Total esophagogastric dissociation - Journal of Pediatric Surgery

Journal of Pediatric Surgery 52 (2017) 260–263

Contents lists available at ScienceDirect

Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg

Total esophagogastric dissociation: single center experience☆,☆☆ Sumita Chhabra a, Anca-Mihaela Nedea a, Lisa Kauffman b, Antonino Morabito a,⁎ a b

Department of Paediatric Surgery, Royal Manchester Children's Hospital, Oxford Road, Manchester M13 9WL, United Kingdom Department of Paediatrics, Royal Manchester Children's Hospital, Oxford Road, Manchester M13 9WL, United Kingdom

a r t i c l e

i n f o

Article history: Received 3 November 2016 Accepted 8 November 2016 Key words: Esophagogastric dissociation Gastroesophageal reflux Severe neurological impairment

a b s t r a c t Aim: Fundoplication fails in approximately 20% of children with severe neurodisability. We aimed to evaluate total esophagogastric dissociation (TOGD) as a primary procedure and as a ‘rescue’ procedure for severely neurologically impaired children with significant swallowing discoordination and gastroesophageal reflux disease. Methods: Casenotes of 40 children with severe neurodisability who underwent TOGD between 2005 and 2015 were retrospectively reviewed. Of these, 33 were primary procedures and 7 were ‘rescue’ procedures following failed fundoplication. Results: Median age at surgery was 3 years 7 months (range 1 month to 13 years 11 months). Preoperatively, all children had symptoms of regurgitation, retching or vomiting and 70% of children had an unsafe swallow. There were 5 early complications related to surgery in 4 children requiring surgical intervention. One child died following relaparotomy for esophagojejunal anastomotic breakdown because of multiorgan failure. Gastrostomy feeding was established by a median of 6 days (range 2 to 25 days) and median hospital stay was 10 days (range 4 to 280 days). There were 5 late complications. Median follow-up was 13 months (range 1 month to 8 years 4 months). All children have had resolution of gastroesophageal reflux. Thirteen percent of children experience bloating or pain on feeding and 26% of children experience retching unrelated to gastroesophageal reflux. There were 8 late deaths unrelated to surgery. Conclusion: TOGD should be considered as a primary and definitive procedure in selected severely neurodisabled children who are at higher risk of failure of fundoplication, recurrent aspiration and a reduced quality of life. Level of Evidence: IV. © 2017 Elsevier Inc. All rights reserved.

Children with severe neurological impairment (NI) have marked pharyngeal neuromuscular incoordination and gastroesophageal reflux disease (GORD) [1,2]. Although laparoscopic or open fundoplication with additional gastrostomy is effective in facilitating nutrition, reducing aspiration pneumonitis and improving quality of life for patients and care givers, there is a significant failure rate in severely neurologically impaired children compared to neurologically normal children [3,4]. The incidence of failure of fundoplication has been reported as

☆ Authors contributions: Miss Sumita Chhabra was the lead author in data collection, data interpretation and writing the manuscript. Dr Anca Nedea has contributed to data collection, data interpretation and writing the manuscript. Dr Lisa Kauffman has a senior role in selecting patients suitable for TOGD with Professor Morabito. She has provided information on all patients and has contributed to the manuscript. Professor Morabito is the senior author as all patients in the study were under his care. The outcomes in the manuscript directly reflect Professor Morabito's work. Professor Morabito has been supervisory to Miss Sumita Chhabra and Dr Anca Nedea in the writing of the manuscript. ☆☆ The authors declare no conflicts of interest. ⁎ Corresponding author at: Royal Manchester Children's Hospital, Oxford Road, Manchester M13 9WL, United Kingdom. Tel.: +44 1617012194; fax: +44 1617012767. E-mail address: [email protected] (A. Morabito). http://dx.doi.org/10.1016/j.jpedsurg.2016.11.018 0022-3468/© 2017 Elsevier Inc. All rights reserved.

18 to 43% [4–7] previously although more recent studies show a failure rate of fundoplication between 14 and 27% in neurologically impaired children, with a reoperation rate of 11–13% [3,8–10]. This remains high, in contrast to total esophagogastric dissociation (TOGD) which has a very low failure rate, with a majority of children having resolution of GOR following dissociation [10–15]. Following the description of TOGD by Bianchi in 1997 as a ‘rescue’ procedure [16], we have adopted a multidisciplinary approach to the management of children with severe NI associated with swallowing incoordination and severe GOR, with consideration of TOGD as a primary procedure in addition to a ‘rescue’ procedure. This allows for a more structured approach to the management of children who are at higher risk of failure of fundoplication, avoiding the complications of redo surgery, to achieve resolution of gastroesophageal reflux. Although there is no strong evidence, such as a randomized control trial, to suggest that TOGD is superior to fundoplication in the treatment of GOR in the severely neurologically impaired child, comparative studies have demonstrated a lower failure rate of TOGD over simple fundoplication and gastrostomy [10–13], in addition to improved nutritional rehabilitation, reduction in respiratory infections and improved quality of life for the child. These results, in addition to our own experience [14,17], have led us to continue to perform TOGD as a primary procedure in selected patients.

