Staged surgical approach of neonates with esophageal atresia and tracheoesophageal fistula from low- and middle-income countries

Neonates born in low- and middle-income countries (LMICs) with esophageal atresia (EA) and tracheoesophageal fistula (TEF) often do not have access to adequate surgical care. We have partnered with the non-profit organization World Pediatric Project (WPP) to facilitate care for such patients. Our protocol included placement of a gastrostomy tube by local surgeons before definitive repair at the Children’s Hospital of Richmond at VCU (CHoR). The outcomes of WPP-sponsored patients with EA and TEF in this program (n = 9) were compared to patients born in the U.S. with the same condition at CHoR (n = 9). The baseline characteristics of the groups aside from the age at admission to CHoR and at day of surgery were comparable (0.5 ± 1.3 days vs. 15.3 ± 11.1 days, p < 0.0001; 1.1 ± 3.9 days vs. 37.3 ± 8.3 days, p < 0.0001, respectively). All patients received definitive repair without mortality; the average length of stay after surgery was lower in the WPP group. There were 4 surgical complications in the U.S. group and 1 in WPP group. All patients were weaned off of TPN and all WPP patients are now on a regular PO diet. Every WPP patient has been seen in follow-up in the U.S. and in their home country. We provided successful multidisciplinary care for neonates with EA and TEF from LMICs with outcomes similar to the neonates with the same abnormality in the U.S. The partnership with WPP has been invaluable as it has provided essential support to identify and manage these patients. IV.


Introduction
The need for global pediatric surgical care is more pressing than ever. Current research estimates that low-and middleincome countries (LMICs) need 143 million surgical procedures annually to save lives and prevent disability, and children under 15 years old comprise 50% of the population in such countries [1,2]. Previously thought to be too expensive and nonessential, pediatric surgery is now considered a high priority as it directly correlates with economic development in LMIC and quality adjusted life years for patients [2]. However, many challenges result in inadequate pediatric surgical care in these countries. LMICs are limited by a shortage of pediatric surgeons (the surgeon to children ratio is 1:6 million in LMICs compared to 1:100,000 in the USA), unwillingness and inability to anesthetize children, and lack of pediatric-specific supplies [2][3][4]. Furthermore, operating room (OR) time is more often allocated toward treating adult patients as they can advocate for themselves in ways children cannot [3]. Children who require surgical interventions subsequently are going untreated and/or dying in LMICs from treatable conditions. This is often the case with infants that have esophageal atresia (EA) and tracheoesophageal fistula (TEF). Historically, without modern surgical intervention, the mortality rate of patients born with esophageal atresia has been recorded to be as high as 61% during 1947-1968 [5]. With the evolution of surgical treatment, the mortality rate has decreased significantly and now is estimated to be less than 10% [6]. Thoracoscopic approach has been shown to have similar rates of complications as the open approach while decreasing time to extubation, first oral feeds and length of stay [7,8]. Today, the total complication rate after surgery is reported to be about 37% [5]. Specifically, the most common complications from EA with distal TEF are reported to be anastomotic stricture requiring intervention (43%) and leak (18%) [9]. Current surgical therapy of this congenital anomaly is effective in allowing the patient to have per oral (PO) diet and if not, enteral nutrition [10].
Despite the advancement in surgical therapy, infants with EA and TEF in LMICs such as St. Lucia and Belize have not had access to adequate surgical care. There is currently minimal specialized pediatric surgical care available in these countries though the formal ratio of pediatric surgeon to children is unknown currently. The Children's Hospital of Richmond (CHoR) at Virginia Commonwealth University and the nonprofit organization World Pediatric Project (WPP) have collaborated to alleviate this disparity. Over the last 15 years, our pediatric surgeons have taken surgical teams to the Caribbean region to provide valuable pediatric surgical care. We previously described this successful partnership in treating 255 pediatric patients from 14 different countries who received a total of 371 complex surgeries [11]. Building on this extensive experience, we have partnered with WPP to provide highly specialized and responsible surgical care to patients with EA and TEF.
