Aliment. Pharmacol. Thevap. (1991) 5, 365-378.
ADONIS
026928139100039A
Fate of orally ingested enzymes in pancreatic insuficiency : comparison of two pancreatic enzyme prepara tions
J.-C. DELCHIER", N. VIDON"", M.-F. SAINT-MARC GIRARDIN', J.-C. SOULE", C. MOULIN", B. HUCHET*" & P. ZYLBERBERG" *Department of Gastroenterology and INSERM Unit 99, H6pifal Henri Mondor, Criteil, and **INSERM Unit 290, H6pifal Suint-lazare, Paris, France Accepted for publication 25 February 1991
SUMMARY
The effect on steatorrhoea of a pH-sensitive enteric-coated pancreatic preparation (Eurobiol 25 000) was compared with a conventional pancreatic enzyme preparation (Eurobiol) in six adult patients with exocrine pancreatic insufficiency. In addition, the fate of orally ingested pancreatic enzymes in the upper digestive tract was evaluated by measuring gastric and duodenal pH, amount of enzymes in the stomach, duodenal enzyme output, and fat absorption at the angle of Treitz for the 4 hours following a standard meal. When compared with placebo, Eurobiol and Eurobiol25 000 reduced daily faecal fat excretion by 24 % (not significant) and 43 % ( P < 0.05), respectively. With the conventional preparation, enzyme output and fat absorption at the duodeno-jejunal flexure were significantly improved ( P < 0.05). Marked inter-individual differences in duodenal enzyme recovery (lipase 3 % to 80%; chymotrypsin 26% to 100%) and, consequently, in the reduction of steatorrhoea (0% to 67%) were observed, with the gastric emptying rate emerging as a key determinant factor. With the enteric-coated Correspondence to: Dr J.-C. Delchier, Unit6 de Recherche INSERM 99, HGpital Henri Mondor 51, Avenue du Mar6chal de Lattre de Tassigny F-94010 Cr6tei1, France. 25 365 BAP 5
366
J.-C. DELCHIER et al. preparation, enzyme output and fat absorption at the duodenojejunal flexure were not significantly improved. Discrepancy between the marked reduction of faecal fat excretion and the low duodenal enzyme recovery could indicate that enzyme delivery from microtablets occurs further down in the small bowel. Efficacy of enteric-coated preparations could be enhanced by adding unprotected enzymes, especially in patients with rapid gastric emptying.
INTRODUCTION Medical treatment of patients with exocrine pancreatic insufficiency is generally unsatisfactory due to persistent steatorrhea despite the administration of large doses of pancreatic enzyme supplements. A large proportion of the enzymes in conventional preparations are inactivated by gastric acidity and the amount reaching the duodenum is not enough to ensure adequate lipolysis.' Eurobiol (Eurorga, Fresnes, France) has been the most commonly used conventional preparation in France or many years. It compares favourably with other commercially available preparations : the recommended dose (5 g) contains a particularly large amount of lipase equalling 100000 Federation Internationale Pharmaceutique (FIP) units.' That is equivalent to 80 000 British Pancreatic Unit (BPU). Its proteolytic enzymes are in an inactive proenzyme form, preventing spontaneous degradation of lipase at 37 0C.3 Duodenal recovery of lipase has been shown to be especially high after ingestion with a meaL4 However, Eurobiol does not fully suppress ~teatorrhoea,~ even with cimetidine adjunctive therapy.6 Moreover, it is not always well accepted by patients because it is presented as a powder and is rather unpalatable. Therefore, an enteric-coated form (Eurobiol 25 000) was developed in order to improve both the effectiveness and the acceptability of substitutive enzyme therapy. Since incomplete correction of steatorrhea with previously marketed gastroprotected enzyme preparations has been r e p ~ r t e d , ~it- ' ~ was thought that a higher amount of pancreatic enzymes had to be provided by dose unit. The target was reached by compressing pancreatin in enteric-coated microtablets, 2 mm in diameter such that 25 000 FIP units of lipase were contained in each 500-mg capsule. As compared to Creon and Pancrease, Eurobiol 25000 contains respectively twice and three times more lipase per capsule. In the present study, we sought to evaluate the efficacy of Eurobiol 25 000 on steatorrhea in patients with severe pancreatic insufficiency and compare duodenal enzyme delivery and absorption of lipids after ingestion of Eurobiol, Eurobiol 25000, or placebo with a standard meal. Particular attention was paid to the influence of pH and gastric emptying on therapeutic efficacy.
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
367
PATIENTS A N D M E T H O D S
Pafienfs Six patients (5 males, I female) with severe pancreatic insufficiency secondary to chronic pancreatitis were studied. The patients averaged 50 years of age (range 29-61 years). The diagnosis of pancreatic insufficiency was based on: (a) an abnormal faecal fat excretion ( > 7.0 g/24 h on a 100 g/day fat intake); (b)a normal d-xylose absorption test; and (c) the presence of at least one of the following criteria : a markedly abnormal BT-PABA test,14radiological evidence of pancreatic calcifications or multiple strictures in the main pancreatic duct, or histological evidence of chronic pancreatitis on surgically resected pancreatic tissue. All three criteria were present in each patient. Five patients had a history of chronic ethanol abuse before onset of pancreatitis, while one patient had a familial chronic pancreatitis. Mean duration of disease was 19 years (range 2-38 years). Four patients presented mild cholestatis and one had histologically proven cirrhosis. Insulindependent and non insulin-dependent diabetes mellitus were present in 3 and 1 patients, respectively. At the time of this study, all 6 patients had been taking oral pancreatic enzyme supplements for more than one year and were in stable metabolic condition.
Infesfinal perfusion sfudies Enzyme preparations, meal and markers. The pancreatic enzyme preparations tested included Eurobiol, a conventional preparation composed of freeze-dried pig pancreas (1dose = 5 g) and Eurobiol 25 000, gelatin capsules containing 500 mg of pH-sensitive, enteric-coated pancreatin microtablets (diameter 2 mm; I dose = 2 capsules); the respective matching placebo consisted of powder of pork fillet (I dose = 5 mg) and gelatin capsules containing 500 mg of enteric-coated microtablets of pork fillet (diameter 2 mm; 1 dose = 2 capsules). Mean ( fS.E.M.) lipase activities in the Eurobiol(5 g powder) and Eurobiol25 000 (2 capsules) doses were 96 776 6961 IU and 76 521 rtr: 4523 IU, respectively, while chymotrypsin activi-9524 IU and 35 320 2794 IU, respectively. ties were 22 229 i The meal consisted of 90 g tenderloin steak, 70 g white bread, 14.5 g olive oil, 90 g pear sherbet, and 190 ml water. Its total caloric value was 490 Kcal, distributed approximately as 5 0 % carbohydrate, 30% fat, and 20% protein. The meal was homogenized for 30 s in a blender, with 30 pCi of I4C-labelledpolyethylene glycol 4000 ([14C]PEG).Its volume, osmolality and pH after blending were 400 ml, 490 mosmol/kg and 5.6, respectively. Unlabelled PEG 4000 (10 g/L) in normal saline was perfused into the duodenum as a duodenal recovery marker at a flow rate of 2 ml/mn.
+
Procedure. The experiments were carried out during a 3-day period under double-blind conditions, with preparations administered in a randomized order. The patients received a detailed explanation of the study and gave their informed
368
J.-C. DELCHIER et d.
1
Collection
PEG 4000 infusion *---
\
:Ample$
'1
nized meal '4c
-
PEG
Duodenojejunal
ll
Figure 1. Diagram of the siting of the tubes within the gastrointestinal tract. Distance of B to C = 20 cm.
