Search tips
Search criteria 


Logo of thijTexas Heart Institute JournalSee also Cardiovascular Diseases Journal in PMCSubscribeSubmissionsTHI Journal Website
Tex Heart Inst J. 2009; 36(5): 425–427.
PMCID: PMC2763446

Repair of Atrial Septal Defects on the Perfused Beating Heart


We present our experience in repairing all varieties of atrial septal defects with the aid of continuous antegrade perfusion of an empty beating heart with normothermic blood.

From September 1999 through December 2008, 266 patients (140 females and 126 males; ages 3–53 yr) underwent atrial septal defect closure by this method. Of these patients, 236 had ostium secundum, 21 had sinus venosus, and 9 had ostium primum defects. Three patients also had rheumatic mitral incompetence requiring mitral valve implantation, and 2 also had mitral stenosis requiring valvuloplasty. Preoperative diagnoses were established by 2-dimensional echocardiography and color-flow Doppler study. The size of atrial septal defects ranged from 2 cm through 4.5 cm. Direct repair was performed in 52 patients, and the rest received an autologous pericardial patch. Normothermic perfusion at 4 to 5 mL/(kg·min) kept the heart beating throughout the procedure. All patients survived the procedure with no complication. Twelve patients with ostium secundum atrial septal defect were extubated on the table and discharged within 24 hours of hospitalization. They are categorized as ambulatory cases. All patients remained in sinus rhythm. One patient with a residual shunt required revision of a patch; postoperative echocardiography showed normal left ventricular function and no residual shunt. Total intensive care unit stay was less than 24 hours for all patients.

The primary aim of the beating-heart technique is to avoid ischemic–reperfusion injury. It is a safe and effective technique for the closure of all varieties of atrial septal defect.

Key words: Cardiopulmonary bypass, congenital heart disease, heart arrest, induced, heart septal defects, atrial/beating-heart surgery, myocardial ischemia/prevention & control, myocardial reperfusion injury/prevention & control

Most of the current myocardial protective approaches that use cold blood to achieve cardioplegia have withstood the test of time, providing safe and effective myocardial protection during various cardiac operations. Nevertheless, all cardioplegic myocardial protective approaches devised to date subject the heart to a period of so-called “mandatory ischemia,” wherein the heart is without circulation. This leads to reperfusion injury when the aorta is declamped.1 Among the efforts to prevent reperfusion injury has been the introduction of off-pump, beating-heart operations, which have enabled major advances in coronary artery bypass grafting.2–4 We ourselves have considered the possibility of closing atrial septal defects (ASDs) on the beating heart, thereby eliminating the ischemic component. We present our experience in closing all types of ASDs with the aid of continuous antegrade perfusion of an empty beating heart, in the absence of systemic hypothermia. The outcome of this prospective study is presented.

Patients and Methods

From September 1999 through December 2008, 266 patients underwent ASD closure on the beating heart. There were 126 males and 140 females, aged from 3 through 53 years. The size of each ASD ranged from 2 cm through 4.5 cm.

Preoperative diagnosis was established with the aid of 2-dimensional echocardiography and color-flow Doppler echocardiography. The size, type, and location of the ASD were ascertained, along with possible associated anomalies. All patients aged 40 years or more underwent coronary angiography to rule out myocardial ischemia. The ASDs, by type, comprised 236 ostium secundum defects, 21 sinus venosus defects (with partial anomalous pulmonary venous connection), and 9 ostium primum defects (all with a cleft mitral valve anterior leaflet). Six patients also had mitral incompetence (4 with rheumatic valves and 2 with cleft anterior mitral leaflets), and 3 had mitral stenosis.

Surgical Technique

The operations are performed through a median sternotomy in the following manner: after heparinization, total cardiopulmonary bypass (CPB) is established by cannulating the ascending aorta and both venae cavae. A needle is inserted into the ascending aorta to facilitate perfusion during the procedure and de-airing upon completion. In cases that involve a concomitant mitral valve procedure, a vent is inserted through the right superior pulmonary vein. The technique of caval cannulation is modified if the patient's anatomy requires it (for example, the presence of sinus venosus ASD or of left superior vena cava). Both cavae are looped. The aorta is cross-clamped and the aortic root is perfused with 4 to 5 mL/(kg·min) normothermic oxygenated blood throughout the procedure; both the caval loops are snugged down before the right atriotomy. Electrocardiography is monitored for any ischemic changes, retraction is kept to a minimum in order to prevent distortion of the aortic root, and an angled suction tip is placed in the coronary sinus ostium in order to keep the operative field bloodless. Simultaneous transesophageal echocardiography is performed in all adults.

