Original Paper
Received: December 3, 2001
Pancreatology 2003;3:36–40
DOI: 10.1159/000069144
An Enteral Therapy Containing Medium-Chain
Triglycerides and Hydrolyzed Peptides Reduces
Postprandial Pain Associated with Chronic
Pancreatitis
Julie C. Shea a Michele D. Bishop b Eliza M. Parker a Andres Gelrud a
Steven D. Freedman a
a Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Mass., and
b Division of Gastroenterology and Hepatology, Mayo Clinic, Jacksonville, Fla., USA
Key Words
Chronic pancreatitis W Enteral therapy, chronic
pancreatitis W Cholecystokinin W Pain, chronic pancreatitis
Abstract
Background/Aim: Pain in patients with chronic pancre-
atitis is difficult to manage. We examined if an enteral
formulation containing medium-chain triglycerides
(MCT) and hydrolyzed peptides would (1) minimally
stimulate the exocrine pancreas by blunting cholecysto-
kinin release and (2) decrease pain in patients with
chronic pancreatitis. Methods: In the first part of the
study, on separate days, 6 healthy controls consumed a
standard enteral formulation, an enteral formulation
containing MCT and hydrolyzed peptides, and a high-fat
meal. Baseline and postprandial plasma cholecystokinin
(CCK) concentrations were analyzed. Subsequently, 8
patients with chronic pancreatitis were enrolled and
instructed to complete a visual analog pain assessment
for a baseline period of 2 weeks followed by three cans
per day of the enteral formulation containing MCT and
hydrolyzed peptides for 10 weeks. Results: Mean CCK
levels for our control subjects were 0.46 B 0.29 pM at
baseline, 10.75 B 0.45 pM in response to the high-fat
© 2003 S. Karger AG, Basel and IAP
ABC 1424–3903/03/0031–0036$19.50/0
Fax + 41 61 306 12 34
E-Mail karger@karger.ch Accessible online at:
www.karger.com www.karger.com/pan
Accepted after revision: July 23, 2002
meal, and 7.9 B 1.25 pM in response to the standard
enteral formulation. Of note, CCK levels were 1.43 B
0.72 pM in response to the enteral supplement contain-
ing MCT and hydrolyzed peptides. In patients with
chronic pancreatitis, the average improvement in pain
scores from baseline to the conclusion of the study was
61.8% (p = 0.01). This corresponded to a clinical improve-
ment in 6 of the 8 patients. Conclusions: A complete
enteral supplement containing MCT and hydrolyzed pep-
tides minimally increases plasma CCK levels. This thera-
py may be effective in reducing postprandial pain associ-
ated with chronic pancreatitis.
Copyright © 2003 S. Karger AG, Basel and IAP
Introduction
Pain associated with chronic pancreatitis is typically
managed with analgesics including narcotics. During
acute flares, some patients may require hospitalization for
intravenous hydration, bowel rest, and parenteral narcot-
ics. The conventional approach has been to ‘rest the pan-
creas’ with the intent that minimal stimulation of the exo-
crine pancreas results in amelioration of pain. However,
this may result in nutritional deficiencies and in the
Steven D. Freedman, MD, PhD
Beth Israel Deaconess Medical Center
Dana 501, 330 Brookline Ave.
Boston, MA 02215 (USA)
Tel. +1 617 667 5576, Fax +1 617 667 0536, E-Mail sfreedma@caregroup.harvard.edu
induction of a catabolic state [1]. This is a significant
problem in the outpatient management of chronic pancre-
atitis. Total parenteral nutrition (TPN) is occasionally
used in the outpatient setting; however, this is both costly
and associated with complications. An alternative to TPN
is the use of oral enteral feedings. However, standard
enteral feedings contain long-chain triglycerides which
stimulate cholecystokinin (CCK) release and may worsen
the pain associated with chronic pancreatitis. The ideal
goal for outpatient management would be an oral enteral
formulation which provides adequate nutrition, yet mini-
mally stimulates secretion from the exocrine pancreas and
hence improves pain.
CCK release is stimulated by the presence of amino
acids, intact proteins, long-chain triglycerides, and fatty
acids in the duodenum [2, 3]. In contrast, medium-chain
triglycerides (MCT) result in minimal stimulation of
CCK [3]. McLaughlin et al. [4] found that administration
of fatty acid emulsions consisting of eleven or fewer car-
bons does not increase the plasma CCK concentration,
while those containing greater than twelve carbons do
increase CCK levels. The effect of hydrolyzed peptides is
currently unknown. Previous studies in both animals and
humans indicate that an elemental diet administered via
the jejunum results in minimal stimulation of the exo-
crine pancreas [5–8]. However, the oral administration of
an elemental diet has not been studied in patients with
chronic pancreatitis. The aim of this study was to deter-
mine (1) if an enteral formulation enriched in hydrolyzed
