Surgical intervention in children with portal hypertension

Over a 5-year period 252 children with bleeding oesophageal varices secondary to portal hypertension were treated by endoscopic injection scletherapy. Surgical treatment was restricted to patients with bleeding complications following sclerotherapy and the occasional problem of symptomatic splenomegaly (2 children). Sclerotherapy failure was due to the development of varices not accessible to sclerotherapy (gastric fundus 7, duodenum 1, jejunum 1, unidentified site in small bowel 1) and to continued bleeding from oesophageal ulcers in 2 children. There were 5 children with Child's grade A cirrhosis. Fifteen children (4.4 % of the total; median age 8.7 years, range 1.5 – 18 years) underwent 16 operations during this period, which included 13 porto-systemic shunts (8 lieno-renal and 5 meso-caval) and 3 oesophageal transections/devascularisations. Sclerotherapy was not attempted in 1 of the 2 children treated for continued bleeding from oesophageal varices. Two lieno-renal shunts thrombosed and 1 of these patients underwent a meso-caval shunt. There were no deaths in this group of patients during a median follow-up of 40 months. One child with biliary atresia has developed a mild intermittent encephalopathy. The use of injection sclerotherapy for bleeding oesophageal varices with surgical intervention limited to patients failing treatment resulted in low morbidity and no mortality in a large group of children with portal hypertension. Correspondence to: N. D. Heaton     

A prospective study of endoscopic esophageal variceal ligation using a multiband ligator

BACKGROUND: Endoscopic variceal band ligation (EVL) is the preferred method of treating variceal hemorrhage in adults. The need to reinsert the endoscope after reloading for each varix ligation has been a drawback. The Saeed multiband ligator allows ligation of multiple varices during a single insertion. The multibander has not been used previously in children. METHODS: Twenty-eight consecutive children were referred to a pediatric liver unit because of esophageal variceal bleeding from 1998 to 2000. Endoscopic variceal band ligation was performed at initial endoscopy and repeated monthly until varices were obliterated or were too small to ligate. RESULTS: Results are expressed as median (range). Age at EVL was 11 years (3 months to 16 years) and weight 30 kg (5.4-63 kg). Portal hypertension was caused by cirrhosis in 15 children. Endoscopic variceal band ligation was performed on 66 occasions with 4 bands applied per session. Ten children had active bleeding at initial endoscopy and all responded to EVL. Interval bleeding developed in 2 children before variceal ablation. Varices were obliterated in 26 of 28 patients after 2 sessions. During the 21-month follow-up (2 months to 3 years), six children have undergone elective liver transplantation and three have had mesoportal bypass procedures. Rebleeding developed in 2 of 26; 1 from recurrent esophageal varices that responded to repeat EVL and 1 from gastric varices. Following variceal ablation, 2-year actuarial variceal recurrence risk was 40%. CONCLUSIONS: Endoscopic variceal ligation is highly effective in obliterating esophageal varices in children. The use of a multibander device for endoscopic variceal ligation is technically feasible and safe even in small children, and its use results in more rapid ablation of esophageal varices.     

Balloon dilatation of anastomotic strictures secondary to surgical repair of oesophageal atresia: a systematic review

Surgical repair of oesophageal atresia may result in anastomotic strictures. These strictures are often treated by balloon dilatation (BD) and currently balloon dilatation (fluoroscopic or endoscopic) is the preferred primary treatment method. Here we review the current evidence of the outcomes of balloon dilatation of anastomotic strictures secondary to surgical repair of oesophageal atresia. We searched the standard databases (January, 1960–May, 2012) to identify all studies that reported outcomes of balloon dilatation of anastomotic strictures secondary to surgical repair of oesophageal atresia in children. Data, reported as median (range), were analysed and compared. Outcomes were success of BD, number of BD sessions, number of oesophageal perforations, need for other surgical interventions and mortality. Five studies were found to be relevant (n?=?139; 81 [58%] male children). The total number of dilatation sessions was 401 (2.9 dilatations per child patient). General anaesthesia was used in two (40%) studies; sedation in a further two (40%) studies and one (20%) study used a combination of both. The size of balloon catheter ranged from 4 mm to 22 mm. Seven perforations were reported (1.8% per dilatation session), of which only one (14%) required surgery. No deaths were recorded. Balloon dilatation for anastomotic strictures post-EA repair is safe, and associated with a low perforation and mortality rates. Most perforations are amenable to conservative management.     