consent. They were advised not to take any drug during the 8 days preceding the study. They were also requested not to smoke during the perfusion study. The evening before the study, a double-lumen duodenal tube was positioned fluorospically with a perfusion site at the ampulla of Vater and a duodeno-jejunal aspiration site at the angle of Treitz (20 cm distal to this point). O n the morning of the study, a double-lumen gastric sump-tube was positioned with its tip in the antrum. A schematic representation of the tubes in the GI tract is shown in Figure 1. The patients remained in a sitting position for the 4-h period of the study. Powdered placebo or Eurobiol was carefully blended with the meal immediately before its intragastric injection by means of the gastric tube. Capsules of placebo or Eurobiol 25000 were taken orally with 100 ml of water just after meal injection (mean & S.E.M. injection time 2 1 7 Ifi: 23 s). The duodenal perfusion of saline solution containing unlabelled PEG 4000 (10 g/L) was started immediately before ingestion at a constant flow rate of 2 ml/min using peristaltic infusion pump. Five minutes after starting the meal injection-the time considered as necessary for a gastric bolus to flow from the pylorus to the ligament of Treitz-continuous aspiration of duodenal contents was begun at a rate of 1 ml/min, with intestinal liquid collected serially in 15-min fractions for 4 h. The first gastric sample was obtained 7.5 min after starting meal ingestion, and gastric samples (8 ml) were taken every 15 min
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
369
thereafter. Gastric samples were withdrawn by syringe after careful mixing by repeated manual aspiration-re-instillation during the two minutes preceding each sampling to ensure that representative samples would be obtained. After 4 h, the entire remaining gastric contents were aspirated. Gastric and duodenal samples were immediately stored on ice, then at - 70 "C until assay. Analytical rnefhods. The concentrations of [14C]PEG, PEG, lipase, chymotrypsin, and total fatty acids were determined in all duodenal and gastric samples. Bile salts were also measured in duodenal samples. Lipase and chymotrypsin concentrations were determined by automatic titration (Metrohm, Herisau, Switzerland) at a constant temperature of 37 OC and 25 O C , respectively; emulsified olive oil and acetyltyrosine ethyl ester were used as substrates, and sodium hydroxide (0.01 M) as titrant. The optimum pH for determination of lipase and chymotrypsin activities were 9 and 7.9, respectively. Lipase and chymotrypsin activities were expressed in International Units (microequivalents of hydrogen ion liberated from substrate .min/mL). Samples of [14C]PEG were analysed by liquid scintillation counting. Unlabelled PEG was measured by the turbidimetric method of Hyden.I7 Bile salts were estimated spectrophotometrically using 3-a-hydroxysteroid dehydrogenase." After hydrolysis, lipids were expressed as total fatty acids extracted using the technique of Blankenhorn & Arhensl' and titrated according to the technique described by Dole.'"
Dafa freafrnenf. Data analysis was based on the theory developed to investigate gastric emptying of liquid and solid liquid meals in man.l5,I6 The principal step in analysis was the measurement of duodenal flow rate from the dilution of PEG 4000 between the ampulla and the angle of Treitz. Duodenal ['4C]PEG output (equivalent to pyloric [''T]PEG output) was estimated from the duodenal [14C]PEGconcentration multiplied by duodenal flow rate. Pyloric flow rate was obtained by dividing duodenal ['4C]PEG output by the concentration of ['4C]PEG in the stomach, measured over the same time interval. Gastric volume was obtained by dividing gastric content in ['4C]PEG (amount injected-amount delivered at the pylorus) by the concentration of [I4C]PEGin the stomach, measured over the same time interval. Amounts of pancreatic enzymes delivered to the ligament of Treitz-expressed in units per min and in units for the whole 4-h postprandial period-were derived from the duodenal flow rate and the concentration of enzyme activities. Recovery of enzymes ingested with each preparation tested was expressed as a percentage and calculated as follows; total enzyme delivered into the duodenum minus total enzyme output on placebo divided by ingested activity (as determined in vitro) minus enzyme activity in gastric samples. The triglyceride absorption rate was calculated from the differences between fat emptied at the pylorus minus fat leaving the duodenum. The amount of pancreatic enzymes present in the gastric juice at each time interval was calculated from the gastric volume and the gastric concentration of enzymes.
370
J.-C. DELCHIER et al.
14C recovery was expressed as the ratio of I4C recovered under 14C injected. Total 14Crecovered was measured by summing the amount in gastric samples, the amount in the final gastric aspiration, and the calculated ['4C]PEG output at the pylorus.
Balance studies. Balance studies were carried out entirely at the metabolic ward of the Henri Mondor Hospital. Under the supervision of a dietician, patients consumed a standard diet containing 100 g fat/day. The dietician ascertained that the test meals were entirely ingested. Faecal sampling was preceded by a 7-day period of equilibration with either Eurobiol(5 g powder), Eurobiol 25 000 (2 capsules/meal) t.d.s. or placebo assigned in random order with a 15-day interval between treatments. Stools were collected for 72 h at the end of each study period. Each stool sample was analysed for faecal fat," with faecal output expressed in g/24 h. Statistical analysis Data were compared by analysis of variance for randomized blocks and by the Dunne t t f- test for mu1tiple comparison.22 RESULTS
Meal studies The mean ( _+ S.E.M.) percentage recovery of ['4C]PEG was 98.7 f 6.2% with placebo, 93.4 1 2 . 6 % with Eurobiol, and 94.8 f2.1% with Eurobiol 25 000. Table 1 shows the mean (kS.E.M.) cumulative amounts of lipase and chymotrypsin delivered to the ligament of Treitz during the 4-h sampling period. A significant difference in lipase activity in the distal duodenal samples emerged between treatment group (F = 5.94, P < 0.05). Mean lipase activity with Eurobiol significantly differed from that with placebo and Eurobiol 25 000 ( P < 0.05). Values with Eurobiol 25000 did not significantly differ from those with placebo. The mean ( &S.E.M.) cumulatative recovery of lipase activity from the ligament of Treitz in 240 min was significantly higher with Eurobiol than with Eurobiol 25 000, 30 k 13Yo US. 1.3 1%( P < 0.05). There was also a significant difference in chymotrypsin activity between Table 1. Mean ( _+ S.E.M.) cumulative postprandial lipase and chymotrypsin enzyme activity delivered to the ligament of Treitz during the 4-h sampling period with placebo, Eurobiol, and Eurobiol 25 000 Enzyme activity delivered with Enzyme
Placebo
Eurobiol
Eurobiol 25 000
Lipase (IU) Chymotryp sin (I U)
5812 5812 1512 If 1162
30 862 15 102 11417 k 2883
6697 5902 2784 f1345
PANCREATIC ENZYMES I N C H R O N I C PANCREATITIS 371 treatment groups (F = 16.26, P < 0.01), with the mean Eurobiol concentration found to be significantly higher than that with placebo and Eurobiol 25 000 (P < 0.01); chymotrypsin concentrations with Eurobiol25 000 and placebo did not differ significantly from each other. Cumulative chymotrypsin recovery was significantly higher with Eurobiol as compared with Eurobiol 25000: 58f 14 us. 3.9+ 1.6% ( P < 0.01). Mean lipase and chymotrypsin delivery rates (IU/min) at the duodeno-jejunal flexure are shown in Figure 2. The pancreatic enzyme delivery rates with placebo and Eurobiol 25000 were low during the entire study period. By contrast, the mean delivery rate with Eurobiol was very high during the first 30 min, then progressively decreased, plateauing at the level observed with placebo at 120 min for lipase and at 180 for chymotrypsin. When expressed as a percentage of ingested amount, the mean ( & S.E.M.) rate
c .-
500.
-
400.
E
\ 3
Q)
u)
.- 300. -1
200. 100,
U.:o 60
30
.-c
90
120 150
180
210 240
200
E
2 .-c g
160 120
5, c b
Figure 2. Mean lipase and chymotrypsin activity delivered to the ligament of Treitz postprandially with placebo (A), Eurobiol (o),or Eurobiol25 000 (0).
O
80
"
40
5 =
1
&...O..
..a.
30
60
80
120 150 180 210 240
Time (min)
372
J.-C. DELCHIER ef al.
of triglycerides emptied from the stomach during the 4-h study period was 71.7 9.9%with placebo, 92.3 & 6.6% with Eurobiol and 70 k 9.1% with Eurobiol 25 000. The mean ( & S.E.M.) amount of triglycerides absorbed in the duodenum was 3.4 f1.4 mmol with placebo, 11.0 k 1.3 mmol with Eurobiol, and 3.3 ? 1.2 mmol with Eurobiol 25 000. The difference between treatment groups was significant (F = 12.50, P < 0.01). Triglyceride absorption with Eurobiol differed significantly from levels measured with placebo ( P < 0.02) and Eurobiol 25 000 (P < 0.01).Values with Eurobiol25 000 did not significantly differ from those with placebo. The fraction of triglycerides absorbed in the duodenum, expressed as a percentage of the amount emptied from the stomach, was 14 I_+ 5.3 %, 40 _+ 7.5 %, and 16 5.7% with placebo, Eurobiol and Eurobiol 25 000, respectively. Analysis of individual results with Eurobiol 25 000 showed that duodenal enzyme delivery time-curves were similar to those observed with placebo with the exception of one patient (M.C.) in whom enzyme activity initially appeared in the duodenum between 60 and 90 min, then from 150 min until the end of the study period. In this patient, lipase and chymotrypsin recovery values were 6 % and 4%, respectively. Similarly, assays of gastric samples revealed activity in only one of the patients (M.C.) which appeared at 90 min and persisted until 210 min. The occurrence of intragastric enzyme activity coincided with a prolonged period of intragastric pH > 5 ; enzyme activity disappeared when pH dropped below 4. Although the shapes of the duodenal enzyme delivery time-curves with Eurobiol were similar in all patients, the amount and duration of enzyme delivery varied considerably. Lipase recovery ranged from 3-80 % and chymotrypsin recovery from 26-100%. Changes in pH, meal volume, and enzyme levels in the stomach were analysed in each patient. Important interindividual variations were noted, accounting for the aforementioned differences in the duodenal enzyme recovery (Figure 3). In patients B.R. and M.C., emptying of gastric contents was especially rapid (time to empty 50% or f50% 25 and 49 min, respectively) and pancreatic enzymes were emptied from the stomach before the gastric pH dropped below 4;in these patients, duodenal enzyme recovery was high (lipase 60% and 80 %, chymotrypsin 102 % and 103 %, respectively). In patient A. R., meal emptying was slower (t50 % 81 min) and gastric pH decreased below 4 within 45 min: no enzyme activity could be measured in the stomach after 60 min and duodenal enzyme recovery was low (lipase 13%, chymotrypsin 34 %). In the three remaining patients (R.E., D. J., D. A.), the gastric emptying rate was particularly low and the meal was incompletely evacuated by the end of the study period. Inactivation of enzymes by the stomach was clearly related to the time course of intragastric pH. In patients D. J. and D. A., gastric pH quickly decreased below 4 and enzyme activity disappeared within 120 min; corresponding duodenal enzyme recovery was very low (lipase 3 % and 9%, chymotrypsin 31% and 26%, respectively). In patient R.E. intragastric pH remained above pH 4 longer than in patients D. J. and D.A. Intragastric lipase activity thereby persisted for a longer period than in the two above-mentioned patients; thereafter gastric pH remained slightly below 4 for the
+
hi,
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
373
- --->-I.-- - ,'--*I- 6
1-r-7
Time ( m i d Figure 3. Gastric pH (-), percentage of remaining meal (---), lipase (-), and chymotrypsin (---) levels in each patient with Eurobiol. Corresponding lipase and chymotrypsin recoveries in duodenum were 60% and 100% in B.R., 13%and 54% in R.E., 3% and 31% in D.J., 9% and 2 1 % in D. A,, 80 % and 100YO in M. C., 13 % and 34 % in A. R.
