Discussion

Many patients with CHD benefit from the presence of appropriate interatrial communication. However, achieving or maintaining this communication after the neonatal period with balloon septostomy is often challenging. Atrial septal stenting has emerged as a critical intervention in managing CHD, particularly for patients with obstructive lesions or complex physiologies that require stable interatrial communication. This technique has demonstrated the ability to rapidly improve systemic oxygenation and reduce atrial pressure gradients, solidifying its role in contemporary paediatric cardiology. Studies by Chowdhury et al. and Awasthy have highlighted its life-saving potential and long-term durability, while Corno et al. emphasised its tailored application to optimise interatrial shunting for individual patients, especially in complex congenital anomalies.^{Reference Awasthy4–Reference Corno, Adebo and LaPar6}

The success of atrial septal stenting relies on meticulous procedural planning and execution. In our study, a procedural success rate of 93.5% was achieved, with only one failure due to stent embolisation. Similar success rates have been reported in other studies; for example, Mainzer et al. documented a 93% success rate (27 out of 29 cases), while Prakoso et al. reported achieving non-restrictive atrial flow in 90.9% of cases, even under high-risk conditions.^{Reference Prakoso, Kurniawati and Siagian7,Reference Mainzer, Goreczny and Morgan8} In our cohort, pre-mounted stents, preferred for their ability to reduce the risk of embolisation, were utilised in most procedures. However, one case involving a 34 mm Cheatham-Platinum stent mounted on a 20 mm balloon resulted in stent displacement into the right atrium. This experience underscores the importance of using pre-mounted stents to enhance procedural success and minimise complications.

Stent size selection is a critical factor for both procedural success and long-term outcomes. For temporary improvement in cyanosis and intracardiac mixing, smaller stents (4–6 mm) are often adequate.^{Reference Sivakumar9} However, larger stents are required for sustained decompression of left atrial pressure. Radityo Prakoso et al. reported an average stent diameter of 8.5 mm, with the size varying based on the patient’s anatomy and specific clinical objectives.^{Reference Prakoso, Kurniawati and Siagian7} In our study, stents with a median diameter of 7.5 mm were used in both age groups, and no significant correlations were observed between stent diameter and patient weight or age. Currently, there is no consensus on the optimal stent size for different scenarios, making the interventionist’s expertise crucial in determining the most appropriate size. Larger multicentre studies are necessary to establish standardised, evidence-based guidelines for stent sizing.

In paediatric populations, expandable stents offer the significant advantage of accommodating growth through subsequent adjustments. This was demonstrated in our cohort, where two patients underwent stent dilation during follow-up to account for growth-related changes. Previous studies, such as those by Quandt et al., highlight the importance of this flexibility for effective long-term management.^{Reference Quandt, Ramchandani and Bhole10} However, determining the optimal timing for redilation can often be challenging. In our practice, we used echocardiographic measurements, specifically gradients of 22 mmHg peak and 14 mmHg mean, combined with the patient’s clinical status to guide the decision for redilation. Similarly, Mainzer et al. identified gradients exceeding 20 mmHg as a clear indicator for redilation, while Sivakumar recommended maintaining interatrial pressure gradients within the range of 5–10 mmHg after stenting to ensure optimal outcomes.^{Reference Mainzer, Goreczny and Morgan8,Reference Sivakumar9}

The choice of tools for septal puncture also impacts procedural success. In our cohort, we used Brockenbrough needles and radiofrequency guidewires in four and two cases, respectively, ensuring precise puncture and stent placement. Sivakumar emphasised the utility of Brockenbrough needles for thickened atrial septum or hypoplastic left heart syndrome. Mainzer et al. and others reported radiofrequency guidewires as beneficial in difficult anatomies.^{Reference Mainzer, Goreczny and Morgan8} These findings underscore the importance of a versatile toolkit to address varied anatomical challenges effectively.

Clinical outcomes in our cohort suggested a trend towards improved oxygenation and relief of atrial level obstruction after interatrial stenting. In a small subgroup of patients undergoing stenting for atrial level mixing, increases in systemic oxygen saturation were observed at the individual patient level. Previous studies have reported similar findings; for instance, Danon et al. reported an average oxygen saturation increase of 11.3% in patients with transposition physiology following stenting.^{Reference Danon, Levi and Alejos11} Similarly, Mainzer et al. observed oxygen saturation levels rising above 70% in critically ill patients, indicating substantial clinical improvement.^{Reference Mainzer, Goreczny and Morgan8} Consistent with these findings, patients in our cohort requiring enhanced atrial mixing exhibited notable increases in oxygen saturation, confirming the procedure’s efficacy in alleviating hypoxaemia in complex congenital settings. However, these results should be interpreted cautiously due to the limited number of patients with complete paired measurements.

Patients with elevated left atrial pressures due to left ventricular dysfunction showed significant symptom resolution, and pulmonary hypertension was successfully prevented in several cases. Radityo Prakoso et al. emphasised that reducing left atrial pressure not only improved cardiac output but also rapidly stabilised haemodynamics.^{Reference Prakoso, Kurniawati and Siagian7} Similarly, Mainzer et al. reported that atrial septal stenting effectively mitigated pulmonary hypertension, thereby reducing the risk of long-term cardiac and pulmonary complications.^{Reference Mainzer, Goreczny and Morgan8} In our cohort, patients with left ventricular dysfunction demonstrated a noticeable decrease in inotropic support requirements and hospital admission rates, underscoring the procedure’s clinical benefits in this population.

