Europace Advance Access originally published online on October 17, 2007
Europace 2007 9(12):1134-1140; doi:10.1093/europace/eum227
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ATRIAL FIBRILLATION
Dynamic multidimensional imaging of the human left atrial appendage
1 Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA; 2 Cardiovascular Institute, University of Pittsburgh, UPMC Presbyterian, B535, Pittsburgh, PA 15213-2582, USA
Manuscript submitted 26 July 2007. Accepted after revision 12 September 2007.
* Corresponding author. Tel: +1 412 647 2762; fax: +1 412 647 7979. E-mail address: schwartzmand{at}upmc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Aims: The human left atrial appendage (LAA) is a region of increasing interest as a target for intervention. We sought to improve insight into the anatomy of this region using computed tomography (CT).
Methods and results: Multidimensional cardiac reconstruction (whole heart and isolated left atrium) from CT images was performed in each of three groups: (i) patients without atrial fibrillation (AF, n =10); (ii) patients with intermittent (paroxysmal) AF (n = 25); (iii) patients with continuous (persistent) AF (n = 10). Indices included LAA morphology, anatomical relationships, dimensions, angulation, and motility. There was substantial interindividual variation in each index. LAA morphologic differences were associated with variations in anatomical relationships. LAA dimensions in AF patients exceeded those in patients without AF, but angulation and motility were similar. The LAA could be subdivided into proximal and distal portions, each of which had distinct morphology and anatomical relationships. Dimensions in men tended to exceed those in women.
Conclusion: Regardless of AF history, there is broad variation in LAA morphology, anatomical relationships, dimensions, angulation, and motility. These observations may have importance for the development of technologies for therapy delivery in this region.
Key Words: Atrial appendage, Atrial fibrillation, Computed tomography
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
The left atrial appendage (LAA) is an important anatomical region, given its association with atrial tachyarrhythmias and thrombi.1
,2 Recent reports have described the development of technologies for therapy delivery this region.3–9 A detailed understanding of LAA anatomy may facilitate such endeavors. Characterization of LAA anatomy has primarily been ex vivo, poorly referenced to contiguous cardiac anatomy, and static.10,11
A growing literature attests to the utility of computed tomographic imaging (CT) for dynamic, multidimensional characterization of cardiac anatomy.12–14 We hypothesized that CT would improve insight into the anatomy of the LAA.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
This was a retrospective analysis of data acquired during the course of standard clinical practice, as is detailed below. Collation of this data was approved by the Institutional Review Board of the University of Pittsburgh Medical Center.
Patients
Three groups were studied:
- No AF group (n = 10): these patients were referred for CT to evaluate for possible coronary atherosclerosis burden. No patient had a history of or symptoms to suggest atrial fibrillation (AF). There were eight men, and the mean age was 52 years (range 35–71 years). Three patients had a history of hypertension. Patients were on a variety of cardiac medications at the time of image acquisition, including beta blockers (3), calcium channel blockers (2), and rennin–angiotensin inhibitors (4). Five patients had CT evidence of significant coronary artery atherosclerosis (calcification).
- Paroxysmal AF group (n = 25): these patients were referred for CT in preparation for catheter ablation of AF.12 Each had intermittent (self-terminating) AF (<24 h/episode by patient description). In the 6 months prior to imaging, AF symptom duration (total duration of symptoms as a proportion of total time) was <30%; no AF symptoms had occurred in any patient for at least 7 days (mean 23 days; range 8–62 days) prior to image acquisition. There were 15 men, and the mean age of the group was 53 years (range 41–68 years). Sixteen patients had a history of hypertension. Patients were on a variety of medications at the time of image acquisition, including beta blockers (5), calcium channel blockers (3), rennin–angiotensin inhibitors (19), propafenone (2), sotalol (1), dofetilide (1), and amiodarone (1). Twelve patients had CT evidence of significant coronary atherosclerosis.
- Persistent AF group (n = 10): these patients were referred for CT in preparation for catheter ablation of AF. Each had non-remitting AF which had been present for at least 6 months prior to imaging. There were nine men, and the mean age of the group was 57 years (range 48–68 years). Eight patients had a history of hypertension. Patients were on a variety of cardiac medications at the time of image acquisition, including beta blockers (5), calcium channel blockers (3), and rennin–angiotensin inhibitors (9). Six patients had CT evidence of significant coronary atherosclerosis.
