Experience of Minimally Invasive Surgery in Neonates with Congenital Malformations in a Tertiary Care Pediatric Hospital.

AIM: The aim of this study is to report the experience with minimally invasive surgery (MIS) in neonates with congenital malformations in a tertiary care pediatric hospital.
MATERIALS AND METHODS: Design: descriptive study. All neonates undergoing MIS from 2013 to 2018 were included in the study. Perinatal data, characteristics of surgery, type and duration of analgesia, postoperative mechanical ventilation duration, postoperative hospitalization, and postoperative morbidity and mortality surgery-related rates were recorded.
RESULTS: Seventy-one neonates were included. Gestational age and weight at surgery ranged from 24 to 41 weeks and from 1350 g to 4830 g, respectively. Procedures performed were esophageal atresia/tracheoesophageal fistula repair, congenital diaphragmatic hernia repair, diaphragmatic plication, fundoplication/gastrostomy, intestinal atresia repair, and pancreatectomy. The median follow-up period was 14 months. Five neonates (7%) were converted to open, for surgical difficulties. Nine (12.6%) neonates had intraoperative complications, with decreased oxygen saturation as the most common complication. The median duration of analgesia and postoperative mechanical ventilation was 3 days in most procedures. The morbidity and mortality rates were 36.6% and 2.8%, respectively.
CONCLUSIONS: In this first experience with MIS in neonates, the duration of analgesia and hospitalization was shorter for some procedures. However, intraoperative and postoperative complications were still high, which was possibly attributed to the learning curve. Thus, it is expected that the frequency of the complications presented in this study will be reduced in future. Copyright:

INTRODUCTION

Neonatal surgery is one of the most complex disciplines of pediatric surgery due to the fact that neonates have incomplete maturation of their organs and are in a period of physiological adaptation, making them vulnerable to the development of complications when undergoing surgical procedures.[12]

Minimally invasive surgery (MIS) is one important advancement in surgery in the last decades, including neonatal surgery.[234] However, neonates have different physiologic and anatomic characteristics that predispose them to greater surgical complications. Hence, serious concerns about the safety and efficacy of MIS in the neonatal stage still exist.[13567]

MIS has potential advantages, including less postoperative pain, shorter hospitalization time, decreased bowel adhesions, better esthetic results, and reduced skeletal deformities.[289]

Considering the development of smaller instruments and training of pediatric and neonatal surgeons, MIS is currently performed in neonates with various congenital anomalies. Thoracoscopic esophageal atresia/tracheoesophageal fistula (EA/TEF) repair,[101112] Bochdalek and Morgagni-type hernia repairs,[1314151617] and laparoscopic repair of various intestinal malformations[181920] have been reported. It has also been used for fundoplication with or without gastrostomy tube insertion[2122] and for diaphragmatic eventration.[2324]

The need for conversion to open surgery varies widely, from 0% to 75%.[59151723] The complication rate in different reports ranges from 0% to 33.3%.[292324] In laparoscopic surgery, it ranges from 0% to 67%.[8] The mortality surgery-related rate ranges from 0% to 7%[2815]

This study is aimed to report our initial experience with MIS in neonates with congenital malformations in a tertiary care pediatric hospital.

MATERIALS AND METHODS

A prospective observational study was conducted in a tertiary-care level neonatal intensive care unit (NICU) of the Pediatric Hospital of the 21st Century National Medical Center of the Mexican Social Security Institute, Mexico City. Seventy-one neonates who underwent laparoscopic or thoracoscopic surgery from January 2013 to December 2018 were included. The following data were recorded: gestational age, birth weight, sex, diagnosis, associated malformations, age and weight at surgery, surgery duration, conversion to open surgery, intraoperative bleeding, intraoperative complications, pre- and post-operative arterial pressure of carbon dioxide (paCO2), postoperative morbidity surgery-related, type and duration of postoperative analgesia, postoperative mechanical ventilation duration, fasting time, postoperative hospitalization, and mortality surgery-related.

Inclusion criteria

Patients experiencing hemodynamic and respiratory stability at the time of surgery, regardless of gestational age, weight, or the need for mechanical ventilatory support before undergoing surgery. No patient had high-frequency ventilation at the time of surgery.

