Computed tomographic pulmonary angiography in the assessment of severity of chronic thromboembolic pulmonary hypertension and right ventricular dysfunction

European Journal of Radiology 80 (2011) e462–e469

Contents lists available at ScienceDirect

European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad

Computed tomographic pulmonary angiography in the assessment of severity of chronic thromboembolic pulmonary hypertension and right ventricular dysfunction Min Liu a,1 , Zhanhong Ma a,∗ , Xiaojuan Guo a,1 , Hongxia Zhang a,1 , Yuanhua Yang b,2 , Chen Wang b,3 a b

Department of Radiology, Chao Yang Hospital, Capital Medical University, No. 8, Gong Ti Nan Road, Beijing 100020, China Respiratory Diseases Research Institute, Chao Yang Hospital, Capital Medical University, Beijing 100020, China

a r t i c l e

i n f o

Article history: Received 20 April 2010 Received in revised form 23 August 2010 Accepted 26 August 2010 Keywords: Chronic thromboembolic pulmonary hypertension Computer tomographic pulmonary angiography Digital subtraction pulmonary angiography Right-heart catheterization

a b s t r a c t Purpose: The aim was to investigate the role of computed tomographic pulmonary angiography (CTPA) in the assessment of severity and right ventricular function in chronic thromboembolic pulmonary hypertension (CTEPH). Materials and methods: Clinical and radiological data of 56 patients with CTEPH January 2006–October 2009 were retrospectively reviewed in the present study. All patients received CTPA with a 64row CT using the retrospective ECG-Gated mode before digital subtraction pulmonary angiography and right-heart catheterization. CTPA findings including Right Ventricular diameter (RVd) and left ventricular diameter (LVd) were measured at the end diastole. CT Pulmonary Artery Obstruction Indexes including Qanadli Index and Mastora Index were used in the assessment of severity of pulmonary arterial obstruction. Hemodynamic parameters and pulmonary hypertension classification were evaluated by right-heart catheterization in all patients. Right ventricular function was measured with echocardiography in 49 patients. Results: Qanadli Index and Mastora Index respectively were (37.93 ± 14.74)% and (30.92 ± 16.91)%, which showed a significant difference (Z = −5.983, P = 0.000) and a good correlation (r = 0.881, P = 0.000). Neither Qanadli nor Mastora Index correlated with pulmonary hypertension classification (r = −0.009, P = 0.920) or New York Heart Association heart function classification (r = −0.031, P = 0.756). Neither Qanadli nor Mastora Index correlated with any echocardiographic right ventricular parameters (P > 0.05), while RVd/LVd by CTPA correlated with echocardiographic right ventricular functional parameters (P < 0.05). Both Qanadli (r = −0.288, P = 0.006) and Mastora Index (r = −0.203, P = 0.032) demonstrated a weakly negative correlation with SPO2 . CTPA findings correlated with hemodynamic variables. Backward linear regression analysis revealed that the RVd/LVd, Right Ventricular Anterior Wall Thickness (RVAWT), Main Pulmonary Artery trunk diameter (MPAd) were shown to be independently associated with mean Pulmonary Artery Pressure (mPAP) levels (model: r2 = 0.351, P = 0.025; RVd/LVd: beta = 11.812, P = 0.000; RVAWT: beta = 2.426, P = 0.000; MPAd: beta = 0.677, P = 0.003). Conclusion: Computed tomographic pulmonary angiography is a valuable tool to evaluate hemodynamics, right ventricular function of CTEPH, but neither Qanadli Index nor Mastora Index can reflect pulmonary arterial obstruction in CTEPH accurately. Crown Copyright © 2010 Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction

∗ Corresponding author. Tel.: +86 10 85231083; fax: +86 10 65951377. E-mail addresses: [email protected] (M. Liu), [email protected] (Z. Ma), [email protected] (X. Guo), [email protected] (H. Zhang), [email protected] (Y. Yang), [email protected] (C. Wang). 1 Tel.: +86 10 85231095. 2 Tel.: +86 10 85231568. 3 Tel.: +86 10 85231000.

Chronic thromboembolic pulmonary hypertension (CTEPH) is an important complication of venous thromboembolism and is caused by incomplete resolution of pulmonary emboli. If left untreated, prognosis in CTEPH is poor and death in most CTEPH patients is caused by a gradual hemodynamic and right ventricular failure. To date, invasive pulmonary digital subtraction angiography (DSA) is considered to be the ‘gold standard’ for the assessment of CTEPH and surgical resectability [1,2] and right ventricular (RV) function is usually assessed qualitatively by echocardiography [3]. Computer tomographic pulmonary angiography (CTPA) of

0720-048X/$ – see front matter. Crown Copyright © 2010 Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2010.08.035

M. Liu et al. / European Journal of Radiology 80 (2011) e462–e469

the pulmonary arteries is less invasive and can display a variety of abnormalities, but its role in the evaluation of RV function and hemodynamic changes remains incompletely understood [4]. The aim of the present retrospective study was to analyze whether CTPA could be used to evaluate severity of CTEPH. 2. Materials and methods 2.1. Patients Eighty-one consecutive patients with confirmed CTEPH in our Respiratory Diseases Research Institute between January 2006 and October 2009 were included in this retrospective study. The diagnosis of CTEPH was in accordance with the guidelines of the American College of Chest Physicians [5] and the European Society of Cardiology [6]. Each patient was functionally classified according to the modified New York Heart Association (NYHA) classification of the World Health Organization [7]. Patients who underwent DSA and right-heart catheterization and CTPA preoperatively were included. Exclusion criteria included the following: (1) patients with acute pulmonary embolism (APE), (2) patients with incomplete clinical or radiographic information and (3) CTPA with poor image quality which cannot be diagnostically used for non-compliance, seriously breathing artifacts or low pulmonary artery contrasted intensity. Written informed consent was obtained from all patients who participated in the study, which was approved by the ethics committee of Chao Yang Hospital. 2.2. CTPA scan protocol and image analysis CTPA was performed with a 64-row multi-detector CT scanner (Lightspeed VCT, GE Healthcare, USA) using the retrospective ECG-Gated mode. The whole chest was craniocaudally scanned from lung apex to the dome of the diaphragm during a single breath-hold. Scan parameters were as follows: tube voltage of 100–140 kV and current of 300–550 mA modulated by personal BMI dose, collimation of 0.6 mm, table speed of 39.37 mm/s, gantry rotation time of 0.8 s, and reconstruction increment of 1 mm. A soft tissue reconstruction kernel was used. A mechanical injector was used for intravenous bolus injection of Iopromide (Ultravist, 370 mg/ml, Bayer Schering Pharma, Germany) about 80–85 ml at a flow rate of 4.5 ml/s. For optimal intraluminal contrast enhancement, the automatic bolus-tracking technique had the region of

