Abstract
Objectives:
Adults with chronic pancreatitis (CP) have a high risk for developing pancreatogenic diabetes mellitus (DM), but little is known regarding potential risk factors for DM in children with acute recurrent pancreatitis (ARP) or CP. We compared demographic and clinical features of children with ARP or CP, with and without DM, in the INternational Study Group of Pediatric Pancreatitis: In Search of a CuRE (INSPPIRE) registry.
Methods:
We reviewed the INSPPIRE database for the presence or absence of physician-diagnosed DM in 397 children, excluding those with total pancreatectomy with islet autotransplantation, enrolled from August 2012 to August 2017. Patient demographics, body mass index percentile, age at disease onset, disease risk factors, disease burden, and treatments were compared between children with DM (n=24) and without DM (n=373).
Results:
24 children (6.0% of the cohort) had a diagnosis of DM. Five of 13 tested were positive for beta cell autoantibodies. The DM group was 4.2 years (95% CI 3.0, 5.4) older at first episode of acute pancreatitis, and tended to more often have hypertriglyceridemia (odds ratio (OR) 5.21 (1.33, 17.05)), coexisting autoimmune disease (OR 3.94 (0.88, 13.65)) or pancreatic atrophy (OR 3.64 (1.13, 11.59)).
Conclusions:
Pancreatic atrophy may be more common among children with DM, suggesting more advanced exocrine disease. However, data in this exploratory cohort also suggest increased autoimmunity and hypertriglyceridemia in children with DM, suggesting that risk factors for Type 1 and Type 2 DM respectively may play a role in mediating DM development in children with pancreatitis.
Keywords: acute pancreatitis, islet, hereditary pancreatitis, pediatric pancreatitis
Introduction
Acute recurrent pancreatitis (ARP) or chronic pancreatitis (CP) result in recurrent injury to the pancreatic parenchyma. As a result, for patients afflicted with ARP and CP, the risk of developing diabetes mellitus (DM) is highly elevated compared to the average population (1–3). For adults with CP, the lifetime risk for glucose intolerance or DM is estimated at 25–80%, depending on the population, disease etiology, and treatments (1). Nearly all epidemiologic research on DM risk in pancreatitis has been based in adult populations, with little research around the risk for DM in children with CP and ARP (4).
Genetic risk factors for pancreatitis are frequent in children with ARP and CP, identified in nearly half of children with ARP and three quarters of those with CP enrolled in the INternational Study group of Pediatric Pancreatitis: In Search of a CuRE (INSPPIRE) (5). In two European series of individuals with hereditary pancreatitis, primarily due to a pathogenic mutation in the Protease Serine 1 (PRSS1) gene encoding cationic trypsinogen, the lifelong risk for developing DM exceeded 60% by 60 years of age, but notably, onset of DM in childhood was rare, with only 4% with DM by age 20 years (6, 7). Thus, even though lifelong risk for DM is high with genetically-mediated pancreatitis, risk for childhood-onset DM is low (7). Other risk factors for DM identified in adults with CP include exocrine pancreatic insufficiency, pancreatic surgery, pancreatic calcifications or atrophy, and overweight/obese BMI (8–10). Whether these factors also convey risk for DM in childhood is unknown.
Using data collected in the INSPPIRE registry, we aimed to describe the burden of DM in children with ARP or CP, and to explore patient and disease characteristics that may differ in those children who developed DM versus those who had not. The latter is important in identifying parameters that should be considered in future studies for progression to DM in children with pancreatitis.
Methods
Subjects and Participating Centers
INSPPIRE is a multi-center, international consortium for the study of ARP or CP in children. Children <19 years of age were considered eligible for inclusion in the study if they met criteria for ARP and/or CP. ARP was defined as 2 or more episodes of acute pancreatitis (AP) confirmed by clinical symptoms and lipase or amylase >3x the upper limit of normal OR imaging features of AP, with episodes separated by ≥1 month. CP was diagnosed if at least one of the following criteria were met: (1) abdominal pain and imaging features of CP; (2) exocrine pancreatic insufficiency and imaging features of CP; or (3) endocrine insufficiency and imaging features of CP. Imaging studies were performed at the discretion of the clinical provider. For the purpose of these analyses, we excluded those patients who had total pancreatectomy (TP) with or without islet autotransplantation (IAT). Patients who have cystic fibrosis included only if they met study entry criteria for ARP or painful CP. All centers obtained local Institutional Review Board (IRB) approval or the equivalent for their country prior to enrolling subjects. All centers met the criteria of the Declaration of Helsinki (https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/). Consent was obtained from the parents of participants less than 18 years and directly from participants 18 years or older. Children gave assent at the age specified by the local IRB.
The INSPPIRE study methodology has been described in detail previously (11). Physician and patients completed questionnaires at baseline and at 1 year longitudinal follow up intervals. Collected data included information on the diagnosis of ARP and CP, disease features, disease burden, and treatments. Diagnosis of DM was assessed by the physician completing the questionnaire based on standard American Diabetes Association (ADA) criteria, specifically any of: (1) fasting glucose ≥126 mg/dL, (2) HbA1c ≥6.5% (for #1 or 2, confirmed on repeat), or (3) random glucose >200 mg/dL with symptoms (12). Details of DM from the physician form included diagnostic tests used to make the diagnosis, date of diagnosis and history of whether any autoantibody testing for type 1 diabetes was performed. Autoantibody testing may include any of the clinically available assays: islet cell antibodies, insulin autoantibodies, glutamic acid decarboxylase antibodies, tyrosine phosphatase (IA-2) antibodies, and Zinc Transporter 8 antibodies. Hypertriglyceridemia was defined by fasting triglyceride level above the upper limit of normal.