All surgeries were performed using basic instruments (Maryland dissector, monopolar energy, atraumatic forceps, and endoscopic scissors). In some cases, an electronic dissection device (Sonicision™ cordless ultrasonic device, Covidien) was used. A 5-mm high-definition camera was used at 30° in all procedures.

Laparoscopic surgeries were performed with patients assuming the supine position using two or three 3-mm work ports. The primary port was inserted with an open technique in the left supra-umbilical region, transcutaneous points were used to expose the surgical area, and 5-0 G sutures with a 3/8 needle (11–13 mm) were used. CO2insufflation was performed with pressures of 4–6 mmHg and flows of 1–3 L/min.

Thoracoscopic surgeries were performed with patients in the supine lateral position of the affected side at 45°. Two or three 3-mm work ports were used. Pulmonary collapse with CO2insufflation was obtained with positive pressures of 4–6 mmHg. In most cases, transcutaneous sutures were used to elevate the adjacent tissues and obtain better visibility. In congenital diaphragmatic hernia (CDH), to prevent organ injury, atraumatic instruments were used gently, and cotton tape pulled by two instruments was used to reduce the spleen back into the abdominal cavity. To perform these procedures, adequate muscle relaxation by the anesthesiologist was necessary. In some patients, a thoracic drainage tube was inserted through the lower port.

In both laparoscopic and thoracoscopic surgeries, the ports were fixed to the skin with transcutaneous sutures to prevent mobilization during the procedure, and the diameter of the insufflation tube was reduced using an extension intravenous tube.

All patients were monitored by electrocardiography, noninvasive blood pressure monitor, pulse oximeter, axillary skin temperature, and end-tidal CO2monitor. Blood gases were obtained during the surgical procedure. All procedures were performed by the same surgeon but with different anesthesiologists.

Statistical analyses

Descriptive statistics were used, with the calculation of frequencies and percentages and measurements of central tendency, median, and range. Wilcoxon or Chi-squared tests were used for comparison between groups.

This study was approved by the Ethics Institutional Review Board (approval number: R-2015-3603-7).

RESULTS

During the study period, 150 neonates who required surgery were admitted to the NICU, of whom 71 met the criteria for MIS (37 girls and 34 boys) and were included in this study. The median gestational age was 36 weeks (range, 24–41 weeks), the median age at surgery was 7 days (range, 2–75 days), and the median weight at surgery was 2500 g (range, 1350-4830 g). Thirty-five neonates (49.3%) were premature.

A total of 46 thoracoscopic and 25 laparoscopic procedures were performed. The thoracoscopic procedures performed were EA/TEF repair, CDH repair, and diaphragmatic plication. The laparoscopic procedures performed were duodenal and jejunal atresia repair, pancreatectomy, Morgagni-type CDH repair, and Nissen fundoplication with or without gastrostomy. The indication for fundoplication was severe gastroesophageal reflux disease with bronchoaspiration.

Forty-one patients (57.7%) had other congenital anomalies or genetic disorders in addition to the diagnosis for which surgery was indicated [Table 1].

Table 1

Associated congenital malformations in the newborns based on the type of surgical procedure

Type of procedure/Type of malformation	n (%)
Thoracoscopic surgery	29/46 (63)
EA/TEF repair	15/22 (68.2)
VACTERL association	9
Ventricular septal defect	2
Down syndrome	1
Craniosynostosis	1
Bronchogenic cyst	1
Multiples anomalies	1
Bochdalek-type CDH repair	7/17 (41.2)
Fryns syndrome	2
Simpson-Golabi-Behmel syndrome	2
Down syndrome	1
Coarctation of the aorta	1
Suspected Okihiro syndrome	1
Diaphragmatic plication	7/7 (100)
Complex congenital heart disease	6
Cystic adenomatoid malformation	1
Laparoscopic surgery	12/25 (44)
Fundoplication/gastrostomy	9/15 (60)
Esophageal atresia type 1	2
VACTERL association	2
Neuronal migration disorder and renal anomalies	1
Mesenchymal hamartoma of the chest wall and rib agenesis	1
Hydrocephalus	1
Dilated myocardiopathy	1
Suspected Smith-Lemli-Opitz syndrome	1
Morgagni-type CDH repair	1/1 (100)
Noonan syndrome	1
Duodenal atresia repair	1/4 (25)
Intestinal malrotation	1
Pancreatectomy	1/3 (33.3)
Ventricular septal defect	1
Total	41/71 (57.7)

