Physiology-Guided Closure of Iatrogenic Atrial Septal Defect After Paravalvular Leak Repair Authors: Yusuke Samejima, Go Hashimoto, Yutsuki Tsukagoshi, Atsuhiro Saito, Sari Waga, Shiho Ide, Kenji Makino, Hiroki Niikura, Nobutaka Ikeda, Hidehiko Hara Published: 2026 ABSTRACT ObjectiveTo describe a physiology-guided approach to evaluating and closing a clinically significant iatrogenic atrial septal defect after transseptal structural heart intervention.Key StepsThe approach included echocardiographic identification of right-to-left shunting, invasive assessment of interatrial pressure gradients and pulmonary vascular load, inhaled nitric oxide as rescue therapy and an adjunctive physiologic maneuver, 3-dimensional assessment of defect morphology, and urgent transcatheter closure.Potential PitfallsInhaled nitric oxide should not be used as a stand-alone determinant of closure. Closure decisions should integrate shunt direction, interatrial pressure gradient, oxygenation, right-sided loading conditions, and whether the defect may serve as left atrial decompression.Take-Home MessagePostprocedural iatrogenic atrial septal defect closure should be individualized according to physiology rather than anatomy alone. Visual SummaryPhysiology-Guided Closure of Iatrogenic Atrial Septal Defect After Paravalvular Leak RepairSchematic summary of the clinical course, imaging findings, invasive hemodynamics, and physiologic decision-making process in a patient who developed refractory hypoxemia after transcatheter paravalvular leak closure. (Left) Clinical course demonstrating development of refractory hypoxemia after paravalvular leak closure, inhaled nitric oxide challenge, subsequent transcatheter iatrogenic atrial septal defect (iASD) closure, and clinical recovery. (Upper Right) Imaging and hemodynamic findings demonstrating predominant right-to-left interatrial shunting with right atrial pressure exceeding left atrial pressure at baseline, followed by improvement in oxygenation and reversal toward predominant left-to-right shunting after inhaled nitric oxide administration. (Lower Right) Proposed physiologic mechanism and clinical interpretation. Elevated right-sided loading conditions promoted pathologic right-to-left shunting through the iASD. Reduction in right-sided loading conditions after inhaled nitric oxide helped characterize dynamic shunt physiology before transcatheter iASD closure. LA = left atrium; NO = nitric oxide; PVL = paravalvular leak; RA = right atrium. Physiology-Guided Closure of Iatrogenic Atrial Septal Defect After Paravalvular Leak Repair Schematic summary of the clinical course, imaging findings, invasive hemodynamics, and physiologic decision-making process in a patient who developed refractory hypoxemia after transcatheter paravalvular leak closure. (Left) Clinical course demonstrating development of refractory hypoxemia after paravalvular leak closure, inhaled nitric oxide challenge, subsequent transcatheter iatrogenic atrial septal defect (iASD) closure, and clinical recovery. (Upper Right) Imaging and hemodynamic findings demonstrating predominant right-to-left interatrial shunting with right atrial pressure exceeding left atrial pressure at baseline, followed by improvement in oxygenation and reversal toward predominant left-to-right shunting after inhaled nitric oxide administration. (Lower Right) Proposed physiologic mechanism and clinical interpretation. Elevated right-sided loading conditions promoted pathologic right-to-left shunting through the iASD. Reduction in right-sided loading conditions after inhaled nitric oxide helped characterize dynamic shunt physiology before transcatheter iASD closure. LA = left atrium; NO = nitric oxide; PVL = paravalvular leak; RA = right atrium. Iatrogenic atrial septal defects (iASDs) are common after transseptal structural heart interventions and are usually clinically tolerated.1, 2, 3 However, closure decisions can be challenging because an iASD may either become pathologic, causing right-to-left shunting and refractory hypoxemia, or provide left atrial decompression in patients with elevated left-sided filling pressures.4, 5, 6 This report describes a stepwise physiology-guided approach to evaluating and closing a clinically significant iASD after transcatheter mitral paravalvular leak (PVL) repair.Take-Home Messages•Not all iASDs after transseptal structural intervention require closure; management should be guided by hemodynamic physiology rather than anatomy alone.