e463

interest positioned at the level of the main pulmonary artery with a threshold of 100 HU pre-defined threshold, and a fixed delay of 5 s was employed for data acquisition. Images were reconstructed from 70% to 85% of RR intervals for making sure the end diastole and then were transferred on AW4.3 work station (GE Healthcare, USA). Images were analyzed in a random order by two experienced thoracic radiologists blinded to the clinical information. Analysis was conducted to assess the image quality, thromboembolic type and burden, hemodynamic change and RV function. Image quality was classified into excellent (clear pulmonary artery without motion artifacts), moderate (minor motion artifacts not hampering the evaluation) and poor (serious motion artifacts or inadequate contrast). CTPA with poor image quality was excluded. The presence of endoluminal clots represented pulmonary embolism. The scoring system was applied to assess totally 31 pulmonary arteries including 5 mediastinal, 6 lobar and 20 segmental arteries [8]. The thromboembolic type was determined by the location of thrombus. (I) Proximal CTEPH only involved main pulmonary artery trunk, right and left pulmonary artery and lobar artery. (II) Distal CTEPH only involved segmental and/or subsegmental artery. (III) Mixed-type CTEPH was defined as thromboemboli involved in both the proximal and peripheral pulmonary arteries. The thromboembolic burden was evaluated with computer tomographic pulmonary artery obstruction index (CTPAOI) using both Mastora Index [8] and Qanadli Index [9]. CTPA variables including Main Pulmonary Artery diameter (MPAd) and Ascending Aortic diameter (AAd, Fig. 1a), Right and Left Pulmonary Artery diameter (RPAd, LPAd, Fig. 1a), Right Atrium transverse diameter (RAd, Fig. 1b), Right and Left Ventricular transverse diameter (RVd, LVd, Fig. 1b) Right Ventricular Anterior Wall Thickness (RVAWT, Fig. 1b), Interventricular Septal Thickness (IVST, Fig. 1b) and Superior Vena Cava maximal diameter (SVCd, Fig. 1b) were assessed on transverse image [10] obtained at mediastinal window settings (level, 50 Hounsfield Units [HU]; width, 350 HU). The ratio of Main Pulmonary Artery diameter and Ascending Aortic diameter (MPAd/AAd) was obtained by computing the ratio between MPAd and AAd in the same transversal level. The ratio of Right and Left Ventricular transverse diameter (RVd/LVd) was obtained by computing the ratio between right and left Ventricular diameter assessed on transverse images at the end diastole. Right Ventricular Outflow Tract diameter (RVOTd, Fig. 1c) and

Fig. 1. Measurement of variables on CTPA: (a) Main Pulmonary Artery diameter (MPAd) and Ascending Aortic diameter (AAd), Right and Left Pulmonary Artery diameter (RPAd, LPAd,) and Superior Vena Cava maximal diameter (SVCd) were assessed on transverse image. (b) Right Atrium transverse diameter (RAd), Right and Left Ventricular transverse diameter (RVd, LVd), Right Ventricular Anterior Wall Thickness (RVAWT), Interventricular Septal Thickness (IVST) were assessed on transverse image at the end diastole. (c) Right Ventricular Outflow Tract diameter (RVOTd) and Right Ventricular Outflow Tract Anterior Wall Thickness (RVOTW) were measured below pulmonary valve on right ventricular outflow tract view obtained by multi-planar reconstruction method.

e464

M. Liu et al. / European Journal of Radiology 80 (2011) e462–e469

Right Ventricular Outflow Tract Anterior Wall Thickness (RVOTW, Fig. 1c) were measured below pulmonary valve on right ventricular outflow tract view obtained by multi-planar reconstruction method.

2.3. Echocardiography A standard echocardiographic examination (GE Vivid Five GE Healthcare, USA) including two-dimensional and M-mode imaging, pulsed and continuous wave Doppler and color Doppler recordings was performed in 3–6 days before pulmonary DSA and right-heart catheterization. Commercially available echocardiographic instruments with 2.5–4.0 MHz transducers were used. The parameters including diameters of main pulmonary artery, diameters of right and left pulmonary artery, transverse diameter of right atrium, transverse diameters of right and left ventricle, right ventricular volume of systolic and diastolic phase, thickness of right ventricular anterior wall and interventricular septum thickness were retrospectively collected in 49 patients.

Fig. 2. Pulmonary thromboendarterectomy specimen.

3. Results 3.1. Patient characteristics

2.4. Digital subtraction pulmonary angiography and right-heart catheterization 56 patients who underwent DSA and right-heart catheterization within 3–5 days after CTPA were included. For DSA (GE LCA PLUS, GE Healthcare, USA), the right and left pulmonary arteries were selectively catheterized by using a 5F pigtail catheter. Contrast bolus consisted in 35 ml Iopromide (Ultravist, 370 mg/ml, Bayer Schering Pharma, Germany) for each of the five series. The flow rate was 15 ml/s. The amount of contrast agent is about 150 ml. Arteriograms were acquired at 1–2 frames per second. Posteroanterior projections of each lung and a right anterior oblique-projection of the right lung and a left anterior oblique-projection of the left lung were obtained. Then, the 8F Swan-Ganz catheter was inserted into the pulmonary artery followed by standard hemodynamic assessment including Pulse Oxygen Saturation (SPO2 ), Central Venous Pressure (CVP), Pulmonary Arterial Systolic Pressure (sPAP), Pulmonary Artery Diastolic Pressure (dPAP), mean Pulmonary Artery Pressure (mPAP), Pulmonary Capillary Wedge Pressure (PCWP). Cardiac Output (CO) was assessed 3–5 times by thermodilution to get mean value and Pulmonary Vascular Resistance (PVR), Cardiac Index (CI), Pulmonary Vascular Resistance Index (PVRI), Systemic Vascular Resistance (SVR), Systemic Vascular Resistance Index (SVRI), Stroke Volume (SV), Stroke Volume Index (SI), Right Ventricular Stroke Work (RVSW) and Right Cardiac Work (RCW) were calculated. All operations were performed by the same surgeon using a technique that has previously been described [11].