EA/TEF: Esophageal atresia/tracheoesophageal fistula; CDH: Congenital diaphragmatic hernia

Table 2 presents the general characteristics of the patients and surgical procedures.

Table 2

General characteristics of newborns and surgical procedures

	n	GA* (weeks)	PA* (days)	WS* (g)	SD* (min)	Bleeding* (ml)	Conversion (%)	SC (%)	Morbidity (%)	Mortality (%)
Thoracoscopic surgery (n=46)
EA/TEF repair	22	36 (34-39)	3 (2-14)	2300 (1700-3150)	150 (90-325)	5 (2-30)	1 (4.5)	4 (18.2)	11 (50)	0
CDH repair	17	37 (33-41)	6 (3-56)	2640 (1980-3980)	180 (120-240)	10 (3-50)	3 (17.6)	3 (17.6)	9 (52.9)	1 (5.8)
Diaphragmatic plication∞	7	32 (29-39)	31 (26-93)	2380 (1350-2900)	120 (60-160)	5 (1-7)	0	1 (14.2)	1 (14.2)	0
Laparoscopic surgery (n=25)
Fundoplication/gastrostomy	15	35 (24-39)	66 (2-140)	2460 (1610-4120)	150 (60-270)	10 (3-60)	1 (6.6)	-	1 (6.6)	0
Duodenal atresia repair**	4	37 (35-40)	6 (2-14)	2425 (1660-3270)	105 (60-120)	5	0	0	2 (50)	0
Pancreatectomy	3	39 (39-40)	41 (41-42)	4765 (4700-4830)	150 (120-180)	30 (30-50)	0	0	0	0
Jejunal atresia repair	2	38 (36-40)	6 (5-7)	2100 (1700-2500)	165 (150-180)	7 (5-10)	0	1 (50)	2 (100)	1 (50)
Morgagni-type CDH repair	1	38	44	3200	120	5	0	0	0	0
Total	71						5 (7.0)	9 (12.6)	26 (36.6)	2 (2.8)

*Expressed as the median and range values, **Kimura procedure was performed in all neonates; in one neonate, Ladd procedure was also performed, ∞Patch closure was performed in three neonates. EA/TEF: Esophageal atresia/tracheoesophageal fistula, CDH: Congenital diaphragmatic hernia, GA: Gestational age, PA: Postnatal age, WS: Weight at surgery, SD: Surgery duration, SC: Surgical complications

Sixty-six procedures (93%) were successfully completed with MIS, and five (7%) neonates required conversion to open procedure.

Nine neonates (12.6%) had intraoperative complications. In thoracoscopic surgery, intraoperative complications were more frequent in EA/TEF and CDH repairs. In laparoscopic surgery, only one complication was observed. Table 3 presents the intraoperative complications and reasons for conversion.

Table 3

Intraoperative complications and the reasons for conversion to open surgery

Procedure	Intraoperative complications (%)	Reasons for conversion (%)
Thoracoscopic
EA/TEF repair	4/22 (18.2)	1/22 (4.5)
	Decreased O2 saturation and bradycardia (4)	Poor visibility (1)*
CDH repair	3/17 (17.6)	3/17 (17.6)
	Decreased O2 saturation and bradycardia (2)	Technical difficulty (2)**
	Gastric and colon perforation (1)	Poor visibility (1)
Diaphragmatic plication	1/7 (14.3)	0
	Pneumothorax (1)
Laparoscopic
Fundoplication/gastrostomy	0	1/15 (6.6)Poor visibility (1)***
Jejunal atresia repair	1/2 (50)	0
	Bradycardia (1)
Duodenal atresia repair	0	0
Pancreatectomy	0	0
Morgagni-type CDH repair	0	0
Total	9/71 (12.6)	5/71 (7.0)