•Inhaled nitric oxide should not be used as a stand-alone determinant of iASD closure, but may provide adjunctive physiologic information in selected patients with dynamic right-to-left shunting. Take-Home Messages •Not all iASDs after transseptal structural intervention require closure; management should be guided by hemodynamic physiology rather than anatomy alone.•Inhaled nitric oxide should not be used as a stand-alone determinant of iASD closure, but may provide adjunctive physiologic information in selected patients with dynamic right-to-left shunting. •Not all iASDs after transseptal structural intervention require closure; management should be guided by hemodynamic physiology rather than anatomy alone. Not all iASDs after transseptal structural intervention require closure; management should be guided by hemodynamic physiology rather than anatomy alone. •Inhaled nitric oxide should not be used as a stand-alone determinant of iASD closure, but may provide adjunctive physiologic information in selected patients with dynamic right-to-left shunting. Inhaled nitric oxide should not be used as a stand-alone determinant of iASD closure, but may provide adjunctive physiologic information in selected patients with dynamic right-to-left shunting. Case Summary A patient in his 60s underwent transcatheter mitral PVL closure for symptomatic heart failure and hemolytic anemia because redo surgery was considered prohibitively high risk. Transseptal puncture was performed through a 7-F femoral venous sheath, followed by exchange to an 8.5-F Agilis steerable sheath (Abbott) for mitral PVL crossing. A paravalvular leak device (PLD occluder 63PLD07T, Occlutech) was delivered using an 8-F Occlutech delivery sheath III because of limited deliverability; balloon dilatation of the interatrial septum was not performed. The patient had undergone mechanical aortic and mitral valve replacement 29 years earlier. His medical history included chronic atrial fibrillation, chronic obstructive pulmonary disease, and symptomatic mitral PVL. Baseline transthoracic echocardiography before PVL closure showed a left ventricular ejection fraction of 53% and severe mitral PVL. The right ventricle was not dilated, but right ventricular systolic function was impaired, with a tricuspid annular plane systolic excursion of 11 mm, fractional area change of 23.2%, and tissue Doppler S′ velocity of 8.4 cm/s. Severe tricuspid regurgitation (TR) with right atrial (RA) enlargement was present; the estimated right ventricular systolic pressure and transtricuspid pressure gradient were 53 mm Hg and 45 mm Hg, respectively. The patient was initially extubated after PVL closure; however, several days later, he developed progressive respiratory failure and refractory hypoxemia requiring reintubation. This deterioration occurred in the setting of aspiration pneumonia, worsening renal failure requiring continuous hemodiafiltration, and hemodynamic instability. Transesophageal echocardiography (TEE) demonstrated an iASD with predominant right-to-left (R-L) shunting on color Doppler imaging (Figure 1A, Video 1). The severe TR jet was directed toward the iASD, which may have promoted R-L shunting. Continuous-wave Doppler showed sustained R-L flow throughout nearly the entire cardiac cycle (Figure 1B). Three-dimensional TEE and multiplanar reconstruction demonstrated a tear-type iASD morphology and were used to assess defect geometry for transcatheter closure (Figure 2).Figure 1Hemodynamically Significant Iatrogenic Atrial Septal Defect With Sustained Right-to-Left Shunting(A) Transesophageal echocardiography (midesophageal bicaval view, color Doppler imaging) demonstrating predominant right-to-left shunting across the iatrogenic atrial septal defect. (B) Continuous-wave Doppler recording across the interatrial septum demonstrating sustained right-to-left flow throughout nearly the entire cardiac cycle, suggesting hemodynamically significant shunting.Figure 2Morphologic Assessment of the Iatrogenic Atrial Septal Defect for Transcatheter Closure Planning(A) Three-dimensional transesophageal echocardiography viewed from the right atrial perspective demonstrating a tear-type or slit-like morphology of the iatrogenic atrial septal defect. (B) Multiplanar reconstruction imaging used to assess defect geometry and procedural feasibility for transcatheter closure. Hemodynamically Significant Iatrogenic Atrial Septal Defect With Sustained Right-to-Left Shunting (A) Transesophageal echocardiography (midesophageal bicaval view, color Doppler imaging) demonstrating predominant right-to-left shunting across the iatrogenic atrial septal