2.5. Statistical analysis All data are expressed as means ± standard deviation, unless otherwise specified. All analyses were performed with a statistical package (SPSS 13; SPSS, Inc., Chicago, III). Comparisons of CTPA findings with hemodynamic data were performed by univariate and multivariate analysis. Differences between variables of CTPA and echocardiography were examined by the paired t-test or Mann–Whitney non-parametric test. The Spearman rank correlation test was used to assess correlations between parameters of CTPA and the hemodynamic parameters and was tested for 2-sided significance. Multivariate analysis was performed by multiple linear stepwise regression analysis. In all cases, a P-value <0.05 was considered statistically significant.

Twenty-five patients with CTEPH were excluded for lack of CTPA images (n = 13) or inferior CTPA image quality (n = 4, 3 patient were excluded for serious breathing artifacts and one for inadequate pulmonary artery contrast) or lack of DSA data (n = 8), so 56 patients including 26 females and 30 males with a mean age of (52.04 ± 13.55) years (17–84 years, median 51 years) were finally included in this present study, in whom 11 patients decided and underwent pulmonary thromboendarterectomy showing the whitish-yellow, fibrotic organized thromboembolus with a thin layer of proximal red appositions, which formed a cast of the pulmonary arterial tree (Fig. 2). Clinical characteristics of patients were collected and shown in Table 1. All patients had clear signs of pulmonary hypertension as revealed by the hemodynamic data in Table 2. CTPA showed Table 1 Baseline clinical information of patients with CTEPH. Baseline parameters

Patients

Age (years) (n = 56) Gender (female/male) (n = 56) Body surface area (m2 ) (n = 56) Body mass index (kg m−1 ) (n = 56) NYHA I II III IV Classification of PAH Mild (25–39 mm Hg) Intermediate (40–69 mm Hg) Severe (≥70 mm Hg) Risk factors History of acute pulmonary embolism History of lower extremity deep venous thrombosis Age >50 years’ old Smoking Obesity History of trauma Hypertension History of splenectomy Hyperlipidemia Osteomyelitis Non-O type blood groups Previous operations Hormone treatment

52.04 ± 13.55 26/30 1.80 ± 0.16 24.41 ± 3.40 0 36 12 8 10 34 12 42 26 38 31 4 24 42 3 37 2 30 32 11

NYHA, New York Heart Association classification; PAH, pulmonary artery hypotension.

M. Liu et al. / European Journal of Radiology 80 (2011) e462–e469

e465

Fig. 3. The thromboembolic type and signs of chronic thrombus: (a) chronic thrombus demonstrating the eccentric filling defect which had the obtuse angle with arterial wall, (b) chronic thrombus demonstrating thickened artery wall, (c and d) chronic thrombus demonstrating intraluminal bands and webs, (e) chronic thrombus demonstrating with intimal irregularities, (f) chronic thrombus demonstrating with complete obstruction.

mixed-type pulmonary embolism in 30 patients and distal pulmonary embolism in 26 patients. No pure proximal pulmonary embolism was noted. Signs of chronic thrombus included the eccentric filling defect which had the obtuse angle with arterial wall (Fig. 3a), thickened artery wall (Fig. 3b), intraluminal bands and webs (Fig. 3c and d), intimal irregularities (Fig. 3e) and complete obstruction (Fig. 3f). 855 of total 1736 arteries in 56 patients including 64 mediastinal arteries, 156 lobar arteries and 635 segmental arteries with chronic thrombus were observed by two radiologists. The concordance rate between two radiologists was 95.1% (kappa value = 0.902 P = 0.000). No significant difference in the diagnosis of pulmonary embolism was observed in two radiologists by McNemar Chi-square test (P = 0.128). 3.2. Computer tomographic pulmonary artery obstruction index In 56 patients with CTEPH, the mean Qanadli Index was (37.93 ± 14.74)% (range: 12.5–87.5%, median: 41.25%) and mean

Mastora Index (30.92 ± 16.91)% (range: 5.81–70.3%, median: 26.81%). Wilcoxon signed ranks test indicated the Qanadli Index was higher than Mastora Index (Z = −5.983, P = 0.000). Spearman correlation analysis indicated there was a good correlation between Qanadli Index and Mastora Index was (r = 0.881, P = 0.000). The mean Qanadli Index was (27.66 ± 18.63)% (range: 12.50–87.50%, median: 22.50%) in the 26 patients with distal pulmonary embolism and (42.09 ± 10.38)% (range: 25.00–67.50%, median: 42.50%) in the 30 patients with mixed-type pulmonary embolism. Wilcoxon signed ranks test showed the Qanadli Index in the patients with distal pulmonary embolism was lower than that in the patients with mixed-type pulmonary embolism (Z = −4.784, P = 0.000). The mean Mastora Index was (17.46 ± 10.30)% (range: 5.81–41.49%, median: 12.90%) in the patients with distal pulmonary embolism and (36.37 ± 16.02)% (range: 10.32–70.30%, median: 35.96%) in the patients with mixed-type pulmonary embolism. Wilcoxon signed ranks test indicated the Mastora Index in the patients with peripheral pulmonary embolism was lower than that

e466

M. Liu et al. / European Journal of Radiology 80 (2011) e462–e469 Table 3 CTPAOI with echocardiographic data by Spearman rank correlation (n = 49).

Table 2 Hemodynamic characteristics of patients with CTEPH. Hemodynamic data (n = 56)

Mean ± S.D.

