Abstract: Closure of congenital atrial communications in the presence of either severe pulmonary arterial hypertension (PAH) with pulmonary-to-systemic (right-to-left) shunting, or severe left ventricular (LV) non-compliance with left-to-right shunting is often considered prohibitive. Thus, the recognition of durable reversibility of these physiologic conditions is crucial. We describe a hemodynamic conundrum in a patient with five septal communications in whom the coexistence of unmasked bidirectional physiologic shunting, severe PAH, and worsening left-sided overload dissuaded initial closure. We report our strategy for hemodynamic evaluation and successful closure of all defects.
J INVASIVE CARDIOL 2015;27(4):E51-E55
Key words: atrial septal defect, pulmonary hypertension
Older patients with atrial septal defects (ASDs) who have right heart diastolic and/or systolic failure, severe pulmonary arterial hypertension (PAH), age-related degeneration in left ventricular (LV) compliance, or a combination thereof, often carry a prohibitive risk of closure.1 However, closure of these defects (even with significant right-to-left [R-L] shunting) can still be considered safe and effective if appropriate physiologic conditions are defined invasively.1,2 Herein, we describe the rare and challenging case of a patient with five separate atrial communications, including a large aneurysmal patent foramen ovale (PFO), who suffered from coexisting severe PAH and severe left atrial (LA) overload with bidirectional interatrial shunts. We provide details of this hemodynamic dilemma and how it was studied sequentially using temporary balloon occlusions (TBOs) and pharmacologic challenges before successful transcatheter closure of all defects was possible using three overlapping nitinol mesh occlusion devices.
A 63-year-old male was referred for evaluation of pulmonary hypertension and progressively worsening dyspnea now occurring at rest, systemic deoxygenation (02 saturation on 4 L 02 = 88%), worsening lower-extremity edema, and New York Heart Association (NYHA) class-IV heart failure symptoms. He also had a history of multiple transient ischemic attacks. Transesophageal echocardiography (TEE) was performed to better define the etiology of his condition (Table 1).
Based on the TEE, his severe PAH was thought likely to be secondary to left-to-right (L-R) shunting at the atrial level as a result of multiple potentially treatable atrial septal communications. The patient underwent complete right and left heart catheterization with intracardiac echocardiography (ICE; Biosense Webster) and his baseline catheterization data are shown in Table 2. Catheterization data revealed markedly elevated right-sided filling pressures, with RV systolic pressure (RVSP) ~2/3 systemic and confirmed precapillary severe PAH (transpulmonary gradient measured 26 mm Hg).
Baseline oxygen saturation data obtained from catheterization revealed systemic desaturation (88%) and net L-R shunting with a pulmonary-to-systemic blood flow ratio (Qp/Qs) of 1.5, indicating masking of a coexisting R-L shunt at the level of the ASDs. Administration of inhaled nitric oxide at 80 ppm for 15 minutes led to a 2-fold increase in L-R shunting (DQp/Qs = 3.1), partial functional removal of the R-L shunt (systemic saturation = 91%), and a favorable hemodynamic response. Thus, we proceeded with simultaneous TBO of the two largest septal defects (posterior and anterior), inflating two 24 mm sizing balloons to their respective “stop-flow” diameters to evaluate the hemodynamic response both with and without vasodilator challenge. Without vasodilator challenge, TBO resulted in a systemic RVSP, symptomatic and significant worsening of both right- and left-sided filling pressures (right and LA pressures, 28 and 25 mm Hg, respectively), and a concomitant >33% drop in cardiac output that persisted for 15 minutes, signifying irreversible PAH. Fortunately, readministration of nitric oxide in combination with TBO achieved a remarkably favorable response (Table 1), as evidenced by a reduction in mean pulmonary artery pressure (mPAP) of >10 mm Hg to <40 mm Hg,1,3 a significant increase in cardiac output (~55%), and a steady increase in mean atrial pressure (MAP), which effectively ruled out fixed PAH. Accordingly, we proceeded with closure of all defects.
