Advances in Vein Therapy

Atrial Septal Angulation Varies Widely in Patients Undergoing Pulmonary Vein Isolation

Erin A. Fender, MD1, Christopher T. Sibley, MD2,3,4, Saman Nazarian, MD, PhD2, Alan Cheng, MD2, David D. Spragg, MD2, Joseph E. Marine, MD2, Ronald D. Berger, MD, PhD2, Hugh Calkins, MD2, Joao A.C. Lima, MD2, Jeffrey A. Brinker, MD2, Charles A. Henrikson, MD2,4

Erin A. Fender, MD1, Christopher T. Sibley, MD2,3,4, Saman Nazarian, MD, PhD2, Alan Cheng, MD2, David D. Spragg, MD2, Joseph E. Marine, MD2, Ronald D. Berger, MD, PhD2, Hugh Calkins, MD2, Joao A.C. Lima, MD2, Jeffrey A. Brinker, MD2, Charles A. Henrikson, MD2,4

Abstract: Purpose. Transseptal puncture (TSP) allows left atrial access for curative procedures. Intracardiac echocardiography (ICE) provides direct visualization of the interatrial septum (IAS), but adds time and expense. We reviewed 100 cardiac multidetector computed tomography (MDCT) scans of patients undergoing AF ablation to determine if the angulation and orientation of the IAS are conserved or variable. Significant variability may suggest a potential role for direct visualization of the IAS during TSP. Methods. We reviewed 100 MDCT scans obtained prior to AF ablation. The IAS plane at the fossa ovalis was identified in axial and coronal images. We measured the angle of the septum relative to an orthogonal plane. Optimal needle orientation was defined as perpendicular to the fossa ovalis. Results. The mean axial plane angle was -60.6 ± 10.6°; range, -29.5° to -88.7°). The mean coronal plane angle was 142.6 ± 9.1°; range, 115° to 162°). The axial angle corresponded to variation in the “clock-face” orientation of the needle during puncture, and was calculated between 4 and 6 o’clock. Coronal plane angulation corresponds to the curvature of the needle tip, which varied by 47°. We found no association between patient characteristics and IAS angle. Conclusion. The septal orientation in the axial plane varied widely and was not predicted by clinical variables such as atrial size or prior valve surgery. The high degree of interpatient variability observed suggests that direct visualization of the septum may be helpful in the performance of TSP.

J INVASIVE CARDIOL 2014;26(3):128-131

Key words: atrial fibrillation ablation, interatrial septum, transseptal puncture, intracardiac echocardiography, cardiac computed tomography, cardiac tamponade


Left atrial access relies on the ability of the proceduralist to safely perform transseptal puncture (TSP). Puncturing the septum requires positioning a Brockenbrough curved needle (Figure 1) perpendicular to the plane of the fossa ovalis (FO), and then advancing through the septum. Accessing the left atrium (LA) is an integral step for many procedures, including ablation of atrial fibrillation (AF), ventricular tachycardia, and left-sided accessory pathways, and for left atrial appendage closure, as well as mitral valve procedures, but introduces risks not commonly seen with right atrial access. It is unclear what proportion of this increased risk can be attributed directly to TSP as opposed to other aspects of the procedure. Data from a 2006 multicenter review of 5520 TSP cases over a 12-year span would suggest an overall complication rate for TSP of 0.74%-0.79%.1 Many advances have been made since this study was published, including the adoption of newer imaging modalities. 

Cardiac multidetector computed tomography (MDCT) is now routinely obtained prior to AF ablation for integration into electroanatomic mapping systems. These three-dimensional (3D) mapping systems have been shown to reduce fluoroscopic exposure2,3 and improve ablation success rates.4,5 Despite these advances, overall AF ablation complication rates remain at approximately 2%-5%.6-8 Cardiac tamponade occurs in approximately 1.3% of cases.6,7 The increased rate of tamponade, relative to other electrophysiology procedures, is partially attributable to TSP and anatomic variations in the interatrial septum.9,10 TSP is a critical step in successful and safe LA access, and attempts to minimize risk have resulted in the creation of virtual reality TSP simulators, which have been shown to significantly improve performance of TSP and minimize errors in trainees.11

Intracardiac echocardiography (ICE) can be used for direct, real-time visualization of the interatrial septum (IAS) during TSP.12-14 However, while the use of ICE during AF ablation has become increasingly widespread, the necessity of ICE for TSP is hotly debated. Some major centers rely on fluoroscopic imaging alone to identify the FO and safely cross the IAS.1,7,15,16 Other centers routinely use ICE to visualize IAS anatomic variations.12-14,17 

The angulation and orientation of the IAS, which determine optimal TSP needle curvature and angulation for puncture, have not previously been described. Our goal was to examine these clinically relevant features using MDCT in this hypothesis-generating study, and ascertain whether orientation of the IAS was preserved or highly individualized.