Statistical Analyses
Summary statistics for demographic, clinical, and treatment variables were computed for DM and non-DM groups, with mean (standard deviation) for continuous variables, median (interquartile range) for ordinal/count/rate variables, and frequency count (percent) for categorical variables. We then considered what variables might be associated with DM; because of the multiple variables examined, the analyses are designed to be exploratory. Comparison of DM and non-DM groups were examined for relevant variables by computing mean difference for age, median difference for ordinal/count/rate variables (i.e. number of ER visits, number of hospitalizations/year, number of days missed school), and odds ratio for categorical variables, with corresponding 95% confidence interval. The coverage of the 95% CIs (i.e. do not include zero for mean or median difference; do not include one for odds ratio) were examined to identify variables that suggest possible differences between groups. For the purpose of these exploratory analyses, p-values were not reported; rather, results should be considered hypotheses generating and need confirmation by future study. All statistical analyses were performed using SAS (version 9.4).
Results
Patient Characteristics
There were 397 children with CP or ARP included in the analyses, including 24 with a diagnosis of DM, 6% of the cohort. Patient demographics are displayed in Table 1. The patients with DM were older at the first episode of AP (12.9±2.6 vs 8.7±4.7, mean difference 4.2 years (3.0, 5.4)). Prevalence of overweight or obese BMI at study entry was similar in the DM and non-DM groups. Of the 24 children with DM, 23 had a history of AP, and 9 (38%) had CP while the other 15 (62%) had ARP alone.
Table 1:
Demographics of the study population
| Variables | Diabetes=Yes (n=24) | Diabetes=No (n=373) | OR or Mean Difference (95% CI) |
|---|---|---|---|
| Female, No. (%) | 16 (67%) | 211 (57%) | 1.53 (0.64, 3.68)a |
| (n=23) | (n=351) | ||
| Ethnicity (Hispanic) | 8 (35%) | 87 (25%) | 1.62 (0.66, 3.95)a |
| Race | (n=20) | (n=330) | |
| White | 16 (80%) | 268 (81%) | 0.93 (0.28, 3.94)b |
| African American | 2 (10%) | 9 (3%) | |
| Asian | 0 (0%) | 19 (6%) | |
| Multi-racial | 1 (5%) | 25 (8%) | |
| Other | 1 (5%) | 9 (3%) | |
| Age at enrollment, years | (n=23) | (n=366) | |
| Mean±SD | 14.7±3.0 | 11.4±4.6 | 3.3 (1.9, 4.6)c |
| Age at first diagnosis AP* | (n=23) | (n=342) | |
| Mean±SD | 12.9±2.6 | 8.7±4.7 | 4.2 (3.0, 5.4)c |
| With CP | 9 (38%) | 150 (40%) | |
| Age at diagnosis of CP | (n=7) | (n=132) | |
| Mean±SD | 14.2±2.4 | 9.9±4.4 | 4.3 (1.0, 7.5)c |
| Time from AP to CP, years** | (n=20) | (n=333) | Hazard Ratio |
| Median (25th-75th percentile) | 4.80 (4.41–5.50) | 4.48 (0.95–14.25) | 0.78 (0.34, 1.77) |
| BMI (percentile) | (n=23) | (n=355) | |
| Underweight (<5) | 0 (0%) | 19 (5%) | |
| Normal (5–<85) | 12 (52%) | 195 (55%) | |
| Overweight (85–<95) | 5 (22%) | 65 (18%) | 1.39 (0.60, 3.24)d |
| Obese (≥95) | 6 (26%) | 76 (21%) |
Abbreviations: AP, acute pancreatitis; BMI, body mass index; CP, chronic pancreatitis; OR, odds ratio
Of those who had an acute pancreatitis attack, 23 in Diabetes and 354 in no Diabetes (12 with missing age data)
median time from Kaplan-Meier curve, with no CP considered as censored with duration of disease as follow-up time
Value is expressed as OR (95% CI) of Diabetes
Value is expressed as OR (95% CI) of Diabetes for White
Value is expressed as mean difference (95% CI) for Diabetes minus No Diabetes
Value is expressed as OR (95% CI) of Diabetes for Overweight/Obese
Of the 24 children with diabetes, the diagnosis of DM was present at study entry in 19 cases based on elevated fasting glucose, elevated hemoglobin A1c (HbA1c) level, and abnormal glucose tolerance test in 4 patients; elevated fasting glucose and elevated HbA1c in 9 patients; elevated HbA1c alone in 5 patients, and abnormal glucose tolerance test alone in 1 patient. In the remaining 5 children, DM was diagnosed in the follow-up interval, after initial enrollment. Only 13 of the 24 children had been tested for beta cell autoantibodies, with positive autoantibodies in 5/13 cases. Diabetes was treated with insulin and non-insulin medications in 5 patients, insulin alone in 15 patients, and no pharmacologic therapies in 2 patients. Treatment regimen was unknown for 2 children.
With regard to the interval between first diagnosis of pancreatitis and diagnosis of DM, 11 (46%) patients were diagnosed with DM prior to pancreatitis diagnosis, 5 (21%) patients had simultaneous diagnosis of DM and pancreatitis (onset of DM within 60 days of pancreatitis diagnosis), and 8 (33%) patients had onset of DM after diagnosis of pancreatitis with a mean duration of 3.6 ± 1.7 years between pancreatitis diagnosis and DM onset. Of note, of the 11 participants with DM preceding their pancreatitis diagnosis, 10 had ARP alone while only 1 had been diagnosed with CP.