rest of the study period and lipase activity decreased faster than chymotrypsin activity. Duodenal lipase recovery in R. E. was markedly lower (13 YO)than chymotrypsin recovery (54 YO). Mean (If: S.E.M.) postprandial gastric and duodenal pH profiles (Figure 4) did not differ according to treatment. Mean gastric emptying rates were also not influenced by enzyme therapy; mean (i S.E.M.) t50% were 112 & 55 min, 110 28 min, and 99 21 min with placebo, Eurobiol and Eurobiol 25 000, respectively. Bile salt concentrations in the duodenum were similar with placebo, Eurobiol and Eurobiol 25 000 during the first 2 h and exceeded 4 mmol, the critical micellar concentration defined by H~ffmann.'~ During the last 2 h, bile salt concentrations remained high with Eurobiol and Eurobiol 25 OOO but declined with placebo. The cumulative amounts of bile salt delivered at the angle of Treitz (mean If: S.E.M.) were 3818 & 788 mmol with placebo, 5725 & 1613 mmol with Eurobiol, and 4917 If: 1518 mmol with Eurobiol 25 000. There were no significant difference between treatment groups.
Faecal balance sfudies Changes in daily faecal output after different treatments are shown in Figure 5. The mean (If: S.E.M.) faecal fat excretion in g/24 h was 42 & 4.5 with placebo, 32 7.8 with Eurobiol, and 24 If: 1.5 with Eurobiol25 000. There was a significant difference
7.-C. DELCHIER et al.
374
30
60
90
120
150
180
210
240
0 8 3
0
5, 4,
Figure 4. Mean gastric and duodenal pH with placebo (A), 30
60
60 S 50 Y
c
0 '
- 40 0 V
0 0
'hx : .0
O
2c
IC _ _ _ _ _ _ _ _ .____._. _ ._. _ .______. _ _........ ...____..____ B. R.;0,D. J.; ~
Placebo
~
Eurobiol
Eur 25000
A, R. E.; 0,squares. A.R.; m, M.C.; 0 , D.A. Rectangles indicate mean kS.E.M.
PANCREATIC ENZYMES I N C H R O N I C PANCREATITIS
375
in faecal fat excretion between different treatment groups (F = 4.15; P < 0.05), and values with Eurobiol 25 000 significantly differed from those with placebo (P < 0.05). Marked interindividual differences were observed with Eurobiol. Daily faecal fat output was not normalized in any patient, regardless of the preparation used.
DISCUSSION The most striking result of the present study was that Eurobiol 25000 did not significantly improve pancreatic enzyme delivery at the ligament of Treitz and triglyceride absorption in the duodenum as compared with placebo. At first sight, these results seem inconsistent with the improvement in steatorrhoea observed in the six patients of the present study (mean reduction 43 %) and with the results of a recently performed study in 41 adult patients with severe exocrine pancreatic insufficiency. In that double-blind randomized steatorrhoea was reduced by 41% ( P < 0.001) in patients receiving Eurobiol 25000 (2 capsules t.d.s.) and increased by 7% (N.S.) in patients on placebo as compared with a 7-day pretreatment baseline period. The reduction of steatorrhoea was similar to that previously reported with other enteric-coated preparation^.^-'^ In the present study, analysis of gastric samples aspirated after administration of Eurobiol25 000 did not reveal any lipase or chymotrypsin activity except in one patient (M.C.) with sustained intragastric pH above 5. This observation confirms the efficiency and pH selectivity of the pH-sensitive enteric coating used for Eurobiol 25 000. Consequently, there may be only two reasons for the absence of enzyme delivery to the ligament of Treitz and the lack of lipolysis in the duodenum: (a) a delay in gastric emptying of microtablets as compared with food and/or (b)a lack of enzyme release form microtablets in the duodenum. Microtablets present in each capsule of Eurobiol 25 000 are 2 mm in diameter. Recently, Meyer ef dz5 established that microspheres > 1.4 mm in diameter were emptied more slowly from the stomach than radiolabelled chicken liver; 78 % and 50% of microspheres measuring I and 2.4 mm in diameter were respectively emptied 210 min after ingestion with a meal. These results were consistent with the observation by Dutta et al.7 of a delay in duodenal delivery of pancreatic enzyme after administration of another pH-sensitive enteric-coated preparation (Pancrease, However, in the Johnson & Johnson, Piscataway, NJ) with a solid-liquid present study with Eurobiol25 000, delayed emptying of microtablets by itself was unlikely to account for the total absence of pancreatic enzyme activity in the duodenum during the entire 4-h postprandial period. Impairment in enzyme release from microtablets in the duodenum was probably the major factor. Indeed, in-vitro studies (Eurorga, unpublished data) have shown that time to complete release of pancreatic enzymes from Eurobiol 25 000 microtablets is 90 min at pH 5 and from 15 to 30 min at pH 6 depending upon ionic strength. Although pH at the ligament of Treitz remained above 6 in most duodeno-
376
J.-C. DELCHIER ef al.
