Clinical Nutrition (1995) 14:249 -253 © Pearson Professional Ltd 1995
The effect of intravenous lipids on gastric emptying in rats subjected to total parenteral nutrition J. J. AZNAR, E. MART[, I. PERKINS, J. VARO, F. SOLANA and V. MONZO Research Centre, Valencia University General Hospital, Tres Cruces s/n, 46014-Valencia, Spain (Correspondence to: IP, Centro de Investigaciones Biotecnol6gicas, Hospital General Universitario, Tres Cruces s/n, 46014 Valencia, Spain) A B S T R A C T - - Total parenteral nutrition (TPN) is indicated in a number of clinical situations involving obstructive disorders of the gastrointestinal tract or when total bowel rest is required. However, little is known of its effects upon the physiology of gastric storage and emptying. A study of the effects of lipid nutrient content of TPN on the gastric emptying of a non-energy liquid test meal in male and female conscious rats (250-280 g) was undertaken. Five experimental groups were set up according to the percentage of total non-protein energy administered in the form of different lipids: A (lipid-free: 100% of non-protein energy as glucose); B (5% long-chain triglycerides (LCT), 95% glucose); C (40% LCT, 60% glucose); D (5% LCT and medium-chain triglycerides (MCT) in 1:1 proportion, 95% glucose); and E (40% LCT and medium-chain triglycerides (MCT) in 1:1 proportion, 60% glucose). Animals were maintained on TPN for 24 h and for 3 and 4 days, after which gastric emptying was measured by the phenol red method. 40% lipid TPN was found to accelerate gastric emptying as infusion was prolonged, although the type of lipid had no effect.
TPN on the GE of a non-energy containing liquid test meal in conscious rats.
Introduction Lipid emulsions are widely used in total parenteral nutrition (TPN) as an energy source and in preventing fatty acid deficiencies (1, 2). Long-chain triglycerides (LCT) were the lipids initially used in TPN (3, 4), although in recent years medium-chain triglycerides (MCT) have been included as a rapidly oxidised energy substrate in mixed LCT/MCT emulsions (5, 6). Gastric emptying (GE) is known to slow down with the inclusion of lipids in the diet (7-11). However, the effects of parenterally administered lipids are less certain. In a review of the literature only one study (12) was found which related TPN lipids to GE, in humans. Tc99 marker monitoring of emptying showed a slight delay in GE (18%) with no changes in either lower esophageal sphincter pressure or gastrointestinal hormone levels. However, in other organ systems the intravenous administration of lipids was generally found to promote motility. Important increases (37%) were noted in human gallbladder motility (13), although the latency period between the start of infusion and contractile manifestation was 26 h, i.e. the effect was not immediate. The aim of the present study was to investigate the influence of lipid nutrient type and concentration in
Materials and methods
Animals Male and female Wistar rats weighing 250-280 g were randomly selected and housed under controlled conditions (25°C, 60-70% relative humidity and 12-h light:darkness cycles). Animal handling at all times abided with European Community ethical standards (86/609/CEE). Food in the form of pellets (A04 Panlab S.L., Barcelona, Spain) and water were provided ad libitum.
Surgical technique The animals were fasted for 24 h before the experiment. Anaesthesia was performed with sodium pentobarbital (30 mg/kg i.p.). The left jugular vein was catheterized with an inert silicon catheter according to a slightly modified version (14) of the technique described by Steiger et al (15). The infusion device was protected by a miniature 249
250 INTRAVENOUS LIPIDS AND GASTRIC EMPTYING
single-channel fluid swivel (# 050-0020) attached to the animal by a rodent jacket (both from Alice King Chatham Medical Arts, Hawthorne, Califomia USA). The swivel head consisted of two interrotating elements and a hollow steel spring through which the catheter was introduced as a protective measure against biting. The swivel was fixed to a rack on the outside of the cage without hindering free movement of the animal. One end of the catheter was linked to the swivel head, and the other was connected to a sonde and syringe containing the feeding formulation. The animals were placed in metabolic cages and allowed to recover for 48 hours wi~ isotonic glucose-saline infusion, after which TPN was initiated, with water (p.o.) ad libitum.