defect. (B) Continuous-wave Doppler recording across the interatrial septum demonstrating sustained right-to-left flow throughout nearly the entire cardiac cycle, suggesting hemodynamically significant shunting. Morphologic Assessment of the Iatrogenic Atrial Septal Defect for Transcatheter Closure Planning (A) Three-dimensional transesophageal echocardiography viewed from the right atrial perspective demonstrating a tear-type or slit-like morphology of the iatrogenic atrial septal defect. (B) Multiplanar reconstruction imaging used to assess defect geometry and procedural feasibility for transcatheter closure. Invasive hemodynamic assessment showed that mean RA pressure exceeded mean left atrial (LA) pressure (32 vs 28 mm Hg), favoring R-L shunting. Pulmonary arterial pressure was 82/32 mm Hg (mean: 50 mm Hg), and pulmonary vascular resistance was 8.0 Wood units. Inhaled nitric oxide (NO) was administered through the ventilator circuit as rescue therapy and as an adjunctive physiologic maneuver to assess dynamic shunt behavior under pulmonary vasodilation. Shortly after NO initiation, oxygenation improved, and TEE demonstrated reversal of shunt direction toward predominant left-to-right flow (Figure 3, Videos 2 and 3). Hemodynamic data are summarized in Table 1.Figure 3Dynamic Reversal of Interatrial Shunting After Inhaled Nitric Oxide Administration(A) Transesophageal echocardiography obtained 2 minutes after initiation of inhaled nitric oxide demonstrating reduction of right-to-left shunting with emergence of left-to-right flow across the iatrogenic atrial septal defect. (B) At 6 minutes after nitric oxide initiation, color Doppler and continuous-wave Doppler imaging demonstrated predominant left-to-right shunting. These findings suggest that reversible right-sided loading conditions contributed to the abnormal shunting physiology.Table 1Hemodynamic Data Before and After Inhaled Nitric Oxide and After Transcatheter iASD ClosureBaselineAfter Inhaled Nitric OxideAfter iASD ClosureRA pressure, mm Hg32 (a/v/m: 60/47/32)—20 (a/v/m: 26/27/20)RV pressure, mm Hg87/14 (EDP: 26)—71/8 (EDP: 18)PA pressure, mm Hg82/32 (mean: 50)76/29 (mean: 47)71/28 (mean: 45)PCWP, mm Hg31/38 (mean: 30)—29/35 (mean: 27)LA mean pressure, mm Hg28—27Aortic pressure, mm Hg96/41—92/60Cardiac output (Fick), L/min6.27——Cardiac index, L/min/m24.09——Qp/Qs0.53——PVR, WU8.0—6.5PVRI, WU·m212.2—10.0Shunt direction on TEEPredominantly right-to-leftReversal toward left-to-rightDefect closedClinical oxygenationRefractory hypoxemiaImprovedImprovedBaseline hemodynamics showed elevated right-sided pressures with right atrial pressure exceeding left atrial pressure, consistent with a condition favoring right-to-left shunting. Inhaled nitric oxide reduced pulmonary artery pressure and was associated with reversal of shunt direction on transesophageal echocardiography. After transcatheter iASD closure, right-sided pressures decreased further, with concomitant clinical improvement in oxygenation.EDP = end-diastolic pressure; iASD = iatrogenic atrial septal defect; LA = left atrium; PA = pulmonary artery; PCWP = pulmonary capillary wedge pressure; PVR = pulmonary vascular resistance; PVRI = pulmonary vascular resistance index; Qp/Qs = pulmonary/systemic flow ratio; RA = right atrium; RV = right ventricle; TEE = transesophageal echocardiography. Dynamic Reversal of Interatrial Shunting After Inhaled Nitric Oxide Administration (A) Transesophageal echocardiography obtained 2 minutes after initiation of inhaled nitric oxide demonstrating reduction of right-to-left shunting with emergence of left-to-right flow across the iatrogenic atrial septal defect. (B) At 6 minutes after nitric oxide initiation, color Doppler and continuous-wave Doppler imaging demonstrated predominant left-to-right shunting. These findings suggest that reversible right-sided loading conditions contributed to the abnormal shunting physiology. Table 1Hemodynamic Data Before and After Inhaled Nitric Oxide and After Transcatheter iASD ClosureBaselineAfter Inhaled Nitric OxideAfter iASD ClosureRA pressure, mm Hg32 (a/v/m: 60/47/32)—20 (a/v/m: 26/27/20)RV pressure, mm Hg87/14 (EDP: 26)—71/8 (EDP: 18)PA pressure, mm Hg82/32 (mean: 50)76/29 (mean: 47)71/28 (mean: 45)PCWP, mm Hg31/38 (mean: 30)—29/35 (mean: 27)LA mean pressure, mm Hg28—27Aortic pressure, mm Hg96/41—92/60Cardiac output (Fick), L/min6.27——Cardiac index, L/min/m24.09——Qp/Qs0.53——PVR, WU8.0—6.5PVRI, WU·m212.2—10.0Shunt direction on TEEPredominantly right-to-leftReversal toward left-to-rightDefect closedClinical oxygenationRefractory