Technique. The large posterior 20 mm defect was crossed first (Figure 1). As our initial plan was to intercalate the large anterior and two small fenestrated central defects between the overlapping atrial discs from a large (22 mm) Amplatzer septal occluder (ASO; St. Jude Medical) and a larger (35 mm) Cribriform ASO, a second 0.35˝ wire was placed across the large PFO from a second vessel sheath within the right groin (Figure 1). Without releasing the posterior device, care was taken to deploy the Cribriform device in an overlapping fashion to capture the right and left ASO atrial discs (Figure 2). Unfortunately, effective nesting of the discs was not possible given the geometry of the largest anterior-superior defect, and upon release of the second device, a significant degree of shunting persisted across this defect (Figure 2). Therefore, resizing of the anterior-superior defect was performed both to determine whether occlusion would result in acceptable reduction in shunting with the other devices in place and to determine at what diameter this would occur (Figure 3). A subtle waist was produced with the 24 mm sizing balloon and sizing with a larger balloon carried the risk of dislodging one or both of the implanted devices. Therefore, oversizing of the third device by 2 mm was predicted to not only cover the defect but also nest the Cribriform device inside the discs of the third ASO. Figure 3 demonstrates final deployment of the third device, sandwiching the two interlacing devices while purposely nesting the Cribriform within the 26 mm ASO for complete closure of all defects including the two fenestrated defects. The post-closure catheterization data are summarized in Table 1.
Medical therapy. The patient was immediately started on the phosphodiesterase-5 inhibitor sildenafil citrate (Pfizer) at a dose of 12.5 mg three times daily. We titrated sildenafil to 25 mg three times daily over the next 24 hours while weaning off the inhaled nitric oxide after 48 hours. Over the next 4 days, the patient’s oxygen saturations remained above 97% on room air with desaturations to no lower than 95% with ambulation and symptomatic improvement from NYHA class-IV symptoms to class I-II (home 02 was discontinued). He was discharged 4 days after defect closures on sildenafil 25 mg three times daily, 81 mg of aspirin, and 75 mg of clopidogrel daily.
Follow-up occurred 2 weeks, 4 weeks, 6 weeks, and 6 months after discharge, during which time his sildenafil was titrated to 50 mg three times daily. His NYHA class I-II symptoms persisted off of home 02. On follow-up, Table 3 demonstrates persistent severe biatrial enlargement, albeit with significant reduction in RV dimensions and RVSP (consistent with invasive hemodynamic data following closure of all defects) and recovery in RV function to super-normal levels by 6 months. Sildenafil was continued at 50 mg three times daily indefinitely.
The presence of large or multiple atrial septal communications in combination with severe PAH generally tends to preclude surgical or transcatheter closure.1 All of these characteristics present in our patient posed both structural and hemodynamic dilemmas. Reticence for closure of these defects is due primarily to the increased risk of hemodynamic unpredictability following removal of the shunt(s). The risk of hemodynamic collapse following shunt closure(s) is particularly high in patients like ours, who had RV enlargement and poor RV systolic function.4 In patients with net pulmonary-to-systemic (R-L) shunting, removal of the relief valve (ASD) for the RV removes the necessary decompression mechanism via the atrial defect(s). This often leads to a pulmonary hypertensive crisis and may ultimately result in death if the relief valve is not restored.4 Likewise, in patients with LV non-compliance and large L-R shunts, removal of the left-sided decompression valve can lead to sudden afterload mismatch, acute cardiogenic pulmonary edema, and hemodynamic collapse. Temporary interruption of shunts using TBO within congenital heart defect(s) in conjunction with concomitant volume and/or pressure loading maneuvers and/or administration of systemic and pulmonary vasoreactive agents has been shown to provide crucial insight into post-closure physiologic conditions.2,5,6