We introduced this partnership in our previous publication in which we describe the logistics and the financial aspects of this program [12]. Our program uses staged repair to treat these patients with EA and TEF; this approach had traditionally been reserved for patients with EA type A with a long gap, premature neonates and those with severe respiratory distress [5,[13][14][15]. We are not aware of any other international referral program with an established protocol to provide proper care for these patients. This study elucidates the feasibility of such a program and the successful outcomes of a staged surgical approach.

Methods
After obtaining the appropriate Institutional Review Board approval (#HM20013739), a retrospective review was performed for control patients with EA and TEF type C who were born in the U.S. and received their care at CHoR from September 2007 to October 2020 (control group) and WPP sponsored patients diagnosed with EA and TEF type C who received their care at CHoR from January 2009 to May 2020 (study group). The following data were obtained from the chart review: demographic information, age at time of transfer and the operation, interventions they had received in their home country, comorbidities, type of definitive surgical intervention, complications, any additional surgeries, mortality, length of stay in the hospital, number of patients on per oral diet or enteral nutrition via a feeding tube, number of days of total parenteral nutrition (TPN), TPN associated complications including cholestasis, postoperative emergency department visits, readmissions or transfers to CHoR, number of postoperative follow-up visits (in the U.S. and home country) and number of years until the last follow-up by the surgical team.
Our protocol for neonates with EA and TEF type C from LMICs has been described by our previous publication [12]. Any referral to WPP is first triaged by a WPP contact in the region consisting of local pediatricians. If the diagnosis is suspected, the head of bed is elevated above 45° and salem sump is left in the proximal esophageal pouch. The patient is then kept nil per os (NPO) and intravenous fluid is started. Depending on the local resources, peripheral parenteral nutrition or TPN is started if central access can be established. The diagnosis is then confirmed with an esophagogram using the barium contrast and arrangements are made with local general surgeons to place a gastrostomy tube for decompression. Typically, a MIC tube (Avanos, Alpharetta, Georgia) provided by the WPP team is placed as a gastrostomy tube using a laparotomy incision; if this is not accessible, a Foley with a bumper is placed instead. One patient also received a jejunostomy tube for enteral access. The pediatric surgeons at CHoR are consulted via telemedicine once the diagnosis is confirmed.
Once the patient is deemed to be stable and all the travel documents have been completed for the patient and the guardian, WPP arranges for the patient to be transferred to the neonatal ICU at CHoR for further care. The patient is transported via commercial or private flight and is accompanied by a family member and a medical provider. The sump in the proximal pouch is kept decompressed by intermittent syringe suction and the baby is transported in a car seat to keep the head elevated.
Primary repair of the EA and TEF is performed once the patient is in an anabolic state. PPI is started postoperatively. Routine esophagogram is obtained on postoperative day 5 before starting nutrition by mouth. If the study is negative for an anastomotic leak, chest tube is discontinued while the patient is started on PO feeds with occupational therapy support. In addition, continuous feeds are started via the gastrostomy tube. Appropriate gastrostomy tube education is provided to the parent while attempting to transition from tube feeds to PO feeds. The patient is discharged when he or she is medically stable, receiving stable enteral nutrition and off TPN.
After discharge, the patient and family member stay in a nearby hospitality house for further follow-up for about a week or two until cleared by the surgeon to return to their home country. WPP sends visiting surgical teams every year to continue to follow-up with patients abroad. The same CHoR pediatric surgeons, in partnership with WPP, travel to the Caribbean at least twice per year to provide follow-up care of these patients in conjunction with their local providers.
Patients diagnosed with EA and TEF in the U.S. and treated at CHoR receive an open or thoracoscopic repair without a gastrostomy tube placement. Similar to the study group patients, a routine esophagogram is performed on postoperative day 5; if negative, the patient is started on nutrition by mouth and the chest tube is removed. TPN is weaned off once adequate PO intake was achieved. Once the patient is on a stable enteral nutrition without TPN and medically stable, he or she is discharged.