Our cohort represents a highly heterogeneous population including patients with hypoplastic left heart syndrome, borderline left ventricle, transposition physiology, myocarditis-related left ventricular dysfunction, pulmonary hypertension, and Fontan failure. These conditions differ fundamentally in pathophysiology, haemodynamic targets, and prognosis. Accordingly, the clinical objective of atrial septal stenting varies substantially across subgroups: improving intracardiac mixing and systemic oxygenation in transposition physiology, decompressing the left atrium and pulmonary venous system in left-sided dysfunction and pulmonary hypertension, and serving as a rescue strategy to augment cardiac output and relieve systemic venous congestion in Fontan failure. Similar diagnosis-specific physiological benefits were reported by Danon et al., who demonstrated a mean increase in systemic oxygen saturation of 11.3% in transposition physiology and a mean reduction of 7.4 mmHg in left atrial pressure in patients with left-sided obstructive lesions.^{Reference Danon, Levi and Alejos11} Contemporary real-world experience further confirms that atrial septal stenting is most often performed in critically ill, high-risk patients, with outcomes predominantly driven by the severity of the underlying disease rather than procedural factors alone. In the largest paediatric series to date, Mainzer et al. reported a procedural success rate of 93% but substantial early mortality in the sickest infants, particularly those weighing ≤3 kg, underscoring the palliative and bridging nature of this intervention.^{Reference Mainzer, Goreczny and Morgan8} Qureshi has similarly emphasised that atrial septal stenting should be regarded as a high-risk palliative strategy rather than a definitive therapy.^{Reference Qureshi and Sivasankaran12} Therefore, although pooling such distinct pathophysiological entities limits direct outcome comparisons, our cohort accurately reflects contemporary clinical practice, in which atrial septal stenting is employed as a stabilising or life-saving intervention across a wide spectrum of complex congenital and acquired heart diseases.

Interatrial septal stenting is typically performed in critically ill patients with advanced haemodynamic compromise and is therefore associated with substantial mortality that is largely driven by the severity of the underlying cardiac disease rather than by procedural factors alone. In previously published series, mortality has consistently been reported as a combination of early post-procedural deaths and late deaths related to disease progression.

Mainzer et al. reported an early mortality rate of 14% in a high-risk paediatric cohort, with deaths predominantly occurring within the first 24 hours in patients with severe clinical instability.^{Reference Mainzer, Goreczny and Morgan8} Importantly, the authors emphasised that most deaths were attributable to the critical pre-procedural condition rather than to traumatic procedural complications. Mortality was particularly high among infants weighing ≤3 kg, reflecting the extreme vulnerability of this subgroup. Danon et al. similarly reported two deaths in a cohort of 13 patients, including a neonate with hypoplastic left heart syndrome and intact atrial septum who died three days after technically successful stent implantation, and a patient with failed Fontan physiology who died following transplantation.^{Reference Danon, Levi and Alejos11} In both cases, stent patency was preserved and mortality was attributed to the underlying disease rather than to the intervention itself. Qureshi has likewise characterised interatrial stenting as a high-risk palliative intervention reserved for patients with failing Fontan circulation or severe pulmonary hypertension, in whom long-term survival is primarily determined by disease trajectory.^{Reference Qureshi and Sivasankaran12}

In our cohort, the overall mortality rate was 37.5%, reflecting the advanced clinical status of the patient population. Deaths were classified according to their temporal relationship with the procedure and underlying cause. Two deaths occurred in the early post-procedural period. One patient with hypoplastic left heart syndrome died on postoperative day 1 despite technically successful stent implantation. A second patient with acute myocarditis requiring extracorporeal membrane oxygenation experienced procedural failure and died on postoperative day one in a salvage setting.

The remaining deaths occurred during follow-up and were related to progression of underlying cardiac disease or non-cardiac complications. One patient with hypoplastic left heart syndrome died two months after the procedure following prolonged ICU stay and inability to be weaned from mechanical ventilation due to advanced cardiac dysfunction. One patient underwent Glenn surgery after interatrial stenting and died one year later from post-infectious respiratory failure while receiving pulmonary arterial hypertension therapy. One patient with borderline left ventricle died one month after the procedure following prolonged ICU stay complicated by respiratory failure. One patient with pulmonary hypertension secondary to borderline left ventricle died from sepsis four months after the procedure.

Consistent with previous reports, early mortality in our cohort occurred in the setting of extreme clinical instability and salvage intervention, whereas late mortality reflected the burden of advanced cardiopulmonary disease and critical illness. No late deaths were directly attributable to procedural complications, and interatrial septal stenting served primarily as a stabilising or bridging intervention in a population with limited therapeutic options.

In long-term follow-up, the majority of our patients successfully transitioned to planned surgical interventions. Mainzer et al. reported an average stent patency duration of 302 days, which was generally adequate to bridge patients to surgery.^{Reference Mainzer, Goreczny and Morgan8} However, some cases required balloon redilation, occurring between 58 and 201 days post-procedure. Similarly, in our study, two patients underwent redilation at 2.5 and 9 months, respectively, during follow-up. Radityo Prakoso et al. reported a 90.9% long-term patency rate, which is consistent with our findings of sustained stent functionality in most patients.^{Reference Prakoso, Kurniawati and Siagian7} The durability of stents was further exemplified by a patient in our cohort with postcapillary pulmonary hypertension who remained asymptomatic for five years post-stenting, demonstrating the procedure’s effectiveness over extended periods.

Nevertheless, complications such as stent thrombosis and venous obstruction, though rare, necessitate careful post-procedural monitoring and anticoagulation. Aspirin therapy was used in all patients in our cohort, effectively preventing stent thrombosis during follow-up.