Almost 90% of patients present with single ASDs, yet the existence of multiple or fenestrated defects has been reported in 10%-13% of patients3 and approximately 10% of patients present with a PFO accompanying an ASD.7 Although the technical difficulty of closing multiple defects may be considered a deterrent to transcatheter treatment, a few studies report satisfactory long-term results using more than one Amplatzer device.3,8,9 There has been demonstrated efficacy for the use of the so-called “overlapping” or “interlacing” techniques for closure of defects separated by less than 5-7 mm; in very contiguous defects, it is recommended to implement the “sandwich” technique, whereby delivery of a smaller device first ensures complete disk opening and correct positioning.3,9
Our patient had five septal fenestrations, including a large aneurysmal PFO in the setting of systemic deoxygenation, LV non-compliance, severe PAH, RV enlargement, and severely depressed RV systolic function at baseline. After demonstrating vasodilator responsiveness with inhaled nitric oxide and improving systemic deoxygenation by reversing his underlying R-L shunt, subsequent TBO led to worsening of his PAH and unmasked coexisting severe LA overload. With lasting improvement in systemic oxygenation and stabilization of right-sided and left-sided hemodynamic conditions during sequential TBO and vasodilator challenge, the decision was made to close all defects. We employed both interlacing and sandwiching techniques, with eventual success using a total of three Amplatzer devices for complete closure of all five communications. We further proved efficacy for this approach as evidenced by both durability of closure and significant hemodynamic and symptomatic improvement. Closure of multiple fenestrations with one or two devices is not uncommon,7,10 and is successful in approximately one-half of these cases.3 Only two reports exist describing the successful use of three Amplatzer closure devices3,9 and no study to date has demonstrated efficacy for the closure of five fenestrations or for the use of more than two of any of the other approved closure systems.11
Whether oral vasodilator therapy would provide effective medical therapy following shunt closure was unknown in our patient at the time of defect closures. Data are limited regarding the safety and efficacy of shunt closure in patients with the diagnosis of severe PAH, regardless of reversibility, and the risk-to-benefit outcomes from closure remain controversial.5 Medical therapy has been standardized for the treatment of severe PAH associated with congenital heart disease.4 Specifically, oral pulmonary vasodilator therapy using the 5-phosphodiesterase inhibitor sildenafil citrate alone or in combination with the oral endothelin receptor antagonist bosentan has been shown to induce favorable remodeling of pulmonary arteriopathies and provide symptomatic improvement.6,12,13 Although few data were available to guide our management of this unique case, we felt that long-term oral medical therapy in the form of oral pulmonary vasodilators would have the potential to improve hemodynamics and symptoms in our patient.
In summary, sequential and combined testing with TBO and vasodilator administration altered the hemodynamics in our patient in a rather unpredictable manner. We demonstrated a rapid and enduring drop in mean pulmonary artery pressure (pulmonary vascular resistance) and left atrial pressure with initial R-L shunt reversal and final shunt removal. Assuming effective vasoreactivity of the pulmonary vascular bed with LV afterload, our decision to close was made with the intent of continuing oral vasodilator therapy to: (1) effectively treat his PAH and prevent further pulmonary vascular remodeling; (2) ameliorate oxygen desaturations; and (3) ultimately improve his symptoms and functional status. This is the first description of transcatheter closure of five atrial communications using three separate ASD occluder devices.
The physiologic principles governing dynamic intracardiac flow in the presence of congenital intracardiac shunts are unpredictable and depend in large part on competition between atrial and ventricular filling pressures. Based on our unique and challenging case and on previous reports,2,3,7,14 TBO with vasodilator challenge should be considered prerequisites for safe and effective repair of ASDs when severe PAH is present and pulmonary vasoreactivity is questionable.
Acknowledgment. We would like to personally thank John Begovich and Brad Smith for their ongoing commitment to our structural heart disease program.
- Steele PM, Fuster V, Cohen M, Ritter DG, McGoon DC. Isolated atrial septal defect with pulmonary vascular obstructive disease — long-term follow-up and prediction of outcome after surgical correction. Circulation. 1987;76(5):1037-1042.