Statistical analysis was performed using the GraphPad Prism program (San Diego, California, USA). The outliers were identified using Grubbs' test and excluded from the statistical analysis. Unpaired T test was performed to compare the baseline characteristics and outcomes of the control and study groups. Statistical significance was determined when p was less than 0.05.

Results
There were nine patients with EA and TEF type C who were born in the U.S. and treated at CHoR from September 2007 to October 2020. There were five male and four female patients. Two patients were admitted from the emergency department while the rest were direct admits to the NICU after birth. Meanwhile, there were nine patients with EA and TEF who were sponsored by WPP and treated at CHoR from January 2009 to May 2020. Six of the patients were male while three were female. Eight patients were from Belize while one patient was from St. Lucia.
The average gestational age of the control and the study groups was comparable (36.6 ± 3.3 weeks vs. 36.4 ± 3.6 weeks, p = NS). The average age of the control was significantly less than the study group at the time of admission to CHoR and at the time of surgery was significantly different (0.5 ± 1.3 days vs. 15.3 ± 11.1 days, p < 0.0001; 1.1 ± 3.9 days vs. 37.3 ± 8.3 days, p < 0.0001, respectively). Of note, the age of the patient with the jejunostomy tube was a statistical outlier (147.7 days at the time of transfer and 154.0 days at the time of surgery) and was therefore excluded.
The number of comorbidities of the patients was comparable in both groups (2.4 ± 1.8 vs. 1.8 ± 2.0, p = NS) and varied widely (range 0-6 comorbidities; Table 1). In particular, while three patients in the study group had no comorbidities, there were others whose conditions ranged from patent foramen ovale with moderate left to right shunting to chronic kidney disease with hydronephrosis.
In regard to the interventions done at the hospitals abroad, all of the patients had a gastrostomy tube placed by the general surgeon for decompression and one patient also had a jejunostomy tube for enteral access due to an anticipated prolonged delay in transfer secondary to severe aspiration pneumonia. Eight of the nine patients had been diagnosed with aspiration pneumonia and were treated with antibiotics before transfer. Two patients arrived requiring oxygen via nasal cannula while the rest were on room air. All patients were on NPO restrictions. Three patients were placed on total or peripheral parenteral nutrition at the outside hospital and the patient with a jejunostomy tube was able to have tube feeds without needing TPN. The average weight at the time of surgery was comparable between the control and study groups (2.6 ± 0.7 kg vs. 3.1 ± 0.8 kg, p = NS; Table 1). All of the patients had a successful primary repair of the EA and ligation of the TEF without any immediate complication by the pediatric surgeons at CHoR. The number of the thoracoscopic cases was comparable between the control and study groups (3 of 9 vs. 4 of 9 cases, p = NS; Table 2). The complication rates associated with the EA and TEF repair were comparable though it was higher in the control group (4 of 9 vs. 1 of 9 patients, p = NS; Table 2). There were three esophageal leaks and one esophageal stricture in the control group; two of these complications required additional surgical repairs while the rest were managed conservatively. One patient in the study group developed an empyema, which was treated conservatively with a drain and antibiotics ( Table 2). Three of the nine patients in the study group had complications associated with the gastrostomy tube placement ranging from stomach dehiscence to cellulitis site at the stoma.
In the immediate postoperative stage, all study group patients received enteral nutrition using the gastrostomy tube while developing oral motor skills. Five of the nine patients were discharged still dependent on their gastrostomy tube for adequate nutrition. There were three patients who had significant reflux symptoms; two patients were started on famotidine which controlled the symptoms while one patient underwent laparoscopic Nissen fundoplication without complication. Meanwhile, four of the nine control group patients required postoperative placement of gastrostomy tube (one patient also had Nissen fundoplication) due to various reasons including poor oral motor skills and severe reflux.