Disease features in children with ARP and CP with or without DM
Risk factors for ARP or CP and comorbidities are displayed in Table 2. Patients with DM commonly had hypertriglyceridemia (23% vs 5% of non-DM, OR 5.21 (1.33, 17.05)), and tended towards increased risk for other autoimmune diseases (17% vs 5%, OR 3.94 (0.88, 13.65)). Patients with DM had an increased prevalence of pancreatic atrophy (50% vs 22% without DM, OR 3.64 (1.13, 11.59)), but pancreas morphology on imaging was otherwise similar in DM and non-DM groups Table 3. Pancreatic calcifications were notably uncommon in children (7% of those with vs 8% of those without DM). Other pancreatitis risk factors were similar between groups, including genetic mutations.
Table 2:
Risk factors for ARP and CP
| Variables | Diabetes=Yes (n=24) | Diabetes=No (n=373) | OR (95% CI) of Diabetes |
|---|---|---|---|
| Family history | |||
| Acute pancreatitis, No./sample sizea (%) | 6/22 (27%) | 84/315 (27%) | 1.03 (0.39, 2.72) |
| Chronic pancreatitis | 1/20 (5%) | 67/319 (21%) | 0.20 (0.00, 1.29) |
| Genetic Risk Factors: | |||
| PRSS1 | 1/17 (6%) | 62/252 (25%) | 0.19 (0.00, 1.29) |
| SPINK1 | 1/15 (6%) | 47/233 (20%) | 0.28 (0.01, 1.96) |
| CFTR | 5/18 (28%) | 78/253 (31%) | 0.86 (0.30, 2.50) |
| CTRC | 1/13 (8%) | 12/158 (8%) | 1.01 (0.02, 8.06) |
| Obstructive factors | 8 (33%) | 114/368 (31%) | 1.11 (0.46, 2.68) |
| Pancreas divisum | 3/22 (14%) | 38/361 (11%) | 1.34 (0.24, 4.88) |
| Sphincter of Oddi disorders | 1/22 (5%) | 8/353 (2%) | 2.05 (0.04, 16.55) |
| Gallstones | 0/23 (0%) | 20/361 (6%) | 0.00 (0.00, 2.56) |
| Pancreatic duct mal-union | 0/22 (0%) | 15/362 (4%) | 0.00 (0.00, 3.72) |
| Post-traumatic pancreatic stricture | 0/23 (0%) | 2/362 (1%) | 0.00 (0.00, 55.50) |
| Preampullary duodenal diverticulum | 0/22 (0%) | 1/364 (0.3%) | 0.00 (0.00, 314.36) |
| Duct obstruction | 0/23 (0%) | 9/364 (2%) | 0.00 (0.00, 6.45) |
| Annular pancreas | 1/22 (5%) | 2/365 (1%) | 8.64 (0.14, 170.00) |
| Choledochal cyst | 0/23 (0%) | 10/366 (3%) | 0.00 (0.00, 5.73) |
| Toxic/Metabolic | 8/23 (35%) | 89/346 (26%) | 1.54 (0.63, 3.76) |
| Alcoholic | 0/23 (0%) | 7/372 (2%) | 0.00 (0.00, 8.93) |
| Active/Passive smoking exposure | 3/21 (14%) | 32/354 (9%) | 1.68 (0.30, 6.21) |
| Hyperlipidemia (above normal limits) | 5/22 (23%) | 17/318 (5%) | 5.21 (1.33, 17.05) |
| Medications | 2/22 (9%) | 37/335 (11%) | 0.80 (0.09, 3.54) |
| Autoimmune pancreatitis | 1/20 (5%) | 10/289 (3%) | 1.47 (0.03, 11.34) |
| Other autoimmune diseases* | 4 (17%) | 17/352 (5%) | 3.94 (0.88, 13.65) |
Includes Chron’s disease, ulcerative colitis, indeterminate colitis IBD-U, hepatitis, and other autoimmune diseases
The differences in sample sizes between rows reflect available data for these parameters
Table 3:
Imaging features of pancreatitis
| Variables | Diabetes=Yes (n=24) | Diabetes=No (n=373) | OR (95% CI) of Diabetes |
|---|---|---|---|
| Imaging studies performed | |||
| ERCP | 7 (29%) | 153 (41%) | 0.59 (0.24, 1.46) |
| CT scan | 10 (42%) | 164 (44%) | 0.91 (0.39, 2.10) |
| MRI | 11 (46%) | 249 (67%) | 0.42 (0.18, 0.97) |
| MRCP | 15 (63%) | 278 (75%) | 0.57 (0.24, 1.34) |
| EUS | 0 (0%) | 38 (10%) | 0.00 (0.00, 1.22) |
| Findings (CT/MRI) | |||
| Focal Acute Pancreatitis | 3/15 (20%) | 43/220 (16%) | 1.28 (0.22, 5.02) |
| Inflammatory changes in Pancreas | 6/15 (40%) | 98/298 (37%) | 1.14 (0.39, 3.31) |
| Pancreatic Gland Enlargement | 5/15 (33%) | 69/267 (26%) | 1.43 (0.37, 4.80) |
| Pancreatic Atrophy Present | 8/16 (50%) | 58/269 (22%) | 3.64 (1.13, 11.59) |
| Calcifications in Pancreas present | 1/15 (7%) | 22/268 (8%) | 0.80 (0.02, 5.75) |
| Duct Irregularities | 2/16 (12%) | 92/267 (34%) | 0.27 (0.06, 1.22) |
| Pancreatic duct dilatation present | 3/16 (19%) | 97/268 (36%) | 0.41 (0.11, 1.46) |
| Lesions present in the pancreas | 0/16 (0%) | 9/272 (3%) | 0.00 (0.00, 7.04) |
| Gallstones/Sludge | 1/16 (6%) | 23/266 (9%) | 0.70 (0.02, 5.01) |