proteins and MCT would result in a minimal increase in
plasma CCK levels, and (2) if such an enteral supplement
might be used to provide nutrition to patients with
chronic pancreatitis without worsening their symptoms or
perhaps even improving postprandial pain.
Patients and Methods
Analysis of CCK Levels in Response to Various Oral Feeds
Institutional review board approval was obtained for research
involving human subjects. A time course was performed to deter-
mine the peak CCK level after the consumption of the enteral formu-
lation containing hydrolyzed peptides and MCT. One healthy con-
trol subject consumed one can of the enteral formulation containing
MCT and hydrolyzed proteins, and CCK levels were determined 30,
60, 90, 120, and 180 min after ingestion. This was repeated on a 2nd
day with the same subject.
Healthy control subjects were subsequently enrolled to assess the
effect of enteral supplements on CCK release. Subjects were fasted
overnight on 3 consecutive days. On each day, baseline blood was
drawn for measurement of basal CCK levels. On 3 separate days,
subjects were given a 30-gram high-fat meal consisting of a quarter
pound hamburger, a complete enteral formulation containing long-
Enteral Therapy in Chronic Pancreatitis
chain triglycerides and intact proteins (Ensure; Abbott Laboratories,
Abbott Park, Ill., USA), and an enteral formulation containing
hydrolyzed peptides (source of protein is whey based) and MCT
(Peptamen; Nestlé Clinical Nutrition, Deerfield, Ill., USA). The
study diet was consumed within 5 min, and phlebotomy was
repeated 60 min later for measurement of CCK levels.
CCK Bioassay
Total CCK plasma concentration was measured by bioassay
using the method of Liddle et al. [3]. In brief, CCK was extracted
from plasma with a Sep-Pak C18 cartridge (Waters, Milford, Mass.,
USA). A dose-response curve was established following incubation of
rat pancreatic acini with known concentrations of CCK-8 (Peninsula
Laboratories, Belmont, Calif., USA). The CCK concentration in the
sample extracts was then determined.
Assessment of Enteral Nutrition and Pain in Patients with
Chronic Pancreatitis
In order to determine if an enteral nutrition supplement rich in
MCT and hydrolyzed protein would result in a decrease in pain, sub-
jects with a confirmed diagnosis of chronic pancreatitis were re-
cruited. Patients were all pancreatic sufficient and had to experience
pain at least three times per week over 2 weeks prior to enrollment in
the study. Patients with alcohol as the etiology had to be abstinent for
at least 6 months. The diagnosis of chronic pancreatitis was made
by clinical features including at least one of the following criteria:
(1) radiographic finding of calcifications of the pancreas; (2) pancre-
atogram displaying ductal changes consistent with chronic pancreati-
tis; (3) histology consistent with chronic pancreatitis, and/or (4) an
abnormal secretin pancreatic function test with a measured peak
bicarbonate level ! 80 mEq/l over a 1-hour collection. Only patients
able to tolerate at least three cans per day of the enteral formulation
were enrolled.
Prior to beginning enteral therapy, patients were asked to com-
plete a daily pain assessment form for 2 weeks to assess baseline pain
scores. The pain assessment was conducted using a visual analog
scale with 0 corresponding to no pain and 10 to severe pain. After
completion of the baseline assessment, patients were instructed to
consume a minimum of three cans per day of the enteral formulation
containing MCT and hydrolyzed proteins (Peptamen). Each 250-ml
can contains 1 cal/cm3 (9.8 g fat/can with 68.9% consisting of MCT).
Patients were allowed to consume other foods, provided that they
consumed no more than an additional 20 g of fat per day. Daily pain
scores were recorded for 10 weeks while on this enteral formulation.
Patients were instructed to record pain scores based on what they
perceived to be related to their pancreatitis.
Statistical Analysis
In the first part of our study, basal CCK levels were calculated as
the mean of the three daily baseline levels. Basal and 60-min CCK
levels are summarized as mean B SD and compared using paired t
tests. In the subjects with chronic pancreatitis, a baseline pain score
was calculated as the mean score over the 2 weeks prior to beginning
the enteral supplement; weekly mean pain scores were calculated
thereafter. The percent change in pain scores between baseline and
week 10 was analyzed using the Wilcoxon signed-rank test. The
change in pain scores over time was assessed separately for each
patient using linear regression analysis; statistically significant
change was determined by comparing the slope with the value of zero
(with zero corresponding to no change).
Pancreatology 2003;3:36–40 37
 1