jejunal samples in our 6 patients, pH in the upper part of the duodenum may well have been below 6, as it has previously been demonstrated in several studie~.'~-~' Moreover, transit time from pylorus to the ligament of Treitz after meal ingestion ~ apparent conhas been calculated to be about 5 min by Malagelada et ~ 1 . ' The tradiction between the absence of enzyme delivery at the ligament of Treitz and the improvement in steatorrhoea with Eurobiol 25000 can only be explained by the release of pancreatic enzymes further down in the small intestine. Comparison of our results with those of the study by Dutta et suggests that pancreatic enzymes are released more slowly from Eurobiol 25000 microtablets than from Pancrease microspheres. A comparative study is clearly needed to draw more definitive conclusions. Absence of lipolysis in the proximal duodenum could play a major role in the pathogenesis of the residual steatorrhoea in patients receiving enteric-coated pancreatic enzyme preparations. A recent study by Zerega ef aL3' seems to substantiate this hypothesis as the continuous instillation of pancreatic enzymes at the ligament of Treitz failed to abolish steatorrhoea in patients with pancreatic insufficiency. A significant increase in pancreatic enzyme delivery to the duodenum with Eurobiol coincided with a marked improvement in duodenal triglyceride absorption. Mean duodenal recovery of lipase was found to be especially high (29%) as compared with other conventional preparation^.^^^^'^^ This is probably related to Eurobiol's galenic form: as a powder, it blends well with food, resulting in concurrent emptying. Mean amounts of lipase reaching the duodenum exceeded the threshold of 10% of the normal value theoretically sufficient to ensure normal l i p ~ l y s i sHowever, .~~ in the present study as in a previous one,s Eurobiol did not normalize daily faecal fat. Comparison of enzyme and meal delivery time courses in duodenum may offer some explanation for this apparently paradoxical result (Figure I): enzymes were only delivered into the duodenum during the first two post-prandial hours, while gastric emptying of the meal took 4 h; the absence of enzymes in the duodenum during the last 2 h was due to intragastric inactivation by acid (Figure 3 ) . Moreover, considerable interindividual variations in lipase duodenal recovery were observed (range 3-80 YO), resulting in differences between patients in the control of steatorrhea. The rate of gastric emptying appeared to be a key factor in duodenal enzyme recovery ; when rapid, pancreatic enzyme inactivation by acid was prevented and duodenal recovery was high (patients B. R. and M. C.),By contrast, when gastric emptying was normal or slow (patients A. R., D. J., J. A., R. E.), intragastric survival and duodenal recovery of pancreatic enzymes depended on individual lcvels of gastric acid and secretion. This was particularly true for lipase. In this regard, results in patient R.E. were noteworthy; intragastric lipase activity decreased faster than chymotrypsin activity resulting in lower duodenal recovery of lipase (13 %) than that of chymotrypsin (54 Yo). Such in-vivo observations accorded with previous in-vitro data.33 neither of the pancreatic In agreement with the study by DiMagno ef enzyme preparations influenced meal emptying. Bile salt concentration in the
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
377
duodenum was unexpectedly lower with placebo than with Eurobiol or Eurobiol 25 000 during the second half of the sampling period. As the gastric emptying rate and duodenal pH were not influenced by treatment, this result suggests an indirect effect of pancreatic extracts on choleresis or gallbladder emptying. In conclusion, results of the present study provide some explanation for the incomplete correction of steatorrhoea observed in patients with severe pancreatic insufficiency treated with conventional as well as with pH-sensitive enteric-coated pancreatic enzyme preparations. When using the conventional preparation, steatorrhoea may persist, despite a theoretically sufficient duodenal recovery of administered enzymes, because part of the meal is delivered into the duodenum without enzymes as a result of enzyme inactivation by intragastric acid. Using the latter preparation, steatorrhoea may remain incompletely corrected, despite effective enzyme protection against acidity, because enzyme release from entericcoated microtablets is too slow for the small intestine to carry out complete lipid absorption. Our results suggest that effectiveness of oral pancreatic enzyme supplementation could be enhanced by preparations combining protected and unprotected enzymes. ACKNOWLEDGEMENTS
The authors wish to thank Laboratoires Eurorga (Fresnes, France) for their financial support. REFERENCES DiMagno E P. Medical treatment of exocrine pancreatic insufficiency. Dig Dis Sci 1982; 27: 481-4. FCdCration Intemationale Pharmaceutique -International commission for the standardization of Pharmaceutical enzymes, Fourth Report. J Mond Pharm; 1968: 3-11. Delchier J C, Moulin C, Dutreuil C, Soul6 J C. Stability of enzyme activities in pancreatic extracts incubated at 37 "C. Dig Dis Sci 1987; 32: 1163A. (Abstr.) Saint-Marc Girardin M F, Moreau J, et al. Duodenal availability of Eurobiol versus placebo in patients with pancreatic insufficiency (abstr). Dig Dis Sci 1986; 31: 347s. Saint-Marc Girardin M F, Cortot A, ThCdore C ef al. Effkacit6 du pancrCas total lyophylisC au cours de la pancrCatite chronique et des rCsections pancrCatiques. Gastroenterol Clin Biol 1987; 11: 430-1. Staub J L, Sarles H, Soul4 J C, Galmiche J P, Capron J P. No effect of cimetidine on the therapeutic response to oral enzymes in
severe pancreatic insufficiency. N Engl J Med 1981; 304: 1364-5. 7 Dutta S K, Rubin J, Harvey J. Comparative evaluation of the therapeutic efficacy of a pH-sensitive enteric-coated enzyme preparation with conventional pancreatic enzyme therapy in the treatment of exocrine pancreatic insufficiency. Gastroenterology 1983; 84: 476-82. 8 Graham D Y. An enteric-coated pancreatic enzyme preparation that works. Dig Dis Sci 1979; 24: 906-9. 9 Halgreen H, Pedersen N T, Worning H. Symptomatic effect of pancreatic enzyme therapy in patients with chronic pancreatitis. Scand J Gastroenterol 1986; 21 : 104-8. 10 Salen G, Prakash A. Evaluation of entericcoated microspheres for replacement therapy in adults with pancreatic insufficiency. Curr Ther Res 1979; 25 : 650-6. 11 Valerio D, Whyte E H A, Schlamm H T, Ruggerio J A, Blackburn G L. Clinical effectiveness of pancreatic enzyme supplement. J Parent Enterol Nutr 1981; 5 : 110-4. 12 Lankisch P G, Lembcke B, Goke B,
378
13
14
15
16
17
18
19
20
21
22 23
24
J.-C. DELCHIER et al.
Creutzfeld W. Therapy of pancreatogenic steatorrhea: does acid protection of pancreatic enzymes offer any advantage? Z Gastroenterol 1986; 24: 753-7. Schneider M V, Knoll-Ruzicka L, Domschke S, Heptner G, Domschke W. Pancreatic enzyme replacement therapy: Comparative effects of conventional and enteric-coated microspheric pancreatic and acid stable fungal enzyme preparation on steatorrhea in chronic pancreatitis. Hepatogastroentero b g y 1985 ; 32 : 97-102. Delchier J C, SoulC J C. BT-PABA with plasma PABA measurements : evaluation of sensitivity and specificity. Gut 1983; 24: 318-25. Malagelada J R, Longstreth G R, Summerskill W H J, Go V L W. Measurement of gastric functions during digestion of ordinary solid meals in man. Gastroenterology 1976; 70: 203-10. Vidon N, Muschart J M, Cosnes J, Ruskone A, Bernier J J. Etude critique de l'estimation de la vidange gastrique par la mkthode de perfusion duodknale dune substance non absorbable h faible dCbit. Gastroenterol Clin Biol 1979; 3: 549-52. Hyden S. A turbidimetric method for the determination of higher polyethylene glycols in biological materials. Ann R Agr Coll (Sweden) 1956; 22: 139-45. Talalay P. Enzymatic analysis of steroid hormones. Methods Biochem Anal 1960; 8: 129-43. Blankenhorn D H, Arhens E H. Extraction, isolation and identification of hydrolytic products of triglyceride digestion in man. J Biol Biochem 1955; 212: 69-81. Dole V P. A relation between nonesterified fatty acids in plasma and the metabolism of glucose. J Clin Invest 1956; 35 : 150-4. Van de Kamer J H, Bokkell T, Huinink H T B, Wyers H A. Rapid method for determination of fat in feces. J Biol Chem 1949; 177: 347-55. Zar J H. Biostatistical analysis. Englewood Cliffs, NJ: Prentice Hall, Inc., 1974. Hoffmann A F. The function of bile salts in fat absorption: The solvent properties of dilute micellar solutions of conjugated bile salts. Biochem J 1963; 89: 57-68. Saint-Marc Girardin M F, Cortot A,
25
26
27
28
29
30
31
32
33
34
Theodore C, ef al. Eurobiol, 25 000 U, nouvel extrait pancrdatique gastroprotkgk: son efficacitk dans la pancrdatite chronique et dans les resections pancreatiques. Gastroenterol Clin Biol 1989; 13 : A185. (Abstract.) Meyer J H, Elashoff J, Porter-Fink V, Dressman J, Amidon C L. Human postprandial gastric emptying of I-3-millimetre spheres. Gastroenterology 1988; 94: 1315-25. DiMagno E P, Malagelada J R, Go V L W, Moertel C G. Fate of orally ingested enzymes in pancreatic insufficiency : Comparison of two dosage schedules. N Engl J Med 1977; 296: 1318-22. Regan P T, Malagelada J R, DiMagno E P, Glanzman S L, Go V L W. Comparative effects of antacids, cimetidine, and enteric coating on the therapeutic response to oral enzymes in severe pancreatic insufficiency. N Engl J Med 1977; 297: 854-8. Dutta S, Russell R M, Iber F L. Influence of exocrine pancreatic insufficiency on the intraluminal pH of the proximal small intestine. Dig Dis Sci 1979; 24: 529-34. Dutta S, Russell R M, Iber F L. Impaired acid neutralization in the duodenum in pancreatic insufficiency. Dig Dis Sci 1979; 24: 77580. Bommelaer G, Benque A, Moreau J, Bouisson M, Ribet A. pH-mCtrie duodenale durant 24 heures dans les pancriatites chroniques. Gastroenterol Clin Biol 1984; 8: 403-6. Zerega J, Lerner f, Meyer J H. Duodenal instillation of pancreatin does not abolish steatorrhea in patients with pancreatic insufficiency. Dig Dis Sci 1988; 33: 1245-9. DiMagno E P, Gos V L W, Summerskill W H J. Relation between pancreatic enzyme outputs and malabsorption in severe pancreatic insufficiency.N Engl J Med 1973; 288 : 813-5. Heizer W D, Cleveland C R, Iber F L. Gastric inactivation of pancreatic supplements. Bull Johns Hopkins Hosp 1965; 116: 26170. DiMagno E P, Clain J E. Chronic pancreatitis in the exocrine pancreas. In: Go V L W, Gardner J D, Brooks F P, et al. eds. Biology, pathobiology and diseases. New York: Raven Press, 1986; 541-75.