Endoscopic injection sclerotherapy with polidocanol for cardiac varices in children and adolescents

BackgroundLike esophageal varices, cardiac varices are often treated with endoscopic variceal ligation (EVL). However, we previously reported that EVL for cardiac varices may be associated with a high risk of rebleeding from the ulcer if the O-ring spontaneously drops off early. The efficacy and safety of para-variceal endoscopic injection sclerotherapy (EIS) with polidocanol for the treatment of cardiac varices in children and adolescents were evaluated.MethodsEleven patients under 18 years of age with portal hypertension who underwent para-variceal EIS with polidocanol for cardiac varices with red signs, which were considered to be at high risk of bleeding, were retrospectively reviewed.ResultsOne session of para-variceal polidocanol-EIS was performed for each of the 11 patients. One patient experienced temporary hypoxia due to aspiration of saliva when the tracheal intubation tube was removed after the procedure but recovered by endotracheal suctioning; there were no other adverse events. In six of the eight cases in which efficacy could be evaluated, eradication of cardiac varices was achieved.ConclusionPara-variceal polidocanol-EIS may be considered instead of EVL for small cardiac varices with red signs in pediatric patients with cardiac varices.     

Gastric transposition for oesophageal replacement

Eighty-three gastric transpositions for oesophageal replacement were carried out over the 15-year period 1981–1995. The vast majority of patients underwent the replacement procedure for oesophageal atresia (OA) (56), 35 following failed or abandoned attempts at primary anastomosis, and 21 for isolated atresia. Caustic oesophageal strictures constituted the next largest group (14). The favoured method of transposition was via the posterior mediastinal route without thoracotomy (n = 50). There were 6 deaths in this series (7.2%); 12% of patients developed anastomotic leaks, all of which healed spontaneously, and 12% developed strictures that responded to dilatation/s alone. Establishing oral feeding was the most common problem, especially in infants with OA who had not been sham-fed. The majority of patients have achieved satisfactory growth at medium-term follow-up and 64% have an excellent and 24% a satisfactory outcome at follow-up of up to 14 years.     

Upper gastro-intestinal tract bleeding in cirrhotic children candidates for liver transplantation

Signs of portal hypertension, history of upper gastro-intestinal tract bleeding episodes and outcome of the latter were recorded in 76 cirrhotic children evaluated for liver transplantation. Fifty-three (70%) had varices and 22 (29%) had experienced upper gastro-intestinal tract bleeding. Of these 22, 19 bled from varices and 3 from ulcers. Non bleeding ulcers were also found in five patients bleeding from varices. Iterative sclerotherapy controlled acute variceal bleeding in all but one patient in whom emergency transplantation was performed. Six of the eight patients with ulcers were successfully treated by the H2 histamine receptor antagonist ranitidine. We conclude that iterative sclerotherapy is efficient to control acute variceal bleeding and prevents recurrent bleeding in children with end-stage liver diseases awaiting liver replacement. Bleeding asymptomatic ulcers are frequent and respond to H2 histamine receptor antagonists.     