hypoxemiaImprovedImprovedBaseline hemodynamics showed elevated right-sided pressures with right atrial pressure exceeding left atrial pressure, consistent with a condition favoring right-to-left shunting. Inhaled nitric oxide reduced pulmonary artery pressure and was associated with reversal of shunt direction on transesophageal echocardiography. After transcatheter iASD closure, right-sided pressures decreased further, with concomitant clinical improvement in oxygenation.EDP = end-diastolic pressure; iASD = iatrogenic atrial septal defect; LA = left atrium; PA = pulmonary artery; PCWP = pulmonary capillary wedge pressure; PVR = pulmonary vascular resistance; PVRI = pulmonary vascular resistance index; Qp/Qs = pulmonary/systemic flow ratio; RA = right atrium; RV = right ventricle; TEE = transesophageal echocardiography. Hemodynamic Data Before and After Inhaled Nitric Oxide and After Transcatheter iASD Closure Baseline hemodynamics showed elevated right-sided pressures with right atrial pressure exceeding left atrial pressure, consistent with a condition favoring right-to-left shunting. Inhaled nitric oxide reduced pulmonary artery pressure and was associated with reversal of shunt direction on transesophageal echocardiography. After transcatheter iASD closure, right-sided pressures decreased further, with concomitant clinical improvement in oxygenation. EDP = end-diastolic pressure; iASD = iatrogenic atrial septal defect; LA = left atrium; PA = pulmonary artery; PCWP = pulmonary capillary wedge pressure; PVR = pulmonary vascular resistance; PVRI = pulmonary vascular resistance index; Qp/Qs = pulmonary/systemic flow ratio; RA = right atrium; RV = right ventricle; TEE = transesophageal echocardiography. Procedural Steps •Step 1: Identify pathologic R-L shunting. TEE confirmed predominant R-L shunting through the iASD, with the severe TR jet directed toward the defect.•Step 2: Define the hemodynamic mechanism. Invasive assessment demonstrated elevated right-sided pressures, RA pressure exceeding LA pressure, and increased pulmonary vascular resistance, consistent with a pressure condition favoring R-L shunting.•Step 3: Assess dynamic physiology. Inhaled NO was used as rescue therapy and an adjunctive physiologic maneuver, not as a stand-alone determinant of closure. The rapid improvement in oxygenation and reversal toward left-to-right flow suggested that the shunt was dynamic and influenced by right-sided loading conditions.•Step 4: Confirm anatomic feasibility. Three-dimensional TEE and multiplanar reconstruction demonstrated a tear-type iASD morphology and were used to assess defect geometry for transcatheter closure (Figure 2).•Step 5: Close the pathologic defect. Because life-threatening hypoxemia due to R-L shunting persisted and ventilator weaning remained difficult, transcatheter iASD closure was performed under TEE guidance. After closure, oxygenation improved promptly, right-sided pressures decreased, and the patient was successfully extubated. Pulmonary hypertension-specific medication was not initiated before closure because the patient was critically ill, leaving no sufficient time window for initiation and titration before urgent iASD closure. •Step 1: Identify pathologic R-L shunting. TEE confirmed predominant R-L shunting through the iASD, with the severe TR jet directed toward the defect. Step 1: Identify pathologic R-L shunting. TEE confirmed predominant R-L shunting through the iASD, with the severe TR jet directed toward the defect. •Step 2: Define the hemodynamic mechanism. Invasive assessment demonstrated elevated right-sided pressures, RA pressure exceeding LA pressure, and increased pulmonary vascular resistance, consistent with a pressure condition favoring R-L shunting. Step 2: Define the hemodynamic mechanism. Invasive assessment demonstrated elevated right-sided pressures, RA pressure exceeding LA pressure, and increased pulmonary vascular resistance, consistent with a pressure condition favoring R-L shunting. •Step 3: Assess dynamic physiology. Inhaled NO was used as rescue therapy and an adjunctive physiologic maneuver, not as a stand-alone determinant of closure. The rapid improvement in oxygenation and reversal toward left-to-right flow suggested that the shunt was dynamic and influenced by right-sided loading conditions. Step 3: Assess dynamic physiology. Inhaled NO was used as rescue therapy and an adjunctive physiologic maneuver, not as a stand-alone determinant of closure. The rapid improvement in oxygenation and reversal toward left-to-right flow suggested that the shunt was dynamic