Imaging
Cardiac CT angiography was performed utilizing a commercial 64-detector scanners (General Electric Healthcare Inc., Milwaukee, WI, USA) and dual barrel injectors (Stellant DTM, Medrad Inc., Pittsburgh, PA, USA). Injection and scanning parameters were as follows: test bolus of 20 cc Iodixanol 320 (General Electric Healthcare Inc.), 20 cc saline flush at 5 cc/s timed for the aortic root; scanning bolus of 50 cc Iodixanol 320, 50 cc mix of 25 cc iodixanol/25 cc saline, 50 cc saline flush at 5 cc/s; kVP 120, mA 450–800 with dose modulation (peak dose 40–80% of R-R interval), 15–18 cm FOV, single breath hold, retrospective ECG-gated (no AF and paroxysmal AF groups) or non-gated (persistent AF group) helical acquisition with 0.63 mm slice thickness, 350 ms rotation time, pitch adjusted to the heart rate.
Data were reconstructed in 0.63 mm axial slices. Coronal, sagittal, and oblique reformats of the volume-rendered whole heart and isolated left atrium were performed. For ECG-gated studies, images were produced at each of 10% increments of the R-R interval; for non-gated studies, a single image was produced. All post-processing was done using commercial software (AdvantageTM version 4.2, General Electric Healthcare Inc.), as described previously.12 Volumes reported herein were obtained from the multidimensional models; linear dimensions were obtained from the multiplanar reformats.
Analysis
Analysis of the image data (described below) was performed by three investigators working independently: with respect to LAA morphology type (below), there was complete agreement among the investigators in all patients. Given the lack of a ‘gold standard,’ the dimensional, angulation, and motility index data generated by each investigator were averaged; the interinvestigator variation in each of these indices was <10%.
|
Whole heart models demonstrated significant interindividual variation in LAA morphology, which impacted on contiguous anatomy. To better illustrate this, we created the following arbitrary classification system based on the orientation and location of the LAA ‘tip,’ defined as the most distal tissue (Figure 1):
- Morphology type I: tip oriented superiorly and approximately parallel to the main pulmonary artery and left cardiac border.
- Morphology type II: tip oriented inferiorly and approximately parallel to the main pulmonary artery and left cardiac border.
- Morphology type III: tip oriented superiorly but turning medially and located between the main pulmonary artery and the left atrial body.
|
Left atrial models were then used to examine the LAA region in greater detail. We found it instructive to divide the LAA into two portions (Figure 2):
- Proximal portion (LAAp): this portion was defined by a non-circular ‘orifice’ formed by a plane perpendicular to the left pulmonary vein-appendage isthmus, and ended as the plane demarcating the transition between smooth and trabeculated endocardial contours. Although the endocardial contour of the LAAp was predominantly smooth, discrete areas of ‘pitting’ (solitary and/or in groups) were common (Figure 1).
- Distal portion (LAAd): this portion was defined by a non-circular ‘orifice,’ which was the same as the plane defining the end of the LAAp. Although also non-circular, it was significantly smaller than the orifice of the LAAp (Figure 3). The axis of the LAAd formed a significant angle with the axis of the LAAp (Figure 2).
|
|
Gated studies, performed in no AF and paroxysmal AF group patients, were analysed at each of three phases of the atrial mechanical cycle (Figure 4):
- Phase I was defined (from among the 10 incremental images spanning the R-R interval, beginning with the peak of the QRS complex) as the image demonstrating the maximum atrial volume. This consistently occurred at 40 or 50% of the R-R interval, which as expected was immediately prior to mitral valve opening.
- Phase II was defined as the last image prior to the onset of atrial contraction. This consistently occurred at 70 or 80% of the R-R interval, and approximated the end of diastasis.
- Phase III was defined as the image demonstrating the minimum atrial volume. This consistently occurred at 90 or 100% of the R-R interval, and approximated the end of atrial contraction.
|
|
Data are reported as mean ± standard deviation, unless otherwise specified. Statistical significance of intergroup differences in continuous variables was assessed using a Kruskal–Wallis analysis of variance on ranks, and in proportions using a Fisher exact test. Intragroup comparisons were performed using a t-test. Correlations were assessed using a Pearson product moment test. For each test, a P-value of <0.05 was used to define significance.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
In each of the study groups, each of the three LAA morphology types was observed. In type I, structures contiguous to the proximal aspects of the appendage included left inferior pulmonary vein, coronary sinus, and LV base. Structures contiguous to the distal aspects of the appendage included pulmonary artery, pulmonic valve, circumflex coronary artery, left anterior descending coronary artery, great cardiac vein, left superior pulmonary vein, and LV. The appendage tip was located in the Trigone of Brocq and Mouchet, the base of which is formed by the great cardiac vein, and the remaining boundaries by the proximal portions of left anterior descending and circumflex coronary arteries.15
In type II, a less intimate relationship with the left pulmonary veins and absence of the tip in the Trigone of Brocq and Mouchet were noted. In type III, more intimate associations between LAA and left atrial body, left main coronary artery, and aortic valve were noted. Occasionally, the tip extended onto the ascending aorta. The prevalence of each LAA morphology type was not significantly different among the groups (no AF group: type I 30%, type II 60%, type III 10%; paroxysmal AF group: type I 20%, type II 68%, type III 12%; persistent AF group: type I 10%, type II 70%, type III 20%.