In regards to parenteral nutrition, all of the nine patients in the control group received preoperative TPN and eight of the nine patients in the study group were placed on preoperative TPN while one patient was able to continue his tube feeds via a jejunostomy tube. Five of the eight patients who required TPN in the study group required additional surgery to place a tunneled central line access. Two patients in each group had TPN associated cholestasis which resolved once they were transitioned to enteral feeds. Postoperatively, all patients had been weaned off their TPN and are on a regular PO diet.
All study group patients required a multidisciplinary team approach. A median of five different medical/surgical teams [3][4][5][6][7][8][9][10] were involved in each patient's care depending on the medical complexity of the patient. Other highly subspecialized pediatric services were also involved including pediatric cardiology and pediatric orthopedic surgery.
Once discharged, the patients in the study group stayed in the U.S. for follow-up care with their primary surgeon (average of 2.3 visits ± 1.2; range 1-4) until they were given clearance by the surgical staff to return to their home country (Table 3). Every patient received appropriate follow-up care by WPP-sponsored visiting surgery teams in conjunction with the pediatricians at their home country (average of 3.5 visits ± 2.7; range 1-7). The average time after discharge from the hospital to the last follow-up by visiting surgery teams or the local physician was 4.0 years ± 3.0 (range 0.9-8.6; Table 3). The current age of the patients ranged from 1 to 12 years (Table 3).

Discussion
While limited in number, we believe the outcomes of this program have been excellent. It is unlikely that the patients would have received adequate multidisciplinary care in their resource-limited home country, given the well-documented limited availability of subspecialized pediatric surgery services in Belize and St. Lucia. Given these circumstances and the reported mortality rate of 61% without modern surgical intervention, we suspect that the value of this collaboration has been immense in improving the life expectancy of these patients [5]. Though definitive repair of EA and TEF during the index surgery is preferred, this approach is not ideal for patients at a facility overseas due to the possibility of significant delays before transfer. Our outcomes suggest that staged repair of EA and TEF is safe and feasible for these patients. The gastrostomy tube allows for decompression of the stomach and prevents aspiration while the logistics are being set up for the patient to be transferred. It is a procedure that the general surgeons in these countries feel comfortable performing; however, there have been complications from the gastrostomy tube placements. One patient required an additional surgery for stomach dehiscence at their native country while 2 were diagnosed with cellulitis that was treated with antibiotics.
Despite these complications, all patients had a good outcome after the repair of their EA and TEF. The complication rates from the repair of EA and TEF between the two groups were comparable though it was lower in the study group. There was one complication in the study group (postoperative empyema) which was managed conservatively during the same admission. As a surveillance for complications, all the study group patients had stayed at a nearby hospitality house for at least few weeks of follow-up with their primary surgeons before returning to their home countries. Major outcomes such as mortality, postoperative nutritional status and overall length of stay were comparable. In fact, the postoperative length of stay was significantly shorter in the study group. This may have been due to the patients in the study group being older and able to be discharged from the NICU earlier as a result.
We want to highlight the importance of postoperative nutrition for these patients. The gastrostomy tube initially placed for decompression in these patients was used for enteral access as the patients learned oromotor skills; in fact, some of the patients were able to be transferred back home while still dependent on tube feeds because they had a stable enteral access. At this point, all patients are on a regular PO diet without TPN dependence. This is critical for these patients as their home conditions are not conducive to care for central line access and TPN.
Given the risk of anastomotic stricture after this type of surgery, it is imperative that WPP patients have long term follow-up. As a result, there has been an extensive effort to provide responsible follow-up care. All patients stayed in Richmond for at least a few weeks to receive outpatient care from specialists until cleared by the team. Once home, the visiting surgical teams from our institution and the local physicians provided follow-up care for all of these patients. The length of follow-up varied based on their progress; we have followed up with patients up to 8 years after discharge. None of the patients have required readmission to their local hospital or CHoR for further care but this can be arranged by WPP, if needed.