S. Chhabra et al. / Journal of Pediatric Surgery 52 (2017) 260–263

261

We have previously published our outcomes following TOGD between 1994 and 2004 [14]. The aims of this study were to outline our approach to the surgical management of the severely neurologically impaired child with reflux, identifying patients suitable for TOGD as a primary procedure, and to evaluate outcomes of both primary TOGD and ‘rescue’ TOGD over the last eleven years.

ensure free access to the stomach and rapid deflation or the gastrostomy is preserved if previously sited. A pyloroplasty is not routinely performed unless the child has previously undergone a fundoplication procedure and there is evidence of vagal nerve injury.

1. Methods Patients who underwent TOGD between 2005 and 2015 were identified from the hospital operative database ORMIS using code G24.5. Casenotes were retrospectively reviewed and data collected on severity of NI, swallowing and feeding status, pre- and postoperative reflux symptoms, preoperative antireflux medication, preoperative investigations, episodes of aspiration pneumonia, failure to thrive, immediate and long-term postoperative complications, duration of hospital stay, time to establishment of gastrostomy feeding and long-term morbidity and mortality.

All patients attended pediatric intensive care (PICU) postoperatively and were managed on the ward when self-ventilating with adequate pain control. The nasojejunal tube and gastrostomy were left on free drainage and gastrostomy feeds were started when aspirates had reduced. Feeds were gradually increased with dietician input in order to establish the child's feeding regime. Pain control was achieved with intravenous morphine and weaned as appropriate. Intravenous antibiotics were administered for 48 hours. Once full gastrostomy feeds were established and pain was adequately controlled with enteral analgesia, the patient was discharged if there were no further clinical concerns.

1.1. Preoperative work-up

2. Results

Patients were referred by the primary pediatric consultant involved in the care of the child, or by surgical colleagues who felt that TOGD would be a better primary option for the child or following failed fundoplication. All patients were reviewed in the joint feeding clinic under the lead surgeon, AM, with a specialist neurodisability pediatric consultant, LK, a special needs community nurse specialist and dietician. Indications for primary TOGD were: (i) severe NI, defined as a Gross Motor Function Classification System level of five [18]; (ii) swallowing difficulties or unsafe swallow (the child is unable to protect the airway and liquids and solids are aspirated into the trachea) (iii) severe gastroesophageal reflux refractory to medical treatment; (iv) recurrent respiratory tract infections or aspiration pneumonia; (v) failure to thrive; (vi) poor quality of life for the children and the carers. ‘Rescue’ TOGD was performed in children who fulfilled the above criteria in addition to failure of previous surgical intervention. At the initial consultation, several surgical options are discussed with parents/carers including the risks of each procedure: gastrostomy alone, feeding jejunostomy, primary fundoplication with gastrostomy and TOGD. During the consultation, it is stressed that the choice of procedure should carry the least risk to the condition of the child with the aim to eliminate the possibility of gastroesophageal reflux. Parents are then reviewed again in order to confirm which procedure they have selected.

A total of 40 children were identified including 27 females and 13 males. Of these, 33 (83%) underwent primary procedures and 7 (18%) underwent ‘rescue’ procedures following failed fundoplication. One child had previously undergone a redo fundoplication and Roux-en-Y jejunostomy. Median age at the time of procedure was 3 years 7 months (range 1 month to 13 years 11 months). A majority of children (63%) had cerebral palsy or global developmental delay as outlined in Table 1. Three children (8%) had specific syndromes and one child (3%) had a chromosomal abnormality.

1.3. Postoperative course

2.1. Preoperative symptoms and feeding Preoperatively, all children had symptoms of regurgitation, retching or vomiting. Of 15 children who underwent videofluoroscopy, 14 children demonstrated an unsafe swallow. A majority of children (95%) had swallowing difficulties and were on antireflux medications. Twenty-eight children (70%) were receiving two or more antireflux agents preoperatively. Thirty-seven children (93%) suffered from recurrent lower respiratory tract infections or had at least one episode of aspiration pneumonia prior to surgery. Thirty-five children (88%) had failure to thrive. Most children were fed via nasogastric or nasojejunal tube (58%) or gastrostomy (34%). One child was fed via gastrojejunostomy and one via jejunostomy. Two children were orally fed (one child failed a trial of nasogastric tube).

