR. The mean value of both readers' results was used for the correlation analysis. A value of p < 0.05 was considered to indicate statistical significance; all p values were two-tailed.
RESULTS
The inter-reader agreement for determining the perfusion pattern on independent reading was moderate with a kappa value of 0.59. The number of patients diagnosed with primary PHT was 30 and 32 with secondary PHT. The common perfusion patterns of primary PHT were DH (n = 15) and NL (n = 12). The perfusion patterns of secondary PHT varied according to the causes (Table 1). The common perfusion pattern of secondary PHT was MG (n = 14). PD ranged from 35.8 ± 22.4% to 57.9 ± 10.0% in secondary PHT. The ICC regarding the perfusion defect was 0.91. Regarding the correlation analysis of all patients, TR Vmax significantly correlated with PD, MPA and APR (Table 2, Fig. 2). However, PD did not correlate significantly with either MPA or APR. Multiple linear regression analysis revealed that MPA and PD were independent determinants of TR Vmax (R2 = 0.31) (Table 3). Mean pulmonary arterial pressure by right-heart catheterization significantly correlated with PD (r = 0.73, p = 0.04), MPA (r = 0.35, p = 0.40), and APR (r = -0.49, p = 0.22).
Fig. 2
Scatter plots show good correlation between TR Vmax (m/s) and CT parameters, including extent of perfusion defect (PD) (%) (A), diameter of main pulmonary artery (MPA) (mm) (B), and aortopulmonary ratio (APR) (C) in all patients of PHT. PHT = pulmonary hypertension, TR Vmax = maximal velocity of tricuspid regurgitation jet
Correlations between TR Vmax (m/s) and CT parameters.
Table 1
Perfusion Pattern and Extent of Mean Perfusion Defect Analysis on DECT and Mean TR Vmax on Echocardiography According to Cause of PHT
Table 2
Pearson Correlation between TR Vmax and CT Parameters
Table 3
Multiple Linear Regression Analysis between TR Vmax and Determining CT Parameters
DISCUSSION
In this study, a different perfusion patterns were observed according to the various causes of PHT thus suggesting the perfusion pattern offers additional information for the PHT cause identification. In addition, the severity of PHT assessed by measuring the extent of the mean perfusion defect on DECT perfusion imaging showed a significant correlation with the TR Vmax. Hoey et al. (21) showed that DECT can offer a "one-stop" assessment of pulmonary vasculature, parenchyma and perfusion in patients with CTEPH. Nakazawa et al. (23) revealed that LPBV imaging using DECT for CTEPH is possible for the evaluation of pulmonary perfusion and is comparable to pulmonary scintigraphy. However, only patients with CTEPH were included in those studies (21, 23). To the best of our knowledge, the present study is the first study to evaluate the differences in perfusion patterns seen on DECT according to the cause of the PHT.
In our study, most of perfusion defects caused by vascular conditions were shown to be wedge-shaped. This pattern is similar to that observed in patients with PTE (16-18). Analysis of specific defect patterns in patients with nonvascular disease causes is complicated owing to the numerous factors that contribute to PHT with complex pathophysiology and physiologic compensation (24, 25). There are several explanations about the pathophysiologic mechanisms of PHT associated with obstructive and restrictive lung diseases. In emphysema, destruction of alveoli and interstitial spaces leads to a loss of pulmonary vessels, causing a decrease in perfusion (25, 26). Physiologic vasoconstriction due to sustained alveolar hypoxia, followed by vascular remodeling and compression of alveolar vessels by increased intra-alveolar pressures, resulting in increased resistance also show an important impact on PHT in COPD (26, 27). PHT in restrictive lung diseases is reported to occur in as many as 46% of the cases, in idiopathic interstitial pneumonia especially (28, 29). Pulmonary artery dilatation may occur as in patients with primary PHT. However, the severity of main pulmonary arterial dilatation seen on CT is not a reliable indicator of PHT in patients with pulmonary fibrosis (30). PHT in patients with fibrotic lung disease is a reflection of pathologic processes such as small-vessel remodeling as well as a consequence of fibrotic obliteration of the pulmonary vasculature (29, 30). In our study, there were 5 congestive heart failures, 2 dilated cardiomyopathy, 1 Eisenmenger syndrome and 1 valvular heart disease in the cardiogenic group. PHT due to left heart disease is a common consequence of chronic left atrial hypertension in heart failure caused by myocardial and valvular heart disease and is known to be the most common cause of PHT. PHT due to left heart disease promote cascade damaging anatomic and functional changes of pulmonary arterial, capillary and venous circulation, finally accelerating right ventricular (RV) dysfunction and failure (31). Several authors showed for patients with left heart disease that there were perfusion defects in the lower zone, symmetrically on the pulmonary ventilation/perfusion nuclear scans, which were caused by redistribution of pulmonary blood flow (32). PHT due to cardiac causes, i.e., Eisenmenger syndrome, are present in patients with congenital heart diseases such as ventricular septal defect, atrial septal defect and patent ductus arteriosus. The exact pathophysiology for the appearance of PHT in patients with heart defects is not known yet (33).
Chae et al. (18) demonstrated that a combined assessment of perfusion and vascular occlusion may be useful in order to gain a better understanding of the pulmonary embolism (PE) severity. In their study, areas were found in which the perfusion defect and vascular obstruction were mismatched. The investigators explained these mismatches may represent disturbances of the microcirculation or normal perfusion in an area of partial occlusion. Their study showed that DECT perfusion imaging can predict the regional perfusion status as well as patient outcomes. However, Chae et al. (18) focused on acute PE patients and a correlation between the perfusion score and the results of other imaging techniques such as perfusion scintigraphy or echocardiography as a reference was not evaluated. Several previous studies reported with controversial results about the relationship between the dilatation of MPA and the severity of PHT (14, 15). Renard et al. (22) reported that DECT demonstrates correlations between the severity of pulmonary arterial obstruction and perfusion impairment in patients with CTEPH. Also these links are affected by the development degree of the systemic collateral supply.