*Pulmonary overdistention, **Patch closure required in two neonates due to a large defect, ***Due to heavy bleeding. EA/TEF: Esophageal atresia/tracheoesophageal fistula, CDH: Congenital diaphragmatic hernia

In the early postoperative period, there was no significant difference between pre- and post-surgical pCO2(P = 0.8). The median of analgesia duration and the median of postoperative mechanical ventilation duration were 3 days in most procedures [Table 4]. Neonates with postoperative morbidity had a longer duration of analgesia (P = 0.02), mechanical ventilation (P = 0.004), fasting time (P = 0.04), and hospital stay (P = 0.04) than neonates with no postoperative morbidity.

Table 4

Postoperative characteristics in thoracoscopic and laparoscopic surgery

	Thoracoscopic surgery (n=46)			Laparoscopic surgery (n=25)
	EA/TEF repair (n=22)	CDH repair (n=17)	Plication (n=7)	Fundoplication/gastrostomy (n=15)	Duodenal repair (n=4)	Pancreatectomy (n=3)	Jejunal repair (n=2)	MDH repair (n=1)
Type of analgesic
Buprenorphine (n)	22	16	5	10	4	3	1	0
Fentanyl (n)	2	5	0	0	0	3	1	0
Morphine (n)	0	2	0	0	1	0	0	0
Ketorolac (n)	16	7	4	9	3	3	1	1
Duration of analgesia (days)	4 (3-5)	4 (1-7)	3 (1-4)	3 (2-4)	3 (3-6)	3 (2-3)	3 (3-5)	4
MV duration (days)	5 (1-25)	7 (1-14)	3 (1-6)	4 (0-7)	2 (1-3)	1	3 (2-4)	6
Preoperative paCO2 (mmHg)	32 (25-64)	39 (21-56)	32 (28-54)	47 (20-77)	33 (14-40)	45 (42-49)	29 (23-35)	30
Postoperative paCO2 (mmHg)	38 (17-61)	48 (26-67)	35 (30-62)	34 (23-60)	42 (30-45)	45 (43-47)	29 (26-32)	26
Fasting time (days)	8 (3-20)	6 (2-16)	1.5 (1-4)	4 (2-6)	12 (7-17)	5	8	5
Hospital stay (days)	14 (3-123)	15 (2-111)	7 (2-9)	8 (5-21)	21 (10-57)	5 (3-7)	10 (8-13)	9

Data are expressed as the median with the range in parenthesis. EA/TEF: Esophageal atresia/tracheoesophageal fistula, CDH: Congenital diaphragmatic hernia, MDH: Morgagni-type diaphragmatic hernia, MV: Mechanical ventilation

The follow-up ranged from 4 months to 4 years (median, 14 months). During the follow-up, 26 (36.6%) neonates developed different types of morbidity surgery-related [Table 5]. The complications were more frequent in thoracoscopic surgery, including EA/TEF repair (54.5%) and CDH repair (52.9%). However, 50% of neonates undergoing EA/TEF repair and 30% undergoing CDH repair, who developed postoperative morbidity had concomitant anomalies, but there was no significant difference when comparing to those who did not have other anomalies (P = 0.86). Some neonates with EA/TEF had VACTERL association, and as part of this association, they had anorectal malformation or duodenal atresia and required abdominal surgery before EA repair. Other neonates required other surgeries for cardiac or renal anomalies. Patients with VACTERL association had a higher frequency of postoperative morbidity related to surgery and a longer hospital stay.