1.2. Operative procedure 2.2. Perioperative course The procedure was carried out by the lead surgeon, AM, in accordance with the original procedure described by Bianchi et al. [14,16,19]. Access is gained via an upper midline abdominal incision and the left lobe of the liver is retracted medially to gain access to the gastroeosophageal junction. For rescue TOGD, the wrap is undone. The esophagus is mobilized preserving the posterior and anterior vagus nerves. The esophagus is transected above the gastroesophageal junction and the gastric end is oversewn. The jejunum is then divided and a 20–30-cm isoperistaltic jejunal loop on a reliable mesenteric vascular pedicle is passed through to the distal esophagus in a Roux-en-Y fashion without tension. This forms the esophagojejunal anastomosis. Passage of the jejunal Roux loop through the transverse mesocolon and behind the stomach during TOGD reduces the risk of volvulus associated with fixed attachment of the small bowel to the abdominal wall. Bowel continuity is established with an end-to-side jejunojejunostomy at approximately 40 cm. The isoperistaltic free loop allows free passage of saliva and other oral intake, and acts as an antireflux mechanism. A gastrostomy tube is placed at the greater curvature of the stomach to

Gastrostomy feeding was established by a median of 6 (range 2 to 25) days. Median hospital stay was 10 (range 4 to 280) days.

Table 1 Underlying conditions in children undergoing TOGD. Underlying condition

Number of patients

Cerebral palsy Global developmental delay/undiagnosed syndrome Neuronal migration disorder Chromosomal abnormality Metabolic disorder Lisencephaly Microcephaly Syndromes Cornelia de Lange Aicardi-Goutières [21] CHARGE

25 6 2 1 1 1 1 1 1 1

262


2.3. Early complications There were 5 early complications (b30 days) related to surgery in 4 children requiring surgical intervention. These comprise a pyloroplasty leak (n = 1), esophagojejunal anastomotic leaks (n = 2) and wound dehiscences (n = 2). One child grew methicillin-resistant Staphylococcus aureus (MRSA) on blood cultures and underwent a laparotomy and revision of esophagojejunal anastomosis following an anastomotic leak at 4 days post surgery. She subsequently underwent a further laparotomy 4 days later because of leakage from the pyloroplasty which required revision. She initially recovered well following this however she developed a colonic perforation secondary to a colonic volvulus 8 years later for which she underwent partial colonic resection and adhesiolysis. One child, who had previously undergone a redo fundoplication and partial lobectomy for a lung abscess, developed a pleural effusion, subhepatic collection and wound dehiscence secondary to esophagojejunal anastomotic break down. This child underwent revision of the anastomosis and drainage of the subhepatic collection on day 9 postoperatively following failed ultrasound drainage. She required total parenteral nutrition and a prolonged course of antibiotics because of growth of Enterobacter from the drain cultures. She remained in PICU for 4 weeks however she has since remained complication free. Two children required formal wound closure following wound dehiscence at day 5 and day 13 postoperatively. One of these children also required repair of iatrogenic injury to the esophagojejunal anastomosis. This child subsequently developed a large ventral hernia and underwent a repair with Strattice™ Reconstructive Tissue Matrix, LifeCell™ six years later. One child died following relaparotomy for esophagojejunal anastomotic breakdown at 8 days post surgery because of septic shock and multiorgan failure. At laparotomy, the front wall of the anastomosis was found to have opened with infarction of the duodenum. Both the proximal and distal limbs of the anastomosis were brought out as defunctioning stomas and a laparostomy bag was placed over the abdomen. 2.4. Late complications There were 5 late complications (after 30 days). In addition to the aforementioned complications, one child developed an esophageal stricture 1 month post surgery requiring balloon dilatation and one child developed a gastric volvulus at 6 months post surgery requiring gastropexy. One child was found to have small bowel herniation in addition to a gastric perforation and abdominal wound dehiscence at 3 months postop. The gastrostomy was re-sited and wound closure was delayed. A chest drain was required for a secondary pleural effusion. The child also underwent a further laparotomy a year later because of stricturing of the jejunal Roux loop which required excision. She has since remained well. 2.5. Long-term outcomes Median follow-up was 13 (range 1 to 100) months. All children have had resolution of gastroesophageal reflux. Thirteen percent of children experience bloating or pain on feeding and 26% of children experience retching unrelated to gastroesophageal reflux. One child has selfinduced vomiting, one child has colonic dysmotility and one child has Barrett's esophagitis. There were 8 late deaths related to preexisting comorbidities. 3. Discussion The results of this study demonstrate that TOGD remains effective in minimizing reflux in severely neurodisabled children in keeping with our previous results [12]. In both studies, children remained reflux free following surgery with a significant decrease in retching. Our two