SPO2 (%) CVP (mm Hg) sPAP (mm Hg) dPAP (mm Hg) mPAP (mm Hg) PCWP (mm Hg) CO (L min−1 ) CI (L min−1 m−2 ) PVR (dyn s cm−5 ) SVR (dyn s cm−5 ) SV (ml) RCW (kg m) RVSW (g m) SI (mL m−2 ) SVRI (dyn s m2 cm−5 ) PVRI (dyn s m2 cm−5 ) RCWI (kg m m−2 ) RVSWI (g m m−2 )

91.88 9.96 89.91 33.40 53.70 10.26 4.02 2.23 954.04 1808.23 49.84 2.37 29.40 27.20 3277.32 1737 1.93 22.57

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

3.90 5.59 23.12 9.66 13.69 3.45 1.20 0.66 455.88 591.37 16.07 0.96 11.32 8.63 978.84 793.85 0.49 20.95

Median (min–max)

Echocardiography

92 (77–99) 8 (−4 to 23) 89 (52–137) 32 (20–63) 52 (33–89) 10 (2–20) 3.80 (1.68–6.9) 2.10 (1.02–4.3) 852 (301–2463) 1683 (815–3240) 48.60 (13.80–94.8) 2.28 (0.53–6.12) 29.19 (4.31–56.7) 25.80 (13.8–52.00) 2981 (1435–5510) 1570 (500–4285) 1.25 (0.32–3.21) 15.90 (2.61–175.9)

RVEF (%)

Pulse Oxygen Saturation (SPO2 ), Central Venous Pressure (CVP), Pulmonary Arterial Systolic Pressure (sPAP), Pulmonary Artery Diastolic Pressure (dPAP), mean Pulmonary Artery Pressure (mPAP), Pulmonary Capillary Wedge Pressure (PCWP). Cardiac Output (CO), Cardiac Index (CI), Pulmonary Vascular Resistance (PVR), Systemic Vascular Resistance (SVR), Stroke Volume (SV), Right Cardiac Work (RCW), Right Ventricular Stroke Work (RVSW), Stroke Volume Index (SI), Systemic Vascular Resistance Index (SVRI), Pulmonary Vascular Resistance Index (PVRI), Right Cardiac Work Index (RCWI), Right Ventricular Stroke Work Index (RVSWI).

in the patients with mixed pulmonary embolism (Z = −4.130, P = 0.000). Among 56 patients with CTEPH, 10 had mild pulmonary hypertension with a mean Qanadli Index of (44.16 ± 11.74)% (range: 20.00–52.50%, median: 50.00%), 34 intermediate pulmonary hypertension with a mean Qanadli Index of (35.80 ± 14.96)% (range: 12.50–87.50%, median: 35.00%) and 12 severe pulmonary hypertension with a mean Qanadli Index of (42.22 ± 16.33)% (range: 12.50–67.80%, median 42.50%). Kruskal–Wallis test indicated significant difference in the Qanadli Index between these groups (2 = 8.616, P = 0.013). In addition, Kendall’s tau-b correlation analysis indicated Qanadli Index was not correlated with the grade of pulmonary arterial hypertension (PAH) (r = −0.009, P = 0.920) and no correlation was observed between the Qanadli Index and the grade of NYHA (r = −0.031, P = 0.756). The mean Mastora Index was 44.58% (range: 11.61–52.50%, median: 50%) in the patients with mild pulmonary embolism, (27.18 ± 15.18)% (range: 8.39–61.62%, median: 23.87%) in the patients with intermediate pulmonary embolism and (34.34 ± 14.69)% (range: 5.81–55.32%, median: 37.57%) in the patients with severe pulmonary embolism. Kruskal–Wallis test demonstrated significant difference in the Mastora Index between these groups (2 = 8.682, P = 0.013). Kendall’s tau-b correlation

r P r P r P r P r P r P

RVEDV (ml) RVESV (ml) IVST (mm) RVd (mm) RVd/LVd

Qanadli Index

Mastora Index

0.075 0.45 0.086 0.388 0.104 0.294 −0.194 0.049 −0.093 0.347 0.122 0.215

0.097 0.327 0.110 0.265 0.002 0.985 −0.193 0.049 −0.116 0.242 0.075 0.447

Qanadli Index and Mastora Index were used as CT Pulmonary Artery Obstruction Indexes (CTPAOI); Right Ventricular Eject Fraction (RVEF); Right Ventricular Eject Diastolic Volume (RVEDV); Right Ventricular Eject Systolic Volume (RVESV); Interventricular Septal Thickness (IVST); Right Ventricular transverse diameter (RVd); the ratio of Right and Left Ventricular transverse diameter (RVd/LVd).

analysis indicated Mastora Index was not associated with the classification of PAH (r = −0.028 P = 0.738) and no correlation was noted between the Mastora Index with the classification of NYHA (r = −0.058, P = 0.561). 3.3. CTPA and echocardiographic variables The Spearman correlation of CTPAOI with echocardiographic right ventricular data summarized in Table 3 showed neither Qanadli Index nor Mastora Index correlates with RV parameters. The Spearman correlation of CTPA variables with echocardiographic data are summarized in Table 4. Those CTPA variables were significantly different with those echocardiographic variables (P < 0.05), but positively correlated with them (P < 0.05). The tricuspid valve pressure gradient (TVPG) was (79.52 ± 22.41) mm Hg, which was positively related with RVd/LVd (r = 0.421, P = 0.000; n = 49). The maximum pulmonary valve velocity (Vmax) was (82.65 ± 35.46) cm s−1 and positively related with RVd/LVd (r = 0.391, P = 0.005; n = 49). The right ventricular enddiastolic volume was (82.10 ± 35.86) ml and was positively related with RVd/LVd (r = 0.671, P = 0.000; n = 49). The right ventricular end-systolic volume was (47.95 ± 24.16) ml, which was positively correlated with RVd/LVd (r = 0.557, P = 0.000; n = 49). The right ventricular ejection fraction (RVEF) was (37.89 ± 11.46) ml with a positive correlation with RVd/LVd (r = 0.592, P = 0.000; n = 49). 3.4. CTPA and hemodynamic change The Spearman correlation of CTPAOI with hemodynamic variables is summarized in Table 5. Both Qanadli Index (r = −0.288, P = 0.006) and Mastora Index (r = −0.203, P = 0.032) showed a

Table 4 Correlation of variables between CTPA and echocardiography (n = 56). Variables

CTPA

MPAd (mm) RPAd (mm) LPAd (mm) MPAd/AAda RVAWT (mm) IVST (mm) RAd (mm) RVd (mm) RVd/LVda