The total duration of bypass ranged from 6 through 71 minutes, and the duration of aortic cross-clamping ranged from 4 through 68 minutes. Direct closure of the ASD was performed in 52 patients, and the rest received an autologous patch of pericardium. Our interventional cardiology colleagues routinely use an Amplatzer® septal occluder device (AGA Medical Corporation; Plymouth, Minn) in ASDs up to 30 or 35 mm in size. Therefore, in each surgical patient, the defect has been at least 30 mm in its largest dimension. Exceptions have included sinus venosus defects that had partial anomalous pulmonary venous connections and defects that had margins too close to the coronary sinus, tricuspid valve, atrioventricular node, or right superior pulmonary vein. All patients so affected underwent patch closure regardless of the actual defect size. Some of the younger patients also underwent surgical repair because cardiac catheterization with the bulky device was considered difficult or unsafe. Mitral valve implantation with total chordal preservation was performed in 3 of 4 patients who had mitral regurgitation, whereas valvuloplasty was performed in patients who had stenosis. Cleft mitral leaflets were repaired with fine interrupted polypropylene suture. All of these procedures were carried out without cardioplegia. De-airing and right atrial closure were done in routine fashion.

During the same study period, 2 other consultant surgeons in our department performed ASD repair in 413 patients under conventional cardioplegic arrest. In neither group was the timing of surgery governed by any specific guidelines.


All patients withstood the procedure well and were extubated within 4 hours after the procedure. Twenty patients with ostium secundum ASD were extubated on the table and were discharged within 24 hours after admission; these cases were categorized as ambulatory.5 No patient required any inotropic support or vasodilation. There were no electrocardiographic changes during or after the procedure. None of the patients required stays in the intensive care unit in excess of 24 hours. Postoperative echocardiography showed normal left ventricular function and no residual shunt across the interatrial septum, except for 1 patient in whom a residual shunt was discovered upon transesophageal echocardiography after decannulation; to make the correction, we resumed CPB but did not cross-clamp the aorta. The procedure's visual field and technical accuracy were never compromised.

A noteworthy finding in this cohort of perfused beating-heart cases was the absence of early postoperative arrhythmias, in comparison with those patients who were operated upon under cardioplegic arrest. We are deliberately deferring the statistical analysis of this phenomenon, because we wish to observe the late outcomes in these 2 groups over a respectable period of follow-up.


Although major technological advances have been made in myocardial protection during cardiac operations over the past decade, perioperative adverse effects caused by myocardial ischemia have not been completely eliminated. Some degree of myocardial stunning occurs even with continuous warm-blood cardioplegia, which is considered the best form of myocardial protection because it keeps the heart in an aerobic state.6 Cardiac dysfunction can be caused by myocardial edema intrinsic to the diastolic state of the arrested heart.7 Therefore, cardioplegic arrest inevitably produces some degree of reperfusion injury. Keeping the heart beating results in less myocardial edema and better myocardial function.

Other advantages of beating-heart ASD repair are the immediate ability to evaluate the severity of any associated mitral or tricuspid valvular insufficiency and to identify iatrogenic conduction injuries. This is particularly important in cases of ostium primum ASDs, wherein the atrioventricular node remains a surgically vulnerable spot. The primary aim of the beating-heart technique is to avoid ischemic–reperfusion injury. It is a safe and effective technique not only for closure of ASDs but also for various other noncoronary procedures, including valve repairs and replacements. Our experience to date suggests that beating-heart ASD closure is safe and requires no compromises in comparison with conventional techniques.


Address for reprints: Amit Banerjee, MS, MCh, Department of Cardiovascular & Thoracic Surgery, Govind Ballabh Pant Hospital, New Delhi 110002, India E-mail: moc.aidni@eejrenab;f5000x#&tima


1. Weman SM, Karhunen PJ, Penttila A, Jarvinen AA, Salminen US. Reperfusion injury associated with one-fourth of deaths after coronary artery bypass grafting. Ann Thorac Surg 2000; 70(3):807–12. [PubMed]
2. Buckberg GD. Oxygenated cardioplegia: blood is a many splendored thing. Ann Thorac Surg 1990;50(2):175–7. [PubMed]
3. Ikonomidis JS, Rao V, Weisel RD, Hayashida N, Shirai T. Myocardial protection for coronary bypass grafting: the Toronto Hospital perspective. Ann Thorac Surg 1995;60(3): 824–32. [PubMed]
4. Mauney MC, Kron IL. The physiologic basis of warm cardioplegia. Ann Thorac Surg 1995;60(3):819–23. [PubMed]
5. Srivastava AR, Banerjee A, Tempe DK, Mishra B, Muppiri V, Narang S, et al. A comprehensive approach to fast tracking in cardiac surgery: ambulatory low-risk open-heart surgery. Eur J Cardiothorac Surg 2008;33(6):955–60. [PubMed]
6. Misare BD, Krukenkamp IB, Lazer ZP, Levitsky S. Recovery of postischemic contractile function is depressed by antegrade warm continuous blood cardioplegia. J Thorac Cardiovasc Surg 1993;105(1):37–44. [PubMed]
7. Mehlhorn U, Allen SJ, Adams DL, Davis KL, Gogola GR, Warters RD. Cardiac surgical conditions induced by beta-blockade: effect on myocardial fluid balance. Ann Thorac Surg 1996;62(1):143–50. [PubMed]

Articles from Texas Heart Institute Journal are provided here courtesy of Texas Heart Institute