Analysis of the left atrial models revealed that in each patient group, total LAA (e.g. LAAp + LAAd) volume comprised 
10% of the total left atrial volume. Dimensions of both LAAp and LAAd tended to be significantly larger in the paroxysmal AF and persistent AF groups than in the no AF group, as did global left atrial volume (Tables 1 and 2). Gated studies, performed in no AF and paroxysmal AF group patients, revealed that at each phase of the atrial mechanical cycle global left atrial volume was significantly larger among paroxysmal AF group than no AF group patients. Although left atrial volume was larger among persistent AF group than paroxysmal AF group patients (the latter assessed for this comparison at maximum volume, phase I), this did not achieve statistical significance. Within groups, interphase changes in dimensions were significant (Tables 1 and 2). Although some motility was apparent in each patient, there was marked interindividual variation. Motility of the LAAp and LAAd were similar. Motility of the LAA as a whole tended to be greater than that of the remainder of the left atrium, although this did not reach statistical significance. Within individuals, the angle between the axes of the LAAp and LAAd (Figure 2) changed notably between phases of the atrial mechanical cycle (Table 2). However, because of baseline differences as well as differences in direction of change between phases among individuals, comparisons of angulation data did not demonstrate significant differences within groups. Similar observations were made in regard to the magnitude of the change in angle between phases. There was no significant change in the location of the appendage tip between phases.
|
|
We were interested to know whether LAA dimensions and motility could be correlated with these indices for the left atrium in its entirety, as well as the influence of gender. Within-group correlations are summarized in Table 3. In general, among paroxysmal AF group patients LAA dimensions and motility were significantly correlated with those of the left atrium in its entirety. These relationships were less consistent among no AF group patients. An additional analysis comparing male and female members within the paroxysmal AF Group was performed, which demonstrated larger dimensions in men (Table 4); however, with the exception of LAAd cross-sectional area, these differences did not reach statistical significance.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Herein, we demonstrate significant LAA variability within and between groups of patients with and without AF. Our observations extend those of previous reports: first, to our knowledge there has been no prior demonstration of the spectrum of LAA anatomical relationships in vivo. Observations of particular interest included contiguity of appendage to pulmonary artery, pulmonary veins, coronary vessels, and aortic root. Secondly, we provide new insight into LAA anatomy. We separated the appendage into two portions, based on smooth (LAAp) or trabeculated (LAAd) endocardial contour. Each portion was morphologically complex. A gradation in dimensions was apparent, with persistent AF, paroxysmal AF and no AF groups in diminishing order. In the paroxysmal AF group, dimensions in men were larger than those in women, albeit not significantly. Our LAAd dimensional data are comparable with those of Heist et al.,16
who performed a morphologic assessment of the LAA based on non-gated images in AF patients using magnetic resonance imaging; these investigators also demonstrated a correlation between LA and LAA. Our LAAd data are also consistent with those of Wongcharoen et al.,17 who analysed computed tomographic images and demonstrated larger dimensions in AF patients, vs. non-AF patients, as well as variability in the spatial relationship between the LAA and left pulmonary veins. Finally, we provide new insight into LAA motility: 3-dimensional quantification of motility is, to our knowledge, unprecedented. As expected, motility was demonstrable during both active and passive phases of the left atrial mechanical cycle, and was apparent in both LAAp and LAAd. LAA motility tended to exceed that of the left atrium as a whole, but it accounted for a minor portion of left atrial volume. Important additional observations concerning LAA motility included marked fluctuations in orifice dimensions and angulation.
