Fat soluble vitamin deficiency is common in children with chronic pancreatitis undergoing TPIAT

Abstract

The prevalence of fat-soluble vitamin (FSV) deficiency in children undergoing total pancreatectomy with islet autotransplantation (TPIAT) for chronic pancreatitis (CP) is unknown. We quantified FSV deficiency in 100 children (age≤18) undergoing TPIAT. FSV levels (vitamins A, E, D) and clinical history were abstracted from medical records. Vitamin A was low in 4% before and 7% at 1 year after TPIAT, vitamin E in 17% and 18%, and vitamin D in 22% and 24% respectively, regardless of pancreatic enzyme or vitamin supplement dosing. Longer duration of CP was associated with pre-TPIAT vitamin D insufficiency (P=0.0002). This remained significant in a multivariate regression model (adjusted P=0.01). On multivariate analysis there were no significant predictors of low FSV levels post-TPIAT. Fat-soluble vitamin deficiencies are common among children undergoing TPIAT and patients who have had longer disease duration may be at increased risk. All children should be monitored for FSV deficiency after TPIAT.

Background

For children with chronic pancreatitis or recurrent acute pancreatitis associated with lifestyle-limiting pain refractory to medical and endoscopic management, total pancreatectomy with islet autotransplantation (TPIAT) has become increasingly used in the past decade for first-line surgical management. While TPIAT can provide pain relief and improved quality of life ^1–5, the procedure results in life-long exocrine pancreatic insufficiency (EPI) requiring pancreatic enzyme replacement therapy (PERT).

Even without undergoing surgical interventions, children with chronic pancreatitis are at high risk of EPI ⁶. If inadequately treated, EPI can lead to malabsorption of fats and fat-soluble vitamins ^7–10. Children with EPI may exhibit symptoms of steatorrhea, loose stools, flatulence, and abdominal pain ^8,11,12. Complications of EPI may include weight loss, poor growth, and deficiencies of the fat-soluble vitamins A, D, E, and K ^8,10,12. Guidelines have been established for PERT and fat-soluble vitamin supplementation and monitoring in children with EPI from chronic or recurrent acute pancreatitis and cystic fibrosis ^8,13–15. However, there are no current publications addressing the prevalence or risk factors for FSV deficiency in children undergoing TPIAT.

Our objectives for the current study were, therefore, to describe the prevalence and factors associated with fat-soluble vitamin deficiencies (vitamins A, D, and E) in a group of children with chronic or recurrent acute pancreatitis undergoing TPIAT. These data are important for informing nutritional surveillance and management of these children before and after TPIAT.

Methods

Study design

We reviewed the records of pediatric (age ≤18) patients who underwent TPIAT at the University of Minnesota from 2010 to 2019 with FSV data available before and at 1 year (range 6–24 months, if 1-year not available) after TPIAT. All patients were participants in our single-center study of TPIAT outcomes, with informed consent or parental consent provided and patient assent obtained as age-appropriate. The University of Minnesota Institutional Review Board approved the study.

Categorization of fat-soluble vitamins

Fat-soluble vitamin levels were abstracted from the medical records. Vitamins A and E were categorized as either sufficient or low based on patient age and laboratory-specific reference ranges (sufficient vitamin A >20 μg/dL for ages 2–12 and >26 μg/dL for ages 13–18; sufficient vitamin E >5.5 μg/mL), and vitamin D was categorized as sufficient (>29 ng/mL), insufficient (20–29 ng/mL), or deficient (<20 ng/mL), according to guidelines from the Endocrine Society ¹⁶. Season of vitamin D measurement was also recorded (summer, fall, winter, spring).

Post-operative vitamin supplement use (from the same time as serum FSV levels) was abstracted from the medical record. Patients were recorded as taking a pancreatic formulation of FSV supplement, an over-the-counter (OTC) supplement (multi-vitamin or vitamin D), or none at all. Dose and type (liquid, chew, tablet) were recorded, and categorized based on the patient’s age and the recommended dose for that age. For example, if a vitamin had recommended dosing of 2 tablets per day for a child over the age of 10, a patient 15-years-old taking 2 per day would be coded as a dose of 1. A 15-year-old taking 1 per day of the same supplement would be coded as a dose of 0.5. Total vitamin D dose incorporated the patient’s multivitamin and any separate vitamin D supplement taken. For intermittent high-dose vitamin D (ie weekly or monthly), total dose was divided over the number of days between doses to calculate dose per day.