Table 5

Postoperative morbidity according to the surgical procedure

Thoracoscopic surgery (n=46)	n (%)	Laparoscopic surgery (n=25)	n (%)
EA/TEF repair	11/22 (50)*	Fundoplication/gastrostomy	1/15 (6.6)
Esophageal stricture	7	Leakage from gastrostomy site	1
GERD	4	Jejunal atresia repair	2/2 (100)
Anastomotic leak	3	Anastomotic leak	1
Sepsis	3	Intestinal obstruction/enterocutaneous fistula	1
Pleural effusion	2	Duodenal atresia repair	2/4 (50)
CDH repair	9/17 (52.9)*	Anastomotic leak	2
Pleural effusion	3	Morgagni-type CDH repair	0/1
Sepsis	3	Pancreatectomy	0/3
Pneumothorax	2
Chylothorax	2
Pneumonia	2
GERD	2
Intestinal perforation and intra-abdominal abscess	1
Diaphragmatic plication	1/7 (14.2)
Recurrence eventration	1
Total	21/46 (45.6)		5/25 (20)

*More than one complication was observed in five patients. In CDH repair, one neonate required the correction of intestinal malrotation at follow-up. EA/TEF: Esophageal atresia/tracheoesophageal fistula, CDH: Congenital diaphragmatic hernia, GERD: Gastroesophageal reflux disease

Most complications related to surgery were observed in the first 4 years (2013–2016) after MIS began, and in the past 2 years (2017–2018), the number of complications diminished. During the latter period, in the EA/TEF repair, the anastomotic leak was no longer observed.

In laparoscopic surgery, two neonates with jejunal atresia repair had complications. One of them had anastomotic leak, peritonitis, and septic shock and eventually died. The other neonate had intestinal occlusion due to multiple adhesions and enterocutaneous fistula and subsequently developed short bowel syndrome.

The long-term esthetic results were adequate, the scars were minimal and almost imperceptible after 3 years. Musculoskeletal deformities were not observed in children undergoing thoracoscopy, except in one neonate with EA/TEF repair with VACTERL association who developed mild scoliosis.

The mortality related to surgery rate was 2.8% (n = 2). One death was observed in a neonate with jejunal atresia repair. The other death was observed in a neonate undergoing CDH repair and diagnosed with Simpson-Golabi-Behmel syndrome who eventually developed severe pulmonary hypertension after surgery.

DISCUSSION

In the hospital where the present study was conducted, MIS in the neonatal stage began in 2013, and we report our initial experience with MIS in 71 neonates.

The lowest weight at surgery was 1350 g in a premature neonate. In the literature, MIS is reported in neonates with weight as low as 1200 g.[4] Currently, patient size is not a factor that contraindicates MIS in neonates. In this study, low weight was not associated with complications (P = 0.71), the same as reported by Kalfa et al.[6] At present, even complex cardiac anomalies are not a contraindication for MIS.[25]

The frequency of complications, both intraoperatively and postoperatively, was high, but within that reported in the systematic review (SR) by Sinha et al.[8] These authors reported complication rates of 67% in EA/TEF repair and 12% in CDH repair versus 18.2% and 17.8%, respectively, in this study.

A high proportion of patients (57.7%) had comorbid conditions, including congenital anomalies, genetic disorders, or complex cardiac diseases, which were mainly observed in neonates undergoing thoracoscopic procedures. Associated malformations have been reported in a few studies.[212]

The conversion rate was lower than that reported by Kalfa et al. (7% vs. 15%, respectively).[6] The causes were mainly due to technical difficulties, primarily the lack of visibility as a result of pulmonary hyperinflation or bleeding, and in two neonates with CDH repair because the defect was very large and required patch placement.

The most frequent intraoperative complications were anesthetically related, mainly in thoracoscopic procedures, due to the lack of tolerance to CO2insufflation, causing a decrease in oxygen (O2) saturation (<89%) and bradycardia, which were managed with increased minute ventilation. This resulted in greater pulmonary insufflation and visibility interference, preventing the surgeon from continuing with the procedure. Considering these complications, the need for conversion to open surgery was observed in two neonates. In the other three neonates, only temporary suspension of the procedure was required, and these complications were treated with ventilation and the intraoperative use of inotropic. Parelkar et al.[9] reported that 50% of their patients developed decreased oxygen saturation, suggesting that pediatric surgeons and anesthesiologists must be widely and skillfully trained in performing neonatal MIS.

