A study of images of Projective Angles of pulmonary veins

European Journal of Radiology 71 (2009) 474–479

A study of images of Projective Angles of pulmonary veins Jue Wang, Zhaoqi Zhang ∗ , Wei Yu, Cuilian Miao, Zixu Yan, Yike Zhao Beijing Anzhen Hospital, Beijing, China Received 18 April 2008; accepted 19 May 2008

Abstract Aims: In images of magnetic resonance and computed tomography (CT) there are visible angles between pulmonary veins and the coronary, transversal or sagittal section of body. In this study these angles are measured and defined as Projective Angles of pulmonary veins. Several possible influential factors and characters of distribution are studied and analyzed for a better understanding of this imaging anatomic character of pulmonary veins. And it could be the anatomic base of adjusting correctly the angle of the central X-ray of the angiography of pulmonary veins undergoing the catheter ablation of atrial fibrillation (AF). Method: Images of contrast enhanced magnetic resonance angiography (CEMRA) and contrast enhanced computer tomography (CECT) of the left atrium and pulmonary veins of 137 health objects and patients with atrial fibrillation (AF) are processed with the technique of post-processing, and Projective Angles to the coronary and transversal sections are measured and analyzed statistically. Result: Project Angles of pulmonary veins are one of real and steady imaging anatomic characteristics of pulmonary veins. The statistical distribution of variables is relatively concentrated, with a fairly good representation of average value. It is possible to improve the angle of the central X-ray according to the average value in the selective angiography of pulmonary veins undergoing the catheter ablation of AF. © 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Projective Angle; Pulmonary vein

1. Introduction In recent years the catheter ablation of atrial fibrillation (AF) has been rapidly developing, and more and more attention has been paid to the morphology of the left atrium and pulmonary veins. In quite a few literatures the left atrium, pulmonary veins and their junctions have been studied [1–5], but angles between the long axis of every pulmonary vein and the coronal, sagittal and transversal sections of the body have not been described in related literatures up to now. In this study these angles are defined as Projective Angles of pulmonary veins (Figs. 1 and 2), which are displayed clearly on images of CEMRA or CECT, either the primary images of the transversal section or the reconstructed 3D or 2D images by applying the technique of post-processing. The scatheless measurement of Project Angles of pulmonary veins has been done on these images (Figs. 1 and 2). The measurement and analysis in this study indicate that these degrees are independent of sex, characteristics of images (CEMRA or CECT) and suffering or not of atrial fibrillation. The result demonstrates that

Project Angle of pulmonary veins is one of the steady imaging anatomic characteristics of pulmonary veins. Their leptokurtic curves of probability density show that Projective Angles have considerably concentrative statistical distribution. There is not much individual variability, and the average value is of good representation. As the experimental sample is great, the average value of experimental samples of this study reflects to a great extent the average value of Projective Angles of pulmonary veins of people, which has a clinic significance of revealing: (1) The Projective Angle of pulmonary veins as one of new imaging anatomic character of pulmonary veins is brought to light. (2) It seems reasonable to adjust the angle of the central X-ray in the fluorography of pulmonary veins undergoing the catheter of ablation of atrial fibrillation according to the Projective Angle of pulmonary veins to improve the conventional method [6].

2. Method 2.1. Object of measurement

∗

Corresponding author. Tel.: +86 1064 444580. E-mail address: [email protected] (Z. Zhaoqi).

0720-048X/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejrad.2008.05.012

87 cases accepting CEMRA are divided into two groups. Group 1 consists of 50 patients (37 men, the mean age of 53 ± 13

W. Jue et al. / European Journal of Radiology 71 (2009) 474–479

475

Fig. 1. Pulmonary veins and their Protective Angles to the coronary section on image of CEMRA. (A) Projective Angles of pulmonary veins to the transversal section. (B) Projective Angle of LSPV to the transversal section on the image of the oblique sagittal section. (C) Projective Angles of pulmonary veins to the coronary and sagittal sections on the image of the transversal section.

years old and the age range of 21–77 years old) with symptomatic drug-refractory paroxysmal AF, the median of episodes of AF is 48 months (2–240 months), while any abnormal shape of the left atrium and any diseases of lungs, mediastinum and heart probably resulting abnormal variation of the left atrium are excluded with the help of clinic, ultrasonic, general radiological and CEMRA examinations. Group 2 is composed of 37

healthy control objects (27 men, the mean age of 38 ± 16 years old and the age range of 25–69 years old). Group 3 includes 50 healthy objects (38 men, the mean age of 57 ± 17 years old and an age range of 17–90 years old) having accepted CECT (Table 1). Groups 2 and 3 consist of individuals without any heart, lungs, pleural and mediastinum diseases proved by clinic, physical examination, ECG, radiology or echocardiogram. The

Fig. 2. Pulmonary veins and their Projective Angles to the coronary section on images of CECT.

