Comparative Study of Thoracic Approaches in Atrial Septal Defect Closure

INTRODUCTION

Midline sternotomy is the incision generally used for heart surgery. Articles that describe alternative approaches appear regularly that are surgically less traumatic and eliminate the risk of sternal instability, in addition to improving cosmetic results.^1-3

In this study we report our initial experience in the closure of atrial septal defects (ASD) in children using right anterolateral thoracotomy, with cannulation of the aorta and both cava veins through the same incision. The results obtained are compared with a control group of patients with ASD closed by midline sternotomy.

MATERIAL AND METHODS

From July 2000 to December 2001, 15 patients (group A) underwent ASD closure via a right anterolateral thoracotomy. The surgical indication was ostium secundum ASD (8 cases), sinus venosus with partial anomalous drainage of pulmonary veins (4 cases), and ostium primum ASD (3 cases). In the same period, 16 patients (group B) underwent surgery using midline sternotomy (12 ostium secundum, 4 ostium primum). Patients were assigned to one approach or another at the discretion of the surgeon (Table 1).

Inclusion criteria

Initially, female patients with complete breast development were chosen. Later, age was not an inclusion criterion. Similarly, the first cases were ostium secundum type ASD, then the indication was expanded to include sinus venosus and ostium primum type ASD.

Exclusion criteria

Age under 2 years; diseases not located in the atrial septum.

Anesthesia

All patients received balanced general anesthesia with propofol, fentanyl, sevoflorane, and ropivacaine. In addition, thoracic epidural anesthesia was associated (puncture level T6-T8) in 10 patients in group A, with intraoperative ropivacaine and postoperative morphine (24-36 h). The epidural catheter was introduced after the patient had been intubated and withdrawn when the patient was discharged from the ICU.

Access

After making the skin incision (Figure 1), the subcutaneous tissue (future mammary gland) and pectoral muscle are dissected en bloc. The latissimus dorsi and serratus muscles do not have to be sectioned. The thorax is opened in the fourth intercostal space. After removing the right lobe of the thymus, the pericardium is opened in front of the phrenic nerve and the incision is prolonged in cephalad direction over the aorta. Exposure of the aorta is facilitated by fixing the pericardium to the second rib, then passing a tape to facilitate aortic traction and manipulation.

Fig. 1. Preparation of patient. Scapular girdle tilted 30-45º. Continuous line: skin incision (sixth space). Dotted line: thoracotomy (fourth space).

Cannulation

The aorta and both vena cava can be cannulated directly (Figure 2) through the thoracotomy. Ostium secundum type ASDs were closed by direct suture or a patch. Sinus venosus ASD was closed with a patch and enlargement of the atriotomy in the superior vena cava with another patch. Ostium primum ASDs were closed with a patch of autologous pericardium and closure of the cleft mitral valve.

Fig. 2. Cannulation through the thoracotomy. RA indicates right atrium; AO, aorta; IVC, inferior vena cava; SVC, superior vena cava. 

Associated procedures

In one patient in group A, a ventricular septal defect was closed (intermediate channel). In group B, a patent ductus arteriosus was ligated in 2 cases. In one patient, a pulmonary commissurotomy was made and in one patients, a small ventricular septal defect was closed.

After completing the correction, ventricular pacemaker leads were placed before unclamping the aorta. The ventilation and defibrillation maneuvers were similar to those used with sternotomy: in every case, the pericardium was closed and two drainages were left in place (intrapericardial and intrathoracic).

RESULTS

The mean duration of the intervention was 152 min with the thoracotomy approach and 137 min with the sternotomy approach. Cardiopulmonary bypass lasted 43.2 min with thoracotomies and 47.4 min with sternotomies, and clamping lasted 25.8 min with thoracotomies and 24.3 min with sternotomies (Table 2). Thirteen of the 15 patients in group A and 2 of the 16 patients in group B were extubated in the operating room. The rest of the patients were extubated within the first 6 h of their stay in the ICU. The total amount of fluid drained was 265 ml (8.4 ml/kg) in group A and 152 ml (9.2 ml/kg) in group B.

The mean stay in the ICU was 2.4 days in group A and 2.23 days in group B. Patients were discharged at 6 days in group A and at 6.87 days in group B.

One patient in each group presented postoperative pericarditis. Specific complications of the approach were one case of transient phrenic paresis in the thoracotomy group and a surgical wound infection in the sternotomy group.

The chest radiograph and echocardiograph made before discharge were acceptable in every patient in both groups.

The preadolescents (5 patients) and parents of younger children who underwent thoracotomy described the cosmetic result as excellent.

DISCUSSION

There is no doubt that midline sternotomy is the most common, convenient, and safe approach, and that it provides better exposure for the cardiac surgeon. Several authors have reported their experience with alternative approaches to sternotomy (less invasive, better cosmetic result, etc.) for the surgical treatment of certain congenital heart diseases:^3-13 all agree that the cosmetic results of the surgical approach are important, especially in girls and women.

In this context, right anterior thoracotomy is particularly attractive for two reasons: it uses a space below the breast for the incision (better cosmetic result), and it provides «easy» access to the right atrium, thus facilitating surgery in conditions that are approached this way. Consequently, surgical correction of ASD, particularly in young women, is the target of this approach. The cosmetic advantage in males is more debatable, considering that hair growth may dissimulate the effects of midline sternotomy.

On the other hand, there is controversy about compromising the future growth of breast tissue in young girls.¹⁴ One French group recommends delaying right thoracotomy until the breasts are fully developed in these patients.¹⁰ However, the real risk of affecting the breast tissue is scant if the skin incision is made below the sixth intercostal space and the subcutaneous tissue is dissected en bloc with the pectoral muscle before attempting thoracotomy in the fourth space.⁹ The possibility of affecting the spine is remote, inasmuch as the anterolateral thoracotomy respects the latissimus dorsi and the spinal muscles.¹⁴ In any case, only prolonged follow-up can provide evidence of cosmetic effects on the breasts or spine.

From a technical point of view, most groups attempt to reproduce the basic steps of sternotomy when using the thoracotomy approach: cannulation of the aorta and two vena cava, clamping and cardioplegia, all through the same incision (Figure 2).

Our work has several biases. In the first place, it is not a randomized study. The distribution in both groups was individualized way at the discretion of the surgeon. The first 3 cases for thoracotomy were selected (ostium secundum type ASD in girls with fully developed breast), then the indications were later extended. For this reason, the mean weight and age of group A is greater, and there are two peak ages (3-4 and 13-14 years). In contrast, 10 of the patients in group B were younger than 4 years. Secondly, several patients in the thoracotomy group were selected by anesthesiologists to undergo general anesthesia in addition to epidural anesthesia and they were extubated in the operating room. This explains the large number of extubations that took place in the operating room in group A.

CONCLUSIONS

Surgical closure of ASDs through a right anterolateral thoracotomy was feasible in our initial series. There were no significant differences in the results compared with midline sternotomy. The instruments used and surgical steps were the usual ones. The cosmetic results of thoracotomy were excellent.

ACKNOWLEDGMENTS

We would like to that Dr. Guinot, of the Department of Plastic and Reconstructive Surgery of the Hospital Universitario Infantil Vall d´Hebron of Barcelona, for his collaboration and advice.

---

Correspondence: Dr. J.M. Gil-Jaurena.
Servicio de Cirugía Cardíaca. Hospital Universitario Vall d´Hebron. Paseo Vall d´Hebron, 119-129. 08035 Barcelona. España.
E-mail: [e-mail]