Intestinal bypass of the oesophagus: 117 patients in 28 years

This study includes 117 patients operated upon in the period from 1970 to 1999. Indications, surgical techniques, complications, and results are reviewed. Indications included: long-gap oesophageal atresia with or without fistula in 81 patients; peptic stenosis in 19; caustic stenosis in 12; oesophageal varices in 2; and 1 case each of oesophageal epidermolysis bullosa, total oesophageal leiomyomatosis, and a non-functioning antiperistaltic retrosternal colic graft operated upon in another hospital. A retrosternal bypass was performed 106 times: 98 first operations and 8 redos; the intrathoracic technique was used 19 times. The left transverse colon was used in 107 cases (85.6%), the right transverse colon in 8 (6.4%), and the ileocecum in 10 (8%). All the intestinal bypasses were placed in the isoperistaltic direction. There were 5 deaths in the first 11 years of our experience; no patient died from 1982 on. Ten complications were treated conservatively (8%): 2 wound infections healed with medical treatment, and 8 leaks of the cervical anastomosis closed spontaneously. The major surgical complications were 8 gangrenous bypassess (6.4%), removed and reopeated about 1 year later utilizing an ileocolic retrosternal graft. Three cases of peptic disease of the colic bypass (2.4%) were successfully treated with the author's technique. Nine patients had minor surgical complications (7.2%): 3 strictures of the oesophagocolic anastomosis in a retrosternal bypass (resected and reoperated) and 6 cases of adhesive occlusion. In our opinion, the best substitute of the oesophagus is the colon, particularly the left transverse segment, which may be placed behind the sternum or in the oesophageal bed, always in the isoperistaltic direction. The low mortality (4%), restricted to the early period of our experience, and few major surgical complications (6.4%) are acceptable considering the importance of the operation, and the long-term results may be considered very satisfactory. Accepted: 8 November 1999     

Oesophageal substitution with free and pedicled jejunum: short- and long-term outcomes

In children, the indications for oesophageal substitution are principally, long gap oesophageal atresia (OA), severe anastomotic disruption following primary repair of OA and severe caustic or peptic strictures. We present an outcome review of eight cases who underwent oesophageal substitution with jejunum at our institution between 1986 and 2001. The purpose of this study was to evaluate our experience with free/pedicled jejunal grafts and its long-term outcome as an oesophageal substitute. Operative and postoperative outcome with free and pedicled jejunal grafts in four cases of pure OA, two cases of OA and distal tracheo-oesophageal fistula (TOF), one patient with a high retrolaryngeal oesophageal web and one case of severe caustic oesophageal stricture. Six patients had an oesophagostomy and a gastrostomy fashioned previously. Eleven free jejunal grafts were performed in six patients (three intraoperative redo interpositions for immediate graft loss, three separate grafts in one patient and two free grafts in two patients). One patient’s pedicled jejunal graft proximally required microvascular anastomosis while the other had a pedicled graft without microvascular anastomosis. Early postoperative complications included four upper anastomotic leaks (three free grafts, one pedicled with microvascular support), pneumothorax requiring prolonged ventilation and Horner’s syndrome. Recurrent laryngeal nerve injury occurred in the patient who had a high retrolaryngeal oesophageal web. During follow up (5–18 years) late complications of upper anastomotic stricture in four patients and graft redundancy with subsequent kinking of the lower anastomosis were observed in one patient. Three patients established a complete oral diet; a further three patients relied on supplemental gastrostomy feeds and one patient is entirely gastrostomy fed. There were two late deaths, one from aspiration and the other from a severe asthmatic attack (5 and 7 months postoperatively, respectively). Our results indicate that there are significant complications related to the use of free jejunal grafts. Early recognition and treatment are of paramount importance in the ultimate achievement of a successful technical outcome.     

Juvenile polyposis in a tropical country

The clinical profile, malignant potential, and management of 17 children with juvenile polyposis (more than five juvenile polyps) were evaluated clinically and endoscopically. Colonoscopy and polypectomy were done three weekly until colonic clearance was achieved, and thereafter two yearly. All polyps were subjected to histological examination. Mean age was 7.7 years, with a male preponderance (3:1). Presentation was with rectal bleeding (94%), pallor (65%), stunted growth (53%), and oedema (47%), and the mean (SD) duration of symptoms was 33 (27) months. None had a positive family history or any congenital anomaly. Two children had six polyps up to the transverse colon; the rest had numerous polyps all over the colon. All children had juvenile polyps on histology and 10 (59%) had adenomatous changes (dysplasia). Total colectomy was done in six for intractable symptoms. Colon clearance was achieved in eight after an average 3.4 polypectomy sessions, and three were still on the polypectomy programme. In conclusion, juvenile polyposis is commonly associated with low grade dysplasia. Serial colonoscopic polypectomy is effective but colectomy is required for intractable symptoms and when clearance of the colon is not possible.     