Another possible intraoperative complication in MIS is the injury of adjacent organs.[8] This complication was observed in a neonate with gastric and colon perforation undergoing CDH repair. In SR by Sinha et al.[8] colon perforation was also reported in a neonate undergoing CDH repair. In diaphragmatic plication, one patient developed pneumothorax immediately after surgery; hence, it is possible that an inadvertent lung injury during the procedure occurred, although other causes could be the incomplete evacuation of insufflated CO2and even barotrauma. A chest tube was placed, and the pneumothorax was resolved. Therefore, it is crucial to completely evacuate the insufflated gas at the end of any thoracoscopic procedure and verify the presence of residual pneumothorax with a chest X-ray.

In CDH repair, the rate of morbidity surgery-related was higher (52.9%) than that reported in SR by Vijfhuize et al.,[13] which ranged from 6.7% to 33.3%. However, the mortality surgery-related rate (5.8%) was similar to that reported by Vijffhuize et al., which ranged from 0% to 7%.

In neonates undergoing duodenal obstruction and jejunal atresia repairs by laparoscopic surgery, the frequency of postoperative complications was high, and they were serious complications. Considering these complications, in the hospital where the study was conducted, laparoscopic repair for intestinal malformations was stopped. In addition, a specific training program in abdominal MIS was conducted. Van der Zee[19] at an early stage reported a high incidence rate (22.7%) of anastomotic leak in duodenal obstruction repair via laparoscopic surgery. Thus, he immediately stopped this procedure, and in a second stage, to improve his surgical technique, he reported that this procedure had no complications.

It has been considered that MIS, due to small incisions, there is less pain. However, the insufflation of a cavity with gas can be painful, which is even more painful when the duration of surgery is extended. In these situations, the postoperative pain may be greater.[26] In this study, in some neonates, the duration of analgesia was prolonged because they developed complications such as pleural effusion, chylothorax, and anastomotic leak, or they required chest tubes for a long time.

Even though neonates, and mainly premature infants, are specifically sensitive to CO2absorption during laparoscopic or thoracoscopic surgery and can easily develop respiratory acidosis,[2728] in our patients there was no difference between the value of pCO2before and after the surgery, and respiratory acidosis development was not reported.

Considering that the physiological and anatomic characteristics of neonates, specifically premature infants, are different from those of other pediatric ages, these patients are more prone to develop complications with MIS. Hence when performing thoracoscopic or laparoscopic procedures in neonates, the participation of a well-trained anesthesiologist is crucial to maintain temperature and pulmonary and cardiac function that can deteriorate during CO2insufflation. Muscle relaxation is also important so that the surgeon can perform the procedures with greater safety. Moreover, pediatric surgeons must have extensive training and experience to identify and resolve complications that may occur during surgery and minimize these complications to the maximum, which is an important objective of MIS. The neonatal team at the NICU is also fundamental for the pre- and post-operative management of these patients. The multidisciplinary approach, including surgeons, anesthesiologists, and neonatologists, has been one of the most significant advancements in recent years, which allows the performance of MIS in neonates.[1]

One of the limitations of this study is that pain assessment was not standardized. Thus, analgesic management was per the attending neonatologist' discretion, and this may partly explain the variability in the type and duration of analgesia.

CONCLUSIONS

In this study assessing the first experience with MIS in neonates with congenital malformations, including complex malformations, some advantages were observed, in some procedures such as diaphragmatic plication, fundoplication, and pancreatectomy, the duration of analgesia and hospital stay, were shorter compared to other procedures, and the best esthetic result was observed in the medium term in all infants. However, overall intraoperative and postoperative complications rates were high, mainly in the first 4 years after MIS was started in neonates. Therefore, the learning curve should not be underestimated since it has been suggested that the duration of this period may be shorter if formal training in MIS is performed by both pediatric surgeons and anesthesiologists, and combined with the development of smaller instruments, may decrease the rate of complications related to surgery. Thus, even children with complex congenital malformations, including congenital heart disease, may benefit from MIS. Furthermore, a longer follow-up period is necessary for evaluating the long-term complications. In addition, long-term follow-up studies that evaluate the effect of pneumoperitoneum with CO2on the neurological development of children undergoing MIS in the neonatal stage are required.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.