and influenced by right-sided loading conditions. •Step 4: Confirm anatomic feasibility. Three-dimensional TEE and multiplanar reconstruction demonstrated a tear-type iASD morphology and were used to assess defect geometry for transcatheter closure (Figure 2). Step 4: Confirm anatomic feasibility. Three-dimensional TEE and multiplanar reconstruction demonstrated a tear-type iASD morphology and were used to assess defect geometry for transcatheter closure (Figure 2). •Step 5: Close the pathologic defect. Because life-threatening hypoxemia due to R-L shunting persisted and ventilator weaning remained difficult, transcatheter iASD closure was performed under TEE guidance. After closure, oxygenation improved promptly, right-sided pressures decreased, and the patient was successfully extubated. Pulmonary hypertension-specific medication was not initiated before closure because the patient was critically ill, leaving no sufficient time window for initiation and titration before urgent iASD closure. Step 5: Close the pathologic defect. Because life-threatening hypoxemia due to R-L shunting persisted and ventilator weaning remained difficult, transcatheter iASD closure was performed under TEE guidance. After closure, oxygenation improved promptly, right-sided pressures decreased, and the patient was successfully extubated. Pulmonary hypertension-specific medication was not initiated before closure because the patient was critically ill, leaving no sufficient time window for initiation and titration before urgent iASD closure. Potential Pitfalls and Current Evidence Most iASDs after transseptal structural heart interventions are clinically insignificant and do not require routine closure; in a randomized trial, closure of persistent iASD after transcatheter mitral valve repair did not improve clinical outcomes.1, 2, 3,7 However, symptomatic iASD has been reported, particularly in patients with elevated right-sided pressures, pulmonary hypertension, or impaired right-sided reserve.4,8 Closure may be considered when an iASD causes clinically significant R-L shunting with refractory hypoxemia, or when substantial left-to-right shunting contributes to right-sided volume overload or heart failure.6,8,9 Conversely, indiscriminate closure should be avoided. In patients with elevated left-sided filling pressures and predominantly left-to-right shunting, an iASD may theoretically serve as a left atrial decompression pathway. In contrast, when RA pressure exceeds LA pressure, as in the present case, R-L shunting causes hypoxemia and does not provide left atrial decompression. Therefore, the decision to close postprocedural iASD should be individualized according to the interatrial pressure gradient, shunt direction, oxygenation, right ventricular function, pulmonary vascular load, and clinical trajectory.6,8 In the present case, immediate post-PVL hemodynamics did not suggest a pressure condition favoring R-L shunting because right-sided pressures decreased after PVL closure and mean RA pressure remained lower than mean LA pressure. However, several days later, aspiration pneumonia, worsening renal failure requiring continuous hemodiafiltration, and hemodynamic instability likely increased right-sided loading conditions and unmasked or exacerbated R-L shunting through the iASD. Severe TR, with the regurgitant jet directed toward the iASD, may have further promoted R-L shunting. Inhaled NO was not used as a stand-alone determinant of iASD closure. The indication for closure was persistent, life-threatening hypoxemia due to R-L shunting. Inhaled NO improved oxygenation and reversed shunt direction toward left-to-right flow, suggesting dynamic shunt physiology influenced by pulmonary vascular loading conditions. Thus, inhaled NO served as rescue therapy and an adjunctive physiologic maneuver that helped characterize the mechanism of hypoxemia and provided a bridge to urgent transcatheter iASD closure. Conclusions In patients with refractory hypoxemia after transseptal structural intervention, iASD should be considered a potential cause of clinically significant R-L shunting. Closure should be individualized according to hemodynamic physiology rather than anatomy alone. Inhaled NO should not be used as a stand-alone determinant of closure, but may provide adjunctive physiologic information and serve as rescue therapy or a bridge to urgent closure in selected critically ill patients. Funding Support and Author Disclosures The authors have reported that they have no relationships relevant to the contents of this paper to disclose.