Caustic ingestion in children—A review

Various domestic or industrial chemicals may cause significant upper aerodigestive tract burns. Preventive measures should be up-scaled, especially in the developing world, to reduce the epidemic of accidental victims, largely unsupervised preschool children. External signs do not predict degree of injury. Non-invasive diagnostic screening includes radio-nuclear imaging, but early oesophago-gastroduodenoscopy remains the standard to predict stricture formation from circumferential submucosal scarring. Serial dilation is the mainstay of oesophageal stricture therapy, with oesophageal replacement reserved for severe refractory strictures. Intra-lesional steroid or mitomycin C may decrease the dilatations required for severe strictures, although long-term effects are unknown. Risk of secondary oesophageal carcinoma mandates long-term surveillance.     

内镜下组织粘合剂注射和套扎术治疗儿童食管胃静脉曲张出血24例临床观察

目的评价内镜下组织粘合剂注射和套扎术（EVL）治疗儿童食管胃静脉曲张（GOV）的临床疗效。 方法分析2009年10月至2013年11月在复旦大学附属儿科医院因GOV出血接受内镜下治疗的24例患儿的临床资料,总结治疗后即时止血率、再出血率及并发症情况。 结果24例患儿平均年龄（8.0±3.2）岁,男15例,女9例,其中1型GOV（GOV1）6例、2型GOV（GOV2）18例,平均随访10.5（1~32）个月。共行内镜下EVL治疗24例,平均每例套扎4（2~7）环;组织粘合剂注射治疗19例,平均每例组织粘合剂注射量为1（0.5~1.5） mL。即时止血率为100%,3个月内再出血率16.7%（4/24）,3个月以上再出血率16.7%（4/24）,其中2例于治疗后3个月内死亡,6例（25%）于随访期间因出血复发而再次行内镜下治疗,平均套扎2.8（2~5）次。术后均未见明显并发症。 结论内镜下组织粘合剂注射和EVL治疗GOV疗效确切,并发症少,是治疗儿童GOV的有效措施。     

Colorectal polyps: an important cause of rectal bleeding.

Seventy one children with rectal bleeding were examined by total fibreoptic colonoscopy. Large bowel polyps were found in 45; 27 (60%) had solitary rectal polyps. Altogether, 83% of resected polyps were juvenile. No complication of colonoscopic polypectomy occurred. New polyps reoccurred in four (9%) treated children.     

Reflux strictures of the oesophagus in children: personal experience with preoperative dilatation followed by anterior funduplication

Introduction Oesophageal surgery for reflux stricture is as challenging in adults as in the paediatric age group. Several management protocols, both medical and surgical, are currently proposed, such as bougienage, funduplication without dilatation, funduplication with pre– and postoperative dilatation, resection and interposition, and pharmacological therapy. However, reported results are not univocal. The aim of this work is to demonstrate that preoperative treatment with H2-antagonist combined with oesophageal dilatation and followed by anterior funduplication (Boix-Ochoa procedure with elongation of intraabdominal segment of the oesophagus) is a long-term, effective treatment for reflux stricture in children. It provides a tension free repair and an adequate protection to reflux, thus, preventing recurrences.Materials and methods In the last five years we observed oesophageal stenosis in 10 out of 49 children, operated for gastroesophageal reflux (mean age 62.9 months, range 12–156 months). All children underwent treatment with H2-antagonist (Ranitidine) and prokinetic agent (Cisapride), followed by oesophageal dilatations (mean 2.8, range 2–4 cycles) with Savary-Gillard dilators. An open anti–reflux procedure was performed (9 Boix-Ochoa and 1 Nissen) on children where a 9 mm endoscope passed easily through the oesophageal lumen. The pre and postoperative evaluation of all patients included symptoms assessment, esophagogram and endoscopy.Results Results were satisfactory in 9 patients. Only one patient where a Nissen wrap was performed, incomplete relaxation was documented radiologically. The patient required several dilatations for residual dysphagia before reaching a symptom free status. All other patients had an average follow-up of 38 months (range, 5 months to 5 years) with relief from dysphagia and no recurrence of stricture. Radiological controls showed good oesophageal lumens, with normally positioned neocardias, opening regularly during barium passage with no sign of reflux. Multiple biopsies from endoscopic controls confirmed complete relief from oesophageal stricture but persistence of Barrett's mucosa.Conclusions Our treatment of choice for reflux stricture is preoperative pharmacological therapy followed by series of dilatation with Savary-Gillard dilators till oesophagus is adequately dilated. Antireflux surgery is mandatory when a stricture is observed. We prefer a Boix-Ochoa funduplication with extensive transhiatal mobilization of thoracic oesophagus. This results in a "tension free" fundoplication even when brachioesophagus is present. The procedure appears to be physiological for pediatric patients and in our hands was free from recurrences.     

