H-type tracheo-oesophageal fistula with congenital oesophageal stenosis

Pediatr Surg Int (2005) 21: 386–387 DOI 10.1007/s00383-004-1343-z

CASE REPORT

R Babu Æ KAR Hutton Æ L Spitz

H-type tracheo-oesophageal fistula with congenital oesophageal stenosis

Received: 18 May 2003 / Accepted: 15 October 2003 / Published online: 18 December 2004 Ó Springer-Verlag 2004

Abstract H-type tracheo-oesophageal fistula (H-TOF) and congenital oesophageal stenosis (COS) are both rare anomalies. Only six cases of H-TOF in association with COS have been described in the literature. Characteristically, diagnosis is delayed, often well beyond the neonatal period. We report a patient who was accurately diagnosed on day 12 of life, and outline several of the key clinical dilemmas encountered in managing a neonate with this combination of rare congenital anomalies. Keywords H-type tracheo-oesophageal fistula Æ Congenital oesophageal stenosis Æ Oesophageal atresia Æ Foregut malformation

Introduction H-type tracheo-oesophageal fistula (H-TOF) is rare [1], comprising only 4% of the oesophageal atresia (OA)/ tracheo-oesophageal fistula (TOF) spectrum [2]. Congenital oesophageal stenosis (COS) caused by heterotopic cartilage and other tracheobronchial remnants in the oesophageal wall is also uncommon [3]. The combined association of H-TOF and COS has been reported in only six previous cases [1, 2, 4–6]. We describe a neonate with H-TOF and a long stenotic distal oesophageal segment that required resection and planned oesophageal replacement. This case report high-

R. Babu Wessex Regional Centre for Paediatric Surgery, Southampton General Hospital, Southampton, SO16 6YD, UK K. Hutton (&) Department of Paediatric Surgery, University Hospital of Wales, Heath Park, Cardiff, CF14 4XW Wales, UK E-mail: Kim.Hutton@cardiffandvale.wales.nhs.uk Fax: +44-2920-746322 L. Spitz Department of Paediatric Surgery, Institute of Child Health and Great Ormond Street Hospital, London, WC1N 1EH, UK

lights the management dilemmas for a neonate affected by this combination of anomalies.

Case report A male infant was born at 38 weeks’ gestation by normal vaginal delivery, with a birth weight of 2,300 g. At 12 h of age he experienced problems with feeding and developed increasing respiratory distress with right upper lobe collapse. Serial chest x-rays over the first few days showed collapse/consolidation changes involving different areas of both lungs. On day 12 of life he had a major desaturation and required mechanical ventilation. A contrast study suggested an H-TOF combined with a 2–3-cm distal oesophageal stricture. A prone tube oesophagogram confirmed the H-TOF at the level of the T2/3 vertebrae and a tight COS distal to the fistula 2.5 vertebral bodies in length (Fig.1). Bronchoscopy showed the presence of an obvious HTOF on the posterior tracheal wall midway between the vocal cords and carina. Oesophagoscopy revealed a normal proximal oesophagus and a tight midoesophageal stenosis. The H-TOF was divided via a cervical incision and a feeding gastrostomy performed. Postoperatively the infant had significant problems clearing oropharyngeal secretions, with continued pooling in the oesophagus above the tight stenosis. A cervical oesophagostomy was performed to divert the secretions and reduce the risk of recurrent aspiration; the distal end of the oesophagus was closed, with a plan for later oesophageal replacement. Full gastrostomy feeds were established successfully, but the patient continued to require respiratory support and developed significant stridor with intercostal recession. Repeat bronchoscopy showed posterior compression of the trachea, and computerised tomography revealed distorted tracheal anatomy related to fluid/ mucous distension of the retained oesophageal segment above the congenital stenotic area. The thoracic oesophageal remnant was resected via a right

387

Fig. 1 Oesophagogram showing H-TOF and COS. The black arrowhead points towards the fistula, and the black arrow shows the stenosis in the lower oesophagus

thoracotomy, and a tracheostomy was performed for continuing airway problems. Subsequent progress was uneventful, and gastric transposition for oesophageal replacement was successfully carried out at 11 months of age. The child is now 2.5 years old and tolerates a normal diet with no respiratory or swallowing difficulties.

Discussion Congenital oesophageal stenosis can be an isolated anomaly related to tracheobronchial remnants, fibromuscular stenosis, or a membranous diaphragm, or it can occur in association with oesophageal atresia or tracheo-oesophageal fistula. In patients with OA and distal TOF, the presence of an associated distal COS is unlikely to be recognised at the time of initial neonatal surgery, and the diagnosis is more usually made on later contrast studies done as a routine or for symptoms of dysphagia, vomiting, failure to thrive, or recurrent respiratory difficulties [4]. The stenosis is often incorrectly labelled a peptic stricture secondary to gastro-oesophageal reflux [6]. Successful management

involves resecting the involved oesophageal segment with end-to-end anastomosis, as dilatation alone is rarely successful [6]. In patients with an H-TOF combined with COS, the diagnosis is also frequently delayed, and the H-TOF may go undetected for many years despite recurrent symptoms. In four previously documented cases the diagnosis was finally established at 18 months [2], 3 years [1], 7 years [6], and 15 [5] years of age. The accurate diagnosis of H-TOF and associated COS in a neonate is unique and presents a number of management dilemmas. Should the lesions be treated concomitantly or as separate surgical procedures? Obviously, the H-TOF requires surgical division, but the stenosis could be managed by dilatation, resection, and end-to-end anastomosis; intestinal patch oesophagoplasty; or oesophageal replacement. Because dilatation of COS has a reported perforation rate of up to 30% [4], we would recommend definitive surgical intervention as the treatment of choice. In our patient we managed the H-TOF as a priority, to avoid repeated aspiration, and performed a feeding gastrostomy. We did not anticipate the problems of retained secretions in the blind oesophageal remnant proximal to the stenosis or the subsequent effect on the airway. We would advocate the following plan of management: If the stenosis is incomplete, the H-TOF should be divided immediately and oesophageal surgery delayed. If the stenosis is short but significant, we would recommend resection and end-to-end anastomosis after dividing the H-TOF. In cases with a long stenosis, we would advise division of the H-TOF, creation of a cervical oesophagostomy, and resection of the distal oesophagus as an initial procedure, with a plan for later oesophageal replacement.

References 1. Dunbar JS (1958) Congenital esophageal stenosis. Pediatr Clin North Am 5:443 2. Homnick DN (1993) H-type tracheoesophageal fistula and congenital esophageal stenosis. Chest 103:308–309 3. Ibrahim NBN, Sandry RJ (1981) Congenital oesophageal stenosis caused by tracheobronchial structures in the oesophageal wall. Thorax 36:465–468 4. Newman B, Bender TM (1997) Esophageal atresia/tracheoesophageal fistula and associated congenital esophageal stenosis. Pediatr Radiol 27:530–534 5. Stephens HB (1970) H-type tracheoesophageal fistula complicated by esophageal stenosis. J Thorac Cardiovasc Surg 59:325– 329 6. Yeung CK, Spitz L, Brereton RJ, Kiely EM, Leake J (1992) Congenital esophageal stenosis due to tracheobronchial remnants: a rare but important association with esophageal atresia. J Pediatr Surg 27:852–855