476

W. Jue et al. / European Journal of Radiology 71 (2009) 474–479

Table 1 Projective Angles of every pulmonary vein on the coronary and transversal section PV

CECT (health)

CEMRA (health)

CEMRA AF

M

F

M

F

M

F

24 24 33 33

8 8 16 16

Project Angle to coronary section

LS LI RS RI

38 38 38 38

12 12 12 12

20 20 27 26

9 7 9 10

Project Angle to the cross-section

LS LI RS RI

21 20 25 24

8 7 10 10

24 24 33 33

8 13 17 16

mediastinum does not bend and the chest is not misshapen. All images are from the database of the Radiologic Department of our hospital. 2.2. Method of radiology The examination of the 87 cases of CEMRA adopts gadolinium-enhanced MRA with a 1.5 T MR imaging system (SONATA, Siemens Medical Systems, Germany). The test-bolus is performed by using 20 ml of contrast agent (gadolinium), which is loaded into the intravenous line and pushed at 5 ml/s. The scanning range includes all the left atrium and the 40 mm length near the left atrium of every pulmonary vein. High signal images are obtained, and images of pulmonary veins and the left atrium are reconstructed from raw data by means of maximum intensity projection (MIP) and multi-planar reconstructions (MPR). Images of the group CECT are obtained from the contrast enhanced CT scan of chest with the multi-detectors CT (SOMATO MDCT, Siemens, Germany). 100 ml of intravenous non-ion contrast medium ULTRAVISION 300 is injected by use of high press injector, and in 20 s the scanning of heart and major vessels is performed, while filling images of heart and major vessels are obtained. 2.3. Measurement of Projective Angles of pulmonary veins With raw data of CEMRA the three dimensions (3D), maximum intensity projection (MIP) and multi-planar reformation (MPR and CPR) are performed to reconstruct images of pulmonary vein and left atria. On the coronary image of CEMRA Projective Angles of pulmonary veins to the trans-vertical section are measured. Analogously on the trans-vertical image Projective Angles of pulmonary veins to the coronary plane are measured (Fig. 2). In the Group of CECT only on the primary transversal images Projective Angles of pulmonary veins to the coronary section are measured. In all the objects any variation with common trunk of pulmonary veins is deleted, so the number of pulmonary veins measured practically probably is smaller than that of objects (Table 1). All measurements are conducted by the use of the measurement software of working sets of MR and CT medical systems.

(k × n)

F

F0.05 (k − 1,n − k)

P

Mean ± S.D.

(6 × 111) (6 × 109) (6 × 135) (6 × 135)

1.8188 0.7788 0.4021 1.3086

4.39 4.39 4.39 4.39

>0.05 >0.05 >0.05 >0.05

15.79 24.17 23.03 28.45

± ± ± ±

5.39 8.34 7.65 9.34

0.46 0.71 0.65 0.80

(4 × 61) (4 × 64) (4 × 85) (4 × 83)

1.6166 0.8440 1.1758 0.3701

8.57 8.57 8.56 8.56

>0.05 >0.05 >0.05 >0.05

30.69 14.78 29.79 16.21

± ± ± ±

12.29 5.55 8.54 5.77

1.05 0.47 0.73 0.49

Sx

2.4. Method of statistical analysis For Projective Angles of every pulmonary vein to the coronary section, all data are divided into 3 groups (CEMRA health, CEMRA AF and CECT), and each group is subdivided into 2 groups (men and women), forming 6 groups of data. Every group of data acts as a single column of data, so the 6 columns of data constitute a single data list. In the list the number of rows is the number of data in the column, which includes the largest of data. For 4 pulmonary veins (the right superior, the right inferior, the left superior and the left inferior) there are 4 data lists as already described. For Projective Angles of every pulmonary vein to the transversal section, all data are divided into 2 groups (CEMRA health and CEMRA AF), and each group is subdivided into 2 groups (men and women). There are 4 groups of data in all. Every group of data acts as a single column of data, so the 4 columns of data constitute a single data list. In the data list the number of rows is the number of data in the column, which includes the largest of data. For 4 pulmonary veins there are 4 data list as already described. Variance analysis of the random sample of the single factor (F-test) is used to compare continuous variables among groups in every data list. P-value of <0.05 is considered statistically significant. While the mean value (average), the standard deviation (S.D.) and the standard error (S.E.) of Projective Angles of pulmonary veins are calculated, continuous data are expressed as mean ± S.D. In the 8 data lists mentioned above the asymmetry coefficient and kurtosis coefficient of curves of distribution of Projective Angles of every pulmonary vein are calculated. Statistical analyses are performed by means of the Clis Medical Statistical Software Ver., 1.0, 2004 (CLIS Software Studio, Hangzhou, Zhejiang, China). 3. Result Result tested by means of variance analysis of the random sample of the single factor (F-test) indicates that there are no significant differences (F < F0.05 , P > 0.05) among the group CEMRA, the group CECT, the group health, the group AF, the group men and the group women for every Projective Angle