Fig. 1. Time course of peak plasma CCK

levels in response to an enteral therapy con- 2
taining MCT and hydrolyzed peptides. The
values are from 1 healthy subject on 2 sepa-
rate days. Values are expressed as mean B
SD.
Fig. 2. Mean plasma CCK concentrations
for healthy controls at baseline (basal) and in
response to a standard enteral supplement
(column A), a high-fat meal (column B), and
an enteral formula containing MCT and hy-
drolyzed peptides (column C). Values are
expressed as mean B SD. * p ! 0.01 com-
pared to basal; ** p ! 0.05 compared to bas-
al, p ! 0.05 compared to A, and p ! 0.0001
compared to B.

Table 1. Patient demographics with a mean of 7.9 B 1.25 pM (fig. 2, column A). In con-
trast, the mean CCK concentration following the liquid
Patient No. Age, years Etiology of pancreatitis
meal containing MCT and hydrolyzed proteins was 1.43
1 29 idiopathic
B 0.72 pM (fig. 2, column C) and was statistically differ-
2 48 alcohol induced ent from the other two feeds (p ! 0.0001 vs. high-fat solid
3 33 idiopathic meal and p = 0.03 vs. liquid meal containing full-length
4 54 idiopathic triglycerides and intact proteins).
5 67 alcohol induced
6 20 idiopathic
7 40 idiopathic
Assessment of Patients with Chronic Pancreatitis
8 43 idiopathic Based on the fact that the enteral formulation contain-
ing MCT and hydrolyzed proteins (henceforth referred to
as the enteral formulation) minimally stimulated CCK
release and, therefore, may minimally stimulate the exo-
Results crine pancreas, the second part of the study was con-
ducted to determine its effect on pain. A total of 17
Analysis of CCK patients were approached for enrollment. Nine were un-
In order to determine when CCK levels peak following able to meet the inclusion criteria which required con-
ingestion of the enteral formulation containing MCT and sumption of three cans of the enteral supplement per day.
hydrolyzed proteins, a time course was examined in 1 Refractory nausea associated with chronic pancreatitis
healthy control on 2 separate days (fig. 1). Peak CCK was the most common cause of being unable to meet the
release was between 60 and 90 min after ingestion with a inclusion criteria. Eight patients with postprandial pain
plasma level at 90 min of 2.3 B 0.14 pM. The time course associated with chronic pancreatitis (6 idiopathic, 2 alco-
of this peak in CCK levels is consistent with values hol induced) who were able to tolerate at least three cans
reported in previously published studies [9]. Therefore, per day were enrolled in this study (table 1). The mean age
for subsequent analysis of peak postprandial CCK levels, of the patients was 42 (range 20–67) years with 3 males
phlebotomy was performed 60 min after consumption of and 5 females. All patients had daily pain for at least the
a liquid meal. preceding 10 months despite a fat-restricted diet (!20 g/
Six healthy control subjects enrolled in the first part of day). The location of the pain for all subjects was epigas-
the study had mean basal CCK levels of 0.46 B 0.29 pM tric radiating to the back. None had a response to high-
(fig. 2, basal). This increased to 10.75 B 0.45 pM 1 h fol- dose pancreatic enzyme supplementation nor to 400 IU of
lowing ingestion of the high-fat solid meal (fig. 2, column vitamin E daily for at least 8 weeks. 2 of the 8 subjects had
B). The liquid meal containing full-length triglycerides failed percutaneous celiac blocks. All subjects had under-
and intact proteins produced a similar increase in CCK gone nutrition counseling.