We note important limitations to these data. First, the AF groups were biased in that they were comprised of patients selected for catheter ablation. In our institution, these patients are relatively young, with preserved LV systolic function and limited comorbidity. As such, these patients may misrepresent community AF and/or LV systolic dysfunction cohorts. Secondly, we have no data to substantiate the reproducibility or temporal stability of our findings. It is possible that, within individuals, significant changes would be observed in association with changes such as those in body position, physical/mental activity, volume status, pharmacotherapy, duration of preceding AF events, and proximity to cardioversion. Indices of motility would be particularly susceptible in this regard. Thirdly, we cannot be sure that filling of the LAA by contrast was uniform between groups. For example, it is possible that blood flow rheologic differences, particularly in the distal portion of the appendage, may have hampered adequate contrast access in some patients. Fourth, our method for classification of LAA morphology, based on orientation and location of the appendage tip, is arbitrary and not necessarily valuable. However, our goal in using this method was not to be encyclopedic, rather to inform as to the spectrum of variation in anatomical relationships with the LAA. Similarly, our assignation of LAA into proximal and distal portions was arbitrary, and also possibly not of value. This demarcation too was motivated by interventional considerations (see below). Fifth, our presentation omitted consideration of structures of potential importance which are known to be consistently contiguous to the appendage, including the left phrenic nerve and left coronary lymphatic channels, as well as those in which contiguity is more variable, such as the sinus node artery.18–20 Finally, we make no attempt to correlate our findings with other dynamic variables of known physiologic significance.21
In summary, we have demonstrated broad interindividual variation in LAA morphology, anatomical relationships, dimensions, angulation, and motility. We believe that these findings may have importance for the development of technologies for interventional procedures being planned for this region. Recent reports have suggested that the LAA region plays a role in initiation and/or perpetuation of atrial tachyarrhythmias, and as such may be of interest for ablative targeting.2 Also of interest is LAA ‘modification’ for prevention of AF-attributable cardioembolism, for which techniques and technologies are developing using endocardial4,6,8,9 or epicardial3,22–24 approaches. Our expectation is that multidimensional imaging such as that provided by CT will come to play an important role in planning, performance, and assessment of these procedures. For example, with respect to planning, preoperative morphology, angulation, and motility information may be useful in selecting approach and device type/size.25 Similarly, integration of the information contained in these images into the operative field may facilitate performance of these procedures.26 Finally, although not addressed in this report, CT imaging may develop as a useful technique for confirming that the desired technical result has been achieved.25
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
J.M.L., O.G., and D.S. have in the past received funding from General Electric Healthcare Inc. for studies pertaining to their clinical imaging devices. No such funding was provided for this study.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Conflict of interest: none declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
[1] Al-Saady N, Obel O, Camm A. Left atrial appendage: structure, function and role in thromboembolism. Heart (1999) 82:547–54.
[2] Yun-Long W, Xue-Bin L, Xin W, Jin-Xin M, Ping Z, Yuan X, et al. Focal atrial tachycardia from the left atrial appendage: electrocardiographic and electrophysiologic characterization and long-term outcomes of radiofrequency ablation. J Cardiovasc Electrophysiol (2007) 18:459–64.[CrossRef][Web of Science][Medline]
[3] Blackshear J, Johnson W, Odell J, Baker V, Howard M, Pearce L, et al. Thoracoscopic extracardiac obliteration of the left atrial appendage for stroke risk reduction in atrial fibrillation. J Am Coll Cardiol (2003) 42:1249–52.
[4] Ostermeyer S, Reisman M, Kramer P, Matthews R, Gray W, Block P, et al. Percutaneous left atrial appendage transcatheter occlusion (PLAATO system) to prevent stroke in high-risk patients with nonrheumatic atrial fibrillation - results from the international multicenter feasibility trials. J Am Coll Cardiol (2005) 46:9–14.
[5] Onalan O, Lashevsky I, Hamad A, Crystal E. Nonpharmacologic stroke prevention in atrial fibrillation. Expert Rev Cardiovasc Ther (2005) 3:619–33.[CrossRef][Medline]
[6] Fountain R, Holmes D, Chandrasekaran K, Packer D, Asirvatham S, Van Tassel R, et al. The PROTECT AF (WATCHMAN left atrial appendage system for embolic PROTECTion in patients with atrial fibrillation) trial. Am Heart J (2006) 151:956–61.[CrossRef][Web of Science][Medline]
[7] Mohrs O, Schraeder R, Petersen S, Scherer D, Nowak B, Kauczor H-U, et al. Percutaneous left atrial appendage transcatheter occlusion (PLAATO): planning and follow-up using contrast-enhanced MRI. Am J Radiol (2006) 186:361–4.