ADONIS
026928139100039A
Fate of orally ingested enzymes in pancreatic insuficiency : comparison of two pancreatic enzyme prepara tions
J.-C. DELCHIER", N. VIDON"", M.-F. SAINT-MARC GIRARDIN', J.-C. SOULE", C. MOULIN", B. HUCHET*" & P. ZYLBERBERG" *Department of Gastroenterology and INSERM Unit 99, H6pifal Henri Mondor, Criteil, and **INSERM Unit 290, H6pifal Suint-lazare, Paris, France Accepted for publication 25 February 1991
SUMMARY
The effect on steatorrhoea of a pH-sensitive enteric-coated pancreatic preparation (Eurobiol 25 000) was compared with a conventional pancreatic enzyme preparation (Eurobiol) in six adult patients with exocrine pancreatic insufficiency. In addition, the fate of orally ingested pancreatic enzymes in the upper digestive tract was evaluated by measuring gastric and duodenal pH, amount of enzymes in the stomach, duodenal enzyme output, and fat absorption at the angle of Treitz for the 4 hours following a standard meal. When compared with placebo, Eurobiol and Eurobiol25 000 reduced daily faecal fat excretion by 24 % (not significant) and 43 % ( P < 0.05), respectively. With the conventional preparation, enzyme output and fat absorption at the duodeno-jejunal flexure were significantly improved ( P < 0.05). Marked inter-individual differences in duodenal enzyme recovery (lipase 3 % to 80%; chymotrypsin 26% to 100%) and, consequently, in the reduction of steatorrhoea (0% to 67%) were observed, with the gastric emptying rate emerging as a key determinant factor. With the enteric-coated Correspondence to: Dr J.-C. Delchier, Unit6 de Recherche INSERM 99, HGpital Henri Mondor 51, Avenue du Mar6chal de Lattre de Tassigny F-94010 Cr6tei1, France. 25 365 BAP 5
366
J.-C. DELCHIER et al. preparation, enzyme output and fat absorption at the duodenojejunal flexure were not significantly improved. Discrepancy between the marked reduction of faecal fat excretion and the low duodenal enzyme recovery could indicate that enzyme delivery from microtablets occurs further down in the small bowel. Efficacy of enteric-coated preparations could be enhanced by adding unprotected enzymes, especially in patients with rapid gastric emptying.
INTRODUCTION Medical treatment of patients with exocrine pancreatic insufficiency is generally unsatisfactory due to persistent steatorrhea despite the administration of large doses of pancreatic enzyme supplements. A large proportion of the enzymes in conventional preparations are inactivated by gastric acidity and the amount reaching the duodenum is not enough to ensure adequate lipolysis.' Eurobiol (Eurorga, Fresnes, France) has been the most commonly used conventional preparation in France or many years. It compares favourably with other commercially available preparations : the recommended dose (5 g) contains a particularly large amount of lipase equalling 100000 Federation Internationale Pharmaceutique (FIP) units.' That is equivalent to 80 000 British Pancreatic Unit (BPU). Its proteolytic enzymes are in an inactive proenzyme form, preventing spontaneous degradation of lipase at 37 0C.3 Duodenal recovery of lipase has been shown to be especially high after ingestion with a meaL4 However, Eurobiol does not fully suppress ~teatorrhoea,~ even with cimetidine adjunctive therapy.6 Moreover, it is not always well accepted by patients because it is presented as a powder and is rather unpalatable. Therefore, an enteric-coated form (Eurobiol 25 000) was developed in order to improve both the effectiveness and the acceptability of substitutive enzyme therapy. Since incomplete correction of steatorrhea with previously marketed gastroprotected enzyme preparations has been r e p ~ r t e d , ~it- ' ~ was thought that a higher amount of pancreatic enzymes had to be provided by dose unit. The target was reached by compressing pancreatin in enteric-coated microtablets, 2 mm in diameter such that 25 000 FIP units of lipase were contained in each 500-mg capsule. As compared to Creon and Pancrease, Eurobiol 25000 contains respectively twice and three times more lipase per capsule. In the present study, we sought to evaluate the efficacy of Eurobiol 25 000 on steatorrhea in patients with severe pancreatic insufficiency and compare duodenal enzyme delivery and absorption of lipids after ingestion of Eurobiol, Eurobiol 25000, or placebo with a standard meal. Particular attention was paid to the influence of pH and gastric emptying on therapeutic efficacy.
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
367
PATIENTS A N D M E T H O D S
Pafienfs Six patients (5 males, I female) with severe pancreatic insufficiency secondary to chronic pancreatitis were studied. The patients averaged 50 years of age (range 29-61 years). The diagnosis of pancreatic insufficiency was based on: (a) an abnormal faecal fat excretion ( > 7.0 g/24 h on a 100 g/day fat intake); (b)a normal d-xylose absorption test; and (c) the presence of at least one of the following criteria : a markedly abnormal BT-PABA test,14radiological evidence of pancreatic calcifications or multiple strictures in the main pancreatic duct, or histological evidence of chronic pancreatitis on surgically resected pancreatic tissue. All three criteria were present in each patient. Five patients had a history of chronic ethanol abuse before onset of pancreatitis, while one patient had a familial chronic pancreatitis. Mean duration of disease was 19 years (range 2-38 years). Four patients presented mild cholestatis and one had histologically proven cirrhosis. Insulindependent and non insulin-dependent diabetes mellitus were present in 3 and 1 patients, respectively. At the time of this study, all 6 patients had been taking oral pancreatic enzyme supplements for more than one year and were in stable metabolic condition.
Infesfinal perfusion sfudies Enzyme preparations, meal and markers. The pancreatic enzyme preparations tested included Eurobiol, a conventional preparation composed of freeze-dried pig pancreas (1dose = 5 g) and Eurobiol 25 000, gelatin capsules containing 500 mg of pH-sensitive, enteric-coated pancreatin microtablets (diameter 2 mm; I dose = 2 capsules); the respective matching placebo consisted of powder of pork fillet (I dose = 5 mg) and gelatin capsules containing 500 mg of enteric-coated microtablets of pork fillet (diameter 2 mm; 1 dose = 2 capsules). Mean ( fS.E.M.) lipase activities in the Eurobiol(5 g powder) and Eurobiol25 000 (2 capsules) doses were 96 776 6961 IU and 76 521 rtr: 4523 IU, respectively, while chymotrypsin activi-9524 IU and 35 320 2794 IU, respectively. ties were 22 229 i The meal consisted of 90 g tenderloin steak, 70 g white bread, 14.5 g olive oil, 90 g pear sherbet, and 190 ml water. Its total caloric value was 490 Kcal, distributed approximately as 5 0 % carbohydrate, 30% fat, and 20% protein. The meal was homogenized for 30 s in a blender, with 30 pCi of I4C-labelledpolyethylene glycol 4000 ([14C]PEG).Its volume, osmolality and pH after blending were 400 ml, 490 mosmol/kg and 5.6, respectively. Unlabelled PEG 4000 (10 g/L) in normal saline was perfused into the duodenum as a duodenal recovery marker at a flow rate of 2 ml/mn.
+
Procedure. The experiments were carried out during a 3-day period under double-blind conditions, with preparations administered in a randomized order. The patients received a detailed explanation of the study and gave their informed
368
J.-C. DELCHIER et d.
1
Collection
PEG 4000 infusion *---
\
:Ample$
'1
nized meal '4c
-
PEG
Duodenojejunal
ll
Figure 1. Diagram of the siting of the tubes within the gastrointestinal tract. Distance of B to C = 20 cm.