38 Pancreatology 2003;3:36–40 Shea/Bishop/Parker/Gelrud/Freedman

Table 2. Response to an enteral therapy
containing MCT and hydrolyzed whey Patient Clinical Mean Mean pain Slope r2 p
protein No. response baseline score at determined
pain score week 10 by regression

1 yes 5.8 2.7 –0.357 0.945 !0.0001

2 yes 8 0 –1.9 0.810 0.0014
3 yes 4 0.5 –0.286 0.595 0.0150
4 yes 3 1 –0.376 0.708 0.0176
5 yes 8 1.75 –0.767 0.503 0.0145
6 yes 2 0.6 –0.154 0.231 0.1342
7 no 5 3.8 –0.151 0.251 0.1167
8 no 8.5 7.3 –0.034 0.053 0.4939

Mean pain scores for the 2-week baseline period and for week 10 are shown. The average
weekly change in pain scores for individual subjects over the course of the study from baseline
to week 10 is represented by the slope as determined by regression analysis. The r2 values of
the regression and the p values comparing the slope to zero are also shown.

The median improvement in pain scores for all pa-
tients from baseline to the conclusion of the study was
68.5% (p = 0.011). 6 of the 8 patients reported improved
pain control, corresponding to decreased narcotic use
over the course of the study. The changes over time of the
pain scores were further analyzed for each individual
patient using regression analysis (table 2). By regression
analysis, 5 of the 8 patients had a statistically significant
decrease in pain scores. 1 of the 3 patients (No. 6) who did
not have a statistically significant response did report an
improvement in pain scores that was corroborated by ces-
sation of narcotics. Another patient (No. 8) had no
improvement in her pain scores, but continued to use the
enteral formulation to maintain her weight.
Discussion
The primary goal of this study was to address one of
the most difficult challenges faced by clinicians treating
outpatients with chronic pancreatitis: pain management.
Although the mechanism by which pain occurs in patients
with chronic pancreatitis is not completely understood,
CCK has been postulated to play a role through hyper-
stimulation of exocrine pancreatic secretion as well as
increasing pancreatic ductal pressures [10]. Several stud-
ies have examined CCK levels in patients with chronic
pancreatitis, with some showing increased basal and post-
prandial levels [11–13], while others found no difference
[14–18]. Garces et al. [19] found that basal as well as
meal-stimulated plasma CCK levels were higher in
chronic pancreatitis patients with pain as compared with
those without pain. Based on the theory that increased
CCK levels following a meal lead to further pain, high
doses of orally administered pancreatic enzyme prepara-
tions have been given to patients with chronic pancreati-
tis. This approach resulted in an improvement in pain
scores, presumably due to negative feedback mechanisms
regulating pancreatic secretions [20–22].
We hypothesized that administration of an enteral for-
mulation that provides nutrition, while minimally in-
creasing CCK levels, may be beneficial in the treatment of
chronic pancreatitis. This would be analogous to the
blunting of CCK release by high doses of exogenous pan-
creatic enzymes yet providing optimum nutrition. Our
data indicate that a complete enteral supplement contain-
ing MCT and hydrolyzed proteins minimally increases
plasma CCK levels in healthy controls as compared with
ingestion of a high-fat meal or a standard enteral supple-
ment. Whether this is related to the enrichment in MCTs
or hydrolyzed peptides or both cannot be ascertained
from this study.
The blunted CCK response to this enteral therapy sug-
gested that it may result in less pain as compared with diets
containing long-chain triglycerides or intact proteins. To
determine if administration of this enteral therapy is asso-
ciated with amelioration of pain, 8 subjects who met the
inclusion criteria were enrolled. Postprandial pain was sig-
nificantly reduced in the majority of the patients enrolled
in this pilot study, although the number of study subjects
was low. All 6 patients in our study who experienced an
improvement in pain reduced or completely discontinued
narcotic use. Although there can be a placebo response seen
in this patient population, these patients had previously

