Role of Pancreatic Divisum in Pediatric Pancreatitis: A Retrospective Observational Study

ABSTRACT

Aim:

The aim of the study was to evaluate the role of pancreatic divisum (PD) in children with pancreatitis.

Methods:

A retrospective observational study was conducted at a pediatric surgery tertiary center which included children aged <18 years who presented with pancreatitis between January 2013 and June 2024. The children with pancreatitis having PD (PD pancreatitis group) were compared to the children with pancreatitis not having PD (non-PD pancreatitis group).

Results:

A total of 275 children with pancreatitis were included in the study and 15 (5.5%) of them had PD. Type 1 (Classical) PD was seen in 11 (69%) of the children and nine (60%) children presented with acute recurrent pancreatitis (ARP). The children with PD pancreatitis presented at a younger age (9.8 ± 4.3 years vs. 12.2 ± 3.7 years; P = 0.02; independent sample t-test) and tend to have a significantly higher incidence of ARP (9 [60%] vs. 58 [22%]; P < 0.001; Chi-square test) compared to the children with non-PD pancreatitis.

Conclusion:

Children with PD present at a younger age with pancreatitis and are at increased risk of developing ARP. This suggests that PD is an independent etiology for developing pancreatitis in children.

INTRODUCTION

Pancreatic divisum (PD) is the most common congenital anomaly of the pancreas, occurring due to defective fusion between the dorsal and ventral pancreatic duct system during the second month of gestation.[1,2] The prevalence of PD varies between 2% and 15%, with low prevalence reported from endoscopic retrograde cholangiopancreatography (ERCP) series compared to magnetic resonance cholangiopancreatography (MRCP) and postmortem studies, possibly due to errors in interpreting pancreatograms, failure to cannulate minor papilla, or referral bias.[1,2] While 95% of PDs are asymptomatic, certain cases of PDs can lead to acute recurrent pancreatitis (ARP) and chronic pancreatitis (CP) due to inadequate drainage of pancreatic secretions and increased intraductal pressure.[1,2,3,4,5] Although the latest systematic review by Gutta et al. failed to define an etiological role of PD in pancreatic disease, it is still a topic of debate whether PD has any role in causation or is just an association.[1,2,3,4,5,6,7]

Pediatric pancreatitis, despite being less common than in adults, the environmental risk factors in the causation are almost negligible.[8] Around 5%–20% of pancreatitis in children is associated with pancreatic anomalies with PD being the most common anomaly.[8] Although pancreatitis in children is considered to be multifactorial, PD is found to be an independent risk factor for ARP and CP in children.[2,8,9,10] PD is also considered to be an additive risk factor for developing pancreatitis in children.[8,11,12] It is essential to identify whether PD is a true causative factor or merely an association in children with pancreatitis to prognosticate and treat pediatric pancreatitis. Hence, a study was conducted to evaluate the role of PD in pediatric pancreatitis. This research seeks to contribute to the ongoing knowledge of the diagnosis, presentation, and therapeutic strategies for PD in children.

METHODS

A retrospective observational study was conducted at a tertiary center after obtaining ethical approval from the Institutional Ethical Committee (No. BMCRI/PS/226/22-23). The study included children under 18 years of age who were admitted to the department of pediatric surgery with a diagnosis of PD between January 2013 and June 2024.

Patient data were retrieved from electronic medical records, which included the child’s age, sex, clinical presentation, laboratory investigations, imaging studies, diagnosis, and treatment. The children with incomplete medical records were excluded from the study. Any associated risk factors for ARP or CP were also documented, which include choledochal cyst, cholelithiasis, choledocholithiasis, pancreatic and/or biliary trauma, annular pancreas, hypertriglyceridemia, chronic medication usage, autoimmune pancreatitis, irritable bowel disease, indeterminate colitis, history of alcoholism or smoking, and family history of pancreatitis.[9]

AP was defined as the presence of at least two of the following three criteria: abdominal pain suggestive of acute pancreatitis, serum lipase and/or amylase ≥3 times than the normal upper limit, and imaging findings suggestive of acute pancreatitis.[13]

ARP was defined as the occurrence of at least two episodes of AP with complete normalization of lipase and amylase between the episodes or complete resolution of pain for at least a month between the episodes of AP.[13]

CP was defined as at least one of the following three criteria: abdominal pain consistent with pancreatic origin with imaging features suggestive of CP, exocrine pancreatic insufficiency with imaging features of CP, and endocrine pancreatic insufficiency with imaging features of CP.[13]

PD was diagnosed on MRCP when there was a defective or failure of fusion between the dorsal and ventral pancreatic ducts. PD was classified as type 1 when there was no communication between the dorsal and ventral duct systems [Figure 1], type 2 when there was an absent ventral duct (Wirsung’s) system [Figure 2], and type 3 when there was a thin duct communicating between the dorsal and ventral duct systems [Figure 3].[14] The management of PD was tailored based on the clinical presentation. The children with ARP and CP were discharged after treatment with pancreatic supplements. Children were followed up at 1 month, 3 months, 6 months, and then yearly after the discharge.

