Singapore Med J 2003 Vol 44(9) : 453-456
O r i g i n a l A r t i c l e
Childhood Acute Pancreatitis in a
S-K Goh, C H Chui, A S Jacobsen
KK Women s and
Children s Hospital
S-K Goh, MA,
C H Chui,
A S Jacobsen,
C H Chui
Tel: (65) 6293 4044
Fax: (65) 6291 0161
Objective: To analyse the cases of acute pancreatitis
presented to a children’s hospital in Singapore.
Methods: Clinical charts of all children, aged under
18 years, who presented to our hospital for the
first time with pancreatitis (ICD search criteria =
577.x) between the period of 1998 and mid-2002
were reviewed. Parameters analysed included
presenting features, aetiology of the acute
pancreatitis, length of hospital stay, complications,
treatment and outcome.
Results: There were 12 cases in the review period,
and the attributable causes in these cases were,
in descending order, trauma, drug-induced,
anatomical anomalies, poisoning and idiopathic.
Of interest were two patients whose pancreatitis
were results of child abuse. The most common
symptoms were abdominal pain (n=11) and
vomiting (n=7), though only five patients localised
the pain to the epigastrium. Abdominal tenderness
could be elicited in all the patients. Eleven had
evidence of acute pancreatitis from computerised
tomography (CT) whilst the twelfth was diagnosed
with ultrasonography. The peak amylase levels
amongst these patients were not high, with a
median of 512.5 U/L. In the acute stage, only one
patient required operative intervention whilst
the remainder were managed conservatively.
The mean length of hospital stay was 12.41 ± 4.54
days. The complications encountered included
pseudocyst formation, ascites, hypocalcaemia,
pleural effusion and coagulopathy.
Conclusions: The diagnosis of acute pancreatitis
in children can be difficult. This is often due to
ambiguous symptoms, signs and laboratory results.
CT and ultrasound are essential investigations in
the diagnosis and subsequent follow-up.
Keywords: acute pancreatitis, children, child
Singapore Med J 2003 Vol 44(9):453-456
The incidence of acute pancreatitis in children is low
when compared to the adult population(1). Alcohol and
gallstones account for 80% of acute pancreatitis in
adults, and available English literature suggests
that the aetiological pattern of childhood acute
pancreatitis is different. We reviewed all the children
who presented with acute pancreatitis to our hospital
for the first time between 1998 to mid-2002, and report
the aetiology, presenting symptoms and signs, clinical
course and outcome.
PATIENTS AND METHODS
The clinical charts of all paediatric patients of
KK Women’s and Children’s Hospital assigned the
ICD code of 577.x between January 1998 to June
2002 were reviewed. The inclusion criteria were (1)
hyperamylasemia associated with severe abdominal
pain or vomiting, (2) radiological evidence of pancreatic
inflammation, or (3) histological or macroscopic evidence
of acute pancreatitis at laparotomy. Only patients who
were newly diagnosed to have acute pancreatitis were
considered for the study.
Twelve patients were identified with acute pancreatitis.
Their age ranged from 3.17 to 15.75 years (Mean
8.96, SD 3.50). The sex ratio was five males to seven
females. The racial composition of the patients was
similar to the local racial composition.
Trauma (n=5) was the leading attributable cause in
this group of patients, accounting for 41.2% (Fig. 1).
The mechanisms of trauma were varied. One patient
was hit by a car on her flank and another was hit
by a 25-inch television which fell through a height of
two feet at home; one girl was stepped upon on the
abdomen by her twin sister whilst playing at home.
Two patients developed acute pancreatitis following
child abuse whereby a seven-year-old boy was
stepped on the abdomen by his step-father and a
nine-year-old boy was repeatedly hit by his mother over
three months. The children who sustained traumatic
pancreatitis were between four and 10 years of age.
There were two patients in whom congenital
causes of acute pancreatitis, namely pancreatic divisum
and choledochal cyst, were found. Two patients had
acute lymphoblastic leukaemia and developed acute
pancreatitis after being started on L-asparaginase.
One patient was diagnosed to have acute pancreatitis
three weeks after ingestion of floor detergent in a
suicide attempt. The remaining two patients had no
identifiable cause for pancreatitis.
Abdominal pain was the most common symptom (n=11),
though only in five patients was the pain localised
in the upper abdomen or epigastrium. Vomiting
was the next common manifestation (n=7). Two
patients had fever. Palpation elicited tenderness in
all 12 patients, and this was localised in the upper
abdominal area in nine of them. One patient had
distended abdomen whilst two had palpable abdominal
mass. (Table I)
Elevated total white cell counts were noted in seven
of the 12 patients (58.3%). Serum amylase levels
were elevated in all the patients, and ranged from
190 to 1370 U/L (Median = 512.5 U/L). One third of
them had amylase levels lower than four times the
upper limit of normal, i.e. 440 U/L. Hypocalcemia was
found in one patient (Ca = 0.719 mmol/L), in whom
the total white count was normal. There were no other
abnormal haematological or biochemical markers in
our group of patients. The Ranson scores(8) for all the
patients were one or less.