| Intraheptic biliary dilatation | 1/16 (6%) | 27/269 (10%) | 0.60 (0.01, 4.20) |
| Changes suggestive of cirrhosis and/or portal hypertension | 2/16 (12%) | 17/266 (6%) | 2.09 (0.21, 10.33) |
| Findings (CT/MRI/ MRCP/ ERCP) | |||
| Pancreatic duct obstruction (stricture) | 1/22 (5%) | 63/321 (20%) | 0.20 (0.00, 1.26) |
| CBD stricture (Intrapancreatic portion) | 0/22 (0%) | 17/334 (5%) | 0.00 (0.00, 2.97) |
| Dilated CBD | 5/22 (23%) | 49/336 (15%) | 1.72 (0.47, 5.15) |
| CBD Stone | 0/22 (0%) | 20/337 (6%) | 0.00 (0.00, 2.49) |
| Findings (CT/MRI/EUS) | |||
| Peripancreatic inflammation/fat stranding | 7/15 (47%) | 90/274 (33%) | 1.79 (0.63, 5.09) |
| Findings (CT/MRI/MRCP/ ERCP /EUS) | |||
| Cysts/Pseudocysts | 4/22 (18%) | 50/338 (15%) | 1.28 (0.30, 4.11) |
| Findings (MRCP/ERCP) | |||
| Main Pancreatic Duct - Abnormal | 7/18 (39%) | 149/291 (51%) | 0.61 (0.23, 1.61) |
| Abnormal Side Branches | 4/16 (25%) | 76/284 (27%) | 0.91 (0.21, 3.13) |
| Intraductal filling defects of Calculi | 2/16 (12%) | 54/290 (19%) | 0.62 (0.07, 2.85) |
ERCP= Endoscopic retrograde cholangiopancreatography; CT = Computed tomography; MRI= Magnetic resonance imaging; MRCP= Magnetic resonance cholangiopancreatography; EUS= Endoscopic ultrasound;
Disease treatment and disease burden in children with ARP and CP with or without DM
Notably, endoscopic procedures and surgical therapies were similar in those with and without DM (Table 4). Pain medication usage in patients was similar between DM and non-DM groups for both provider reported use and patient self-report. Those children with DM were more likely to report any level of constant pain (45% vs 22% of those without DM, OR 5.05 (1.30, 16.27)). Exocrine pancreatic insufficiency trended towards more common among those with DM (36% vs 17%, OR 2.81 (0.97, 7.54)) (Table 5).
Table 4:
Treatments administered for pancreatitis
| Variables | Diabetes=Yes (n=24) | Diabetes=No (n=373) | OR (95% CI) of Diabetes |
|---|---|---|---|
| Medications (from provider response) | |||
| Pain medications | 4/17 (24%) | 94/279 (34%) | 0.61 (0.19, 1.91) |
| Medical therapies | 11/23 (48%) | 150/371 (40%) | 1.35 (0.58, 3.14) |
| Pancreatic enzymes | 9/23 (39%) | 121/371 (33%) | 1.33 (0.56, 3.15) |
| PERT for pain or recurrent pancreatitis | 9/23 (39%) | 103/365 (28%) | 1.64 (0.69, 3.89) |
| Vitamins/anti-oxidants | 1/22 (5%) | 28/363 (8%) | 0.57 (0.01, 3.83) |
| Steroids | 1/22 (5%) | 7/359 (2%) | 2.39 (0.05, 20.05) |
| Octreotide | 0/22 (0%) | 6/365 (2%) | 0.00 (0.00, 11.12) |
| Medical PPI | 6/19 (32%) | 52/307 (17%) | 2.26 (0.67, 6.73) |
| Medications (from patient response) | |||
| Pain medications | 8/23 (35%) | 140/342 (41%) | 0.77 (0.32, 1.86) |
| Pancreatic enzymes | 8/23 (35%) | 95/341 (28%) | 1.38 (0.57, 3.36) |
| Vitamins/anti-oxidants | 10/22 (45%) | 131/344 (38%) | 1.36 (0.57, 3.22) |
| Steroids | 0/23 (0%) | 9/346 (3%) | 0.00 (0.00, 6.12) |
| Medication helpful in treating pancreatitis | |||
| (from patient response of those using medication) | |||
| Pain medications helpful | 5/7 (71%) | 91/107 (85%) | 0.44 (0.07, 5.04) |
| Pancreatic enzymes helpful | 2/2 (100%) | 46/51(90%) | -- |
| Vitamins/anti-oxidants helpful | 0/4 (0%) | 13/55 (24%) | 0.00 (0.00, 4.00) |
| Steroids helpful | -- | 2/2 (100%) | -- |
| Procedures | |||
| Any ERCP | 7/21 (33%) | 141/369 (38%) | 0.81 (0.32, 2.05) |
| Biliary sphincterotomy | 2/21 (10%) | 51/364 (14%) | 0.65 (0.07, 2.81) |
| Pancreatic sphincterotomy | 5/21 (24%) | 83/365 (23%) | 1.06 (0.30, 3.15) |
| Pancreatic duct stent | 3/21 (14%) | 66/365 (18%) | 0.76 (0.14, 2.70) |
| Biliary stent | 0/21 (0%) | 10/366 (3%) | 0.00 (0.00, 6.30) |
| Pancreatic duct stone removal | 1/21 (5%) | 40/366 (11%) | 0.41 (0.01, 2.69) |
| Surgeries | |||
| Surgical therapies | 3/23 (13%) | 57/373 (15%) | 0.83 (0.15, 2.94) |
| Cholecystectomy | 2/23 (9%) | 35/372 (9%) | 0.92 (0.10, 4.02) |