Categorization of exocrine pancreatic insufficiency

Fecal elastase was used, where available, to determine presence of preoperative EPI, with EPI defined as <200 μg/g of stool. Postoperatively, all patients were on pancreatic enzyme replacement therapy (PERT; due to total pancreatectomy). The lipase units per meal of PERT and patient weight at the time of post-TPIAT FSV levels were used to calculate the dose in lipase units per kilogram of body weight per meal (units/kg/meal).

Comparisons with normative data

The National Health and Nutrition Examination Survey (NHANES) has been collecting data on dietary habits of Americans, including children, since the 1970’s. As part of the effort to monitor nutrition trends and inform public health policies, it collects longitudinal and cross-sectional data on reported nutrient consumption and (less frequently) biochemical measures of nutrient intake ¹⁷. This publicly available data was used to estimate normative prevalence of FSV deficiencies among children in the United States for descriptive comparison with children with CP.

Characterizing predictors of FSV deficiency in children undergoing TPIAT

To determine the pre-operative factors associated with FSV deficiencies (as previously defined), we performed multiple logistic regression from the following candidate predictors: disease duration, disease etiology (obstructive, hereditary, or idiopathic), BMI percentile for age, presence of EPI, prior pancreas surgery, and (for vitamin D analysis) season of vitamin D measurement. To determine the post-operative factors associated with FSV deficiencies, we performed multiple logistic regression from the following candidate predictors: BMI percentile, PERT dose (u/kg/meal), vitamin type (OTC, pancreatic, or none), and vitamin dose (as previously defined). For vitamin D, the daily dose of vitamin D was included in the model. The logistic regression models for both pre- and post-operative vitamin D included only two categories, normal and low (insufficient and deficient combined), due to low numbers in the deficient group (n=3).

Surgical technique of TPIAT in children

Our institutional criteria for TPIAT in children have been published previously ⁵. Our group and others have also previously described the surgical techniques of pediatric TPIAT ⁵, and children in this cohort were treated similarly. All surgical procedures were performed by the same primary surgeon (SC). Briefly, the pancreas and duodenum are mobilized with special attention paid to maintaining blood supply to the pancreas for as long as possible, to minimize islet ischemia. Once the pancreas has been removed, it is transported to the islet lab where beta islets are isolated as previously described ⁴ and subsequently infused into the portal vein. The GI tract is then reconstructed with Roux-en-Y hepaticojejunostomy and duodenojejunostomy.

Post-TPIAT screening for and management of FSV deficiency

At our institution, we recommend screening FSV before TPIAT and at 3, 6, and 12 months, then yearly, after TPIAT. Lifelong multivitamin supplementation is recommended for all patients, preferably with a pancreatic-specific formulation that delivers FSV in a water soluble form for better absorption. Pancreatic enzyme replacement therapy is based on patient weight and symptoms, with a typical dose of 1,000 to 2,000 lipase units/kg/meal.

Statistical analyses

All statistical analyses were performed in SAS (SAS Institute Inc., Cary, NC). Categorical variables were compared with Chi squared or Fisher’s exact tests. Logistic and linear regression were used to model predictors of categorical and continuous outcomes, respectively. A P-value of 0.05 was considered significant.

Results

Patient demographics

A total of 100 patients were included in the study. Mean age was 12 ±4.2 years, and 60% of patients were female. Most (86%) had hereditary pancreatitis, mean disease duration was 5.6 (standard deviation 3.9) years, and 60% had baseline EPI evidenced by a fecal elastase level <200 μg/g of stool.

Prevalence and predictors of FSV deficiency before and after TPIAT

FSV deficiencies were common before and after TPIAT (Table 1, contrasted with NHANES data).

Table 1.

Prevalence (percent) of fat-soluble vitamin deficiency pre- versus post-TPIAT (n = number with data available), with P-value from Fisher’s exact test comparing pre- to post-. Vitamin D categories of insufficient and deficient were combined for testing. NHANES = National Health and Nutrition Examination Survey in the United States, data from 2005–2006 biochemical indicators of micronutrient deficiency report.