[8] Fountain R, Holmes D, Hodgson P, Chandrasekaran K, Van Tassel R, Sick P. Potential applicability and utilization of left atrial appendage occlusion devices in patients with atrial fibrillation. Am Heart J (2006) 152:720–3.[CrossRef][Web of Science][Medline]
[9] Sick P, Schuler G, Holmes D, Turi Z, Yakubov S, Kalbfleisch S, et al. Stroke prevention technology for patients with non-valvular AF—worldwide experience with the new Watchman LAA system. J Am Coll Cardiol (2006) 47:36A.
[10] Ernst G, Stollberger C, Abzieher F, Veit-Dirscherl W, Bonner E, Bibus B, et al. Morphology of the left atrial appendage. Anat Rec (1995) 242:553–61.[CrossRef][Medline]
[11] Veinot J, Harrity P, Gentile F, Khanderia B, Bailey K, Eickholt J, et al. Anatomy of the left atrial appendage: a quantitative study of age related changes in 500 autopsy hearts: implications for echocardiographic examinations. Circulation (1997) 96:3112–5.
[12] Schwartzman D, Lacomis J, Wigginton W. Characterization of left atrium and distal pulmonary vein morphology using multidimensional computed tomography. J Am Coll Cardiol (2003) 41:1349–57.
[13] Cronin P, Sneider M, Kazerooni E, Kelly A, Scharf C, Oral H, et al. MDCT of the left atrium and pulmonary veins in planning radiofrequency ablation for atrial fibrillation: a how-to guide. Am J Radiol (2004) 183:767–78.
[14] Lemola K, Sneider M, Desjardins B, Case I, Han J, Good E, et al. Computed tomographic analysis of the anatomy of the left atrium and esophagus. Implications for left atrial catheter ablation. Circulation (2004) 110:3655–60.
[15] El-Maasarany S, Ferrett C, Firth A, Sheppard M, Henein M. The coronary sinus conduit function: anatomical study (relationship to adjacent structures). Europace (2004) 7:475–81.[CrossRef][Web of Science]
[16] Heist E, Refaat M, Danik S, Holmvang G, Ruskin J, Mansour M. Analysis of the left atrial appendage by magnetic resonance angiography in patients with atrial fibrillation. Heart Rhythm (2006) 3:1313–8.[CrossRef][Web of Science][Medline]
[17] Wongcharoen W, Tsao H-M, Wu M-H, Tai C-T, Chang S-L, Lin Y-J, et al. Morphologic characteristics of the left atrial appendage, roof, and septum: implications for ablation of atrial fibrillation. J Cardiovasc Electrophysiol (2006) 17:951–6.[CrossRef][Web of Science][Medline]
[18] Miller A. Lymphatics of the Heart (1982) New York: Raven Press.
[19] Sanchez-Quintana D, Cabrera J, Climent V, Farre J, Weiglein A, Ho S. How close are the phrenic nerves to cardiac structures? Implications for cardiac interventionalists. J Cardiovasc Electrophysiol (2005) 16:309–13.[CrossRef][Web of Science][Medline]
[20] Busquet J, Fontan F, Anderson R, Ho S, Davies M. The surgical significance of the atrial branches of the coronary arteries. Int J Cardiol (1984) 6:223–36.[CrossRef][Web of Science][Medline]
[21] Agmon Y, Khanderia B, Gentile F, Seward J. Echocardiographic assessment of the left atrial appendage. J Am Coll Cardiol (1999) 34:1867–77.
[22] Kamohara K, Fukamachi K, Ootaki Y, Akiyama M, Zahr F, Kopcak MJ, et al. A novel device for left atrial appendage exclusion. J Thorac Cardiovasc Surg (2005) 130:1639–44.
[23] Jayakar D, Gozo F, Gomez E, Carlos C. Use of tissue welding technology to obliterate left atrial appendage—novel use of Ligasure. Interact Cardiovasc Thorac Surg (2005) 4:372–3.
[24] Gillinov A, Pattersson G, Cosgrove D. Stapled excision of the left atrial appendage. J Thorac Cardiovasc Surg (2005) 129:679–80.
[25] Schwartzman D, Katz W, Smith A, Anderson W. Malpositioning of a left atrial appendage occlusion device? A case with implications for percutaneous transcatheter left atrial appendage occlusion device therapy. Heart Rhythm (2007) 4:648–50.[CrossRef][Web of Science][Medline]
[26] Sra J. Registration of three dimensional left atrial images with interventional systems. Heart (2005) 91:1098–104.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. Schwartzman and J. L. Williams Electroanatomic properties of pulmonary vein antral regions enclosed by encircling ablation lesions Europace, April 1, 2009; 11(4): 435 - 444. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||