Postoperatively, a chest radiograph should be performed to rule out pneumothorax, and an echocardiogram should be confirmed to confirm occlusion of flow [21–24]. Close hemodynamic monitoring is crucial in the postoperative period, due to the abrupt hemodynamic changes associated with surgical ligation. Following ligation, there is an increase in diastolic arterial pressure and associated increase in MAP compared to preoperative levels [16]. However, there is commonly a transient decrease in systolic arterial pressure, likely associated with delayed left ventricular adaptation to the abrupt increase in afterload [16]. Postoperative MAP frequently fails to rise to normal levels in the immediate postoperative period. Some studies suggest a transient decrease in postoperative cerebral perfusion, further highlighting the need to maintain adequate MAPs [16]. Some studies also suggest increased oxygenation index following ligation [16]. The risk of post-ligation cardiorespiratory decompensation likely depends on the size of the defect, the time of closure, and the gestational age of the infant. Careful titration of fluids and vasopressors must be employed to maintain adequate perfusion while avoiding pulmonary fluid overload.

The rate of intraoperative mortality is exceedingly low, with the overall rate reported to be <1 %. There are no intraoperative deaths reported in recent literature of thoracoscopic PDA repair, and two reported during open repair. Intraoperative hemorrhage is also uncommon, with rates estimated between 0.5 and 1 % and even lower in recent series [17–19, 22–24]. Injury to surrounding structures, such as the pulmonary artery and descending aorta, is exceedingly rare, usually resulting in death [25–27].

More common complications include injury to the recurrent laryngeal nerve, which are reported to range between 0.5 and 6 %, with no significant difference noted between VATS and open approaches [17–19, 21, 23]. Over half of these are transient injuries with return of function [18, 19]. Rates of pneumothorax are between 0.5 and 6 %, and rates of chylothorax are estimated between 0.5 and 3 %, with no difference between minimally invasive and open repair [17–19, 21, 23]. Reported incidence of residual flow in the duct ranges between 0.5 % and 6 % [17–19, 21, 23], with a trend toward higher incidence in open repair that reaches statistical significance in some studies [17]. In contemporary series, conversion from thoracoscopic to open repair is between 0.5 and 4 %, with the most common cause reported to be poor exposure due to inability to retract the lung safely. Missed concurrent abnormalities are reported in less than 1 % of cases, with the most common abnormality being coarctation of the aorta [21, 23].

In the Johns Hopkins University series of 25 T-PDA ligations in very low-birth-weight infants, there have been no intraoperative deaths, 1 conversion to open procedure, 1 recurrent laryngeal nerve injury, and no other immediate morbidity.

Postoperative mortality approaches 15 % in some series, with the greatest risk factors being prematurity and low birth weight [14, 18, 23]. Sepsis is the leading cause of in-hospital death, frequently with associated necrotizing enterocolitis. Other causes are chronic respiratory disease and multisystem organ failure [14, 18, 23]. In our series we have had 3 deaths postoperatively all from sepsis related to NEC.

Thoracoscopic PDA ligation offers many of the advantages of a minimally invasive surgical approach that have been previously reported as well as a potential reduction in post thoracotomy scoliosis by avoiding rib spreading, muscular division, and minimization of the risk of nerve injury or rupture of intercostal ligaments. T-PDA ligation is effective and can be safely performed in very low-birth-weight infants. Exposure and magnification of important anatomic landmarks are superior compared to open techniques. This factor results in shorter operative time. In addition to the enhanced cosmesis, this minimally invasive approach may reduce postoperative pain, pulmonary morbidity, and the future incidence of chest wall and spine deformities in children.