Infusion groups The rats were randomly divided into 5 groups (A to E) (n>30 in each group) as a function of TPN composition (Table 1). Group A (100% non-protein energy provided in the form of glucose) was used as the control. All 5 nutritional formulae were isoenergetic (800 kJ/kg/day) and isonitrogenous (1.36 g Nz/kg/ day: Synthamin 14, Baxter, Valencia, Spain), and were prepared under laminar flow and sterile conditions. The proportion of lipids (0%, 5% and 40% of total non-protein energy) and type (LCT (Intralipid, Pharmacia AB, Sweden) or combined MCT/LCT (Lipofundin MCT/LCT, B. Braun Medical, Barcelona, Spain)) differed across groups (Table 1). Electrolyte and vitamin concentrations were identical in all 5 groups (Table 2). Each group was divided into three subgroups, corresponding to three different infusion times (24 h, 3 days and 5 days). We determined gastric emptying in at least 10 rats in each subgroup (i.e. a total of at least 30 rats from each TPN group). The TPN formulae were continuously administered by a Harvard
Table 1 Characteristics of the different parenteral nutrition formulae used
Group A (control) B C D E
100% 95% 60% 95% 60%
5% 40% 2.5% 20%
Values correspond to percentage of total non-protein energy. All 5 parenteral nutrition formulae are isoenergetic (800 kJ/kg/day) and isonitrogenous (1.36 g N2/kg/day). LCT: long-chain triglycerides, MCT: medium-chain triglycerides.
Table 2 Main electrolytes and other micronutrients provided daily by TPN Component Na ÷ K÷ Ca 2+ Mg 2÷ PO4HzC1Multivitamins * Oligoelements **
Concentration 30.88 25.31 6.60 4.23 12.69 29.61
mEq/1 mEq/1 mEq/l mEq/l mmol/1 mEq/1
* Soluvit and Vitalipid (Pharmacia Iberia, Spain). ** Addamel N (Pharrnacia Iberia, Spain)
Apparatus pump (Model 2265) with 50 ml syringes at an infusion rate of 2 ml/h. The animals were weighed before and after infusion. Nitrogen balance and diuresis were recorded at 24-h intervals to follow-up the condition of the animal. Nitrogen balance was estimated as 24-h urea excretion in urine, compared with net nitrogen provided by TPN (16).
Measurement of gastric emptying Gastric emptying was evaluated following the method described by Scarpignato, based on a suspension of 50 mg phenol red in 100 ml aqueous carboxymethyl cellulose (1.5%) (,17). The non-energy containing liquid test meal was administered orally, with the animal in a vertical position, through an orogastric cannula (1.5 ml). After 20 min the rats were sacrificed by cervical dislocation. A midline laparotomy was performed to expose the stomach; the pylorus and cardias were rapidly clamped, and the stomach was removed and homogenized together with its contents in 50ml NaOH (0.1N) for 5min. l m l of 20% trichloroacetic acid was added to a 10 ml sample of the homogenate to preciptate the protein. Following centrifugation at 5000 rpm for 10 min, the supematant was removed. A 5 ml sample was added to an equal volume of borate buffer (pH 10) to return the solution to the phenol red colour pH range (the medium had initially been acidified by the addition of trichloroacetic acid). Posteriorly phenol red content was quantified spectrophotometrically at 560 nm (Syva S-III). GE was expressed as the percentage of dye emptied after 20 min with respect to the amount administered at time zero.
Statistical analysis All experimental data exhibited a normal distil-
CLINICAL NUTRITION 251
bution (Kolmogorov-Smirnov) and variance equality (Bartlett). A one-way analysis of variance (ANOVA) was performed using mean values (+ SEM) for comparing the different experimental groups and controls. The F-Scheff6 test was applied for multiple comparisons. ANOVA was performed for comparing the daily analytical results between groups. Statistical significance was considered when tx = 0.05.
