Thoracoscopy Versus Open Surgery for Persistent Ductus Arteriosus and Vascular Ring Anomaly in Neonates and Infants

Introduction

Thoracoscopic surgery in neonates and infants for aortic anomalies, such as persistent ductus arteriosus (PDA) or vascular ring (VR), is widely used now with the miniaturization of the instrument, either for cardiothoracic and general pediatric surgery.^1,2 The pioneers were Laborde and Burke in the 90's.^3–7 More than 20 years later, many pediatric centers offer this opportunity to the patients.

The numerous advantages of thoracoscopy tend nowadays to be demonstrated by numerous reports, regarding decreased postoperative pain, shorter hospital stay, decreased incidence of chest wall deformity (including scoliosis and breast deformity), lower incidence of postthoracotomy pain syndrome, and better cosmetic results.^1,8,9

However, there is still a need for report series comparing thoracoscopy with the standard thoracotomy regarding effectiveness and morbidity, especially in neonates and premature babies for PDA closure. The aim of our study was to report our thoracoscopic experience of general pediatric surgeons for surgery of such pathology (PDA+VR) and to compare our results to open surgery regarding the short-term outcome.

Patients and Methods

Surgical technique

The thoracoscopic approach for both pathologies used three trocars with optic port placement above the tip of the scapula (Fig. 1): 5 mm/30° trocar for the camera, and 3 or 5 mm for two operator trocars. For PDA surgery, we also used a fourth 3 mm trocar placed in the inferior part of the thorax to push away down the lung. Insufflation pressure was around 5 mmHg, without single lung ventilation. PDA closure was realized with a titan clip, and the VR was divided after intracorporeal knot with nonabsorbable suture. We did not leave any thoracic drain for PDA patients.

OPEN IN VIEWER

FIG. 1.

Ports placement for both persistent ductus arteriosus and vascular ring ligation.

Results

Patients operated on by thoracoscopy by the same operator (AB) included 13 PDA (median age and weight at surgery: 34 days and 1800 g) and 11 VR (median age and weight at surgery: 8 months and 7000 g).

All PDA patients operated on by thoracoscopy were premature neonates (except one born at 37 WG), with a median gestational age at birth of 29.5 WG (min–max: 25–37) and a median weight at birth of 1255 g (min–max: 615–2520), and all were operated before 2 months of life. The smallest baby weighed 795 g, but was converted to open surgery because of the lack of working space.

The 22 VR patients presented various clinical signs at diagnosis, with no difference between the thoracoscopic group and the historical series operated on by thoracotomy. Results are presented in Table 1. One patient was diagnosed prenatally with an abnormal karyotype 47XXX. The anatomical description was assessed in all cases by contrast-enhanced computed tomography (CT) scanner: Complete double aortic arch (N = 12), right aortic arch with arterial ligament (Neuhauser anomaly) (N = 10). We retrieved a slight difference of age at surgery in the VR group, the rarity of the disease having prevented us to match the two groups for age and weight at surgery (Table 2).

Operative time between the two surgical techniques was not different for both PDA and VR. Two cases of PDA and one case of VR required conversion for anesthetic reasons and lack of working space at the beginning of the learning curve. Residual shunt was not detected in any PDA patients on immediate postoperative cardiac ultrasound.

Recurrent nerve palsy systematically assessed by fibroscopy occurred in three cases of PDA, two cases by thoracotomy, and one by thoracoscopy. They were all transient and not retrieved at 6 months follow-up. Chylothorax occurred in two cases of VR both by thoracoscopy, resolving without reoperation. One case of VR operated by thoracotomy required reoperation for persistent tracheal compression. These small numbers prevented us from performing any further statistical analysis.

Regarding the immediate postoperative course of PDA, we failed to demonstrate a statistical trend neither for thoracic tube duration, nor for mechanical ventilation and/or switch from high-frequency oscillation to conventional ventilation for PDA. Results regarding the use of analgesia were not comparable since the protocol changed between these two periods with a trend to decrease the morphinic requirement. Results are presented in Tables 2 and 3.