Eleven patients had evidence of pancreatic inflammation
on CT whilst one patient had ultrasonography done
only. Some of the patients had ultrasonography
performed subsequent to CT for ease of follow-up on
the progression of the disease. The most common
radiological finding was swelling of the pancreas with
hypoechogenicity which was diagnostic of pancreatitis
(n=7). Pseudocysts were picked up in five of the
patients, of which four were patients who had traumatic
pancreatitis. Other findings included free fluid
(n=4), laceration or disruption of pancreas (n=3).
Mildly dilated pancreatic duct was found in one
patient, who subsequently was demonstrated to
have pancreatic divisum on magnetic resonance
One patient had to undergo a distal pancreatectomy
due to finding of complete transection at exploratory
laparotomy. The remainder recovered with conservative
management of bowel rest and nasogastric aspiration
(n=8), intravenous antibiotics (n=9), parenteral
octreotide (n=6), total parenteral nutrition (n=3).
The two patients in whom pancreatic divisum and
choledochal cyst were found underwent an endoscopic
retrograde cholangiopancreatography with sphincterotomy
and resection of choledochal cyst respectively. The
average length of hospital stay for the acute pancreatitis
in all patients was 12.41 ± 4.54 days.
Mortality and morbidity
Two patients had recurrences, numbering one to two
episodes each, of acute pancreatitis. One patient had
pleural effusion and pulmonary consolidation with
coagulopathy for which fresh frozen plasma was given.
Another patient developed mild pleural effusion which
resolved spontaneously. There was no mortality.
Table I. Distribution of clinical presentation.
Abdominal distension1 (8.3%)
Abdominal mass2 (16.7%)
Fig.1 Distribution of patients by aetiology.
454 : 2003 Vol 44(9) Singapore Med J
Singapore Med J 2003 Vol 44(9) : 455
tract, especially in the setting of severe trauma or
child abuse(10). All of our patients had definite changes
consistent with acute pancreatitis on ultrasonography
or CT. A study by Cox et al. showed a poor correlation
between serum amylase and ultrasonographic
evidence of pancreatitis in children(7).
Radiological techniques together with the clinical
assessments were able to provide us with conclusive
diagnoses of acute pancreatitis in all our children. The
presence of a markedly raised serum amylase level
would further confirm the diagnosis whilst a mildly
raised serum amylase level would usually confuse the
picture. We propose that such mildly raised serum
amylase level should not exclude acute pancreatitis
Acute pancreatitis is usually managed non-
operatively. The aim of management is to rest the
pancreas whilst providing nutritional support, anti-
microbial therapy and to manage complications when
they occur. Half of our patients received intravenous
octreotide in the course of their treatment with no
adverse effects. Octreotide, a long-acting somatostatin,
has been advocated for use in the treatment of acute
pancreatitis and pancreatic pseudocyst in adult(13).
It is postulated to work by decreasing pancreatic
secretion. Its role and safety in children has not yet
Two observations suggest that acute pancreatitis
in this group of patients tends to be mild and has good
prognosis: All the children in our series had Ranson
scores of one or less and there was no mortality.
Paediatric pancreatitis, though rare, does exist in our
population. As the presenting symptoms, clinical signs
and laboratory results are often ambiguous, the diagnosis
of pancreatitis can be delayed. Thus, a high index of
suspicion is essential when evaluating children with
abdominal pain, especially if the history is suggestive
of home accident or abuse. Radiological evaluation
is essential in the diagnosis of pancreatitis and the
1. Greenfeld JI, Harmon CM. Acute pancreatitis. Curr Opin Pediatr
2. Haddock G, Coupar G, Youngson GG, MacKinlay GA, Raine PA.
Acute pancreatitis in children: a 15-year review. J Pediatr Surg 1994;
3. Lopez MJ. The changing incidence of acute pancreatitis in children:
a single-institution perspective. J Pediatr 2002; 140:622-4.
4. Vane DW, Grosfeld JL, West KW, Rescorla FJ. Pancreatic disorders in
infancy and childhood: experience with 92 cases. J Pediatr Surg 1989;
5. Weizman Z, Durie PR. Acute pancreatitis in childhood. J Pediatr 1988;
6. Lerner A, Branski D, Lebenthal E. Pancreatic diseases in children.
Pediatr Clin North Am 1996; 43:125-56.