38.74 26.33 24.42 1.25 7.28 9.73 70.80 50.85 1.76

Echocardiography ± ± ± ± ± ± ± ± ±

5.10 3.90 1.91 0.57 1.95 1.98 17.03 8.59 0.52

34.88 23.66 22.72 1.16 6.31 10.85 57.01 52.54 2.17

± ± ± ± ± ± ± ± ±

5.17 3.88 1.97 0.18 1.87 2.77 11.30 8.18 0.70

t (P)

r (P)

t = 3.579 (P = 0.003) t = 5.601 (P = 0.000) t = 5.120 (P = 0.000) Z = −2.768 (P = 0.006) t = 3.755 (P = 0.001) t = −2.281 (P = 0.029) t = 3.943 (P = 0.000) t = −4.56 (P = 0.000) Z = −5.882 (P = 0.000)

r = 0.503 (P = 0.005) r = 0.390 (P = 0.041) r = 0.607 (P = 0.002) r = 0.507 (P = 0.003) r = 0.455 (P = 0.007) r = 0.395 (P = 0.021) r = 0.456 (P = 0.007) r = 0.673 (P = 0.000) r = 0.708 (P = 0.000)

a Differences between variables of CTPA and echocardiography were examined by Mann–Whitney non-parametric test. Main Pulmonary Artery diameter (MPAd), Right Pulmonary Artery diameter (RPAd), Left Pulmonary Artery diameter (LPAd), the ratio of Main Pulmonary Artery diameter and Ascending Aortic diameter (MPAd/AAd), Right Ventricular Anterior Wall Thickness (RVAWT), Interventricular Septal Thickness (IVST), Right Atrium transverse diameter (RAd), Right Ventricular transverse diameter (RVd), the ratio of Right and Left Ventricular transverse diameter (RVd/LVd).

SCVd

RVd/LVd

RVd

RAd

ISVT

RVOTd

RVOTW

RVAWT

LPAd

RPAd

MPd/AAd

Significant correlations printed in bold letters.

SV

0.193 0.054 −0.003 0.974 0.239 0.017 0.394 0.000 0.247 0.013 0.323 0.001 0.111 0.273 0.477 0.000 −0.113 0.264 −0.162 0.107 −0.272 0.006 −0.157 0.122 0.022 0.826 −0.293 0.003 0.146 0.143 0.211 0.034 0.044 0.661 0.179 0.071 −0.157 0.114 0.236 0.017 −0.163 0.101 −0.189 0.057 −0.411 0.000 −0.258 0.009

RVSW RCW

0.276 0.005 0.041 0.682 0.366 0.000 0.450 0.000 0.281 0.005 0.266 0.008 0.124 0.221 0.389 0.000 −0.216 0.031 −0.165 0.100 −0.278 0.005 −0.209 0.039

SI

CTPA has progressively been established as a first line test in the APE diagnostic algorithm and CTPA parameters are important for predicting the outcome of patients presenting with APE [12]. Reichelt et al. [13] compared CTPA and DSA in diagnosis of CTEPH, the results show CTPA is an accurate and reliable noninvasive alternative to conventional DSA in the diagnostic workup in patients with CTEPH. Our results indicated the accuracy of CTPA in the diagnosis of CTEPH was also favorable, which was consistent with previously reported by Reichelt et al. [13]. Qanadli Index [8] and Mastora Index [9] were originally designed to quantify arterial obstruction with helical CT in APE,

Table 6 CTPA variables and hemodynamic data.

4. Discussion

−0.029 0.773 0.100 0.319 −0.005 0.958 −0.060 0.548 −0.034 0.737 −0.121 0.224 −0.052 0.605 −0.095 0.343 0.078 0.434 0.018 0.861 0.276 0.005 0.046 0.651

SVRI sPAP

dPAP

mPAP

PCWP

CO

weakly negative correlation with SPO2 , whereas their correlations with other hemodynamic variables were not significant (P > 0.05). The univariate correlations of CTPA with hemodynamic variables are summarized in Table 6. RVd/LVd correlated with all hemodynamic variables except SPO2 . Mean PAP showed a correlation with MPAd, RPAd and LPAd, the ratio of Main Pulmonary Artery diameter and Ascending Aortic diameter (MPAd/AAd), RVAWT, IVST, RVd/LVd, SVCd. By backward linear regression analysis, RVd/LVd/RVAWT, MPAd were shown to be independently associated with mPAP levels (model: r2 = 0.351, P = 0.025; RVd/LVd: beta = 11.812, P = 0.000; RVAWT: beta = 2.426, P = 0.000; MPAd: beta = 0.677, P = 0.003). PVR correlated with MPAd, MPd/AAd, RVOTd, RAd, RVd/LVd,SVCd. By backward linear regression analysis, MPAd, MPd/AAd and RVd/LVd were shown to be independently associated with PVR levels. (model: r2 = 0.397, P = 0.021; MPAd: beta = 31.907, P = 0.001; MPd/AAd: beta = 24.285, P = 0.003; RVd/LVd: beta = 297.318, P = 0.009).

0.052 0.606 0.139 0.164 0.002 0.986 0.017 0.868 0.061 0.546 −0.025 0.806 0.008 0.935 −0.012 0.908 0.126 0.208 0.108 0.281 0.290 0.003 0.091 0.370

SVR CI

PVR

a Qanadli Index and Mastora Index weakly correlated with SPO2 ; Qanadli Index and Mastora Index were used as CT Pulmonary Artery Obstruction Indexes (CTPAOI); Pulse Oxygen Saturation (SPO2 ), Central Venous Pressure (CVP), Pulmonary Arterial Systolic Pressure (sPAP), Pulmonary Artery Diastolic Pressure (dPAP), mean Pulmonary Artery Pressure (mPAP), Pulmonary Capillary Wedge Pressure (PCWP), Cardiac Output (CO), Cardiac Index (CI), Pulmonary Vascular Resistance (PVR), Systemic Vascular Resistance (SVR), Stroke Volume (SV), Right Cardiac Work (RCW), Right Ventricular Stroke Work (RVSW).