| Celiac nerve block | 0/23 (0%) | 1/372 (0.3%) | 0.00 (0.00, 307.3) |
| Cyst/pseudo-cyst operation | 1/23 (4%) | 12/372 (3%) | 1.36 (0.03, 10.03) |
| Lateral pancreaticojejunostomy | 0/23 (0%) | 14/372 (4%) | 0.00 (0.00, 3.95) |
| Partial pancreatectomy | 0/23 (0%) | 1/372 (0.3%) | 0.00 (0.00, 307.3) |
Table 5:
Burden of disease in patients ARP and CP with and without DM
| Variables | Diabetes=Yes (n=24) | Diabetes=No (n=373) | OR or Median Difference (95% CI) |
|---|---|---|---|
| Pattern of abdominal pain | (n=22) | (n=345) | |
| 0) no abdominal pain | 2 (9%) | 42 (12%) | |
| 1) usually pain free; episodes of mild-moderate pain | 3 (14%) | 54 (16%) | |
| 2) constant mild-moderate pain | 1 (5%) | 17 (5%) | |
| 3) usually pain free; episodes of severe pain | 3 (14%) | 127 (37%) | |
| 4) constant mild-moderate pain+ episodes of severe pain | 8 (36%) | 86 (25%) | |
| 5) constant severe pain | 5 (23%) | 19 (6%) | 5.05 (1.30, 16.27)a |
| Constant Pain score | (n=22) | (n=327) | |
| Median (IQR) | 0 (0–54) | 0 (0–0) | |
| Range | 0–100 | 0–100 | |
| With any level of constant pain | 10 (45%) | 71 (22%) | 3.00 (1.25, 7.24)b |
| Episodic Pain score | (n=21) | (n=314) | |
| Median (IQR) | 56 (38–88) | 61 (3–80) | −5 (−28, 18)c |
| Range | 0–100 | 0–100 | |
| With any level of episodic pain | 17 (81%) | 237 (75%) | |
| Number of ER visits, lifelong (average*/yr) | (n=12) | (n=191) | |
| Median (IQR) | 2.2 (0.9–4.6) | 1.2 (0.3–2.4) | 1.0 (−1.0, 3.0)c |
| Range | 0–13.9 | 0–14.9 | |
| Number of ER visits, past year | (n=19) | (n=326) | |
| Median (IQR) | 2 (1–3) | 2 (0–3) | 0 (−0.7, 0.7)c |
| Range | 0–4 | 0–30 | |
| Number of hospitalizations, lifelong (average*/yr) | (n=13) | (n=193) | |
| Median (IQR) | 2.0 (0.9–4.1) | 1.0 (0.4–2.0) | 1.0 (−1.0, 2.9)c |
| Range | 0–13.9 | 0–14.9 | |
| Number of hospitalizations, past year | (n=19) | (n=330) | |
| Median (IQR) | 2 (1–2) | 1 (0–2) | 1.0 (0.3, 1.7)c |
| Range | 0–4 | 0–23 | |
| Days missed school past month | (n=20) | (n=275) | |
| Median (IQR) | 2 (0–5.5) | 1 (0–6) | 1.0 (−1.8, 3.8)c |
| Range | 0–22 | 0–45 | |
| Abdominal pain related to pancreatitis | 17/21 (81%) | 290/361 (80%) | 1.04 (0.33, 4.38)d |
| Exocrine insufficiency | 8/22 (36%) | 59/349 (17%) | 2.81 (0.97, 7.54)d |
| Number of attacks (attacks/yr) –parent response | (n=20) | (n=257) | |
| Median (IQR) | 1.9 (0.9–3.5) | 1.5 (0.8–2.8) | 0.4 (−0.8, 1.6)c |
| Range | 0–13.9 | 0–27.3 |
average lifelong events per year was computed for those that had condition for >1yr
Value is expressed as OR (95% CI) for constant severe pain
Value is expressed as OR (95% CI) for any pain
Value is expressed as median difference (95% CI) for Diabetes minus No Diabetes
Value is expressed as OR (95% CI)
Discussion
DM is a well-described consequence of pancreatitis in adults, but features of and risk factors for DM in children with pancreatitis have been lacking in the medical literature to date. The INSPPIRE registry represents a unique opportunity to describe the burden of DM in children with pancreatitis. In this large, multi-center, international cohort of children with CP and ARP, 6% of children studied had been diagnosed with DM. Children with either CP or ARP were at risk for DM, with about 40% of children in the registry with a diagnosis of DM having a diagnosis of CP and the remainder, about 60%, having ARP. Other than an increased prevalence of pancreatic atrophy in the DM group, pancreatitis risk factors, imaging features, and treatment history were similar between the DM and non-DM groups. Five children with DM also had beta cell antibodies, and DM was more common in those children who also had hypertriglyceridemia and tended to be more common in those with other autoimmune disorders, suggesting classic risk factors for metabolic syndrome or type 1 DM, respectively, in addition to the underlying pancreatic disease, may play an important role in mediating risk for DM in children with ARP or CP.