It is widely accepted that Doppler echocardiography is an essential component of the diagnostic algorithm of PHT as it makes it possible to confirm the diagnosis. In daily practice TR Vmax is a useful indicator to measure the RV systolic pressure, which is the most accurate echocardiographic method and thought to be an equivalent to pulmonary artery systolic pressure in the absence of pulmonary stenosis (10). It can be calculated from the systolic transtricuspid gradient using the Bernoulli equation (34). Thereafter the estimated right atrial pressure will be added (6, 34). Several studies reported that TR Vmax had a strong positive correlation (0.57 to 0.93) with pulmonary artery systolic pressure measured in right-heart catheterization and the Doppler-estimated pulmonary artery systolic pressure had a high sensitivity (0.79-1.00) and specificity (0.6-0.98) to detect PHT (6-10).
Our results demonstrate the close relationship between TR Vmax and the CT parameters (including PD, MPA, and APR). However, the PD and the MPA did not correlated with each other. Both parameters were also found to be independent determinants of TR Vmax on multiple linear regression analysis. If we assume that a perfusion defect reflects microcirculation alteration and that dilatation of the main pulmonary artery indicates structural arterial change, we might consider that these two elements contribute in different ways to the severity of PHT.
Our study had several limitations. First, the study was retrospective with a relatively small sample size. So it was not possible to avoid bias and the general population cannot be reflected. However, we were able to show the significant correlation between TR Vmax and the CT parameter in this preliminary study. Therefore further prospective studies with larger sample sizes will be required. Second, the CT protocol was fixed with a contrast agent of 100 mL and a scan delay of 20 seconds regardless of individual patient characteristics such as weight or Rt. heart function. Third, we used TR Vmax as a reference standard for PHT rather than the direct measurement of pulmonary arterial pressure in right-heart catheterization. However, right-heart catheterization is not routinely performed in clinical practice and the reliable accuracy and feasibility of TR Vmax as a substitute for direct measurement has been reported in many published studies (6-9, 33, 35). Fourth, the technical limitations of DECT were a limitation factor as due to the smaller detector width of the 80 kV system, data displayed inside the limited FOV (260 mm) were used for dual-energy reconstruction only. Nevertheless, the FOV included almost all lungs in most of our patients, and in some large patients the far peripheral portion was excluded from evaluation. The iodine enhancement seen on the perfusion map does not reflect all the diverse aspects of pulmonary perfusion, such as regional flow, transit time and regional blood volume. In addition, the systemic blood supply from the bronchial artery or from other collateral vessels cannot be excluded from the perfusion images. However, this is the innate nature of perfusion CT and our purpose was to evaluate perfusion defects in the clinical setting. Finally, interpretation of a perfusion pattern and its extent may be subjective, although our study indicated reliable agreement between two readers.
In conclusion, different perfusion patterns are noted on DECTA of PHT according to the types and causes. The extent of the perfusion defect and the degree of dilatation of the main PA correlate well with the TR Vmax on the echocardiography. The severity of the PHT is more accurately and completely explained by combining the information regarding the perfusion defect and the vascular diameter change.
This study was supported by a grant of the Korean Health Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (A111599).
References
-
Powe JE, Palevsky HI, McCarthy KE, Alavi A. Pulmonary arterial hypertension: value of perfusion scintigraphy. Radiology 1987;164:727–730.
-
-
McLaughlin VV, Archer SL, Badesch DB, Barst RJ, Farber HW, Lindner JR, et al. ACCF/AHA 2009 expert consensus document on pulmonary hypertension a report of the American College of Cardiology Foundation Task Force on Expert Consensus Documents and the American Heart Association developed in collaboration with the American College of Chest Physicians; American Thoracic Society, Inc.; and the Pulmonary Hypertension Association. J Am Coll Cardiol 2009;53:1573–1619.
-
-
Fink C, Johnson TR, Michaely HJ, Morhard D, Becker C, Reiser M, et al. Dual-energy CT angiography of the lung in patients with suspected pulmonary embolism: initial results. Rofo 2008;180:879–883.
-
-
Chae EJ, Seo JB, Jang YM, Krauss B, Lee CW, Lee HJ, et al. Dual-energy CT for assessment of the severity of acute pulmonary embolism: pulmonary perfusion defect score compared with CT angiographic obstruction score and right ventricular/left ventricular diameter ratio. AJR Am J Roentgenol 2010;194:604–610.
-
-
Nakazawa T, Watanabe Y, Hori Y, Kiso K, Higashi M, Itoh T, et al. Lung perfused blood volume images with dual-energy computed tomography for chronic thromboembolic pulmonary hypertension: correlation to scintigraphy with single-photon emission computed tomography. J Comput Assist Tomogr 2011;35:590–595.
-
-
Pierson DJ. Pathophysiology and clinical effects of chronic hypoxia. Respir Care 2000;45:39–51.discussion 51-53.
-
-
Galiè N, Torbicki A, Barst R, Dartevelle P, Haworth S, Higenbottam T, et al. The Task Force on Diagnosis and Treatment of Pulmonary Arterial Hypertension of the European Society of Cardiology. Guidelines on diagnosis and treatment of pulmonary arterial hypertension. Eur Heart J 2004;25:2243–2278.
-
MeSH Terms
Figures
1 / 2
Tables
1 / 3