- Javed U, Levisman J, Rogers JH. A tale of two balloons: assessment of hemodynamics with atrial septal defect temporary balloon occlusion. J Invasive Cardiol. 2012;24(5):248-229.
- Butera G, Romagnoli E, Saliba Z, et al. Percutaneous closure of multiple defects of the atrial septum: procedural results and long-term follow-up. Catheter Cardiovasc Interv. 2010;76(1):121-128.
- Dimopoulos K, Peset A, Gatzoulis MA. Evaluating operability in adults with congenital heart disease and the role of pretreatment with targeted pulmonary arterial hypertension therapy. Int J Cardiol. 2008;129(2):163-171. Epub 2008 Mar 26.
- Huang ZW, Fan ZX, Sun JT, et al. The short- and medium-term results of transcatheter closure of atrial septal defect with severe pulmonary arterial hypertension. Heart Vessels. 2012;27(6):603-609. Epub 2011 Sep 17.
- Lewis GD, Bloch KD, Semigran MJ. Pulmonary thromboembolism superimposed on a congenital ventricular septal defect in a 50-year-old man inhaled nitric oxide and sildenafil to the rescue. Cardiol Rev. 2004;12(4):188-190.
- Khositseth A, Cabalka AK, Sweeney JP, et al. Transcatheter Amplatzer device closure of atrial septal defect and patent foramen ovale in patients with presumed paradoxical embolism. Mayo Clin Proc. 2004;79(1):35-41.
- Cao Q, Radtke W, Berger F, Zhu W, Hijazi ZM. Transcatheter closure of multiple atrial septal defects. Initial results and value of two- and three-dimensional transoesophageal echocardiography. Eur Heart J. 2000;21(11):941-947.
- Mahadevan VS, Gomperts N, Haberer K, et al. Transcatheter closure of atrial septal defects with multiple devices in adults: procedural and clinical outcomes. Int J Cardiol. 2009;133(3):359-363. Epub 2008 May 21.
- Awad SM, Garay FF, Cao QL, Hijazi ZM. Multiple Amplatzer septal occluder devices for multiple atrial communications: immediate and long-term follow-up results. Catheter Cardiovasc Interv. 2007;70(2):265-273.
- Freixa X, Garceau P, Asgar AW. First experience with the new GORE septal occluder for the closure of multiple atrial septal defects. Catheter Cardiovasc Interv. 2013;81(7):1238-1242.
- Lim ZS, Salmon AP, Vettukattil JJ, Veldtman GR. Sildenafil therapy for pulmonary arterial hypertension associated with atrial septal defects. Int J Cardiol. 2007;118(2):178-182.
- Zeng WJ, Lu XL, Xiong CM, et al. The efficacy and safety of sildenafil in patients with pulmonary arterial hypertension associated with the different types of congenital heart disease. Clin Cardiol. 2011;34(8):513-518. Epub 2011 Jun 15.
- Opotowsky AR, Landzberg MJ, Kimmel SE, Webb GD. Trends in the use of percutaneous closure of patent foramen ovale and atrial septal defect in adults, 1998-2004. JAMA. 2008;299(5):521-522.
From 1the Division of Cardiovascular Medicine, University of Louisville of School of Medicine and Jewish Hospital, Heart and Lung Institute, Louisville, Kentucky; and 2the Division of Cardiology, Johns Hopkins University, Baltimore, Maryland.
Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.
Manuscript submitted June 17, 2014, provisional acceptance given July 23, 2014, final version accepted September 4, 2014.
Address for correspondence: Michael P. Flaherty, MD, PhD, FACC, FSCAI, Assistant Professor of Medicine, Physiology & Biophysics, Division of Cardiovascular Medicine, University of Louisville School of Medicine, Rudd Heart and Lung Center, 201 Abraham Flexner Way, Suite 800, Louisville, KY 40202. Email: email@example.com