In the United States, the crude prevalence of DM in children is estimated at 1.82 cases per 1,000 youth (13). In contrast, we observed 24 children (6%, or equivalent to 60 out of 1,000 youth) diagnosed with DM. These findings clearly support the hypothesis that, as in adults, children with CP or ARP have an increased risk for DM, and is similar to that reported in smaller series of children with CP elsewhere (14, 15). It is important for gastroenterologists, endocrinologists, primary care physicians, and other providers who encounter children with CP or ARP to be aware of this increased risk and screen these children appropriately for DM. Annual screening for DM may include fasting glucose and hemoglobin A1c levels, with or without oral glucose tolerance testing (16). Notably, 46% of children had DM preceding the diagnosis of CP and ARP. It is possible that these children had subclinical pancreatitis that was undiagnosed at an earlier stage. Alternatively, one could postulate that there is a shared pathology underlying both DM and pancreatitis, or that the DM is actually increasing the risk for pancreatitis later.
This study provides novel exploratory data on potential risk factors for DM in children with pancreatitis, and sets the stage for future studies on mechanisms underlying DM in pediatric pancreatitis. While we have previously reported on pancreatitis features in this cohort of patients, and other previous series have reported the occurrence of DM in CP and beta cell dysfunction (without DM) in ARP, this is the first report to consider potential risk factors for DM in children with pancreatitis (14, 15, 17–19). Other studies to date on risks for DM in the setting of CP have focused nearly entirely in adult populations—adults may have different causes of pancreatitis (environmental vs genetic) and different co-morbidities and older age that impact DM risk.
Similar to findings in adults with CP and DM, children with DM in our series had features suggestive of more severe fibrosis, including increased odds of pancreatic atrophy and suggestion of increased risk of pancreatic exocrine insufficiency. One of the primary mechanisms leading to development of pancreatogenic DM is bystander destruction of the islets (and beta cells) as pancreatic parenchyma is replaced by fibrosis in CP. However, we did not find an association of other pancreatitis disease features classically associated with pancreatogenic DM in adults including pancreatic calcifications or prior surgical intervention with DM in our cohort, although these features are also less common in children than adults (8, 10, 20). It is possible that children with these features have a higher risk of progression to DM as they enter adulthood.
The only pancreatic morphologic finding associated with DM in our cohort was pancreatic atrophy. Pancreatic atrophy has been well-described in insulin-dependent DM, particularly type 1 DM (21–25). Thus, while the atrophy in our DM cases could be reflecting more severe fibrosis and islet loss related to more advanced pancreatitis, conversely atrophy could also be a consequence of the underlying DM. Similarly, exocrine pancreatic insufficiency, which was two-fold higher in the DM group (36% vs 17%), may be a consequence of CP, or associated with insulin-dependent DM (26, 27).
A particularly novel observation in this study, which has not been previously reported, is the presence of beta cell antibodies in children with CP and DM and suggestion for underlying autoimmunity. Type 1 DM is an autoimmune disorder against the insulin-producing beta cells of the pancreas (28, 29). In addition to a trend towards increased odds for other autoimmune disorders, 5 of 13 tested were positive for beta cell autoimmunity (38%). The latter is particularly notable as these findings suggest that some of the children with DM in this cohort may have beta cell autoimmunity, more similar to a Type 1 DM phenotype, rather than classic Type 3c DM. In the setting of genetic predisposition for type 1 DM, repeated episodes of inflammation may trigger onset of beta cell autoimmunity. Mechanistically, inflammatory cytokines and reactive oxygen species in pancreatic tissue may trigger post-translational protein modifications of beta cell antigens that alter the immunogenicity of these antigens, thereby increasing the risk for loss of self-tolerance (30). To this point, investigators in Europe have recently identified an increased prevalence of beta cell autoimmunity in patients with cystic fibrosis (CF) who develop CF related diabetes (CFRD), a different form of pancreatogenic diabetes. Onset of CFRD occurred at a younger age in patients with beta cell autoantibodies (31). Glutamic acid decarboxylase antibodies have also been reported in a subset of patients with CP but without DM during evaluation for TPIAT (32). Nearly one-fourth of children with DM had hypertriglyceridemia, which is one component of metabolic syndrome and associated with insulin resistance (33, 34). In the setting of metabolic syndrome, loss of functional beta cell mass could increase the risk for progression to DM in individuals with insulin resistance as they are no longer able to compensate for insulin resistance by increasing insulin secretion. In our cohort, overweight and obesity by BMI percentile was not different in children with or without DM, in contrast to recent data in adults in the North American Pancreatitis Study cohort, in which odds for DM in those with CP was significantly increased in the setting of obesity (8). While the greater prevalence of hypertriglyceridemia could be a manifestation of metabolic syndrome in the DM group, an alternative explanation is that DM, particularly if poorly controlled, increases the risk for hyperlipidemia (35). We lack data in our cohort that show whether hypertriglyceridemia preceded the onset of DM or vice versa.
We did not find an association of DM with PRSS1 gene mutations, despite a high lifetime risk of DM in hereditary pancreatitis, but prior case series suggest a much older median age of onset (37–51 years). (6, 7). Thus, presumably our children with PRSS1 mutations in this series remain at high lifetime risk for development of DM, but onset is more often in adulthood.
While this represents the largest and only multi-center international cohort to describe DM in children with ARP and CP, our findings are based on a relatively small number of children with DM and intended to generate hypothesis for future studies. In this retrospective registry, we considered the potential associations of many patient and disease variables with development of DM. As noted in the methodology, because of the multiple variables explored, we intentionally did not present p-values in our statistical approach for DM risk factors. The findings in this preliminary report should be considered exploratory and require confirmation in future studies. However, we suggest these putative risk factors, including beta cell autoimmunity and hypertriglyceridemia, should be considered in future biorepository or clinical studies.