W. Jue et al. / European Journal of Radiology 71 (2009) 474–479

477

Table 2 Type of distribution of Projective Angles of every pulmonary vein Pulmonary veins

g1

Projective Angles to the coronary section

LS LI RS RI

0.3391 0.6004 0.1745 2.0448

Projective Angles to the transversal section

LS LI RS RI

0.3513 −0.05 0.1202 0.6557

Type Positively skewed Positively skewed Positively skewed Positively skewed Positively skewed Negatively skewed Positively skewed Positively skewed

g2

Type

1.6076 1.2264 0.3999 8.1282

Leptokurtic Leptokurtic Leptokurtic Leptokurtic

0.4455 −0.1616 −0.4846 2.0520

Leptokurtic Platykurtic Platykurtic Leptokyrtic

5. Discussion

Fig. 3. Distribution of Projective Angle of pulmonary vein. The leptokurtis and positively skewed curve of distribution of Projective Angle of pulmonary vein.

of pulmonary vein (Table 1). According to this result objects of all groups are merged, and mean values (average), standard deviation (S.D.) and standard error (S.E.) of every Projective Angle of pulmonary veins are calculated (Table 1). Because kurtosis coefficient >0, all Projective Angles of pulmonary veins are leptokurtic, except Projective Angles to the transversal plane of the left inferior and right superior pulmonary veins are platykurtic distribution. The symmetry of curves of distribution of all Projective Angles of pulmonary veins are positive skewness distribution (skewness coefficient > 0), except that Projective Angle to the transversal plane of the left inferior pulmonary vein are the negative skewness distribution (Table 2; Fig. 3). 4. Conclusion The Projective Angle of pulmonary vein is the true imaging anatomic structure of pulmonary vein, and longstanding atrial fibrillation does not influence the degrees. We suggest that before the catheter ablation the CEMRA of the left atrium and pulmonary veins be done in advance to measure every Projective Angle of pulmonary veins. During the angiography of pulmonary veins the angle of the central X-ray can be adjusted according to these angles and the conventional method will be improved. Yet if a patient is in no condition, the mean value (average) of Projective Angle can be used too.

The purpose of this study is to explore Projective Angles of pulmonary veins as one of the morphological characters of pulmonary vein and their statistical distribution. Quite a few literatures deal with imaging morphology of pulmonary veins [1,3,4], but heretofore Projective Angles have not been described, and this study is to add some important information to the knowledge of imaging anatomy of pulmonary vein. The study educes three important conclusions: First, Projective Angles of pulmonary veins are independent of sex and the character of images (MR or CT). This shows that Projective Angles of pulmonary veins are true and steady imaging anatomic character. It is especially important that symptomatic drug-refractory paroxysmal AF has no effect on Projective Angles of pulmonary veins, therefore the results of this study can be applied to the patients undergoing the catheter ablation of pulmonary veins. Second, the distribution curves of Projective Angles of pulmonary veins in the main show signs of leptokurtic (Table 2; Fig. 3), the standard deviation (S.D.) is smaller; while the sample is great enough, the standard discrepancy (Sx–) is even smaller. The 95% confidence interval of the average of the collectivity (Projective Angles of crowd) is only average ± 1◦ or so (Table 1), and according to the statistical principle the average of the sample of this study represents the average of collectivity to a considerable extent. And third, the leptokurtic indicates that variables of Projective Angles of pulmonary veins are centered around the average more closely than mesokurtic. In common clinics for a patient who is selected at random, his/her Projective Angles of pulmonary veins are approximately equal to the average of the sample in this study. On the other hand, in the positively skewed distribution, because the mode is smaller than the average, it is not likely that Projective Angles of pulmonary

Table 3 The suggesting angle of the central X-ray (vertical to the ground) according to Projective Angles of pulmonary veins Pulmonary vein

The angle of the central X-ray vertical to the ground

The left superior pulmonary vein The left inferior pulmonary vein The right superior pulmonary vein The right inferior pulmonary vein

RAO 16◦ + tail 59◦ (90–31◦ ) LAO 24◦ + cranial 75◦ (90–15◦ ) LAO 23◦ + tail 60◦ (90–30◦ ) RAO 28◦ + cranial 74◦ (90–16◦ )

