Case Study

A Novel Way of Managing an Aortic Root Perforation at Transseptal Puncture

Kevin A. Michael, MBChB, MD, Damian P. Redfearn, MD, Adrian M. Baranchuk, MD, Rob G. Dhillon*, MD, Benedict M. Glover, MD, Heart Rhythm Service, Queen`s University, Kingston General Hospital, Kingston, Ontario; *Imaging Services, Kingston General Hospital, Kingston, Ontario


Kevin A. Michael, MBChB, MD, Damian P. Redfearn, MD, Adrian M. Baranchuk, MD, Rob G. Dhillon*, MD, Benedict M. Glover, MD, Heart Rhythm Service, Queen`s University, Kingston General Hospital, Kingston, Ontario; *Imaging Services, Kingston General Hospital, Kingston, Ontario


The transseptal puncture (TS) is very much a routine procedure to the interventional electrophysiologist. The following case illustrates an inadvertent puncture into the aortic root at TS, as well as measures that can be taken to avoid a residual aorto-atrial shunt and the possible need for surgical repair.

Case Presentation

A 69-year-old male was considered for a redo left atrial ablation for longstanding persistent atrial fibrillation (AF). The left atrium (LA) was severely dilated (LA volume index = 64.8 ml/m2). The initial procedure was suboptimal with failure to isolate all pulmonary veins. Therefore, it seemed reasonable to commit to a repeat ablation after recurrence of AF. In the interim, the patient had been compliant with continuous positive airway pressure (CPAP) therapy for his obstructive sleep apnea, as well as with propafenone and bisoprolol. The second ablation was attempted on uninterrupted warfarin with a therapeutic INR.

The procedure was performed under conscious sedation. The TS was attempted with a 71 cm extra sharp Brockenbrough (BRK Transseptal Needle, XS Series, St. Jude Medical) needle through an 8.5 French, 63 cm SL1 sheath (St. Jude Medical). A guidewire was advanced to the left subclavian vein and replaced with the BRK needle under fluoroscopy. The sheathed needle was withdrawn under fluoroscopy in a conventional LAO projection until a posterior movement (or drop) was observed toward the fossa ovale (FO). The location was noted to be posterior to the His catheter, which provided a surrogate marker for the location of the aortic root with RAO and LAO projections. An injection of contrast following puncture through the BRK needle suggested inadvertent puncture into the aortic root; withdrawal of the BRK needle confirmed arterial flow from the dilator. Thus, the tip of the dilator was determined to be within the aortic root (Figure 1 and video).

The dilator and introducer were not advanced any further, and the patient remained pain free and hemodynamically stable. A ViewFlex Xtra intracardiac echocardiography (ICE) catheter (St. Jude Medical) was introduced into the right atrium (RA) through an additional femoral vein puncture. The site of the aortic puncture was visualized with the aid of color flow Doppler, demonstrating a puncture into the non-coronary cusp (NCC) of the aortic root (Figure 2 and video). There was no communication into the pericardial space, and therefore, no pericardial effusion. The cardiothoracic surgeon was notified, and a thoracotomy tray was prepared in readiness.

A guidewire was then advanced into the dilator, and the passage of the wire was visualized on fluoroscopy and on ICE, confirming the location of the puncture (Figure 3 and video). The tip of the dilator was then advanced over the wire up to the point of the site of puncture into the aortic root, and the wire was withdrawn with pressure applied to the puncture site using the tip of the SL1 dilator. 

Next, 20 ml of prothrombin complex concentrate as well as 10 mg of intravenous vitamin K were administered to reverse the anticoagulant effect of coumadin. Labetalol 10-20 mg bolus doses were given to initially lower the blood pressure from 180/110 mmHg to a systolic blood pressure of 110/80 mmHg, together with sodium nitroprusside at an infusion rate of 0.25-0.3 mcg/min. Thereafter, labetalol was titrated (0.5-8 mg/min) to maintain a systolic blood pressure of 100 mmHg (intra-arterial pressure) and sodium nitroprusside was stopped.

Constant manual pressure on the puncture site was achieved through the SL1 dilator and sheath compressed against the puncture site after withdrawal of the guidewire, but not protruding into the aortic sinus of Valsalva, as visualized by ICE. The flow was monitored with color Doppler on ICE. The leak from the NCC to the RA was observed to diminish to minor blushes during systole, until complete resolution over 30 minutes.

Compressive pressure was eventually reduced, and the puncture site was carefully monitored. No further flow through the puncture site was observed.