consent. They were advised not to take any drug during the 8 days preceding the study. They were also requested not to smoke during the perfusion study. The evening before the study, a double-lumen duodenal tube was positioned fluorospically with a perfusion site at the ampulla of Vater and a duodeno-jejunal aspiration site at the angle of Treitz (20 cm distal to this point). O n the morning of the study, a double-lumen gastric sump-tube was positioned with its tip in the antrum. A schematic representation of the tubes in the GI tract is shown in Figure 1. The patients remained in a sitting position for the 4-h period of the study. Powdered placebo or Eurobiol was carefully blended with the meal immediately before its intragastric injection by means of the gastric tube. Capsules of placebo or Eurobiol 25000 were taken orally with 100 ml of water just after meal injection (mean & S.E.M. injection time 2 1 7 Ifi: 23 s). The duodenal perfusion of saline solution containing unlabelled PEG 4000 (10 g/L) was started immediately before ingestion at a constant flow rate of 2 ml/min using peristaltic infusion pump. Five minutes after starting the meal injection-the time considered as necessary for a gastric bolus to flow from the pylorus to the ligament of Treitz-continuous aspiration of duodenal contents was begun at a rate of 1 ml/min, with intestinal liquid collected serially in 15-min fractions for 4 h. The first gastric sample was obtained 7.5 min after starting meal ingestion, and gastric samples (8 ml) were taken every 15 min
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
369
thereafter. Gastric samples were withdrawn by syringe after careful mixing by repeated manual aspiration-re-instillation during the two minutes preceding each sampling to ensure that representative samples would be obtained. After 4 h, the entire remaining gastric contents were aspirated. Gastric and duodenal samples were immediately stored on ice, then at - 70 "C until assay. Analytical rnefhods. The concentrations of [14C]PEG, PEG, lipase, chymotrypsin, and total fatty acids were determined in all duodenal and gastric samples. Bile salts were also measured in duodenal samples. Lipase and chymotrypsin concentrations were determined by automatic titration (Metrohm, Herisau, Switzerland) at a constant temperature of 37 OC and 25 O C , respectively; emulsified olive oil and acetyltyrosine ethyl ester were used as substrates, and sodium hydroxide (0.01 M) as titrant. The optimum pH for determination of lipase and chymotrypsin activities were 9 and 7.9, respectively. Lipase and chymotrypsin activities were expressed in International Units (microequivalents of hydrogen ion liberated from substrate .min/mL). Samples of [14C]PEG were analysed by liquid scintillation counting. Unlabelled PEG was measured by the turbidimetric method of Hyden.I7 Bile salts were estimated spectrophotometrically using 3-a-hydroxysteroid dehydrogenase." After hydrolysis, lipids were expressed as total fatty acids extracted using the technique of Blankenhorn & Arhensl' and titrated according to the technique described by Dole.'"
Dafa freafrnenf. Data analysis was based on the theory developed to investigate gastric emptying of liquid and solid liquid meals in man.l5,I6 The principal step in analysis was the measurement of duodenal flow rate from the dilution of PEG 4000 between the ampulla and the angle of Treitz. Duodenal ['4C]PEG output (equivalent to pyloric [''T]PEG output) was estimated from the duodenal [14C]PEGconcentration multiplied by duodenal flow rate. Pyloric flow rate was obtained by dividing duodenal ['4C]PEG output by the concentration of ['4C]PEG in the stomach, measured over the same time interval. Gastric volume was obtained by dividing gastric content in ['4C]PEG (amount injected-amount delivered at the pylorus) by the concentration of [I4C]PEGin the stomach, measured over the same time interval. Amounts of pancreatic enzymes delivered to the ligament of Treitz-expressed in units per min and in units for the whole 4-h postprandial period-were derived from the duodenal flow rate and the concentration of enzyme activities. Recovery of enzymes ingested with each preparation tested was expressed as a percentage and calculated as follows; total enzyme delivered into the duodenum minus total enzyme output on placebo divided by ingested activity (as determined in vitro) minus enzyme activity in gastric samples. The triglyceride absorption rate was calculated from the differences between fat emptied at the pylorus minus fat leaving the duodenum. The amount of pancreatic enzymes present in the gastric juice at each time interval was calculated from the gastric volume and the gastric concentration of enzymes.
370
J.-C. DELCHIER et al.
14C recovery was expressed as the ratio of I4C recovered under 14C injected. Total 14Crecovered was measured by summing the amount in gastric samples, the amount in the final gastric aspiration, and the calculated ['4C]PEG output at the pylorus.
Balance studies. Balance studies were carried out entirely at the metabolic ward of the Henri Mondor Hospital. Under the supervision of a dietician, patients consumed a standard diet containing 100 g fat/day. The dietician ascertained that the test meals were entirely ingested. Faecal sampling was preceded by a 7-day period of equilibration with either Eurobiol(5 g powder), Eurobiol 25 000 (2 capsules/meal) t.d.s. or placebo assigned in random order with a 15-day interval between treatments. Stools were collected for 72 h at the end of each study period. Each stool sample was analysed for faecal fat," with faecal output expressed in g/24 h. Statistical analysis Data were compared by analysis of variance for randomized blocks and by the Dunne t t f- test for mu1tiple comparison.22 RESULTS
Meal studies The mean ( _+ S.E.M.) percentage recovery of ['4C]PEG was 98.7 f 6.2% with placebo, 93.4 1 2 . 6 % with Eurobiol, and 94.8 f2.1% with Eurobiol 25 000. Table 1 shows the mean (kS.E.M.) cumulative amounts of lipase and chymotrypsin delivered to the ligament of Treitz during the 4-h sampling period. A significant difference in lipase activity in the distal duodenal samples emerged between treatment group (F = 5.94, P < 0.05). Mean lipase activity with Eurobiol significantly differed from that with placebo and Eurobiol 25 000 ( P < 0.05). Values with Eurobiol 25000 did not significantly differ from those with placebo. The mean ( &S.E.M.) cumulatative recovery of lipase activity from the ligament of Treitz in 240 min was significantly higher with Eurobiol than with Eurobiol 25 000, 30 k 13Yo US. 1.3 1%( P < 0.05). There was also a significant difference in chymotrypsin activity between Table 1. Mean ( _+ S.E.M.) cumulative postprandial lipase and chymotrypsin enzyme activity delivered to the ligament of Treitz during the 4-h sampling period with placebo, Eurobiol, and Eurobiol 25 000 Enzyme activity delivered with Enzyme
Placebo
Eurobiol
Eurobiol 25 000
Lipase (IU) Chymotryp sin (I U)
5812 5812 1512 If 1162
30 862 15 102 11417 k 2883
6697 5902 2784 f1345
PANCREATIC ENZYMES I N C H R O N I C PANCREATITIS 371 treatment groups (F = 16.26, P < 0.01), with the mean Eurobiol concentration found to be significantly higher than that with placebo and Eurobiol 25 000 (P < 0.01); chymotrypsin concentrations with Eurobiol25 000 and placebo did not differ significantly from each other. Cumulative chymotrypsin recovery was significantly higher with Eurobiol as compared with Eurobiol 25000: 58f 14 us. 3.9+ 1.6% ( P < 0.01). Mean lipase and chymotrypsin delivery rates (IU/min) at the duodeno-jejunal flexure are shown in Figure 2. The pancreatic enzyme delivery rates with placebo and Eurobiol 25000 were low during the entire study period. By contrast, the mean delivery rate with Eurobiol was very high during the first 30 min, then progressively decreased, plateauing at the level observed with placebo at 120 min for lipase and at 180 for chymotrypsin. When expressed as a percentage of ingested amount, the mean ( & S.E.M.) rate
c .-
500.
-
400.
E
\ 3
Q)
u)
.- 300. -1
200. 100,
U.:o 60
30
.-c
90
120 150
180
210 240
200
E
2 .-c g
160 120
5, c b
Figure 2. Mean lipase and chymotrypsin activity delivered to the ligament of Treitz postprandially with placebo (A), Eurobiol (o),or Eurobiol25 000 (0).
O
80
"
40
5 =
1
&...O..
..a.
30
60
80
120 150 180 210 240
Time (min)
372
J.-C. DELCHIER ef al.
of triglycerides emptied from the stomach during the 4-h study period was 71.7 9.9%with placebo, 92.3 & 6.6% with Eurobiol and 70 k 9.1% with Eurobiol 25 000. The mean ( & S.E.M.) amount of triglycerides absorbed in the duodenum was 3.4 f1.4 mmol with placebo, 11.0 k 1.3 mmol with Eurobiol, and 3.3 ? 1.2 mmol with Eurobiol 25 000. The difference between treatment groups was significant (F = 12.50, P < 0.01). Triglyceride absorption with Eurobiol differed significantly from levels measured with placebo ( P < 0.02) and Eurobiol 25 000 (P < 0.01).Values with Eurobiol25 000 did not significantly differ from those with placebo. The fraction of triglycerides absorbed in the duodenum, expressed as a percentage of the amount emptied from the stomach, was 14 I_+ 5.3 %, 40 _+ 7.5 %, and 16 5.7% with placebo, Eurobiol and Eurobiol 25 000, respectively. Analysis of individual results with Eurobiol 25 000 showed that duodenal enzyme delivery time-curves were similar to those observed with placebo with the exception of one patient (M.C.) in whom enzyme activity initially appeared in the duodenum between 60 and 90 min, then from 150 min until the end of the study period. In this patient, lipase and chymotrypsin recovery values were 6 % and 4%, respectively. Similarly, assays of gastric samples revealed activity in only one of the patients (M.C.) which appeared at 90 min and persisted until 210 min. The occurrence of intragastric enzyme activity coincided with a prolonged period of intragastric pH > 5 ; enzyme activity disappeared when pH dropped below 4. Although the shapes of the duodenal enzyme delivery time-curves with Eurobiol were similar in all patients, the amount and duration of enzyme delivery varied considerably. Lipase recovery ranged from 3-80 % and chymotrypsin recovery from 26-100%. Changes in pH, meal volume, and enzyme levels in the stomach were analysed in each patient. Important interindividual variations were noted, accounting for the aforementioned differences in the duodenal enzyme recovery (Figure 3). In patients B.R. and M.C., emptying of gastric contents was especially rapid (time to empty 50% or f50% 25 and 49 min, respectively) and pancreatic enzymes were emptied from the stomach before the gastric pH dropped below 4;in these patients, duodenal enzyme recovery was high (lipase 60% and 80 %, chymotrypsin 102 % and 103 %, respectively). In patient A. R., meal emptying was slower (t50 % 81 min) and gastric pH decreased below 4 within 45 min: no enzyme activity could be measured in the stomach after 60 min and duodenal enzyme recovery was low (lipase 13%, chymotrypsin 34 %). In the three remaining patients (R.E., D. J., D. A.), the gastric emptying rate was particularly low and the meal was incompletely evacuated by the end of the study period. Inactivation of enzymes by the stomach was clearly related to the time course of intragastric pH. In patients D. J. and D. A., gastric pH quickly decreased below 4 and enzyme activity disappeared within 120 min; corresponding duodenal enzyme recovery was very low (lipase 3 % and 9%, chymotrypsin 31% and 26%, respectively). In patient R.E. intragastric pH remained above pH 4 longer than in patients D. J. and D.A. Intragastric lipase activity thereby persisted for a longer period than in the two above-mentioned patients; thereafter gastric pH remained slightly below 4 for the
+
hi,
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
373
- --->-I.-- - ,'--*I- 6
1-r-7
Time ( m i d Figure 3. Gastric pH (-), percentage of remaining meal (---), lipase (-), and chymotrypsin (---) levels in each patient with Eurobiol. Corresponding lipase and chymotrypsin recoveries in duodenum were 60% and 100% in B.R., 13%and 54% in R.E., 3% and 31% in D.J., 9% and 2 1 % in D. A,, 80 % and 100YO in M. C., 13 % and 34 % in A. R.