Figure 1.

Magnetic resonance cholangiopancreatography image showing type 1 pancreatic divisum. The main pancreatic duct (yellow arrow) is seen joining the duodenum at minor the duodenal papillae, whereas the accessory pancreatic duct (green arrow) is seen joining the duodenum at the major duodenal papillae along with the common bile duct (red arrow)

Figure 2.

Magnetic resonance cholangiopancreatography image showing type 2 pancreatic divisum. The main pancreatic duct (yellow arrow) is seen joining the duodenum at the minor duodenal papillae, whereas the common bile duct (red arrow) joining the duodenum at the major duodenal papillae. The accessory pancreatic duct is absent

Figure 3.

Magnetic resonance cholangiopancreatography image showing type 3 pancreatic divisum. The main pancreatic duct (yellow arrow) is seen joining the duodenum at the minor duodenal papillae, whereas the accessory pancreatic duct (green arrow) joining the duodenum at the major duodenal papillae along with the common bile duct (red arrow). A thin duct (blue arrow) is seen communicating between the main and the accessory pancreatic ducts

The children with pancreatitis having PD (PD pancreatitis group) were compared to the children with pancreatitis not having PD (non-PD pancreatitis group).

Statistical analysis

Statistical analysis was executed using SPSS Statistics, Version 27.0 (IBM Corp., Armonk, NY, USA). Categorical variables were expressed as numbers and percentages, whereas continuous variables were presented as mean ± standard deviation or median (range). Categorical variables were compared using the Chi-square test or Fisher’s exact test, and continuous variables were compared using the independent sample t-test or the Mann–Whitney U-test. A P < 0.05 was considered statistically significant.

RESULTS

A total of 275 children presented with pancreatitis during the study period and 15 (5.5%) of them were found to have PD. A child with cholelithiasis was found to have PD incidentally. Eleven (69%) were boys and seven (47%) children belonged to the 10–15-year age group. The mean age of the children was 9.8 ± 4.3 years and nine (60%) children presented with ARP. Fourteen (93%) children presented with pain abdomen, while 12 (80%) had vomiting as presenting symptoms. Type 1 (Classical) PD was seen in 11 (69%) of the children [Table 1]. Associated risk factors included two cases of choledochal cyst and a case of choledocholithiasis.

Table 1.

Clinical characteristics of children with pancreatic divisum

Demography	n (%)
Age group (years)
<5	3 (19)
5–10	4 (25)
10–15	8 (50)
>15	1 (6)
Mean age (years)	9.8±4.3
Boys	11 (69)
Presentation
Asymptomatic	1 (6)
Acute pancreatitis	1 (6)
Acute recurrent pancreatitis	9 (56)
Chronic pancreatitis	5 (31)
Diameter of main pancreatic duct (mm)
<5	11 (69)
5–10	5 (31)
>10	0
Types of PD
Type 1	11 (69)
Type 2	1 (6)
Type 3	4 (25)

PD: Pancreatic divisum

Out of nine children with ARP, one was lost to follow-up, three are having episodes of pancreatitis awaiting ERCP, and four are asymptomatic for more than a year. One child stopped pancreatic enzyme supplementation, taking ayurvedic medication, and is pain-free for more than 7 months.

Out of five children with CP, two underwent modified Puestow’s procedure, two underwent ERCP with sphincterotomy, and one child is on close follow-up. All these five children are asymptomatic and are on pancreatic enzyme supplements.

A child with malrotation and para duodenal hernia underwent Ladd’s procedure and repair of the defect. The child with cholelithiasis underwent cholecystectomy, whereas the child with choledocholithiasis underwent ERCP, extraction of calculi, and stenting of the common bile duct. Two children with choledochal cysts underwent laparoscopic excision of the choledochal cyst with hepatico-duodenostomy. All these children are asymptomatic at present. The follow-up ranged from 9 months to 12 years.

The children with PD pancreatitis presented at a younger age (9.8 ± 4.3 years vs. 12.2 ± 3.7 years; P = 0.02; independent sample t-test) compared to children with non-PD pancreatitis. The children with PD tend to have a significantly higher incidence of ARP (9 [60%] vs. 58 [22%]; P < 0.001; Chi-square test) while the children with non-PD pancreatitis were found to have a higher incidence of acute pancreatitis (95 [37%] vs. 1 [7%]; P = 0.02; Fisher’s exact test). There was no difference between the children with PD and the children without PD concerning age distribution, sex, size of main pancreatic duct, associated comorbidities, or family history of pancreatitis [Table 2].