Longitudinal studies will be needed to define who is at risk for progression to DM among children with ARP and CP. In addition, our analyses rely upon physician report and patient report from standard clinical care. Hypertriglyceridemia was defined by triglyceride levels above the upper limit of normal for the reference lab, so while this is a metabolic abnormality, mild elevations can increase pancreatitis risk (36). In the ARP cohort, it is possible that some patients have progressed to CP not yet evident on imaging, and we lack data on severity of AP; necrotic episodes of AP, although rare in children, may convey a higher risk of DM due to loss of pancreatic parenchyma (37). Likewise, we do not have data on whether HbA1c and fasting glucose were measured in the non-DM group and cannot entirely exclude mild undiagnosed DM in the non-DM controls, introducing risk for selection bias. Not all children with DM were tested for beta cell autoimmunity. Because of the clinical nature of the testing, in those who were tested we lack details on which antibodies and how many antibodies were positive.
In conclusion, we report a higher prevalence of DM than expected in the general population in youth with ARP or CP, with 6% of children with ARP or CP also diagnosed with DM. Children with DM more often have pancreatic atrophy, but also show evidence of beta cell autoimmunity and hypertriglyceridemia, suggesting that risk factors for Type 1 and Type 2 DM, respectively, may play a role in mediating DM development in this cohort of children with pancreatitis and warrant further investigation.
WHAT IS KNOWN
Adults with chronic pancreatitis (CP) and acute recurrent pancreatitis (ARP) are at high risk for developing diabetes mellitus (DM).
Little is known regarding risk of and risk factors for DM in children with CP and ARP.
WHAT IS NEW
The INSPPIRE registry is the first multi-center, international study to explore diabetes risk in a geographically diverse cohort of children with ARP and CP. We observed a 6% prevalence of DM, about 30-fold higher than expected in the general pediatric population.
DM may be common in children with other features of autoimmunity or with hyperlipidemia, suggesting risk features of type 1 or type 2 diabetes warrant future investigation in this population.
Funding Source:
Research reported in this publication was supported by National Cancer Institute and National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health under award numbers R21 DK096327 (Uc), U01 DK108334 (Uc).
Conflicts of Interest: Dr. Mark Lowe is on the Board of Directors of the National Pancreas Foundation; receives royalties from Millipore Inc and UpToDate. Drs. Tanja Gonska and Michael Wilschanski received a research grant from Vertex Pharmaceuticals. Dr. Sohail Husain owns equity in PrevCon. Dr. John Pohl is on the speaker’s bureau for Medical Education Resources, Inc.; Dr. Melena Bellin has served as a consultant for AbbVie Inc, NovoNordisk, and ARIEL Precision Medicine, and receives research funding from ViaCyte. Dr. Aliye 4 Uc is a member of American Board of Pediatrics, Subboard of Pediatric Gastroenterology. The other authors declare no conflicts of interest.
References
- 1.Hart PA, Bellin MD, Andersen DK, et al. Type 3c (pancreatogenic) diabetes mellitus secondary to chronic pancreatitis and pancreatic cancer. Lancet Gastroenterol Hepatol. 2016;1(3):226–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Uc A, Andersen DK, Bellin MD, et al. Chronic Pancreatitis in the 21st Century - Research Challenges and Opportunities: Summary of a National Institute of Diabetes and Digestive and Kidney Diseases Workshop. Pancreas. 2016;45(10):1365–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Das SL, Singh PP, Phillips AR, et al. Newly diagnosed diabetes mellitus after acute pancreatitis: a systematic review and meta-analysis. Gut. 2014;63(5):818–31. [DOI] [PubMed] [Google Scholar]
- 4.Raman VS, Loar RW, Renukuntla VS, et al. Hyperglycemia and diabetes mellitus in children with pancreatitis. J Pediatr. 2011;158(4):6126–e1. [DOI] [PubMed] [Google Scholar]
- 5.Kumar S, Ooi CY, Werlin S, et al. Risk Factors Associated With Pediatric Acute Recurrent and Chronic Pancreatitis: Lessons From INSPPIRE. JAMA pediatrics. 2016;170(6):562–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rebours V, Boutron-Ruault MC, Schnee M, et al. The natural history of hereditary pancreatitis: a national series. Gut. 2009;58(1):97–103. [DOI] [PubMed] [Google Scholar]
- 7.Howes N, Lerch MM, Greenhalf W, et al. Clinical and genetic characteristics of hereditary pancreatitis in Europe. Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association. 2004;2(3):252–61. [DOI] [PubMed] [Google Scholar]
- 8.Bellin MD, Whitcomb DC, Abberbock J, et al. Patient and Disease Characteristics Associated With the Presence of Diabetes Mellitus in Adults With Chronic Pancreatitis in the United States. Am J Gastroenterol. 2017;112(9):1457–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Wakasugi H, Funakoshi A, Iguchi H Clinical assessment of pancreatic diabetes caused by chronic pancreatitis. Journal of gastroenterology. 1998;33(2):254–9. [DOI] [PubMed] [Google Scholar]
- 10.Malka D, Hammel P, Sauvanet A, et al. Risk factors for diabetes mellitus in chronic pancreatitis. Gastroenterology. 2000;119(5):1324–32. [DOI] [PubMed] [Google Scholar]