Enteral Therapy in Chronic Pancreatitis Pancreatology 2003;3:36–40 39

failed conventional therapies for pain management, in-
cluding administration of high doses of pancreatic en-
zymes, antioxidants, a low-fat diet, and celiac plexus nerve
blocks. This, combined with the greater than 10-month
history of daily pain, would suggest that a placebo response
is unlikely and allows each subject to serve as its own con-
trol. A larger, randomized control trial is needed to con-
firm these results. A crossover study would be impractical
because of the potential long-lasting effects of nutrition as
well as increasing endogenous glutathione stores.
One patient enrolled in this study with alcohol-induced
chronic pancreatitis (No. 2) required narcotics for pain
management and had symptomatic distal common bile
duct and duodenal stenoses for which he was referred for
a Whipple procedure. He was pain free by the 4th week of
the study. The patient has continued using the enteral for-
mulation and remains pain free without narcotics after 2
years of follow-up and has not required further treatment
including surgical or endoscopic therapies.
In summary, our data indicate that an enteral formula
containing hydrolyzed peptides and is enriched in MCT
minimally increases plasma CCK levels. Whether this is
responsible for the improvement in pain scores in patients
with chronic pancreatitis cannot be ascertained with cer-
tainty. This enteral formulation provides increased
amounts of cysteine which may increase glutathione pro-
duction and improve antioxidant function. Antioxidants
have been reported to provide benefit in pain manage-
ment of pancreatitis in a case report [23], although all our
subjects had been treated with vitamin E without im-
provement. Based on the current study, a trial of this form
of enteral therapy containing MCT and hydrolyzed pro-
teins may be warranted for patients with chronic pancre-
atitis in whom the pain is principally postprandial with
minimal nausea.
Acknowledgements
We appreciate the nutritional counseling for the patients pro-
vided by Kathianne Sellers, RD, and the advice on analysis of statis-
tics from Meredith Regan, ScD. Support for this work came from a
grant from Nestlé Clinical Nutrition, Inc.