rest of the study period and lipase activity decreased faster than chymotrypsin activity. Duodenal lipase recovery in R. E. was markedly lower (13 YO)than chymotrypsin recovery (54 YO). Mean (If: S.E.M.) postprandial gastric and duodenal pH profiles (Figure 4) did not differ according to treatment. Mean gastric emptying rates were also not influenced by enzyme therapy; mean (i S.E.M.) t50% were 112 & 55 min, 110 28 min, and 99 21 min with placebo, Eurobiol and Eurobiol 25 000, respectively. Bile salt concentrations in the duodenum were similar with placebo, Eurobiol and Eurobiol 25 000 during the first 2 h and exceeded 4 mmol, the critical micellar concentration defined by H~ffmann.'~ During the last 2 h, bile salt concentrations remained high with Eurobiol and Eurobiol 25 OOO but declined with placebo. The cumulative amounts of bile salt delivered at the angle of Treitz (mean If: S.E.M.) were 3818 & 788 mmol with placebo, 5725 & 1613 mmol with Eurobiol, and 4917 If: 1518 mmol with Eurobiol 25 000. There were no significant difference between treatment groups.
Faecal balance sfudies Changes in daily faecal output after different treatments are shown in Figure 5. The mean (If: S.E.M.) faecal fat excretion in g/24 h was 42 & 4.5 with placebo, 32 7.8 with Eurobiol, and 24 If: 1.5 with Eurobiol25 000. There was a significant difference
7.-C. DELCHIER et al.
374
30
60
90
120
150
180
210
240
0 8 3
0
5, 4,
Figure 4. Mean gastric and duodenal pH with placebo (A), 30
60
60 S 50 Y
c
0 '
- 40 0 V
0 0
'hx : .0
O
2c
IC _ _ _ _ _ _ _ _ .____._. _ ._. _ .______. _ _........ ...____..____ B. R.;0,D. J.; ~
Placebo
~
Eurobiol
Eur 25000
A, R. E.; 0,squares. A.R.; m, M.C.; 0 , D.A. Rectangles indicate mean kS.E.M.
PANCREATIC ENZYMES I N C H R O N I C PANCREATITIS
375
in faecal fat excretion between different treatment groups (F = 4.15; P < 0.05), and values with Eurobiol 25 000 significantly differed from those with placebo (P < 0.05). Marked interindividual differences were observed with Eurobiol. Daily faecal fat output was not normalized in any patient, regardless of the preparation used.
DISCUSSION The most striking result of the present study was that Eurobiol 25000 did not significantly improve pancreatic enzyme delivery at the ligament of Treitz and triglyceride absorption in the duodenum as compared with placebo. At first sight, these results seem inconsistent with the improvement in steatorrhoea observed in the six patients of the present study (mean reduction 43 %) and with the results of a recently performed study in 41 adult patients with severe exocrine pancreatic insufficiency. In that double-blind randomized steatorrhoea was reduced by 41% ( P < 0.001) in patients receiving Eurobiol 25000 (2 capsules t.d.s.) and increased by 7% (N.S.) in patients on placebo as compared with a 7-day pretreatment baseline period. The reduction of steatorrhoea was similar to that previously reported with other enteric-coated preparation^.^-'^ In the present study, analysis of gastric samples aspirated after administration of Eurobiol25 000 did not reveal any lipase or chymotrypsin activity except in one patient (M.C.) with sustained intragastric pH above 5. This observation confirms the efficiency and pH selectivity of the pH-sensitive enteric coating used for Eurobiol 25 000. Consequently, there may be only two reasons for the absence of enzyme delivery to the ligament of Treitz and the lack of lipolysis in the duodenum: (a) a delay in gastric emptying of microtablets as compared with food and/or (b)a lack of enzyme release form microtablets in the duodenum. Microtablets present in each capsule of Eurobiol 25 000 are 2 mm in diameter. Recently, Meyer ef dz5 established that microspheres > 1.4 mm in diameter were emptied more slowly from the stomach than radiolabelled chicken liver; 78 % and 50% of microspheres measuring I and 2.4 mm in diameter were respectively emptied 210 min after ingestion with a meal. These results were consistent with the observation by Dutta et al.7 of a delay in duodenal delivery of pancreatic enzyme after administration of another pH-sensitive enteric-coated preparation (Pancrease, However, in the Johnson & Johnson, Piscataway, NJ) with a solid-liquid present study with Eurobiol25 000, delayed emptying of microtablets by itself was unlikely to account for the total absence of pancreatic enzyme activity in the duodenum during the entire 4-h postprandial period. Impairment in enzyme release from microtablets in the duodenum was probably the major factor. Indeed, in-vitro studies (Eurorga, unpublished data) have shown that time to complete release of pancreatic enzymes from Eurobiol 25 000 microtablets is 90 min at pH 5 and from 15 to 30 min at pH 6 depending upon ionic strength. Although pH at the ligament of Treitz remained above 6 in most duodeno-
376
J.-C. DELCHIER ef al.