Endoscopic sclerotherapy of esophageal varices in infants and children

Thirty children, aged 7 months to 13 years, with bleeding esophageal varices were managed by endoscopic sclerotherapy (EST). Of these children, 73.3% had extrahepatic portal vein obstruction (EHPVO), 16.6% had cirrhosis of the liver, and 10% had noncirrhotic portal fibrosis. EST was done with fiberoptic pediatric upper gastrointestinal endoscopes and a flexible sclerotherapy needle under sedation with intravenous diazepam and pentazocine. The sclerosants used were ethoxysclerol 1% and absolute alcohol. Ten injections were given to control active variceal bleeding, and 145 injections were given on planned basis at 2-3-week interval. An average of five injections was required to obliterate the esophageal varices. In 90% of cases, an avariceal state was achieved; 10% had decreased size and number of varices. Emergency EST was effective to control bleeding in all sessions. Complications, including retrosternal discomfort, esophageal ulceration, stricture formation, and aspiration pneumonia, occurred in 60, 20, 20, and 6.6% of cases, respectively; complications were managed conservatively. Strictures were dilated with Eder-Puestow's olive dilators. Recurrence of esophageal varices, gastric varices, and rebleeding was seen in 13.3, 13.3, and 10% of cases, respectively. Shunt surgery was performed in 13.3% cases, who had developed gastric varices and were having EHPVO.     

Corrosive strictures of the oesophagus in children

In a series of 142 children treated for corrosive burns at the University of Cape Town teaching hospitals between 1957 and 1990, 55 developed strictures of the oesophagus. Prograde dilatations with or without a guiding trans-stricture string proved successful in 22 children while 33 had an oesophageal bypass procedure. The complications following dilatations, particularly the 10 perforations of the oesophagus, are reviewed. Three of these later responded successfully to further dilatations. Since 1969, a single-staged left colon interposition was employed as the procedure of choice. The operative technique of the left colon interposition is described and the early and late complications following the bypass procedures are detailed and discussed, stenosis and strictures at the upper anastomosis being the most significant.     

Bleeding oesophageal varices with long term follow up.

Complete long term follow up was obtained in 27 children who had bled from oesophageal varices. Most presented with haematemesis or melaena at an average age of 5.2 years in the portal vein thrombosis group (20 children) and 9.5 years in the intrahepatic group (7 children). All had splenomegaly. Only 6 of 20 children with portal vein thrombosis had a possible precipitating factor. A total of 182 admissions for bleeding are reported, in 68 of which injection sclerotherapy was used to control bleeding. Control rate with injection sclerotherapy was 97%. Shunts performed below age 10 years were associated with a high thrombosis rate. A conservative approach to bleeding varices in children is recommended with transfusion, pitressin, and injection sclerotherapy. Oesophageal transection may have a role in the emergency management of the few children in whom bleeding is not controlled by injection sclerotherapy.     

Bleeding oesophageal varices with long term follow up

Complete long term follow up was obtained in 27 children who had bled from oesophageal varices. Most presented with haematemesis or melaena at an average age of 5.2 years in the portal vein thrombosis group (20 children) and 9.5 years in the intrahepatic group (7 children). All had splenomegaly. Only 6 of 20 children with portal vein thrombosis had a possible precipitating factor. A total of 182 admissions for bleeding are reported, in 68 of which injection sclerotherapy was used to control bleeding. Control rate with injection sclerotherapy was 97%. Shunts performed below age 10 years were associated with a high thrombosis rate. A conservative approach to bleeding varices in children is recommended with transfusion, pitressin, and injection sclerotherapy. Oesophageal transection may have a role in the emergency management of the few children in whom bleeding is not controlled by injection sclerotherapy.     