studies appear to be the largest series of TOGD to date. Molinaro et al. describe their experience of TOGD as a primary procedure for 14 severely neurologically impaired children between 2005 and 2011. They have reported 3 postoperative complications: bleeding from the anastomosis; anastomotic stricture; dumping syndrome. There were no deaths and all patients were GORD free following dissociation. The severity of the complications and how they were managed are not outlined clearly and it is thus difficult to draw any comparison in addition to the small study population [13]. Other studies date up to 2006 and are summarized in a systematic review by Peters et al. [15]. There do not appear to be any more recent studies from which we can compare our outcomes. The reoperation rate in our cohort of children was 15.7%. Reoperation rates in TOGD appear to be higher than that for fundoplication [3,8–10] however these do not relate to failure of the procedure and instead correlate with complications related to the procedure. There were 16% early complications and 13% late complications in our study. We acknowledge that the rate of esophagojejunal leak (7.5%) may contribute to TOGD not being widely accepted as a primary antireflux procedure. This is however comparable to rates of esophagojejunal anastamotic leakage in the adult population following gastrectomy for esophageal carcinoma with reported incidences between 2.9 and 9% [20]. Conversely, it is evident that TOGD has a much lower incidence of failure in comparison to fundoplication [10,12–14]. Although outcomes following primary open and laparoscopic fundoplication are equivocal [22], there are many studies which show a higher failure rate of fundoplication in children with NI [3–6,8,9]. While fundoplication with additional vagotomy and pyloroplasty (FVP) is not commonly performed, rates of failure in this patient group are significantly lower than fundoplication alone with 95% success rate reported by Horwood et al. [23]. Of 75 neurologically impaired children who underwent FVP, only 4% required further antireflux surgery in contrast to 18% (10 of 57 children) in the simple fundoplication group. This has been attributed to antroduodenal dysmotility related to impaired innervation of the foregut with improvement in gastric emptying following FVP. In this study, the severity of neurological dysfunction in the selected FVP group is not clear and this may contribute to the favorable outcomes. Failure of redo fundoplication is also significantly associated with NI in addition to open surgery at initial fundoplication [24]. Not only are the risks of failure reduced with primary TOGD in children with severe NI, the risk of vagal nerve injury and difficult dissection associated with rescue TOGD is also reduced [19]. Other problems with fundoplication in children with NI include retching, gas bloat, dysphagia and dumping syndrome [4,24]. In our study, only 13% of children had symptoms of pain or bloating on feeding and 26% of children had retching following surgery. Although TOGD is good at minimizing GOR, it does not eliminate the swallowing difficulty, aspiration of saliva and centrally mediated ‘retching’ that is often distressing to the family. In a recent study comparing laparoscopic fundoplication (LapFundo) with TOGD in severely neurodisabled children, TOGD was associated with a decreased use of postop antireflux medication and significantly lower rates of gas bloating with an improvement in the child's enjoyment of life. LapFundo was associated with a shorter duration of surgery, decreased postoperative intensive care requirement, decreased length of hospital stay and quicker time to establish feeds [10]. It is important to point out that children with severe neurological impairment do have significant comorbidities and thus have a reduced lifetime expectancy. Neurological status and presence of a gastrostomy have been shown to be significantly associated with decreased survival following fundoplication. Although mortality following TOGD was low in our series (2.6%), the child's comorbidities and life expectancy should be considered during the preoperative evaluation period [25]. In conclusion, the surgical management of children with severe NI and GOR should be tailored to the individual child in view of improving


the overall quality of life of the child and life expectancy, in addition to the treatment of symptoms. All options should be explored and discussed in order for the parents/carers to make an informed decision. TOGD is more effective in eliminating gastroesophageal reflux than fundoplication however potential complications should be balanced against the risks of failure of fundoplication. TOGD should be considered as a primary and definitive procedure in selected severely neurodisabled children who are at higher risk of failure of fundoplication, recurrent aspiration and a reduced quality of life.

