Minimally invasive surgery (MIS) has become more common in the pediatric population, and this approach has been extended to patients with congenital diaphragmatic hernia (CDH) [1]. In addition to reducing pain, surgical stress and length of hospital stay, and encouraging early recovery, MIS for CDH repair has been thought to be beneficial over conventional open surgery, especially because of reduced duration of postoperative mechanical ventilation, less need for narcotics, and lower incidence of skeletal deformities [2-5]. Many reports comparing MIS and open surgery for CDH repair have been published. However, some issues about the operative outcome and patient selection criteria of MIS for CDH repair still remain controversial. Some reports have suggested better outcomes with MIS [4-6], but others have been more cautious, especially in terms of the recurrence rate of the hernia [2,3,7-10]. To achieve a safe procedure and acceptable results in terms of recurrence, recent reports have suggested some selection or exclusion criteria for performing MIS for CDH repair [2-5,10,11]. However, no consensus exists for the selec tion criteria of MIS for CDH repair. The purpose of this study was to compare the outcomes of MIS with open surgery in patients with CDH and to identify the clinical factors that determine the selection criteria of MIS for CDH repair.

Retrospective analysis of the medical records was performed for patients with CDH who were operated on at a single children's hospital between January 2008 and Decem ber 2012. Patients who underwent repeated surgery for a recurrent hernia after the first correction in other hospitals were excluded. The clinical data of the patients were reviewed, including patient characteristics, operative approach, intra operative findings, and postoperative outcomes. Data of pa tients who underwent open surgery were compared with those of patients who underwent MIS for CDH. The patient's charac teristics included sex, gestational age, Appearance, Pulse, Grimace, Activity, Respiration (APGAR) score at 5 minutes after birth, birth weight, age at surgery, delayed presentation (detected more than 30 days after birth), location of the CDH, and combined anomalies. The data for the operative approach and intraoperative findings included the type of surgery, operative time, size of diaphragmatic defect, presence of a her nial sac, use of a patch, and open conversion rate. The pos toperative outcomes included postoperative ventilation duration, time to initiation of enteral feeding, length of hospital stay, complication, recurrence, and mortality. All MIS approaches were performed by three experienced pediatric surgeons. The Institutional Review Board of our hospital approved this study (4-2012-0918).

Since the first MIS repair of CDH in 1995 [12], many studies have reported comparable outcomes of MIS with open surgery to repair a CDH [4,5,10,13-17]. However, some concerns remain about the high recurrence rate of CDH after MIS repair [2,3,7-9]. The reported recurrence rate following MIS for CDH repair ranges from 0% to 39% with an overall recurrence rate of 12.3% [3-11,14,16,17]. To improve the recurrence rate or to complete successful MIS for CDH repair, a few reports have suggested selection criteria or exclusion criteria for MIS [2-5,10,11]. In one recent study, systematic quality improvement was conducted to reduce the recurrence rate following MIS for CDH [8]. In most reports, patients selected for MIS were hemodynamically stable and tolerated with the use of a ventilator without the need for ECMO. Severe pulmonary hypoplasia causing respiratory distress and pulmonary hypertension should also be considered contraindications for MIS [15,16]. A few reports of low recurrence rates have a common approach to MIS for CDH repair; they performed MIS only in patients with a small diaphragmatic defect using a primary repair and converted to open repair if a large defect was found during MIS, necessitating a patch repair [4,11,16]. In this study, all patients were stabilized preoperatively with ventilator care with or without high-frequency oscillatory ventilation or nitric oxide inhalation. ECMO was not needed in either group. Because primary repair of a large defect is difficult with MIS, we converted to the open method for a patch repair. These points could be considered selection criteria for MIS for CDH repair. For patients with preoperative stable hemodynamics, delayed presentation, or without comorbidity, MIS could be suggested as the primary treatment of choice. However, if a large diaphragmatic defect is found endoscopically during MIS and patch repair is necessary, open conversion should be considered to reduce the high recurrence rate following MIS for CDH. This idea is supported by a recent prospective, multicenter study in which the open patch repair group has a lower recurrence rate of 1.6% compared to the higher recurrence rate of 8.8% in the MIS patch repair group [7]. In addition, if a large diaphragmatic defect is observed, MIS repair should not continue. Although patch repair of a large defect is technically possible, the additional time required for patch repair using MIS can increase the risk to the patient during the course of the operation [15-17]. From these selection criteria of MIS, excluding a patient with a large diaphragmatic defect who was converted to open patch repair, no recurrence of CDH was reported in this study.

Among postoperative outcomes, time to initiation of enteral feeding and length of the hospital stay were much shorter with MIS compared with open surgery. Gourlay et al. [5] reported shorter postoperative days until full enteral feeding, leading to a shorter hospital stay and much lower hospital charges. Because of these successful results and financial benefits, more efforts are needed to select appropriate candidates for MIS.

The overall incidence rate of the postoperative complications in this study was 25.7%, which is comparable to what has been reported in the literature [2,5,8]. Although the rate of postoperative complications was not different between the groups, one patient in the MIS group suffered postoperative subcutaneous emphysema because of pneumothorax during thoracoscopy. The severity of the complications was also not different. Esophageal injury occurred in the MIS group that converted to open surgery, and bowel perforation and major vessel bleeding occurred in the open surgery group. In a recent systematic review, no evidence was found for a decreased risk of postoperative complications with MIS [2], and this study supports that finding.

One of the major limitations of this study is the retrospective nature. The size of the diaphragmatic defect was measured during the operation and noted in the medical chart. However, in eight of thirty-five patients (22.9%), no data about the size of the diaphragmatic defect were found. The second limi tation in this study is that it was not a randomized study, and selection bias between the two groups may have existed. The selection bias for patient grouping and preference of the surgeon for the surgical approach may have affected the results. In particular, patients who had combined anomalies or a large diaphragmatic defect necessitating a patch repair were included only in the open surgery group, which may have created a selection bias between the groups. Postoperative outcomes are affected by disease severity including pulmonary dysplasia, pulmonary hypertension, prematurity, or combined anomaly [18]. In this study, although no significant difference was found in the postoperative complication rate or mortality between the two groups, a larger diaphragmatic defect size in the open surgery group could induce pulmonary dysplasia or pulmonary hypertension, making the postoperative course poorer. Moreover, combined anomalies that were pre sent only in the open surgery group or the more frequent delayed presentation in the MIS group may have affected the postoperative outcome. Lastly, the follow-up duration was not long enough to evaluate differences in long-term effects between the two groups. Long-term follow-up is needed to compare late outcomes such as morbidity, recurrence rate, and functional results following both surgical approaches.

In conclusion, to achieve early enteral feeding and a short hospital stay without recurrence, MIS repair of CDH can be suggested as the treatment of choice in neonates with stable hemodynamics and without combined anomalies, or in infants with delayed presentation of CDH, if the diaphragmatic defect is small enough to be repaired primarily without the necessity of a patch repair.