0.197 0.031 0.354 0.000 0.112 0.261 0.120 0.231 0.053 0.598 −0.028 0.783 0.281 0.004 0.073 0.467 0.288 0.003 0.104 0.299 0.467 0.000 0.352 0.000

RVSW (g m)

PVRI

RCW (kg m)

0.250 0.011 0.380 0.000 0.123 0.216 0.194 0.051 0.101 0.313 0.047 0.639 0.351 0.000 0.148 0.137 0.299 0.002 0.140 0.159 0.359 0.000 0.351 0.000

SV (ml)

−0.061 0.545 −0.342 0.000 0.151 0.129 0.008 0.935 −0.053 0.596 0.016 0.872 −0.193 0.052 −0.027 0.786 −0.277 0.005 −0.241 0.015 −0.438 0.000 −0.395 0.000

SVR (dyn s cm−5 )

0.027 0.788 −0.261 0.008 0.191 0.055 0.154 0.123 0.052 0.600 0.110 0.273 −0.116 0.245 0.099 0.321 −0.259 0.008 −0.183 0.066 −0.433 0.000 −0.365 0.000

PVR (dyn s cm−5 )

−0.066 0.507 −0.041 0.679 −0.069 0.488 0.114 0.254 0.060 0.551 −0.068 0.496 −0.136 0.172 0.190 0.056 0.131 0.189 0.223 0.024 0.232 0.019 0.195 0.051

CI (L min−1 m−2 )

0.267 0.007 0.228 0.021 0.332 0.001 0.307 0.004 0.259 0.009 0.122 0.222 0.261 0.008 0.320 0.001 0.164 0.099 0.049 0.627 0.220 0.030 0.215 0.031

CO (L min−1 )

0.014 0.889 0.090 0.369 −0.141 0.158 −0.088 0.379 0.044 0.659 −0.148 0.138 0.061 0.542 0.003 0.980 0.451 0.000 0.217 0.028 0.299 0.002 0.490 0.000

PCWP (mm Hg)

CVP

mPAP (mm Hg)

−0.203 0.032a −0.132 0.187 −0.069 0.490 −0.124 0.215 −0.130 0.195 0.038 0.704 0.094 0.348 0.068 0.495 −0.157 0.115 −0.156 0.118 0.067 0.500 −0.043 0.672 −0.032 0.753

−0.396 0.000 −0.003 0.976 −0.423 0.000 −0.342 0.001 0.095 0.372 −0.203 0.055 −0.001 0.992 −0.063 0.558 0.034 0.751 −0.142 0.181 −0.054 0.614 −0.109 0.314

dPAP (mm Hg)

Mastora Index

−0.288 0.006a −0.012 0.907 −0.055 0.586 −0.107 0.283 −0.110 0.272 −0.008 0.939 0.046 0.644 0.017 0.866 −0.108 0.281 −0.131 0.189 −0.022 0.827 −0.068 0.500 −0.130 0.196

SPO2

sPAP (mm Hg)

Qanadli Index

r P r P r P r P r P r P r P r P r P r P r P r P

CVP (mm Hg)

r P r P r P r P r P r P r P r P r P r P r P r P r P

MPd

SPO2 (%)

0.307 0.002 0.267 0.007 0.241 0.015 0.391 0.000 0.273 0.005 −0.024 0.809 0.249 0.012 0.213 0.032 0.208 0.036 0.112 0.261 0.257 0.012 0.286 0.004

CTPAOI

Spearman correlation (n = 56)

Variables

Hemodynamics variables

0.174 0.080 0.227 0.022 0.224 0.023 0.266 0.007 0.240 0.015 0.160 0.108 0.263 0.008 0.352 0.000 0.111 0.269 −0.003 0.977 0.243 0.020 0.112 0.269

Table 5 CTPAOI and hemodynamic variables in CTEPH.

e467

−0.043 0.674 −0.349 0.000 0.117 0.245 0.080 0.429 −0.072 0.478 0.156 0.122 −0.195 0.052 0.090 0.374 −0.197 0.050 −0.248 0.013 −0.452 0.000 −0.321 0.001

M. Liu et al. / European Journal of Radiology 80 (2011) e462–e469

e468

M. Liu et al. / European Journal of Radiology 80 (2011) e462–e469

and indices were shown correlation with right ventricular function [14–17]. Until now, there still a index has not been designed to quantify arterial obstruction with helical CT in chronic PE, so Qanadli and Mastora Index were used to quantify arterial obstruction in chronic PE in the first time. In this study, the Qanadli Index was higher than Mastora Index regardless of type of pulmonary embolism. Furthermore, Qanadli Index was strongly related with Mastora Index. These findings were consistent with results previously reported in APE [18,19]. Difference comes from the calculation methods of two indices. The Qanadli Index emphasizes on the location of embolus and the severity of pulmonary embolism is classified into only two grades. Mastora Index takes the severity of pulmonary embolism (five grades) into account. Because the hemodynamics related with the extent of pulmonary arterial obstruction, it seems more reasonable than that in Qanadli Index. In fact, the pathophysiologic process of CTEPH includes not only pulmonary arterial obstruction, the incomplete resolution and organization of pulmonary thromboemboli, but also the vessel remodeling, release of vasoconstrictive substance, reflex vasoconstriction over time, resulting in a vasculopathy of precapillary vessels similar to the arteriopathy found in patients with idiopathic pulmonary hypertension [20,21]. Although Qanadli Index was strongly related with Mastora Index in CTPEH, Qanadli Index nor Mastora Index did correlated with hemodynamic parameters except a weakly negative correlation with SPO2 from right-heart catheterization. It suggests neither Qanadli Index nor Mastora Index can accurately reflect the severity of pulmonary artery obstruction in CTEPH. As CTEPH progresses, the RV enlarge and its function may become depressed. Qanadli Index nor Mastora Index did correlate with right ventricular function obtained from echocardiography. These findings are not consistent with previously reported in APE [14–17,22–25], suggesting the deteriorated right ventricular function as a result of chronic pulmonary embolism not only depend on the actual severity of pulmonary arterial obstruction but also are associated with the complicated pathophysiology secondary to chronic pulmonary embolism [26]. It suggests neither Qanadli Index nor Mastora Index can accurately reflect the alteration of right ventricular function in CTEPH. With development of multi-detector row spiral CT, CTPA can display all morphological features of right-heart adaptation, so it could provides an objective tool in the evaluation of right cardiac function. At present, CTPA is routinely reformed with helical scanning mode, but this scanning mode produces too more cardiac motion artifacts to clearly observe the cardiac R–R wave interval and measure parameters. The retrospective ECG-Gated scanning mode allows for image reconstruction in any phase of the cardiac cycle. Thus, end-systolic and end-diastolic images can be produced to assess ventricular volumes and function. In our research, with retrospective ECG-Gated scanning mode, the dilated pulmonary artery, enlarged right ventricle, contracted left ventricle and displaced ventricular septum, thickened right ventricular anterior wall, as well as dilated superior vena cava can be clearly observed and at same time the maximal diameter of RV and LV can be measured at the diastolic phase. The diameters of main pulmonary artery, right pulmonary artery and left pulmonary artery obtained from transverse image of CTPA in the ventricular diastolic phase were greater than those from echocardiography. This difference between CTPA and the echocardiographic estimation is attributed to inherent differences of the two imaging modalities and the measured position. Although significant difference exists between CTPA and echocardiography, better correlation has been found between these two methods, suggesting CTPA not only displays the location and burden of embolus but presents RV function as echocardiography [3]. Our results also showed the RVd/LVd was related with TVPG, maximum