References [1] Andrew MJ, Parr JR, Sullivan PB. Feeding difficulties in children with cerebral palsy. Arch Dis Child Educ Pract Ed 2012;97:222–9. [2] Sullivan PB, Lambert B, Rose M, et al. Prevalence and severity of feeding and nutritional problems in children with neurological impairment: Oxford feeding study. Dev Med Child Neurol 2000;41:674–80. [3] Knatten CK, Kvello M, Fyhn TJ, et al. Nissen fundoplication in children with and without neurological impairment: a prospective cohort study. J Pediatr Surg 2016;25: 212–8. [4] Pearl RH, Robie SH, Ein B, et al. Complications of gastroesophageal antireflux surgery in neurologically impaired versus neurologically normal children. J Pediatr Surg 1990;25:1169–73. [5] Rice H, Seashore JH, Touloukian RJ. Evaluation of nissen fundoplication in neurologically impaired children. J Pediatr Surg 1991;26:697–701. [6] Mauritz FA, van Herwaarden-Lindeboom MY, Stomp W, et al. The effects and efficacy of antireflux surgery in children with gastroesophageal reflux disease: a systematic review. J Gastrointest Surg 2011;15:1872–8. [7] Martinez DA, Ginn-Pease ME, Caniano DA. Sequelae of antireflux surgery in profoundly disabled children. J Pediatr Surg 1992;27:267–73. [8] Rossi V, Mazzola C, Leonelli L, et al. Long-term outcome and need of re-operation in gastro-esophageal reflux surgery in children. Pediatr Surg Int 2016;32:277–83. [9] Livingston MH, Shawyer AC, Rosenbaum PL, et al. Fundoplication and gastrostomy versus percutaneous gastrojejunostomy for gastroesophageal reflux in children with neurological impairment: a systematic review and meta-analysis. J Pediatr Surg 2015;50:707–14.

263

[10] Lansdale N, McNiff M, Morecroft J, et al. Long-term and ‘patient-reported’ outcomes of total esophagogastric dissociation versus laparoscopic fundoplication for gastroesophageal reflux disease in the severely neurodisabled child. J Pediatr Surg 2015;50:1828–32. [11] Goyal A, Khalil B, Choo K, et al. Esophagogastric dissociation in the neurologically impaired: an alternative to fundoplication? J Pediatr Surg 2005;40:915–9. [12] Gatti C, Federici di Abriola G, et al. Esophagogastric dissociation versus fundoplication: which is best for severely neurologically impaired children? J Pediatr Surg 2001;36:677–80. [13] Molinaro F, Bindi E, Cerchia E, et al. Esophagogastric dissociation reduces the re-operation rate for persistent gastroesophageal reflux in severely neurologically impaired children. Pediatr Surg Int 2014;30:997–1001. [14] Morabito A, Lall A, Piccolo RL, et al. Total esophagogastric dissociation: 10 years' review. J Pediatr Surg 2006;41:919–22. [15] Peters RT, Goh YL, Veitch JM, et al. Morbiditiy and mortality in total esophagogastric dissociation: a systematic review. J Pediatr Surg 2012;48:707–12. [16] Bianchi A. Total esophagogastric dissociation: an alternative approach. J Pediatr Surg 1997;32:1291–4. [17] Zaidi T, Sudall C, Kauffmann L, et al. Physical outcome and quality of life after total esophagogastric dissociation in children with severe neurodisability and gastroesophageal reflux, from the caregiver's perspective. J Pediatr Surg 2010;45:1772–6. [18] Palisano R, Rosenbaum P, Walter S, et al. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol 1997;39:214–23. [19] Lall A, Morabito A, Bianchi A. “Total gastric dissociation (TGD)” in difficult clinical situations. Eur J Pediatr Surg 2006;16:396–8. [20] Carboni F, Valle M, Federici O, et al. Esophagojejunal anastomosis leakage after total gastrectomy for esophagogastric junction adenocarcinoma: options of treatment. J Gastrointest Oncol 2016;7:515–22. [21] Stephenson JBP. Aicardi-Goutières syndrome (AGS). Eur J Paediatr Neurol 2008;12: 355–8. [22] McHoney M, Wade AM, Eaton S, et al. Clinical outcome of a randomized controlled blinded trial of open versus laparoscopic nissen fundoplication in infants and children. Ann Surg 2011;254:209–16. [23] Horwood JF, Calvert W, Mullassery, et al. Simple fundoplication versus additional vagotomy and pyloroplasty in neurologically impaired children – a single centre experience. J Pediatr Surg 2015;50:275–9. [24] Pacilli M, Eaton S, Maritsi D, et al. Factors predicting failure of redo nissen fundoplication in children. Pediatr Surg Int 2007;23:499–503. [25] Wockenforth R, Gillespie CS, Jaffray B. Survival of children following nissen fundoplication. Br J Surg 2011;98:680–5.