What is the infection risk of oesophageal dilatations?

Oesophageal dilatation is the most widely used treatment option for the management of oesophageal strictures. Complications include bleeding, a slight increase in body temperature, thoracic or abdominal pain, oesophageal perforation, brain abscess and bacteraemia. We performed a prospective study to evaluate the frequency of postdilatation bacteraemia in nine patients subjected to a total of 50 dilatations. Bacteraemia was detected in 36 cases (72%), In all but three cases, however, it was transient and not associated with fever or other clinical complications. The organisms most commonly responsible (64%) were alpha-haemolytic streptococci (Streptococcus viridans), probably originating as contaminants from the oropharynx and oesophagus and introduced into the bloodstream during dilatation. Despite the relatively low incidence of bacteraemia-related postdilatation complications, the potential severity of such complications argues for the use of antibiotic prophylaxis as a routine measure prior to oesophageal dilatation. Conclusion Oesophageal dilatation is associated with a high incidence of bacteraemia. The organisms most commonly responsible were alpha-haemolytic streptococci. We recomend the use of antibiotic prophylaxis as a routine measure prior to oesophageal dilatation. Received: 23 September 1997 / Accepted: 3 March 1998 and in revised form: 24 February 1998     

Treatment of oesophageal strictures in children: a comparison of fluoroscopically guided balloon dilatation with surgical bouginage

Purpose. To compare the technical feasibility and procedural complications of fluoroscopically guided balloon dilatation with conventional surgical bouginage for the treatment of oesophageal strictures in children.¶Materials and methods. A retrospective analysis of 125 balloon dilatations in 37 children with oesophageal strictures of varying aetiology. Twenty-four of the 37 children also underwent 88 procedures of surgical bouginage and comparison was made between the methods.¶Results. Fluoroscopic balloon dilatation had fewer technical failures (0/125 vs 4/88, P < 0.02) and fewer iatrogenic perforations (2/125 vs ¶5/88 P = 0.1) than surgical bouginage.¶Conclusion. Fluoroscopically guided balloon dilatation is safer and has fewer technical failures than surgical bouginage and should be considered the first line of treatment for oesophageal strictures in children.     

Upper gastrointestinal bleeding: Etiology and management

Upper gastrointestinal bleeding is a potentially fatal condition at times due to loss of large volumes of blood. Common sources of upper gastrointestinal bleeding in children include mucosal lesions and variceal hemorrhage (most commonly extra hepatic portal venous obstruction) and, in intensive care settings infections and drugs are other etiological factors associated with bleeding. Massive upper Gl bleeding is life threatening and requires immediate resuscitation measures in the form of protection of the airways, oxygen administration, immediate volume replacement with ringer lactate or normal saline, transfusion of whole blood or packed cells and also monitoring the adequacy of volume replacement by central venous lines and urine output. Upper Gl endoscopy is an effective initial diagnostic modality to localize the site and cause of bleeding in almost 85–90% of patients. Antacids supplemented by H2-receptor antagonists, proton pump inhibitors and sucralfate are the mainstay in the treatment of bleeding from mucosal lesion. For variceal bleeds, emergency endoscopy is the treatment of choice after initial haemodynamic stabilization of patient. If facilities for endoscopic sclerotherapy (EST) are not available, pharmacotherapy which decreases the portal pressure is almost equally effective and should be resorted to. Shunt surgery is reserved for patients who do not respond to the above therapy. β blockers combined with sclerotherapy have been shown to be the most effective therapy in significantly reducing the risk of recurrent rebleeding from varices as well as the death rates, as compared to any other modality of treatment. Based on studies among adult patients, presence of shock, co-morbidities, underlying diagnosis, presence of stigmata of recent hemorrhage on endoscopy and rebleeding are independent risk factors for mortality due to upper Gl bleeding. Rebleeding is more likely to occur if the patient has hematemesis, liver disease, coagulopathy, hypotension and or anemia. There is a great need for conducting therapeutic trials as well as identifying predictors of outcome of upper Gl bleeding in children to develop evidence based management protocols.