This study was approved by the institutional review board at Johns Hopkins Hospital. In this single-center retrospective analysis, we reviewed 100 MDCT scans obtained prior to AF ablation. MDCTs were used because they are routinely obtained prior to AF ablation, a 0° plane is easily defined, and 3D imaging permits quantification of the spatial orientation of the IAS. We enriched the study population with 8 complicated patients in whom increased IAS angulation variability might be anticipated. Four of these patients experienced a procedural complication, 2 patients had mitral valve replacement, 1 patient had hypertrophic cardiomyopathy, and 1 patient had a prior atrial septal defect repair. The other 92 patients were sequential ablation cases with MDCTs that were adequate for analysis.

Contrast-enhanced MDCT imaging was performed using a 64-detector MDCT scanner (Aquilion; Toshiba America Medical Systems). Tube voltage was 120 kV, with tube current 250-400 mA, dependent on body habitus. Slice collimation was 64 x 0.5 mm or 32 x 1 mm. Iopamidol contrast (Isovue 370; Bracco Diagnostics, Inc) was administered using a power injector at 4-5 mL/s, followed by a normal saline chase bolus. Multisegment reconstruction was performed using retrospective electrocardiographic gating at atrial end-systole (40%-50% RR interval). The effective spatial resolution was 1 x 0.35 x 0.35 mm. Image analysis was performed by a single cardiologist with advanced level III training in cardiac CT, blinded to clinical characteristics of the patients, using commercially available postprocessing software (Vitrea; Vital Images).  

The IAS is a planar structure, which requires 2 angles in perpendicular planes to define its 3D orientation in space. We identified the IAS plane at the FO in atrial end-systolic axial and coronal images, and measured the angle of the IAS in each image set relative to a perpendicular plane. We defined a line parallel to the CT table as 0° in the axial plane (Figure 2), and the patient’s feet as 0° in the coronal plane (Figure 3). As the TS needle is optimally positioned orthogonally to the IAS in the FO at the time of puncture, we subsequently calculated the ideal orientation of the needle for TSP. The orientation of the FO in the axial plane corresponds to the rotation of the TS needle catheter required to face the IAS, and conventionally is described in reference to a clock face. Using a line horizontal to the CT table as our 0° reference point; 0° corresponds to 3 o’clock, -90° to 6 o’clock, and -180° to 9 o’clock. The orientation of the FO in the coronal plane corresponds to the angulation of the TSP needle required for perpendicular positioning within the FO.

Measurements are reported as continuous variables and summarized as mean ± standard deviation and absolute range.


Patient baseline characteristics are presented in Table 1. The mean IAS angle in the axial plane was -60.6° ± 10.6° and an observed range of -29.5° to -88.7°. Using the clock-face orientation commonly employed to define TS puncture, the proper needle angle varied between 4 and 6 o’clock, with a mean proper target of 5 o’clock. In the coronal plane, the mean IAS angle was 142.6 ± 9.1°, with a range of 115° to 162°. The proper angle in the coronal plane, which corresponds to the curvature of the TS needle tip, was found varied by 47°. 

Axial and coronal angles were then graphed in order to look for clusters of measurements, and ascertain if patients within these clusters shared clinical features (Figure 4). We did not identify any relationship between the selected patient characteristics described in Table 1 and the measured IAS angles by univariate linear regression. Additionally, the patients who enriched the population did not cluster in one part of the graph nor provide the outliers.


We found significant interpatient variation in the angulation of the IAS in both the axial and coronal planes. While the mean IAS angle in the axial plane was -60.6°, there was a significant range, with observed angles measuring between -29.5° to -88.7°. Based on the observed axial IAS angulations, the proper orientation of the TS needle varied by almost 60°. Additionally, the mean angle in the coronal plane was 142.6 ±  9.1°. The coronal angulation predicts the optimal curvature of the TS needle, which varied by over 45°, with a range of 115° to 162°.