Catheters were withdrawn from the heart, but central access was maintained and intra-arterial monitoring was continued. An indwelling urine catheter was left in place to monitor renal perfusion via urine output. Labetalol was continued in order to maintain a systolic pressure of ≤100 mmHg. Since no access was obtained into the LA, heparin was not administered and no left atrial ablation was undertaken, making re-anticoagulation less urgent. Warfarin was held for 48 hours. A transthoracic echocardiogram the next morning showed no evidence of an aorto-atrial fistula.


The overall complication rate of TS was 0.74% in a multicenter Italian survey comprising of 5,520 TS procedures.1 Factors that increase the risk of complications include operator experience, older patients, larger atrial size, and patients having had prior TS procedures. Anatomical aneurysmal characteristics, a fibrous inter-atrial septum, and aortic root dilatation are also factors. In their series, Bayrak and colleagues showed only three major TS complications in 205 patients undergoing a PVI.2 In their assessment, when inexperienced trainees used transesophageal echocardiography (TEE) to assess the need for repositioning the TS needle, the most common malposition was with the BRK needle directed toward the aortic root. This occurred more often than a more posteriorly orientated needle that potentially could perforate the posterior wall of the LA.

More often, an inadvertent BRK needle entry into the aortic root has no significant sequalae but may occasionally lead to an aorto-atrial shunt.3 It is therefore best not to advance the dilator or sheath until the chamber accessed by the needle is clearly confirmed. This may be achieved by echocardiography, intracardiac echo (ICE), TEE, or by pressure monitoring of the needle. Inserting a pigtail catheter into the aortic root or an electrophysiology catheter positioned at the maximal His deflection provides anatomical markers on fluoroscopy to guide the puncture. The main technique employed in most electrophysiology laboratories is a reliance upon the His signal as a surrogate locator for the aortic root, the behavior of the needle at the drop, and the characteristic contrast staining of the FO. 

The overall low complication rate in prior studies suggests that this is adequate, but in inexperienced hands and particularly in training centers, complication rates are higher.4 The complications incurred at TS puncture are also potentially life-threatening.

Saliba et al advocated the advantages of using ICE during TS puncture in their review on the topic. They emphasize the additional usefulness in monitoring for early tamponade or for thrombus, char formation, and microbubbles during ablation.5 

The sequence of cardiac CT images taken prior to the procedure (Figure 4) helps one appreciate the limited target area for a safe transseptal puncture and the proximity to the aortic sinus of Vasalva. There was an increase in thickness of the anterior limbus surrounding the FO. This is more commonly a consequence of lipomatous infiltration, thus resulting in posterior displacement of the membranous interatrial septum, i.e., the anticipated “safe area” for a TS puncture. By inference, one can assume that the attempted puncture was therefore too anterior and superior. The standard BRK needle and sheath orientation to 4 or 5 o’clock was not adequate in this patient. Presumably, a posteriorly turned needle (facing 6 o’clock) would have been more appropriate. One can appreciate that these nuances of the anatomy of the interatrial septum are beyond the resolution of standard fluoroscopy, and best appreciated with the assistance of ICE or TEE. 


The risk of an aortic root puncture at TS puncture is uncommon. Despite this, if the complication does occur, one should avoid the temptation to immediately withdraw the sheath and needle. The dilator and sheath can be used as a means of compressing the puncture site. The use of ICE or TEE can avoid such a complication, but also can be used to visualize the perforation with color Doppler, and thus monitor the effectiveness of the compression. 

Disclosures: The authors have no conflicts of interest to report regarding the content herein.   

Editor’s Note: This article underwent peer review by one or more members of EP Lab Digest®’s editorial board.


  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. Bayrak F, Chierchia GB, Namdar M, et al. Added value of transesophageal echocardiography during transseptal puncture performed by inexperienced operators. Europace. 2012;14(5):661-665.
  3. Roelke M, Smith AJC, Palacios IF. The technique and safety of transseptal left heart catheterization: the Massachusetts General Hospital experience with 1,279 procedures. Cathet Cardiovasc Diagn. 1994;32(4):332-339.
  4. Filgueiras-Rama D, Torres-Alba F, Castrejón-Castrejón S, et al. Utility of intracardiac echocardiography for catheter ablation of complex cardiac arrhythmias in a medium-volume training center. Echocardiography. 2014;32(4):660-670.
  5. Saliba W, Thomas J. Intracardiac echocardiography during catheter ablation of atrial fibrillation. Europace. 2008;10(Suppl 3):iii42-47.
  6. Tsang MY, Hagler DJ, Dearani JA, Rihal CS, Anavekar NS. Aorto-right atrial fistula: a rare complication of trans-septal puncture and catheter ablation for atrial fibrillation. Eur Heart J Cardiovasc Imaging. 2014;15(11):115.