- 11.Morinville VD, Lowe ME, Ahuja M, et al. Design and implementation of INSPPIRE. J Pediatr Gastroenterol Nutr. 2014;59(3):360–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2018. Diabetes Care. 2018;41(Suppl 1):S13–s27. [DOI] [PubMed] [Google Scholar]
- 13.Liese AD, D’Agostino RB Jr., Hamman RF, et al. The burden of diabetes mellitus among US youth: prevalence estimates from the SEARCH for Diabetes in Youth Study. Pediatrics. 2006;118(4):1510–8. [DOI] [PubMed] [Google Scholar]
- 14.Poddar U, Yachha SK, Borkar V, et al. Clinical profile and treatment outcome of chronic pancreatitis in children: a long-term follow-up study of 156 cases. Scand J Gastroenterol. 2017;52(6–7):773–8. [DOI] [PubMed] [Google Scholar]
- 15.Chowdhury SD, Chacko A, Ramakrishna BS, et al. Clinical profile and outcome of chronic pancreatitis in children. Indian pediatrics. 2013;50(11):1016–9. [DOI] [PubMed] [Google Scholar]
- 16.Rickels MR, Bellin M, Toledo FG, et al. Detection, evaluation and treatment of diabetes mellitus in chronic pancreatitis: recommendations from PancreasFest 2012. Pancreatology. 2013;13(4):336–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Puttaiah Kadyada S, Thapa BR, Kaushal K, et al. Incomplete functional and morphological recovery after acute and acute recurrent pancreatitis in children. Journal of gastroenterology and hepatology. 2018. [DOI] [PubMed] [Google Scholar]
- 18.Uc A, Zimmerman MB, Wilschanski M, et al. Impact of Obesity on Pediatric Acute Recurrent and Chronic Pancreatitis. Pancreas. 2018;47(8):967–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lin TK, Abu-El-Haija M, Nathan JD, et al. Pancreas Divisum in Pediatric Acute Recurrent and Chronic Pancreatitis: Report From INSPPIRE. Journal of clinical gastroenterology. 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hardt PD, Bellin MD, Andersen DK, et al. Type 3c (pancreatogenic) diabetes mellitus secondary to chronic pancreatitis and pancreatic cancer. Lancet Gastroenterol Hepatol. 2016;in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Regnell SE, Peterson P, Trinh L, et al. Pancreas volume and fat fraction in children with Type 1 diabetes. Diabetic medicine : a journal of the British Diabetic Association. 2016;33(10):1374–9. [DOI] [PubMed] [Google Scholar]
- 22.Campbell-Thompson ML, Kaddis JS, Wasserfall C, et al. The influence of type 1 diabetes on pancreatic weight. Diabetologia. 2016;59(1):217–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Li X, Campbell-Thompson M, Wasserfall CH, et al. Serum Trypsinogen Levels in Type 1 Diabetes. Diabetes Care. 2017;40(4):577–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Goda K, Sasaki E, Nagata K, et al. Pancreatic volume in type 1 and type 2 diabetes mellitus. Acta Diabetol. 2001;38(3):145–9. [DOI] [PubMed] [Google Scholar]
- 25.Williams AJ, Thrower SL, Sequeiros IM, et al. Pancreatic volume is reduced in adult patients with recently diagnosed type 1 diabetes. J Clin Endocrinol Metab. 2012;97(11):E2109–13. [DOI] [PubMed] [Google Scholar]
- 26.Philippe MF, Benabadji S, Barbot-Trystram L, et al. Pancreatic volume and endocrine and exocrine functions in patients with diabetes. Pancreas. 2011;40(3):359–63. [DOI] [PubMed] [Google Scholar]
- 27.Mohapatra S, Majumder S, Smyrk TC, et al. Diabetes Mellitus Is Associated With an Exocrine Pancreatopathy: Conclusions From a Review of Literature. Pancreas. 2016;45(8):1104–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Atkinson MA, Eisenbarth GS, Michels AW Type 1 diabetes. Lancet. 2014;383(9911):69–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Greenbaum C, Lord S, VanBuecken D Emerging Concepts on Disease-Modifying Therapies in Type 1 Diabetes. Curr Diab Rep. 2017;17(11):119. [DOI] [PubMed] [Google Scholar]
- 30.James EA, Pietropaolo M, Mamula MJ Immune Recognition of beta-Cells: Neoepitopes as Key Players in the Loss of Tolerance. Diabetes. 2018;67(6):1035–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Konrad K, Kapellen T, Lilienthal E, et al. Does beta-Cell Autoimmunity Play a Role in Cystic Fibrosis-Related Diabetes? Analysis Based on the German/Austrian Diabetes Patienten Verlaufsdokumentation Registry. Diabetes Care. 2016;39(8):1338–44. [DOI] [PubMed] [Google Scholar]
- 32.Kizilgul M, Wilhelm J, Dunn TD, et al. The prognostic significance of glutamic acid decarboxylase antibodies in patients with chronic pancreatitis undergoing total pancreatectomy with islet autotransplantation. Diabetes & Metabolism. 2018. [in press]. [DOI] [PubMed] [Google Scholar]
- 33.Titmuss AT, Srinivasan S Metabolic syndrome in children and adolescents: Old concepts in a young population. Journal of paediatrics and child health. 2016;52(10):928–34. [DOI] [PubMed] [Google Scholar]
- 34.Viner R, White B, Christie D Type 2 diabetes in adolescents: a severe phenotype posing major clinical challenges and public health burden. Lancet. 2017;389(10085):2252–60. [DOI] [PubMed] [Google Scholar]
- 35.Dunn FL Management of hyperlipidemia in diabetes mellitus. Endocrinol Metab Clin North Am. 1992;21(2):395–414. [PubMed] [Google Scholar]
- 36.Pedersen SB, Langsted A, Nordestgaard BG Nonfasting Mild-to-Moderate Hypertriglyceridemia and Risk of Acute Pancreatitis. JAMA internal medicine. 2016;176(12):1834–42. [DOI] [PubMed] [Google Scholar]
- 37.Vipperla K, Papachristou GI, Slivka A, et al. Risk of New-Onset Diabetes Is Determined by Severity of Acute Pancreatitis. Pancreas. 2016;45(1):e14–5. [DOI] [PMC free article] [PubMed] [Google Scholar]