Pancreatitis is uncommon in the Singaporean
paediatric population. Only 12 new cases have been
documented in this hospital within the 31/2-year
review period. This is not different from the experience
of other institutions(2-5).
Unlike acute pancreatitis in the adult population,
paediatric pancreatitis is rarely attributable to biliary
stone disorders or alcoholism. Instead, the leading
causes of pancreatitis tend to be trauma, infections,
drugs, congenital disorders (e.g. pancreatic divisum,
choledochal cyst and cystic fibrosis)(6). This is
consistent with our finding that 75% of our patients
reviewed had acute pancreatitis due to trauma,
drugs or congenital abnormalities. In particular,
trauma accounted for five out of the 12 patients.
The observation that home accidents and child
abuse contributed to four cases suggests a need for
the clinician to be mindful of the possibility of non-
accidental injury when handling children presenting
with abdominal discomfort.
The diagnosis of acute pancreatitis in children
frequently requires a high index of suspicion. In
adults, the diagnosis of acute pancreatitis is classically
based on the finding of elevated serum amylase of
three to four times the normal value in the presence of
tell-tale clinical symptoms of severe epigastric pain
radiating to the back and vomiting(7,8).
Based on our experience, four out of the 12 of
our patients had serum amylase levels that were lower
than 440 U/L, which was four times the upper limit of
the amylase level. Furthermore, contrary to cases of
acute pancreatitis in the adult population where the
majority had amylase levels of greater than 1,000,
only four of our patients had amylase levels this
deranged. This observation could be explained by
the lack of cases attributable to gallstones and
alcoholism, which in the adult population, is known to
result in extremely high amylase levels(9). In traumatic
pancreatitis, it is also known that hyperamylasemia
may not occur until 12 hours after the acute event(10).
In addition, children with acute pancreatitis often do
not manifest classical symptoms of severe epigastric
pain radiating to the back. Whilst almost all of our
patients presented with abdominal pain and vomiting,
the intensity and location of the abdominal pain
varied greatly. The diagnostic difficulty in childhood
acute pancreatitis is illustrated by the fact that the
diagnoses of pancreatitis is often delayed(11).
Radiological studies are accurate in the diagnosis
of pancreatitis. Ultrasonography has a reported accuracy
of 80% in the diagnosis of acute pancreatitis(12). CT is
also extremely accurate, and has the added advantage
of being able to assess other solid organ and gastrointestinal
456 : 2003 Vol 44(9) Singapore Med J
7. Cox KL, Ament ME, Sample WF, Sarti DA, O Donnell M, Byrne WJ.
The ultrasonic and biochemical diagnosis of pancreatitis in children.
J Pediatr 1980; 96:407-11.
8. Russell RCG. Chapter 55: Pancreas. Bailey and Love s Short Practice
in Surgery. 23rd edition. Edited by Russell RCG, Williams NS, Bulstrode
CJK. Published by Edward Arnold; 23rd edition (May 15, 2000).
9. Reber HA and Way LW. Chapter 27: Pancreas. Pp. 567-94. Current
Surgical Diagnosis & Treatment. Edited by Way LW. Published by
Aplleton & Lange. 10th edition.
10. Tolia V, Patel AS, Amundson GM. Pancreatic fracture secondary to
child abuse: the role of computed tomography in its diagnosis. Clin
Pediatr (Phila) 1990; 29:667-8.
11. Lam JP, Eunson GJ, Munro FD, Orr JD. Delayed presentation of
handlebar injuries in children. BMJ 2001; 322:1288-9.
12. Synn AY, Mulvihill SJ, Fonkalsrud EW. Surgical disorders of the
pancreas in infancy and childhood. Am J Surg 1988; 156:201-5.
13. Gullo L, Barbara L. Treatment of pancreatic pseudocysts with
octreotide. Lancet 1991; 338:540-1.
Second Asian Regional IOF Conference on Osteoporosis
16-18 January 2004, Hong Kong
Venue: Hong Kong Convention and Exhibition Centre
1 Harbour Road, Wanchai, Hong Kong
Organised by : International Osteoporosis Foundation (IOF)
Co-Hosts: Jockey Club Centre for Osteoporosis Care and Control of
Chinese University of Hong Kong (JCCOCC)
Asian Pacific Osteoporosis Foundation (APOF)
Theme: Bisphosphonates Treatment
Epidemiology, Pathogenesis, Diagnosing and Monitoring Osteoporosis
Hormonal Therapy and Management
Pharmacological Management on Osteoporosis
Prevention of Osteoporosis
For more information, please contact:
The Federation of Medical Societies of Hong Kong
4/F., Duke of Windsor Social Service Building
15 Hennessy Road
Tel: (852) 25278898
Fax: (852) 28667530