References

1 Rose P, Fraine E, Hunt LP, Acheson DW, Bra-
ganza JM: Dietary antioxidants and chronic
pancreatitis. Hum Nutr Clin Nutr 1986;40:
151–164.
2 Go VLW, Hoffman AF, Summerskill WHJ:
Pancreozymin bioassay in man based on pan-
creatic enzyme secretion: potency of specific
amino acids and other digestive products. J
Clin Invest 1970;49:1558–1564.
3 Liddle RA, Goldfine ID, William JA: Bioassay
of plasma cholecystokinin in rats: effects of
food, trypsin inhibitor, and alcohol. Gastroen-
terology 1984;87:542–549.
4 McLaughlin J, Grazia Luca M, Jones MN,
D’Amato M, Dockray GJ, Thompson DG: Fat-
ty acid chain length determines cholecystoki-
nin secretion and effect on human gastric mo-
tility. Gastroenterology 1999;116:46–53.
5 McArdle AH, Echave W, Brown RA, Thomp-
son AG: Effect of Elemental Diet on Pancreatic
Secretion. Am J Surg 1974;128:690–692.
6 Bodoky G, Harsanyi L, Pap A, Tihanyi T,
Flautner L: Effect of enteral nutrition on exo-
crine pancreatic function. Am J Surg 1991;161:
144–148.
7 Cassim MM, Allardyce DB: Pancreatic secre-
tion in response to jejunal feeding of elemental
diet. Ann Surg 1974;180:228–231.
8 Grant JP, Davey-McCrae J, Snyder PJ: Effect
of Enteral Nutrition on Human Pancreatic Se-
cretions. JPEN 1987;11:302–304.
9 Ling PR, Sheikh M, Boyce P, Keane-Ellison M,
Thibault A, Burke P, Freedman S, Bistrian BR:
Cholecystokinin (CCK) Secretion in Patients
with Severe Short Bowel Syndrome (SSBS).
Dig Dis Sci 2001;46:859–864.
10 Go VL, Dimagno EP, Gardner JD, Lebenthal
E, Reber HA, Scheele GA, editors: The Pan-
creas: Biology, Pathobiology, and Disease,
ed 2. New York, Raven Press, 1993.
11 Becker HD, Werner M, Schafmayer A: Release
of radioimmunologic cholecystokinin in hu-
man subjects. Am J Surg 1984;147:124–128.
12 Schafmayer A, Becker HD, Werner M, Folsch
UR, Creutzfeldt W: Plasma cholecystokinin
levels in patients with chronic pancreatitis. Di-
gestion 1985;32:136–139.
13 Gomez Cerezo J, Codoceo R, Fernandez Calle
P, Molina F, Tenias JM, Vazquez JJ: Basal and
postprandial cholecystokinin values in chronic
pancreatitis with and without abdominal pain.
Digestion 1991;48:134–140.
14 Funakoshi A, Nakano I, Sinozaki H, Ibayashi
H, Tateishi K, Hamaoka T: Low plasma chole-
cystokinin response after ingestion of a test
meal in patients with chronic pancreatitis. Am
J Gastroenterol 1985;80:937–940.
15 Bozkurt T, Alder G, Koop I, Koop H, Turner
W, Arnold R: Plasma CCK levels in patients
with pancreatic insufficiency. Dig Dis Sci
1988;33:276–281.
16 Nustede R, Kohler H, Folsch UR, Schafmayer
A: Plasma concentrations of neurotensin and
CCK in patients with chronic pancreatitis with
and without enzyme substitution. Pancreas
1991;6:260–265.
17 Gielkens HAJ, Eddes EH, Vecht J, van Oos-
tayen JA, Lamers CB, Masclee AA: Gallblad-
der motility and cholecystokinin secretion in
chronic pancreatitis: relationship with exocrine
pancreatic function. J Hepatol 1997;27:306–
312.
18 Nakamura T, Tando Y, Yamada N, Imamura
K, Ishii M, Terada A, Takebe K, Kaji A, Wata-
nabe T, Suda T, Koide M, Otsuki M: Meal-
related changes in plasma CCK bioactivity in
patients with chronic pancreatitis. Acta Gas-
troenterol Belg 1998;61:400–406.
19 Garces MC, Gomez-Cerezo JG, Alba D, Codo-
ceo R, Vazquez-Munoz E, Arnalich F, Barbado
FJ, Vazquez JJ: Relationship of basal and post-
prandial Intraduodenal bile acid concentra-
tions and plasma cholecystokinin levels with
abdominal pain in patients with chronic pan-
creatitis. Pancreas 1998;17:397–401.
20 Isaksson G, Ihse I: Pain reduction by an oral
pancreatic enzyme preparation in chronic pan-
creatitis. Dig Dis Sci 1983;28:97–102.
21 Slaff J, Jacobson D, Tillman CR, Curington C,
Toskes P: Protease-specific suppression of pan-
creatic exocrine secretion. Gastroenterology
1984;87:44–52.
22 Slaff J, Wolfe MM, Toskes P: Elevated fasting
cholecystokinin levels in pancreatic exocrine
impairment: Evidence to support feedback reg-
ulation. J Lab Clin Med 1985;105:282–285.
23 Braganza JM, Thomas A, Robinson A: Antioxi-
dants to treat chronic pancreatitis in child-
hood? Case report and possible implications
for pathogenesis. Int J Pancreatol 1988;3:209–
216.

40 Pancreatology 2003;3:36–40 Shea/Bishop/Parker/Gelrud/Freedman