Surgery-related mortality was null for both PDA and VR in both groups. Median follow-up was of 6 months.

Discussion

In this study, we provided a comparative study of both VR and PDA treatments by open or thoracoscopic surgery. Our results, despite the small number of patients, tended to show that the mini-invasive approach in experienced hands is as safe and effective than the open approach. It provides a better cosmetic result and a minimal risk for thoracic sequelae, which can occur after thoracotomy, especially in preterm babies.

Thoracoscopic surgery in neonates and infants for aortic anomalies are mostly performed around the world by cardiothoracic surgeons. Large series of PDA in children have already been published since many years by them, fewer regarding neonates,^10–13 and small series of VR.^3,4,14,15 However, pediatric thoracoscopic surgery is currently spreading throughout the general pediatric surgery community and tends to become the gold standard, especially in experimented teams, such as S. Rothenberg's department. ¹⁶ Indeed, this technique applied to aortic malformations requires an experienced pediatric laparoscopic surgeon, familiar with suturing, intracorporeal knots, and adequate using of clips. They have demonstrated that this technique, as surgeons become more comfortable with the approach, is safe and effective, minimizing physiologic and cosmetic adverse effects. The learning curve allowed them to use laparoscopy even in patients less than 4 kg in the last half of their series, with a minimum weight of 2 kg, and a mean weight of 8.5 kg. This is the main difference with our series, which included mostly premature babies, but with similar results in terms of success and conversion rate. They mentioned only 3 failures of PDA ligation among 78 patients, due to a torn ductus and 2 recurrences, respectively, treated by surgery and cardiac catheterization. We had no recurrence, but had to convert the smallest baby (795 g) due to lack of working space. Regarding VR treatment, they did not mention any complication, such as chylothorax, like us. We unfortunately did not manage to avoid this complication in two laparoscopic cases despite the magnification of the laparoscope, although both patients recovered without requiring surgery. Similarly, all our patients were clinically improved at short-term evaluation (6 months) by the VR surgery.

Due to the relative rarity on these cardiovascular malformations, and the skills required to perform laparoscopic surgery on these patients, very few comparative studies with open surgery are available in the literature.^3,17 It is therefore difficult to prove to the community that it is worth to acquire these skills, in terms of potential risk-to-benefit ratio for the patient. Indeed, our statistical analysis is based on a very small series, but it stills appears that we did not have major complications, nor high rate of recurrence. Postoperative morphinic requirement has been difficult to compare in our study since the 20-year period have seen so many changes in the postoperative pain control in neonates. This is even more comprehensible in the PDA group, where no drain was left progressively over the period after the chest closure, in both open and thoracoscopic techniques. This also might have an effect on the duration of the mechanical ventilation requirement, although we failed to demonstrate any statistical result. In such premature babies with potential risk of pulmonary dysplasia, a long-term evaluation on a larger series is needed. The operative time, in experienced hands nowadays, is quite even, compared with older series, ³ and the complications tend to decrease compared with open surgery.

Thoracotomy in neonates and especially in preterm babies may present some immediate complication such as ribs fractures and late complications such as chest wall deformities and/or scoliosis. We think that in these particular surgical indications, thoracoscopic approach should be the preferable approach to minimize these sequelae and offer a better cosmetic result.

More reports are needed regarding the possibilities of robotic surgery in infants for PDA and VR treatments,^18,19 and due to the lack of space within the chest cavity in these small infants, miniaturizing of surgical ports and instruments remains mandatory.

Conclusion

Our short-term results in thoracoscopic PDA closure and VR anomalies treatment in neonates and infants is comparable to our experience in open surgery. Thoracoscopy in neonates and premature babies provides a better cosmetic result and decreases the risk of chest wall deformities, which can occur after thoracotomy. A larger series is mandatory to confirm these preliminary results.