FSV status	Pre-TPIAT % (n)	Post-TPIAT % (n)	P-value for difference	Prevalence from NHANES
Vitamin A (low)	4% (99)	7% (94)	0.09	1%
Vitamin E (low)	17% (99)	18% (98)	0.60	2%
Vitamin D (insufficient)	22% (99)	24% (98)	0.87	9–24%
Vitamin D (deficient)	5% (99)	6% (98)

Pancreatitis disease duration was associated with low (insufficient or deficient) vitamin D levels. On average, each year of pancreatitis pain conferred 26% increased odds of having vitamin D insufficiency or deficiency (odds ratio 1.26, 95% confidence interval 1.05–1.50, adjusted p=0.01, Figure 1a). The threshold for low (≤29 ng/mL) vitamin D level by simple linear regression was seven years of pancreatitis pain (average reduction in vitamin D level 0.98 ng/mL for each year of pain, standard error = 0.25, p=0.0002, Figure 1b).

Figure 1.

(a) Years of pancreatitis pain by vitamin D category. Lines within boxes represent the mean, and error bars represent the range. Association adjusted for BMI percentile, prior pancreas surgery, presence of EPI defined as fecal elastase <200, pancreatitis etiology (obstructive, hereditary, or idiopathic), and season of vitamin D measurement. (b.) Simple linear association between years of pancreatitis pain and preoperative vitamin D level. Dashed line at vitamin D = 30 ng/mL (normal).

Other variables evaluated including disease duration, BMI percentile, presence of EPI, disease etiology, prior pancreas surgery, and season of vitamin D measurement were not associated with pre-TPIAT Vitamin A, E, or D levels.

In logistic regression analyses, post-TPIAT vitamin A, E, or D categories were not associated with any of the following: BMI percentile, PERT dose (lipase units/kg/meal), vitamin dose (including daily dose of vitamin D), or vitamin type. Of note, most patients were taking a pancreatic formulation multivitamin; only three patients were taking an OTC multivitamin, and four patients were taking no multivitamin.

Discussion

In our group of pediatric patients undergoing TPIAT, fat-soluble vitamin deficiency was common compared with published prevalence from NHANES. Low vitamin A levels were present in approximately one out of every 14 TPIAT patients, low vitamin E in almost one out of five, and low vitamin D in one out of every four after surgery, despite universal administration of PERT post-TPIAT and pancreatic-specific vitamin supplementation in most. Vitamin D insufficiency/deficiency appeared to be associated with duration of pancreatitis, perhaps due to more compromised pancreatic function or other nutritional compromise associated with pancreatitis. Our data support that children undergoing TPIAT should be screened and treated for FSV deficiency before and after surgery. Our current practices include putting all pre-TPIAT patients with EPI on FSV supplementation. Dosing follows recommendations by the Cystic Fibrosis Foundation ¹⁴ which are based on patient age. We recommend screening for FSV deficiency at the time of TPIAT evaluation, and if levels are low additional supplementation is added. After TPIAT, our patients are screened for FSV deficiencies regularly over the first year, and then annually thereafter. Supplementation dosing recommendations are the same as pre-TPIAT.

Guidelines from the INSPPIRE consortium and those from the North American Society for Pediatric Gastroenterology, Hepatology & Nutrition (NASPGHAN) recommend routine monitoring of fat-soluble vitamins in children with pancreatitis ^6,7. Our results support the importance of continuing this monitoring for children after TPIAT (even though they no longer have active pancreatitis). Pancreatectomy results in complete exocrine insufficiency and increases the risk of fat and fat-soluble vitamin malabsorption. Even with appropriate pancreatic enzyme replacement therapy dose and FSV supplementation, fat-soluble vitamin insufficiencies were common in our patients.

Limitations of our study include the single-center, retrospective design. Our quaternary referral center closely manages our chronic pancreatitis patients with PERT, so despite lack of association between enzyme dose and vitamin levels in our patients, presumably risk of vitamin deficiencies would be higher if patients were not taking enzymes as directed or were significantly under-dosed. Future directions for this work include prospective collection of nutritional data in chronic pancreatitis and TPIAT patients via the Prospective Observational Study of TPIAT (POST; R01DK109124) multi-institutional study, which will provide longitudinal data from all age groups including children. In a preliminary abstract from POST, vitamin D deficiency was more common in adults than children ¹⁸. This multicenter study will evaluate nutritional outcomes before and after TPIAT in a larger cohort.

This is the first report in the literature characterizing fat-soluble vitamin levels in pediatric patients with chronic pancreatitis before and after TPIAT. Pediatric patients undergoing TPIAT are at high risk for fat-soluble vitamin deficiencies, and vitamin levels should be closely monitored, even while patients are maintained on appropriate pancreatic enzyme replacement therapy and multivitamin administration.