The partnership with WPP has been invaluable in the coordination of care for these patients. They provide essential logistical support in caring for pediatric surgical patients from Belize and St. Lucia in the identification as well as triage of patients. They have a strong relationship with the pediatricians from these countries and serve as a bridge between the primary care physicians and the specialists here at CHoR. Time is of essence in treating this condition as malnutrition may quickly result in mortality. In this program, the median age of the transferred patient was 3.1 weeks. The patient with the jejunostomy tube was an outlier (21.1 weeks at the time of transfer) as he had stable enteral access and could be transferred after recovering from a prolonged illness due to pneumonia and sepsis. This highlights the agency's efficiency in organizing the triage and transfer of the patient. WPP can arrange the logistics of travel for both the patient and the guardian and allow the guardian to stay with the patient until they are medically cleared to return to their home country. After discharge, WPP continues to follow the patients and arrange for longitudinal follow-ups. This program will continue to evolve; for example, jejunostomy tube placement is now being considered before transfer to CHoR. The gastrostomy tubes placed at the OSH are used for decompression as these patients are prone to life threatening aspiration events via the fistula. There was one patient who had a jejunostomy tube placed in addition to the gastrostomy tube due to a delay in transfer and inaccessibility of TPN; he was able to avoid a prolonged period of TPN as he had stable enteral access throughout his hospital stay. We recognize that jejunostomy tube may be necessary if there is a prolonged delay in transfer as it provides essential enteral nutrition and reduces potential complications of TPN and central lines.

Limitations
We recognize the importance of building a sustainable surgical capacity in these countries. In addition to providing medical care, surgery teams from WPP travel to these countries to provide important surgical education for the local general surgeons. However, surgical education is not the only hindrance to countries with limited resources; these countries often lack other essential resources such as ICU support, pediatric anesthesiology and pediatric nursing. As a result, building surgical capacity to a self-sustainable level will be a long and arduous process. There are currently capacity building efforts underway to support the development of advanced pediatric services in Belize including comprehensive neonatal care, a NICU training program and a pediatric anesthesia training program.
Meanwhile, the need for complex medical care continues to exist and this program aims to address this need. WPP relies on its referral process with local pediatrician partners to identify those children for whom there is a not an option for local care. WPP explores every local and regional resource within its network's capacity at the time of the child's referral before referring to its U.S. partners. We highlight this program not as a solution to end the global healthcare disparity but as an adjunct to address the dire situation of those whose lives are in danger without intervention. In December of 2020, the first case of EA and TEF repair in Belize was performed by a volunteering Cuban surgeon and we believe it needs to become the standard of care for these patients [16].
We also acknowledge that this study only highlights single institutional experience. WPP's model for mobilizing children to advanced centers for critical care is increasingly focused on growing a network of partnerships that allow children to be treated closer to home. Dozens of children with high-acuity surgical needs in various medical specialties receive care through regional partnerships in Latin America and the Caribbean each year through the WPP network. This network currently includes advanced referral centers in the Cayman Islands, Barbados, Martinique, Mexico, Colombia, Honduras, Guatemala and the Dominican Republic. This paper is a limited review of select cases that were brought to one partner U.S. hospital system. WPP's program model employs different modes of service delivery in response to the needs expressed by the partners in the countries where WPP works.
Lastly, another limitation in this study is that there are likely patients with this condition who may not have survived before referral to WPP. WPP's system of patient intake and placement should not be misunderstood as a complete system of surveillance for the incidence of EA with TEF in the countries where WPP works. To our knowledge, since the inception of this program, there was one patient who passed away from hypoxia before being referred to CHoR in March 2012. We did not include this infant in our study due to lack of clinical documentation regarding the circumstances of the mortality. We emphasize the importance of building capacity in these countries while providing much needed medical care of these neonates.

Conclusion
Through partnership with WPP and local physicians, we have been able to provide successful and responsible surgical care for patients with EA and TEF born in countries without appropriate personnel and resources. We recognize the importance of investing in building surgical capacity in these countries and we will continue to partner with local physicians in these LMICs to build a sustainable solution.