Table 2.

Pancreatic divisum pancreatitis versus nonpancreatic divisum pancreatitis

Parameters	PD with pancreatitis (n=15)	Non-PD pancreatitis (n=260)	P
Age group (years)
<5	3 (20)	23 (9)	0.15*
5–10	4 (27)	58 (22)	0.75*
10–15	7 (47)	132 (51)	0.79*
>15	1 (7)	47 (18)	0.48*
Mean age	9.8±4.3	12.2±3.7	0.02@
Boys	10 (67)	157 (60)	0.63#
Presentation
Acute pancreatitis	1 (7)	95 (37)	0.02*
Acute recurrent pancreatitis	9 (60)	58 (22)	<0.001#
Chronic pancreatitis	5 (33)	107 (41)	0.36#
Diameter of main pancreatic duct (mm)
<5	10 (67)	183 (70)	0.78#
5–10	5 (33)	64 (25)	0.45#
>10	0	13 (5)	1*
Associated risk factors
Total	3 (20)	20 (8)	0.12*
Choledochal cyst	2 (13)	9 (3)	0.11*
Choledocholithiasis	1 (7)	2 (1)	0.16*
Cholelithiasis	0	9 (3)	1*
Family history of pancreatitis	0	2 (1)	1*

*Fisher’s exact test, ^@Independent sample t-test, ^#Chi-square test. PD: Pancreatic divisum

DISCUSSION

Pancreatitis is a major health concern among children and adolescents worldwide.[15] The prevalence and incidence rates show a rising trend, particularly affecting regions with lower sociodemographic indices.[15] Mortality due to pancreatitis is notably higher in children under 5 years of age and 15–18 years of age.[15] The etiology and pathogenesis of pancreatitis in children differ significantly from that of adults. Gallstones and alcohol abuse account for 70%–80% of cases of pancreatitis in adults.[16] In children, pancreatitis has a more diverse etiology which includes congenital anomalies, genetic mutations, infections, biliary pathology, trauma, autoimmune disorders, metabolic disorders, drug induced, and systemic diseases.[8,9,10,17]

Congenital pancreatic anomalies such as PD, choledochal cyst, and annular pancreas are frequently implicated in Pediatric pancreatitis, with PD contributing to the majority of the cases secondary to congenital anomalies.[8,9,10,17] Genetic factors, including mutations in CFTR, SPINK1, and PRSS1, play a central role in the development of ARP and CP in children.[8,9,10,17] Viral infections caused by the Epstein–Barr virus, mumps, cytomegalovirus, and human immunodeficiency virus can also lead to pancreatitis in children.[8,9,10,17] In addition, drugs such as corticosteroids, valproic acid, and L-asparaginase can induce pancreatitis in children.[8,9,10,17] Despite comprehensive evaluation, a significant proportion of pediatric pancreatitis cases remain idiopathic, although they may have undetected anatomic or genetic causes.

The fundamental pathogenesis of pancreatitis in children as well as adults involves premature intra-acinar activation of trypsinogen to trypsin, leading to pancreatic inflammation and autodigestion.[17] However, the cascade is initiated by genetic or anatomic factors in children, whereas in adults, the cascade is activated by mechanical obstruction or lifestyle-related factors.[16,17] Understanding these distinctions in the etiopathogenesis of pancreatitis in children is essential to tailor the diagnostic modalities, management approaches, and effective prevention strategies.

Whether PD has any role in causation or is just an association with pancreatitis is the subject of ongoing debate.[1,2,3,4,5,6,7] While some studies argue that PD is just an association with idiopathic pancreatitis, many studies suggest PD as a causative factor. Several studies have demonstrated an increased incidence of PD in patients with idiopathic pancreatitis, which is supported by studies establishing the benefit of therapeutic interventions in patients with PD.[2,3,4,5,7,9,10,18,19] Conversely, a few studies challenge the role of PD in the development of pancreatitis.[11,12] However, most of these studies are based on PD in adults. Hence, a retrospective study was conducted to evaluate the role of PD in pediatric pancreatitis.