pulmonary valve velocity, right ventricular end-diastolic volume and RVEF, which suggested CTPA, might play an important role in the evaluation of pathophysiological changes of RV in patients with CTEPH. Right-heart catheterization has been widely used in the evaluation of hemodynamics of patients with PAH [27]. The parameters of hemodynamics obtained from heart catheterization can precisely assess the severity of PAH playing important roles in guiding clinical treatment and predicting the prognosis [28,29]. To date, heart catheterization still cannot be replaced by any other approach but invasiveness is a major weakness of heart catheterization and skillfulness. Previous studies [30,31] have reported a correlation between the ratio of the main pulmonary artery/ascending aorta diameters and the pressure measurement by right-heart catheterization. This is the first systematic study in CTEPH demonstrated that CTPA parameters correlated with hemodynamic and RV functional parameters, which suggested CTPA could be used to evaluate the hemodynamic change (such as pulmonary artery pressure, pulmonary vascular resistance) and their severity. Based upon our observation, RVd/LVd correlating with all hemodynamic parameters except SPO2 suggested the potential usefulness of RVd/LVd as a non-invasive marker of the hemodynamic severity and RV dysfunction. RVd/LVd, RVAWT and MPAd are the independent factors correlated with mean pulmonary arterial pressure, which was consistent with the study of Collomb et al. [25]. This result implied the detection of RVd/LVd, the thickness of right ventricular anterior wall and the diameter of main pulmonary artery by CTPA make the assessment of pulmonary artery pressure. Clinically, pulmonary vascular resistance has been applied to evaluate the changes in pulmonary vascular histology and to predict the post-operative occurrence and development of pulmonary hypertension crisis [32]. MPAd, MPd/AAd and RVd/LVd were shown to be the independent factors correlated with PVR, which suggested the primary assessment of PVR can be performed using MPAd, MPd/AAd and RVd/LVd. From the results from CTPA, the patient care could efficiently be done. There are some limitations of this study. The main limitation is its retrospective observational design with relative small number, so although CTPA variables correlate with the hemodynamic data, the correlations need to be further verified. Although CTPA appears to be helpful in preoperative evaluating hemodynamic characteristics of patients with CTEPH, it is unclear whether operability and surgical success, defined as mortality and/or improvement of PVR, can be predicted by CTPA parameters with sufficient accuracy because only 11 patients underwent PEA and neither follow-up CTPA nor right-heart catheterization was obtained. Future studies are warranted on the role of CTPA to identify ‘high risk’ CTEPH patients and its relation to post-operative hemodynamic outcome, RV failure and mortality. Moreover, although CTPA with retrospective ECG-Gated scanning mode could more clearly demonstrate interventricular septal thickness and location, and the ventricular morphology than those with helical scanning mode, the higher radiation dose with retrospective ECG-Gated scanning mode than that of helical mode is the other limitation. So the retrospective ECG-Gated scanning mode is currently not recommended as a routine CTPA protocol. CTPA with prospective ECG-Gated mode scanning at diastolic phase is suspected to get clearly ventricular diastolic image with lower radiation dose and needs a further verification. In summary, our results suggest that CTPA could be important to evaluate the hemodynamic and RV overload in patients with CTEPH, which is important for patient care, but neither Qanadli Index nor Mastora Index can reflect pulmonary arterial obstruction in CTEPH accurately. A new index is needed to reflect the hemodynamics such as mPAP, PVR and RV function.