jejunal samples in our 6 patients, pH in the upper part of the duodenum may well have been below 6, as it has previously been demonstrated in several studie~.'~-~' Moreover, transit time from pylorus to the ligament of Treitz after meal ingestion ~ apparent conhas been calculated to be about 5 min by Malagelada et ~ 1 . ' The tradiction between the absence of enzyme delivery at the ligament of Treitz and the improvement in steatorrhoea with Eurobiol 25000 can only be explained by the release of pancreatic enzymes further down in the small intestine. Comparison of our results with those of the study by Dutta et suggests that pancreatic enzymes are released more slowly from Eurobiol 25000 microtablets than from Pancrease microspheres. A comparative study is clearly needed to draw more definitive conclusions. Absence of lipolysis in the proximal duodenum could play a major role in the pathogenesis of the residual steatorrhoea in patients receiving enteric-coated pancreatic enzyme preparations. A recent study by Zerega ef aL3' seems to substantiate this hypothesis as the continuous instillation of pancreatic enzymes at the ligament of Treitz failed to abolish steatorrhoea in patients with pancreatic insufficiency. A significant increase in pancreatic enzyme delivery to the duodenum with Eurobiol coincided with a marked improvement in duodenal triglyceride absorption. Mean duodenal recovery of lipase was found to be especially high (29%) as compared with other conventional preparation^.^^^^'^^ This is probably related to Eurobiol's galenic form: as a powder, it blends well with food, resulting in concurrent emptying. Mean amounts of lipase reaching the duodenum exceeded the threshold of 10% of the normal value theoretically sufficient to ensure normal l i p ~ l y s i sHowever, .~~ in the present study as in a previous one,s Eurobiol did not normalize daily faecal fat. Comparison of enzyme and meal delivery time courses in duodenum may offer some explanation for this apparently paradoxical result (Figure I): enzymes were only delivered into the duodenum during the first two post-prandial hours, while gastric emptying of the meal took 4 h; the absence of enzymes in the duodenum during the last 2 h was due to intragastric inactivation by acid (Figure 3 ) . Moreover, considerable interindividual variations in lipase duodenal recovery were observed (range 3-80 YO), resulting in differences between patients in the control of steatorrhea. The rate of gastric emptying appeared to be a key factor in duodenal enzyme recovery ; when rapid, pancreatic enzyme inactivation by acid was prevented and duodenal recovery was high (patients B. R. and M. C.),By contrast, when gastric emptying was normal or slow (patients A. R., D. J., J. A., R. E.), intragastric survival and duodenal recovery of pancreatic enzymes depended on individual lcvels of gastric acid and secretion. This was particularly true for lipase. In this regard, results in patient R.E. were noteworthy; intragastric lipase activity decreased faster than chymotrypsin activity resulting in lower duodenal recovery of lipase (13 %) than that of chymotrypsin (54 Yo). Such in-vivo observations accorded with previous in-vitro data.33 neither of the pancreatic In agreement with the study by DiMagno ef enzyme preparations influenced meal emptying. Bile salt concentration in the
PANCREATIC ENZYMES IN C H R O N I C PANCREATITIS
377
duodenum was unexpectedly lower with placebo than with Eurobiol or Eurobiol 25 000 during the second half of the sampling period. As the gastric emptying rate and duodenal pH were not influenced by treatment, this result suggests an indirect effect of pancreatic extracts on choleresis or gallbladder emptying. In conclusion, results of the present study provide some explanation for the incomplete correction of steatorrhoea observed in patients with severe pancreatic insufficiency treated with conventional as well as with pH-sensitive enteric-coated pancreatic enzyme preparations. When using the conventional preparation, steatorrhoea may persist, despite a theoretically sufficient duodenal recovery of administered enzymes, because part of the meal is delivered into the duodenum without enzymes as a result of enzyme inactivation by intragastric acid. Using the latter preparation, steatorrhoea may remain incompletely corrected, despite effective enzyme protection against acidity, because enzyme release from entericcoated microtablets is too slow for the small intestine to carry out complete lipid absorption. Our results suggest that effectiveness of oral pancreatic enzyme supplementation could be enhanced by preparations combining protected and unprotected enzymes. ACKNOWLEDGEMENTS
The authors wish to thank Laboratoires Eurorga (Fresnes, France) for their financial support. REFERENCES DiMagno E P. Medical treatment of exocrine pancreatic insufficiency. Dig Dis Sci 1982; 27: 481-4. FCdCration Intemationale Pharmaceutique -International commission for the standardization of Pharmaceutical enzymes, Fourth Report. J Mond Pharm; 1968: 3-11. Delchier J C, Moulin C, Dutreuil C, Soul6 J C. Stability of enzyme activities in pancreatic extracts incubated at 37 "C. Dig Dis Sci 1987; 32: 1163A. (Abstr.) Saint-Marc Girardin M F, Moreau J, et al. Duodenal availability of Eurobiol versus placebo in patients with pancreatic insufficiency (abstr). Dig Dis Sci 1986; 31: 347s. Saint-Marc Girardin M F, Cortot A, ThCdore C ef al. Effkacit6 du pancrCas total lyophylisC au cours de la pancrCatite chronique et des rCsections pancrCatiques. Gastroenterol Clin Biol 1987; 11: 430-1. Staub J L, Sarles H, Soul4 J C, Galmiche J P, Capron J P. No effect of cimetidine on the therapeutic response to oral enzymes in
severe pancreatic insufficiency. N Engl J Med 1981; 304: 1364-5. 7 Dutta S K, Rubin J, Harvey J. Comparative evaluation of the therapeutic efficacy of a pH-sensitive enteric-coated enzyme preparation with conventional pancreatic enzyme therapy in the treatment of exocrine pancreatic insufficiency. Gastroenterology 1983; 84: 476-82. 8 Graham D Y. An enteric-coated pancreatic enzyme preparation that works. Dig Dis Sci 1979; 24: 906-9. 9 Halgreen H, Pedersen N T, Worning H. Symptomatic effect of pancreatic enzyme therapy in patients with chronic pancreatitis. Scand J Gastroenterol 1986; 21 : 104-8. 10 Salen G, Prakash A. Evaluation of entericcoated microspheres for replacement therapy in adults with pancreatic insufficiency. Curr Ther Res 1979; 25 : 650-6. 11 Valerio D, Whyte E H A, Schlamm H T, Ruggerio J A, Blackburn G L. Clinical effectiveness of pancreatic enzyme supplement. J Parent Enterol Nutr 1981; 5 : 110-4. 12 Lankisch P G, Lembcke B, Goke B,
378
13
14
15
16
17
18
19
20
21
22 23
24
J.-C. DELCHIER et al.
Creutzfeld W. Therapy of pancreatogenic steatorrhea: does acid protection of pancreatic enzymes offer any advantage? Z Gastroenterol 1986; 24: 753-7. Schneider M V, Knoll-Ruzicka L, Domschke S, Heptner G, Domschke W. Pancreatic enzyme replacement therapy: Comparative effects of conventional and enteric-coated microspheric pancreatic and acid stable fungal enzyme preparation on steatorrhea in chronic pancreatitis. Hepatogastroentero b g y 1985 ; 32 : 97-102. Delchier J C, SoulC J C. BT-PABA with plasma PABA measurements : evaluation of sensitivity and specificity. Gut 1983; 24: 318-25. Malagelada J R, Longstreth G R, Summerskill W H J, Go V L W. Measurement of gastric functions during digestion of ordinary solid meals in man. Gastroenterology 1976; 70: 203-10. Vidon N, Muschart J M, Cosnes J, Ruskone A, Bernier J J. Etude critique de l'estimation de la vidange gastrique par la mkthode de perfusion duodknale dune substance non absorbable h faible dCbit. Gastroenterol Clin Biol 1979; 3: 549-52. Hyden S. A turbidimetric method for the determination of higher polyethylene glycols in biological materials. Ann R Agr Coll (Sweden) 1956; 22: 139-45. Talalay P. Enzymatic analysis of steroid hormones. Methods Biochem Anal 1960; 8: 129-43. Blankenhorn D H, Arhens E H. Extraction, isolation and identification of hydrolytic products of triglyceride digestion in man. J Biol Biochem 1955; 212: 69-81. Dole V P. A relation between nonesterified fatty acids in plasma and the metabolism of glucose. J Clin Invest 1956; 35 : 150-4. Van de Kamer J H, Bokkell T, Huinink H T B, Wyers H A. Rapid method for determination of fat in feces. J Biol Chem 1949; 177: 347-55. Zar J H. Biostatistical analysis. Englewood Cliffs, NJ: Prentice Hall, Inc., 1974. Hoffmann A F. The function of bile salts in fat absorption: The solvent properties of dilute micellar solutions of conjugated bile salts. Biochem J 1963; 89: 57-68. Saint-Marc Girardin M F, Cortot A,
25
26
27
28
29
30
31
32
33
34
Theodore C, ef al. Eurobiol, 25 000 U, nouvel extrait pancrdatique gastroprotkgk: son efficacitk dans la pancrdatite chronique et dans les resections pancreatiques. Gastroenterol Clin Biol 1989; 13 : A185. (Abstract.) Meyer J H, Elashoff J, Porter-Fink V, Dressman J, Amidon C L. Human postprandial gastric emptying of I-3-millimetre spheres. Gastroenterology 1988; 94: 1315-25. DiMagno E P, Malagelada J R, Go V L W, Moertel C G. Fate of orally ingested enzymes in pancreatic insufficiency : Comparison of two dosage schedules. N Engl J Med 1977; 296: 1318-22. Regan P T, Malagelada J R, DiMagno E P, Glanzman S L, Go V L W. Comparative effects of antacids, cimetidine, and enteric coating on the therapeutic response to oral enzymes in severe pancreatic insufficiency. N Engl J Med 1977; 297: 854-8. Dutta S, Russell R M, Iber F L. Influence of exocrine pancreatic insufficiency on the intraluminal pH of the proximal small intestine. Dig Dis Sci 1979; 24: 529-34. Dutta S, Russell R M, Iber F L. Impaired acid neutralization in the duodenum in pancreatic insufficiency. Dig Dis Sci 1979; 24: 77580. Bommelaer G, Benque A, Moreau J, Bouisson M, Ribet A. pH-mCtrie duodenale durant 24 heures dans les pancriatites chroniques. Gastroenterol Clin Biol 1984; 8: 403-6. Zerega J, Lerner f, Meyer J H. Duodenal instillation of pancreatin does not abolish steatorrhea in patients with pancreatic insufficiency. Dig Dis Sci 1988; 33: 1245-9. DiMagno E P, Gos V L W, Summerskill W H J. Relation between pancreatic enzyme outputs and malabsorption in severe pancreatic insufficiency.N Engl J Med 1973; 288 : 813-5. Heizer W D, Cleveland C R, Iber F L. Gastric inactivation of pancreatic supplements. Bull Johns Hopkins Hosp 1965; 116: 26170. DiMagno E P, Clain J E. Chronic pancreatitis in the exocrine pancreas. In: Go V L W, Gardner J D, Brooks F P, et al. eds. Biology, pathobiology and diseases. New York: Raven Press, 1986; 541-75.