Contrary to our expectations, these measurements were not correlated with patient characteristics. We hypothesized that the measurements from patients who experienced no complications, and with normal left atrial diameters, would cluster near the mean observed angles. We also anticipated that the 8 patients selected to enrich the study population would make up the outliers in the data set. However, no such relationship was observed. Within the limits of this study, we did not observe a correlation with clinical variables such as procedural complications, age, gender, left ventricular ejection fraction, coronary artery disease, or LA size. We found a random distribution of the observed angle measurements, which failed to correlate with any clinical parameters. Further study with larger populations is needed to confirm these findings.

Despite technological advances, the risk of life-threatening complications during atrial fibrillation ablation remains between 1%-2%.18 Cardiac tamponade accounts for a substantial portion of this risk, and may sometimes be a result of TSP. Some efforts to make AF ablation safer have focused on TSP,16,19 including the development of virtual reality TSP simulators.11 ICE has been adopted by some centers as it allows for validation of proper needle position within the FO, visualization of septal tenting, confirmation of successful LA access, and rapid recognition of pericardial effusion. The drawbacks to ICE are the additional expense, procedural time, and the operator-dependent nature of ICE. The impact of ICE on ablation efficacy and incidence of pulmonary vein stenosis has been described by several studies.20,21 The effect of ICE on rates of tamponade has yet to be examined by large prospective trials. One prospective study of 53 patients undergoing AF ablation utilizing ICE reported no complications, and identified 4 patients in whom the tented FO abutted the LA wall, which required redirection of the TS needle to avoid puncture of the lateral LA wall.17 These findings suggest that using ICE during TSP may decrease rates of tamponade; however, no studies of ICE have been powered appropriately to answer this question.

Our review of 100 MDCTs demonstrated significant inter-patient variability in IAS anatomy. This may suggest that a “standardized” approach to TSP could have a higher incidence of perforation than when puncture is guided by ICE. However, many high-profile centers routinely perform TSP without ICE guidance, and have extremely low procedural complication rates. While it is speculated that ICE may reduce rates of tamponade, this has yet to be demonstrated by a large, prospective trial. Further inquiry into the role of ICE for TSP will provide further insight to this question.

Study limitations. This study has several limitations; notably, it is a single-center, retrospective review. The lack of randomization, and enrichment of the population with 8 complicated patients, may have skewed the measurements and resulted in overestimating IAS variability. Several clinical features that may impact IAS 3D spatial orientation were not considered, including height, body surface area (BSA), lung disease, and severe valvular disease. Given the rapid integration of MDCT-based electroanatomic systems, and the additional expense of ICE catheters, a prospective validation of these findings in a larger, randomly selected cohort is warranted.  


The ready availability of high-quality anatomic imaging from cardiac MDCT permits accurate characterization of IAS anatomy. In theory, MDCT could be utilized to anticipate catheter orientation and needle selection, thereby minimizing the need for expensive additional equipment, such as ICE and alternative transseptal needles. However, there are significant advantages to ICE that cannot be reproduced by MDCT, namely, verification of proper TSP position, and real-time assessment for procedural complications. The high degree of interpatient anatomic variability observed in this study, and previously described in others,9,10 reinforces the need for further studies on the role of advanced imaging modalities such as ICE during TS puncture.  


