Background

MitraClip (MC) therapy is an established therapy for mitral regurgitation (MR) in high-risk patients. MC therapy requires transseptal puncture and produces an iatrogenic atrial septum defect (iASD). The indication and timing for iASD closure is not yet known . The aim of this study was to assess the incidence and specific indications for iASD closure after MC in a single-center population. 

Material and Methods

A total of 756 patients with moderate-to-severe or severe MR underwent MC at our center between 2009 and 2017. Invasive hemodynamics and echocardiographic parameters were assessed before and after MC implantation. Retrospectively all iASD closures after MC, either acute or during follow-up, were investigated.

Results

ASD closure was performed in 15 patients out of 756 patients (1.98%, mean age of 73 years, 13 male patients, 8 degenerative MR and 7 functional MR). 12 patients received closure immediately after MC, while 3 patients received iASD closure in interval (median 202±103 days). The iASDs had a median size of 112±54.5 mm². 

iASD shunt direction was evaluated by transosophageal echocardiography: 8 patients (53%) showed a right-to-left (R-t-L) shunt, while 4 patients (27%) showed a left-to-right (L-t-R) shunt and 3 patients (20%) showed a bidirectional shunt. 

14 of 15 (93%) iASD closures showed acute success defined as the correct device deployment in absence of a relevant shunt and without major complication. In one case the occluder (24mm) embolized into the left atrium. The occluder was retrieved by the use of a snare catheter and a larger occluder (30mm) was used for final closure. The 30mm occluder also embolized the day after, consequently the patient underwent open surgical closure of the iASD.

The indication for immediate closure of iASD during MC was desaturation leading to hemodynamic instability (independent of shunt direction) in 5 patients, presence of a relevant R-t-L shunt in 5 patients, presence of a relevant bidirectional shunt in one patient and the excessive size of the ASD after rupture of the septum in one patient. Relevance of the shunt was defined by operator`s choice. All patients who received iASD closure due to hemodynamic relevance were stabilized after iASD closure. 

Among the patients who received iASD closure in interval, the indication was a relevant R-t-L shunt in one patient and a relevant bidirectional shunt with L-t-R dominance in two patients. 

The patient who was suffering from R-t-L-shunt 5 days after MC originally showed a not relevant L-t-R at the end of the procedure, but had experienced a shunt reversal in the meantime.

Both other patients showed a bidirectional shunt, which was evaluated as not being relevant at the end of the MC procedure. During the follow up period they developed a progressive right heart failure due to a L-t-R dominant bidirectional shunt and consequently underwent iASD closure 202 and 315 days after MC. 

Conclusion

This is the first report describing iASD closure after MC in a large single-center population. iASD closure might be necessary directly after MC due to hemodynamic instability, relevant shunting or in interval due to progressive heart failure in the presence of a relevant iASD. Moreover, this analysis demonstrates that iASDs, which are evaluated as hemodynamically irrelevant at the end of MC can become hemodynamically relevant during the follow up period and consequently deserve particular attention and repetitive evaluation.