The frequency of PD in idiopathic pancreatitis varies from study to study. A few studies indicate that the frequency of PD in patients with idiopathic pancreatitis is the same as that in the general population or cases of alcoholic CP.[11,20,21] However, all these studies involved adults and are small-numbered studies. A large pediatric pancreatitis study from the INSPIRRE database reported a 14.5% prevalence of PD in children with ARP and CP.[9] In addition, Gonoi et al. conducted a community-based as well as in-hospital pancreatitis patients study and found that the prevalence of all idiopathic pancreatitis, ARP, and CP was significantly higher in people with PD compared with those without PD.[7] This finding is also supported by many other studies including Indian studies.[18] In our study, the prevalence of PD was 5.5% of children with pancreatitis. The prevalence increased to 8% when we considered only children with ARP and CP, which is higher than that of the general population (1.5% to 2%) in India.[1,4,6,7,18] The relatively lower prevalence of PD in our series may be attributed to the sole reliability of MRCP for the diagnosis, as well as geographical variation in the prevalence of PD.[1,4,6]

Boys constituted 60% of PD cases in our study group, unlike the INSPIRRE study, where females were predominant.[9] Similarly, studies by Wen et al. and Lucidi et al. reported a higher prevalence of PD in girls with ARP.[2,22] The role of sex in influencing pancreatitis remains unclear.[19] The mean age of our study population was 9.8 ± 4.3 years which is similar to that of Wen et al.[2]

In our study, 87% of children with PD developed either ARP or CP. Previous studies have also reported an increased prevalence of ARP and CP in the PD population.[1,2,10] The obstruction of the duct at the level of minor papilla leads to increased pressure in the dorsal pancreatic duct (23.7 ± 1.3 mm Hg) compared to the ventral pancreatic duct (10.8 ± 1.9 mm Hg), leading to ARP and CP.[1,9] Classical (type 1) PD was the most common type in our study, similar to other studies by Stimec et al. and MacCarty et al.[23,24]

Several diagnostic modalities have been used in diagnosing PD. Although ERCP with ductography and minor papilla sphincterotomy is considered the gold standard for the diagnosis and treatment, it is invasive, not easily available, costly, and sometimes, technically not possible in younger children.[25,26]

MRCP is a noninvasive diagnostic modality with a sensitivity ranging from 60% to 73% in diagnosing PD.[27,28] Secretin-enhanced MRCP increases the sensitivity of MRCP to 83%–86% and specificity to 97%–99%.[29,30] However, the availability of the same along with the need for sedation is an issue. Endoscopic ultrasound has a sensitivity of 87%–95% in diagnosing PD, but its availability and cost remain major concerns.[30] In our study, all cases of PD were diagnosed by MRCP.

In our study, we found that the children with PD pancreatitis tend to present at a younger age compared to the children with non-PD pancreatitis. This evidence was supported by Lucidi et al., who reported a lower mean age at diagnosis in children with PD (5.7 ± 4.5 years) compared to those without PD (8.8 ± 5.1 years).[22] Similarly, a study by Neblett and O’Neill found that children with PD developed an early onset of episodic pain and vomiting, with a mean age of 6 years.[31] Other studies, including Klein et al. and Muzaffar et al., also support the finding of earlier presentations in children with PD.[5,32]

In a few studies, for instance, Bertin et al. found that the incidence of PD in idiopathic pancreatitis was the same as that in alcohol-induced pancreatitis and also demonstrated associated PRSS 1, SPINK 1, and CFTR gene mutations could be the reason for pancreatitis.[11] Garg et al. found that the patients with idiopathic pancreatitis with PD had a higher incidence of SPINK-1 gene mutation and that is the reason for pancreatitis.[12] However, both these studies were conducted in adults and smaller studies. A large pediatric study by INSPIRRE, which included 52 children with PD pancreatitis and 307 with non-PD pancreatitis found that there was no increased incidence of genetic mutation in children with PD.[9] We have recently started conducting genetic analysis on children with PD, as gene mutations may predispose to early onset of the disease. Hence, a comprehensive analysis of these children is required to aid in the proper diagnosis and treatment and also to provide new insights for a better understanding of this condition.

In our study, we observed a higher incidence of PD in children with ARP and CP compared to the general population. Children with PD also presented earlier and had a higher prevalence of ARP compared to children with non-PD pancreatitis. These findings suggest that PD has a causative role in the development of ARP and CP in children.

The limitations of our study include its retrospective nature, small sample size, and absence of genetic analysis. In addition, we did not explore therapeutic interventions in cases of PD with nondilated pancreatic ducts.

CONCLUSION

Children with PD present most commonly with ARP, with pain abdomen and vomiting being the most common symptoms. The majority of these children had type 1 PD. Children with PD pancreatitis present at a younger age and tend to have an increased risk of developing ARP compared to children with non-PD pancreatitis, suggesting that PD is an independent risk factor for developing pancreatitis in children.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.