  1. De Ponti R, Cappato R, Curnis A, et al. Trans-septal catheterization in the electrophysiology laboratory: data from a multicenter survey spanning 12 years. J Am Coll Cardiol. 2006;47(5):1037-1042.
  2. Estner HL, Deisenhofer I, Luik A, et al. Electrical isolation of pulmonary veins in patients with atrial fibrillation: reduction of fluoroscopy exposure and procedure duration by the use of a non-fluoroscopic navigation system (NavX). Europace. 2006;8(8):583-587.
  3. Sporton SC, Earley MJ, Nathan AW, Schilling RJ. Electroanatomic versus fluoroscopic mapping for catheter ablation procedures: a prospective randomized study. J Cardiovasc Electrophysiol. 2004;15(3):310-315.
  4. Martinek M, Nesser HJ, Aichinger J, Boehm G, Purerfellner H. Impact of integration of multislice computed tomography imaging into three-dimensional electroanatomic mapping on clinical outcomes, safety, and efficacy using radiofrequency ablation for atrial fibrillation. Pacing Clin Electrophysiol. 2007;30(10):1215-1223.
  5. Bella PD, Fassini G, Cireddu M, et al. Image integration-guided catheter ablation of atrial fibrillation: a prospective randomized study. J Cardiovasc Electrophysiol. 2009;20(3):258-265. 
  6. Calkins H, Kuck KH, Cappato R, et al. 2012 HRS/EHRA/ECAS expert consensus statement on catheter and surgical ablation of atrial fibrillation: recommendations for patient selection, procedural techniques, patient management and follow-up, definitions, endpoints, and research trial design. Europace. 2012;14(4):528-606. 
  7. Cappato R, Calkins H, Chen SA, et al. Updated worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation. Circ Arrhythm Electrophysiol. 2010;3(1):32-38. 
  8. Hoyt, H, Bhonsale A, Chilukuir K, et al. Complications arising from catheter ablation of atrial fibrillation: temporal trends and predictors. Heart Rhythm. 2011;8(12):1869-1874.
  9. Reig J, Mirapeix R, Jornet A, Petit M.  Morphologic characteristics of fossa ovalis as an anatomic basis for transseptal catheterization. Surg Radiol Anat. 1997;19(5):279-282.
  10. Ho SY. Embryology and anatomy of the atrial septum. In: Transseptal Catheterization and Interventions. Thakur R, Natale A, eds. Minnesota: Cardiotext 2010:11-26. 
  11. De Ponti R, Marazzi R, Ghiringhelli S, Salerno-Uriarte JA, Calkins H, Cheng A. Superiority of simulator-based training compared with conventional training methodologies in the performance of transseptal catheterization. J Am Coll Cardiol. 2011;58(4):359-363.
  12. Saliba W, Thomas J. Intracardiac echocardiography during catheter ablation of atrial fibrillation. Europace. 2008;(10 Suppl 3):iii42-iii47. 
  13. Ren JF, Marchlinski FE. Utility of intracardiac echocardiography in left heart ablation for tachyarrhythmias. Echocardiography. 2007;24(5):533-540.
  14. Saad EB, Costa IP, Camanho LE. Use of intracardiac echocardiography in the electrophysiology laboratory. Arq Bras Cardiol. 2011;96(1):e11-e17.
  15. Cheng A, Calkins H. A conservative approach to performing transseptal punctures without the use of intracardiac echocardiography: stepwise approach with real-time video clips. J Cardiovasc Electrophysiol. 2007;18(6):686-689.
  16. Tzeis S, Andrikopoulous G, Deisenhofer I, Ho SY, Theodorakis G. Transseptal catheterization: considerations and caveats. Pacing Clin Electrophysiol. 2010;33(2):231-242.
  17. Daoud EG, Kalbfleisch SJ, Humm JD. Intracardiac echocardiography to guide transseptal left heart catheterization for radiofrequency catheter ablation. J Cardiovasc Electrophysiol. 1999;10(3):358-363.
  18. Calkins H. Catheter ablation to maintain sinus rhythm. Circulation. 2012;125(11):1439-1445.
  19. Hanaoka T, Suyama K, Taguchi A, et al. Shifting of puncture site in the fossa ovalis during radiofrequency catheter ablation: intracardiac echocardiography guided transseptal left heart catheterization. Jpn Heart J. 2003;44(5):673-680.
  20. Ren JF, Marchlinski FE, Callans DJ, Zado ES. Intracardiac doppler echocardiographic quantification of pulmonary vein flow velocity: an effective technique for monitoring pulmonary vein ostia narrowing during focal atrial fibrillation ablation. J Cardiovasc Electrophysiol. 2002;13(11):1076-1081.
  21. Marrouche NF, Martin DO, Wazni O, et al. Phased-array intracardiac echocardiography monitoring during pulmonary vein isolation in patients with atrial fibrillation: Impact on outcome and complications. Circulation. 2003;107(21):2710-2716.

From the 1Department of Medicine and 4Knight Cardiovascular Institute, Oregon Health and Science University, Portland, Oregon, 2Division of Cardiology, Johns Hopkins Hospital, Baltimore, Maryland, and 3National Institutes of Health Clinical Center, Bethesda, Maryland.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Henrikson reports a minor research grant from Biosense Webster; Dr Nazarian reports research grant and consultant fees from Biosense Webster. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript submitted October 24, 2013, provisional acceptance given November 11, 2013, final version accepted November 12, 2013.

Address for correspondence: Charles Henrikson, MD, Knight Cardiovascular Institute, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, Portland OR 97239. Email: