Findings of Paediatric Upper Gastrointestinal Endoscopy at a Tertiary Care Centre

Introduction: Endoscopy is an important diagnostic tool for gastrointestinal disorders in children. Practitioners still debate the need and utility of endoscopy in children and this important evaluation is potentially underutilized. There is a paucity of paediatric endoscopy data in Nepal. This report describes an initial three-year experience on profile of upper gastrointestinal endoscopy at Tribhuvan University Teaching Hospital (TUTH). Material and Methods: This retrospective study was conducted by reviewing hospital records for children undergoing endoscopy for various indications at TUTH during November 2011 to October 2014 (3 years). Results: We performed 362 endoscopies during this period. Mean age of patients was 9.49 years (range 2 months to 18 years) and 199 (54.9%) patients were under 10 years of age. Common indications were upper gastrointestinal bleeding (27.6%), recurrent abdominal pain (26.2%), acute abdominal pain (24.6%), recurrent vomiting (5.2%), and caustic ingestion (3.6%). An abnormality was detected in 159 (42.2%) patients. Varices (16.6%), gastritis/duodenitis (14.9%), ulcer (3.3%), oesophagitis (1.9%) were common diagnoses. Etiology of upper gastrointestinal bleed was identified in 65% of patients. Majority of patients with recurrent abdominal pain had negative results (80%). Failure to thrive constituted a very small proportion of children (0.2%) evaluated. All children received sedation/ analgesia and tolerated the procedure well. Conclusion: Paediatric upper gastrointestinal endoscopy can be safely performed across wide age range with proper sedation/ analgesia. Endoscopic diagnosis may improve management and children requiring such evaluation should be timely referred. Paediatricians should be encouraged to refer patients with growth failure early to ensure proper management.


Introduction
E ver since Hirschowitz introduced fi berop c endoscopes in the 1970s, understanding of the paediatric gastrointes nal disorders has become more and more broad 1,2 .Upper gastrointes nal endoscopy (UGIE) has served as a basic tool for diagnosis and therapy in paediatric gastrointes nal disorders in much of the developed world right a er this discovery.It has also fostered development of paediatric gastroenterology as an important discipline in the medical diversity 1,2,3,4,5 .Endoscopic techniques have become the most cost-eff ec ve and effi cient methods in many common paediatric gastrointes nal disorders with technological advance in this fi eld.
In the developing world however, use of gastrointes nal endoscope in paediatrics is s ll in infancy.Lack of awareness of its applica on, limited paediatric friendly endoscopy suites/endoscopes and lack of adequately trained physicians who can safely perform the procedure are the greatest hurdles 6,7 .Paediatric gastroenterology services have just started to be recognized as important subspecialty in Nepal and we are s ll far from u liza on of technical advance with this procedure.Tribhuvan University Teaching Hospital (TUTH) serves as a referral hospital and has recently established a separate paediatric endoscopy suite with child friendly scope.It is one of very limited paediatric service available in the country.Majority of endoscopies performed in this centre are actually referrals from diff erent medical centres across the country.Referrals to this service however are limited and the service is s ll underu lized.
We conducted this study with the objec ve of dissemina ng the informa on on common indica ons and fi ndings of UGIE among all paediatricians of the country.

Material and Methods
This is a retrospec ve study conducted in Department of Paediatrics at TUTH.TUTH is a referral hospital and pa ents arrive to this hospital from all over the country.It has a separate paediatric endoscopic suite with child friendly paediatric video gastro-duodenoscope.We performed upper gastrointes nal (UGI) endoscopy for all pa ents referred by paediatricians.Informed consent was obtained from parents/pa ents for the procedure a er careful explana on of procedure details and poten al complica ons.All children underwent the procedure under ketamine and midazolam seda on/anaesthesia in accordance to their comfort level and coopera on for the procedure.Lignocaine gargle was applied for some adolescents to reduce the dose of intravenous anaesthe cs.A paediatrician monitored the child during the procedure without any other responsibili es while the endoscopist performed the procedure.Trained paediatricians performed all endoscopies.Endoscopy fi ndings were recorded, biopsies taken for histopathological examina on and serological tests were sent wherever necessary.A er the procedure, pa ents stayed in observa on un l the reversal of eff ect of anaesthesia and the child accepted oral fl uids without problems.We recorded all informa on collected electronically and in the pa ent fi le.
We conducted this retrospec ve study by reviewing the hospital and endoscopy room records and extrac ng the data.We reviewed all the pa ents, who underwent upper gastrointes nal (UGI) endoscopy from October 2011 to October 2014 over a period of three years.The data was analyzed using SPSS (Version 20) sta s cal program.We calculated the frequencies of various indica ons, endoscopic fi ndings, and histopathological fi ndings for UGI endoscopy.

Results
During the study period, 362 pa ents underwent UGI endoscopy.Mean age of pa ents undergoing the procedure was 9.49 years (Range: 2 months to 18 years) with median of 10 years (standard devia on σ=3.68).Majority of them were male (Fig 1 ) and children under the age of 10 years cons tuted 55% of pa ents undergoing UGI endoscopy.(Fig 2) Table 1 summarizes the various indica ons for UGI endoscopy.Major indica ons in our service were UGI bleed, recurrent abdominal pain, acute abdominal pain and recurrent vomi ng.
Table 2 describes the endoscopic fi ndings in all pa ents.Normal fi ndings were observed in majority of pa ents (57.7%).Varices (16.6%),Gastri s/Duodeni s (14.9%),Ulcer disease (3.3%) and oesophagi s (1.9%) were among the most commonly iden fi ed abnormali es.Upper gastrointes nal obstruc ons in the form of strictures were also iden fi ed during UGI endoscopy.Congenital obstruc ons were all gastric outlet obstruc ons without any mucosal endoscopic fi ndings.Acquired strictures were all post caus c inges on strictures and ulcera ve/infl ammatory narrowing of gastric outlet.We also analyzed the endoscopic fi ndings of pa ents undergoing the procedure for specifi c indica ons.UGI bleeding was the most common indica on in our service.Varying severity of UGI bleeding was the primary indica on for endoscopy in 100 pa ents but 35% of them had a nega ve endoscopy.Varices (50%) and gastri s/ duodeni s (15%) in various forms were the most common endoscopically recognized abnormali es for bleeding.Colonoscopy was not performed in any pa ent.All pa ents with obscure UGI bleed were inac ve (i.e.no evidence of ongoing bleed).
Abdominal pain either acute or recurrent, was other most common reasons for endoscopy referrals.Majority of pa ents in these categories had normal fi ndings.Among 95 children with recurrent abdominal pain, 76 (80%) had normal endoscopic fi ndings.Gastri s/duodeni s 10 (10.5%), duodenal and antral ulcers 7 (7.4%) were some of the lesions iden fi ed in children with recurrent abdominal pain.Similarly, among 89 children with acute abdominal pain, normal endoscopic result was obtained in 62 (69.7%) pa ents.Gastri s /duodeni s 17 (19.1%),ulcers 4 (4.5%),oesophagi s 2 (2.2%) were the most common endoscopically determined ae ologies for acute abdominal pain.
Endoscopy was nega ve for all fi ve pa ents who underwent UGI endoscopy for evalua on of lower GI bleed.Endoscopy revealed persistent varices for all pa ents who underwent evalua on following postsplenectomy and devasculariza on surgery for extrahepa c portal venous obstruc on (EHPVO).
Failure to thrive is one of the most common reasons to perform paediatric endoscopy in children in order to obtain ssue samples for diagnosis of celiac disease.However, only a small propor on of pa ents were referred to our endoscopy service for this indica on.Duodenal biopsies from each of these pa ents revealed histopathological fi ndings consistent with celiac disease and corroborated with high serological ters for ssue transglutaminase (tTG) an bodies confi rming diagnosis of celiac disease.
Pa ents tolerated seda on/ anesthesia with ketamine and midazolam administered to their level of comfort.Only two pa ents developed transient deteriora on in oxygen satura on that improved on supplemental oxygen.

Discussion
Upper gastrointes nal endoscopy is an important diagnos c and therapeu c tool for gastrointes nal disorders in children 8 .There is a rising trend of diagnos c and therapeu c u lity of UGI endoscopy in children in parallel with increasing spectrum of its indica ons in the western world 9 .However, this diagnos c and therapeu c modality is underu lized in majority of the underdeveloped world due to lack of manpower, equipment and diffi cul es of invasive procedure in children.The scenario is rapidly changing in many areas of the developing world although Nepal has lagged behind in this development.We have very limited adult endoscopy service in the country and paediatric services were almost nonexistent un l the recent past 10 .Many paediatricians in Nepal therefore are not aware of the availability of services and many needy pa ents get treatment on presump ve grounds alone in absence of endoscopic diagnosis.Tribhuvan University Teaching Hospital (TUTH) is one of the major referral centres in the country and has recently started this important service.This study intends to present the experience of the paediatric endoscopy suite of TUTH to make paediatricians of the country aware on availability of service in the country and share our experience on common indica ons and fi ndings of paediatric endoscopy in Nepalese children.
Indica ons for UGI endoscopy from developing and developed countries are reportedly diff erent.Growth failure is the most common indica on for UGI endoscopy in children in the west; in contrast, literature from most of the developing countries report abdominal pain as the most common indica on for UGI endoscopy 11 .Abdominal pain, acute or recurrent accounted for 50.8% of all endoscopies performed in our pa ents.Similar fi gures are reported in other developing countries 11,12 .Studies from specialized gastroenterology centre or developed world however report much lower propor ons of children undergoing endoscopy primarily for abdominal pain 6 .
In our study, 26.2% children underwent the procedure for recurrent abdominal pain and 24.6% were scoped for acute abdominal pain.Yield of endoscopic diagnosis in children with recurrent abdominal pain is consistently low in published literature although results are be er with acute abdominal pain 13,14 .In our series, 80% pa ents with recurrent abdominal pain had nega ve UGI endoscopy.A systema c review in this subject showed a low diagnos c yield of UGIE in children with recurrent abdominal pain and concluded that u lity of the procedure for change in treatment, quality of life, improvement of abdominal pain, and cost-eff ec veness is unknown 14 .Although diagnos c yield has varied across studies, UGIE is s ll valuable for specifi c diagnosis and has therapeu c implica ons in many children with chronic abdominal pain 15 .
The commonest indica on for pa ents undergoing UGI endoscopy in our study was upper GI bleed (27.6%) followed by recurrent abdominal pain (26.2%).Endoscopy remains the principal method used to iden fy the source of bleeding and to perform therapeu c procedures if necessary 16 .Oesophageal varices (50%) were the most commonly iden fi ed lesions leading to UGI bleed among pa ents who underwent endoscopy for the UGI bleed.High diagnos c yield of varices has also been reported in other studies 17,18 and probably represent the group of pa ents catered by the endoscopic service rather than the actual prevalence of varices responsible for UGI bleed 19 .Endoscopic services in this country are s ll in infancy and our endoscopy service was the fi rst paediatric service to serve children with GI bleeding.Majority of pa ents with minor or self-limi ng bleeds obviously were not advised endoscopy.In addi on, high incidence of portal hypertension is known to exist in underdeveloped and poorer areas of the world 20 and there is marked diff erence in ae ologies of upper GI bleeding in children in reports from the developed and developing countries 17,19,21,22 .However, 35% of pa ents undergoing endoscopy in our series did not show an evident bleeding.Giannakopoulos et al reported a poor diagnos c yield of endoscopy performed a er an episode of non-variceal bleeding; an evident bleeding source was discovered in only 5 % of such endoscopies 23 .A high rate of nega ve endoscopies has also been reported in other studies that looked at all ae ologies of GI bleed 6,24,25 .
Growth failure/ Failure to thrive is major indica on for endoscopy in children to obtain ssue for histology of the duodenum as this is the most preferred and accepted diagnos c test for celiac disease 26 .Celiac disease has been iden fi ed as one of the most common ae ology for growth failure in young children 27 .A rising trend of celiac disease diagnosis has been reported in Indian studies 28,29 .Majority of Nepalese popula on has Indo-Aryan ancestry and closely resembling dietary prac ce.Prevalence of celiac disease in Nepal is therefore likely to match north Indian popula on due to iden cal gene c and dietary risk factors.However, celiac disease is reported in Nepalese children infrequently.This is refl ected in number of pa ents referred for endoscopy in our series.Only 0.6% of pa ents in this series underwent endoscopy with primary indica on of failure to thrive.The latest Nepal Demographic and Health Survey reported under fi ve stun ng and was ng rate of 41% and 11% respec vely 30 .Majority of these children certainly have chronic malnutri on as the primary cause; however, a substan al number of these children may have hidden celiac disease.Celiac disease has long-term health implica ons and its associa on with many other chronic diseases are emerging 31,32 .Overt celiac disease is readily diagnosed and managed; however, there should be a high index of suspicion in children with growth failure without overt manifesta ons 33 .Successive na onal surveys have shown improvement in nutri onal status of children in the country and paediatricians are more likely to iden fy and treat growth failure of organic causes rather than nutri onal defi ciencies in future.A high index of suspicion, performing serological tests and early referral to endoscopy services should improve growth and health status of many of these children with celiac disease.
Poten al serious airway problems such as laryngospasm, par al airway obstruc on, apnea, and stridor are reported during endoscopic procedure with use of ketamine and midazolam 34 .We performed the procedure safely without major problem in all pa ents under intravenous seda on/ anaesthesia with ketamine and midazolam.In all procedures, however, physician administering the seda on/ anaesthesia was only responsible for pa ent monitoring and airway management should such need arise.Such procedures are considered rela vely safe if performed in accordance to the guidelines 35 .

Conclusion
Upper GI endoscopy can be performed as a safe procedure in paediatric pa ents without complica ons under adequate seda on with appropriate monitoring and use of child friendly endoscopes in a paediatric suite.Paediatricians should be aware of its diagnos c u lity in their scope of prac ce and should use this service more frequently in diagnosis and management of common childhood ailments.Management of growth failure in Nepalese children should be revisited with emerging literature in Indian subcon nent.There is a need to expand this important service across the country.

Fig 1 :Fig 2 :
Fig 1: Showing gender distribu on of children who underwent endoscopy

Table 1 :
Indica ons for endoscopy in study popula on

Table 2 :
Findings of UGI endoscopy in study popula on