M. Liu et al. / European Journal of Radiology 80 (2011) e462–e469

Conflict of interest None of the authors has any conflict of interest to disclose. All authors have participated sufficiently in any submission to take public responsibility for its content. The article I submitted to the journal for review is original, has been written by the stated authors and has not been published elsewhere. This article is not currently being considering for publication by any other journal and will not be submitted for such view while under review by the journal. The article contains no libellous or other unlawful statement and does not contain any materials that violate any personal or proprietary rights of any other person or entity. Acknowledgements The authors thank Youmin Guo M.D. for helpful comments regarding hemodynamics, Xiao Xia Peng Ph.D. for advice on data management and for valuable suggestions about statistical analysis. This research is supported by the National 11th FiveYear Researching Foundation (No. 2006BAI01A06) and the Youth Researching Foundation of Chao Yang Hospital (No. 2009-22). References [1] Auger WR, Channick RN, Kerr KM, et al. Evaluation of patients with suspected chronic thromboembolic pulmonary hypertension. Semin Thorac Cardiovasc Surg 1999;11(2):179–90. [2] Jamieson SW, Kapelanski DP, Sakakibara N, et al. Pulmonary endarterectomy: experience and lessons learned in 1500 cases. Ann Thorac Surg 2003;76(5):1457–62. [3] Raisinghani Ajit, Ben-Yehuda Ori. Echocardiography in chronic thromboembolic pulmonary hypertension. Semin Thorac Cardiovasc Surg 2006;18(3):230–5. [4] Fedullo PF, Auger WR, Kerr KM, Rubin LJ. Chronic thromboembolic pulmonary hypertension. N Engl J Med 2001;15(345(20)):1465–72. [5] McGoon M, Gutterman D, Steen V, et al. Screening, early detection, and diagnosis of pulmonary arterial hypertension: ACCP evidence-based clinical practice guidelines. Chest 2004;126(1 Suppl.):14S–34S. [6] Galie N, Torbicki A, Barst R, et al. Guidelines on diagnosis and treatment of pulmonary arterial hypertension. The task force on diagnosis and treatment of pulmonary arterial hypertension of the European Society of Cardiology. Eur Heart J 2004;25(24):2243–78. [7] Simonneau G, Galie N, Rubin LJ, et al. Clinical classification of pulmonary hypertension. J Am Coll Cardiol 2004;16(43(12 Suppl. S)):5S–12S. [8] Qanadli SD, El Hajjam M, Vieillard-Baron A, et al. New CT index to quantify arterial obstruction in pulmonary embolism: comparison with angiographic index and echocardiography. AJR Am J Roentgenol 2001;176(6):1415–20. [9] Mastora I, Remy-Jardin M, Masson P, et al. Severity of acute pulmonary embolism: evaluation of a new spiral CT angiographic score in correlation with echocardiographic data. Eur Radiol 2003;13(1):29–35. [10] Coulden Richard. State-of-the-art imaging techniques in chronic thromboembolic pulmonary hypertension. Proc Am Thorac Soc 2006;3(7):577–83. [11] Tscholl D, Langer F, Wendler O, et al. Pulmonary thromboendarterectomy: risk factors for early survival and hemodynamic improvement. Eur J Cardiothorac Surg 2001;19(6):771–6.

e469

[12] Ghuysen A, Ghaye B, Willems V, et al. Computed tomographic pulmonary angiography and prognostic significance in patients with acute pulmonary embolism. Thorax 2005;60(11):956–61. [13] Reichelt A, Hoeper MM, Galanski M, Keberle M. Chronic thromboembolic pulmonary hypertension: evaluation with 64-detector row CT versus digital substraction angiography. Eur J Radiol 2009;71(1):49–54. [14] Wang Jianguo, Zhu Li, Liu Min, et al. Diagnostic value of CT pulmonary angiography for right ventricular dysfunction secondary to acute pulmonary embolism. J Pract Radiol 2008;24(9):1205–9. [15] Wang Jianguo, Wang Chen, Guo Youmin, et al. A evaluation of pulmonary artery obstruction degree and right ventricular function of patients with pulmonary embolism by CTPA. Chin J Med Imaging 2009;17(2):81–5. [16] Wang JG, Zhu L, Liu M, et al. The value of CT pulmonary angiography to the diagnosis of right ventricular dysfunction due to acute pulmonary embolism: compared with ultrasonographic cardiography. JNMU 2008;22(4):234–7. [17] Wang JG, Guo XJ, Liu M, et al. The dynamic study of the pulmonary artery obstruction degree and the right ventricular function in massive pulmonary embolism on CT pulmonary angiography. JNMU 2008;22(3):188–92. [18] Bankier AA, Janata K, Fleischmann D, et al. Severity assessment of acute pulmonary embolism with spiral CT: evaluation of two modified angiographic scores and comparison with clinical data. J Thorac Imaging 1997;12(2):150–8. [19] Engelke C, Rummeny EJ, Marten K. Acute pulmonary embolism on MDCT of the chest: prediction of cor pulmonale and short-term patient survival from morphologic embolus burden. AJR Am J Roentgenol 2006;186(5):1265–71. [20] Dartevelle P, Fadel E, Mussot S, et al. Chronic thromboembolic pulmonary hypertension. Eur Respir J 2004;23(4):637–48. [21] Sacks RS, Remillard CV, Agange N, et al. Molecular biology of chronic thromboembolic pulmonary hypertension. Semin Thorac Cardiovasc Surg 2006;18(3):265–76. [22] Van der Meer RW, Pattynama PM, van Strijen MJ, et al. Right ventricular dysfunction and pulmonary obstruction index at helical CT: prediction of clinical outcome during 3-month follow-up in patients with acute pulmonary embolism. Radiology 2005;235(3):798–803. [23] Pech M, Wieners G, Dul P, et al. Computed tomography pulmonary embolism index for the assessment of survival in patients with pulmonary embolism. Eur Radiol 2007;17(8):1954–9. [24] Mansencal N, Joseph T, Vieillard-Baron A, et al. Diagnosis of right ventricular dysfunction in acute pulmonary embolism using helical computed tomography. Am J Cardiol 2005;15(95(10)):1260–3. [25] Collomb D, Paramelle PJ, Calaque O, et al. Severity assessment of acute pulmonary embolism: evaluation using helical CT. Eur Radiol 2003;13(7):1508–14. [26] Moser KM, Bloor CM. Pulmonary vascular lesions occurring in patients with chronic major vessel thromboembolic pulmonary hypertension. Chest 1993;103(3):685–92. [27] Guillinta P, Peterson KL, Ben-Yehuda O. Cardiac catheterization techniques in pulmonary hypertension. Cardiol Clin 2004;22(3):401–15. [28] Riedel M, Stanek V, Widimsky J, Prerovsky I. Longterm follow-up of patients with pulmonary thromboembolism. Late prognosis and evolution of hemodynamic and respiratory data. Chest 1982;81(2):151–8. [29] Lewczuk J, Piszko P, Jagas J, et al. Prognostic factors in medically treated patients with chronic pulmonary embolism. Chest 2001;119(3):818–23. [30] Kuriyama K, Gamsu G, Stern RG, Cann CE, Herfkens RJ, Brundage BH. CT-determined pulmonary artery diameters in predicting pulmonary hypertension. Invest Radiol 1984;19(1):16–22. [31] Ng CS, Wells AU, Padley SP. A CT sign of chronic pulmonary arterial hypertension: the ratio of main pulmonary artery to aortic diameter. J Thorac Imaging 1999;14(4):270–8. [32] Heath D, Edwards JE. The pathology of hypertensive pulmonary vascular disease; a description of six grades of structural changes in the pulmonary arteries with special reference to congenital cardiac septal defects. Circulation 1958;18(4 Part 1):533–47.