478

W. Jue et al. / European Journal of Radiology 71 (2009) 474–479

Fig. 4. The adjust of the angle of the central X-ray according to the Projective Angle of pulmonary vein in the selecting fluoroscopic angiography of pulmonary vein. (A) Improved method suggested by this study: the angle of the central X-ray = the Projective Angle of the pulmonary vein. (B) The conventional method: the angle of central X-ray is always fixed on LOA or ROA 30 or 60◦ .

veins of a object selected at random are somewhat smaller than average. The understanding of degrees and the distribution of Projective Angles of pulmonary veins have certain clinical significance. In conducting the catheter ablation of AF, it is necessary to carry out the selecting fluoroscopic angiography of pulmonary veins to measure the size of pulmonary veins in order to select equal instruments. In fluoroscopic angiography of pulmonary veins the routine angle of the central X-ray is adjusted at LAO or RAO 30◦ and 60◦ , once more cranial angle of 20◦ [6]. In fact these angles are just rough approximations and are quite different from Projective Angles measured in this study. So we suggest that the angle of the central X-ray should be adjusted according to the averages of Projective Angles of pulmonary veins instead of the conventional method above mentioned. Taking account of the precision of practical operation, degrees of angles are exacted at 1◦ (Table 3). So it is possible to ensure the central X-ray to be vertical to the long axis of shot pulmonary vein by all means (Fig. 4), in order to avoid distortion of images and ensure the veracity of measurement from the perspective of anatomy and geometry. If a patient undergoes primarily the contrast enhanced magnetic resonance angiography (CEMRA) before the catheter ablation, every Projective Angle of pulmonary veins will be measured directly and according to these Projective Angles the angle of the central X-ray can be adjusted. Apart from these in recent years there have been quite a few productions about the angiography of heart and main blood vessels by use of CEMRA demonstration. As “one station” heart examination a series of examinations of heart, for example, the

morphology and function of heart can be accomplished in a single examination. Data obtained from CEMRA can offer certain important information, including the Projective Angle of pulmonary veins for catheter ablation of the atrial fibrillation (AF). It is by far the most important to follow-up the patient after the catheter ablation by means of CEMRA to explore the presence or the absence of postoperative stenosis of pulmonary veins. CEMRA has no X-ray radiation, the contrast medium is secure, and disturbances from pulmonary artery can be eliminated completely by means of the technique of the post-proceeding of images. So it should be a satisfactory imaging means of preoperative study and postoperative follow-up checkups for catheter ablation of pulmonary veins. It is certain that the medical expense might rise because of CEMRA. For poor patients in developing countries and areas it should be satisfactory to adjust the angle of the central X-ray according to the mean value (average) of Projective Angle of pulmonary veins. 6. Study limitation In this study, different measurement results from different angles of central X-ray have not been compared in practice. The authors have to study this problem in other subject. Funding The study is funded by Radiologic Department of Anzhen Hospital, Beijing, China.

W. Jue et al. / European Journal of Radiology 71 (2009) 474–479

Conflict of interest None declared. References [1] Kato Ritsushi, Lickfett Lars, Meininger Glenn, et al. Pulmonary vein anatomy in patients undergoing catheter ablation of atrial fibrillation: lessons learned by use of magnetic resonance imaging. Circulation 2003;107(April):2004–10. [2] R. Verlato, P. Turrini, M.S. Baccillieri, et al. Pulmonary veins: anatomy and transcatheter electrical isolation: 7.1. Variability of Pulmonary Vein Anatomy in Patients Ablated for Atrial Fibrillation Europace 2005; 7:S11.

479

[3] Wood MA, Wittkamp M, Henry D, et al. A comparison of pulmonary vein ostial anatomy by computerized tomography, echocardiography, and venography in patients with atrial fibrillation having radiofrequency catheter ablation. Am J Cardiol 2004;93:49–53. [4] Ector Joris, De Buck Stijn, Adams Jef, et al. Cardiac three-dimensional magnetic resonance imaging and fluoroscopy merging: a new approach for electroanatomic mapping to assist catheter ablation. Circulation 2005;112(December):3769–76. [5] Wittkampf Fred HM, Vonken Evert-Jan, Derksen Richard, et al. Pulmonary vein ostium geometry: analysis by magnetic resonance angiography. Circulation 2003;107(January):21–3. [6] Lin Wei-Shiang, Prakash VS, Tai Ching-Tai, et al. Pulmonary vein morphology in patients with paroxysmal atrial fibrillation initiated by ectopic beats originating from the pulmonary veins: implications for catheter ablation. Circulation 2000;101(March):1274–81.