Results The results relating to GE are shown in Table 3. After TPN for 24 hours, GE was found to be similar in all groups. Differences in the amount and type of lipid did not significantly modify short-term emptying. After 3 days of TPN, GE was found to be greater in the high lipid groups (C and E) than in the control groups (by 17.3 and 11.9%, respectively), although the differences were not significant. The other groups showed almost no difference compared with the control groups. After 5 days of TPN, GE in Groups C and E was significantly greater than in the controls given only glucose as energy source (p < 0.05). Differences averaged 36.1 and 33.6%, respectively (Table 3). These increases in GE were found to be timedependent within each group. In the animals given 40% of total non-protein energy as lipids, GE was found to be significantly accelerated after 5 days compared to the values at 24 h. The effects of lipids in TPN were therefore not immediately apparent and the longer term effects were dependent upon the duration of TPN and on the proportion of lipids administered. When GE between groups on TPN with similar lipid percentages but of different type (LCT versus LCT/MCT) was compared, no significant differences were found. Without correcting for non-quantifiable losses, the results showed no significant changes in nitrogen balance in the groups. This was slightly positive,
with respect to nitrogen provided in the TPN - as reflected by 24-h urinary urea output. The differences between groups were likewise non-significant. Nitrogen balance could not be used to establish the superiority of any single TPN composition. However, the inclusion of MCT appeared to favour anabolism, since nitrogen balance in these groups was nonsignificantly more positive than in the other groups. No significant differences in diuresis were noted between groups or TPN times. Diuresis was increased with respect to normal reference values in animals not subjected to TPN. This may be attributed to the volume of liquid required for parenteral nutrition in the rat (48 ml/day) to cover all nutritional requirements.
Discussion The results showed that TPN with a low (5%) proportion of total non-protein energy given as LCT or LCT/ MCT failed to modify GE after any of the three infusion periods compared to the control rats. A statistically significant increase in the GE of a non-energy containing liquid test meal was only obtained when the TPN lipid content represented 40% of the total non-protein energy (Groups C and E) and was given for 5 days. After 3 days, although an increase was observed, this was not statistically significant. This effect, whereby lipids in TPN accelerate GE, contrasts with the known emptying retarding effect of orally administered lipids (7-11). The results reported by Casaubon et al in humans (12), differ from ours. When monitoring GE with Tc99, the authors observed a slight (18%) delay in emptying, with no modifications in lower esophageal sphincter pressure or in gastrointestinal hormone levels. This discrepancy could be due to the different species studied, and possibly to the acute nature of the human experiments, with GE evaluated after only 12 hours of infusion. Moreover, the fipid concentrations infused were considerably lower than those that yielded significant results in our study.
Mean values of gastric emptying (_+ SEM) in rats subjected to TPN (800 kJ&g/ day; 1.36 gN2/kg/day)
Gastric emptying (%)
Group 24 h A (control) B C D E
49.8_+ 6.8 (n 49.1+ 8.0 (n 37.1 + 5 . 1 ( n 36.6 -+ 9.7 (n 43.3 + 8.2 (n
3 days = = = = =
10) 10) 10) 10) 10)
53.1 _+5.3 ( n = 52.8 + 5.1(n = 62.3_+ 4.3 (n = 51.7 _+7.9 (n = 59.4 + 7.9 (n =
5 days 10) 11) 10) 12) 11)
56.8_ 6.1(n = 13) 57.9_+ 5.2 (n = 14) 77.3 ± 3.3" (n = 15) 60.8 _+6.6 (n = 12) 75.9 _+2.8* (n = 15)
Values are mean + SEM * p < 0.05 with respect to the controls (Group A)
252 INTRAVENOUSLIPIDSAND GASTRICEMPTYING The discrepancies in the effects of orally or parentally administered lipids suggests different u n d e r l y i n g m e c h a n i s m s of action. The mediators in the effects of orally administered lipids are k n o w n , and are ascribed to chemoreceptors in the proximal b o w e l (18--21). However, little has b e e n established regarding the m e c h a n i s m s i n v o l v e d in the effect of parenterally administered lipids o n gastric emptying. A n u m b e r of studies suggest that i n t r a v e n o u s l y infused lipids exert an effect o n the motility of different tissues, including the h u m a n gallbladder (13). A given substance also frequently exerts different effects d e p e n d i n g on the w a y in which it is administered, for example, oral lipids inhibit G E and generate diarrhea through an osmotic process that is probably mediated b y an overproduction of prostaglandins (22). In dogs the oral administration of lipids increases total thoracic duct l y m p h v o l u m e - an effect not observed with parenteral dosing (23). It has b e e n postulated that hormones, in particular cholecystokinin and motilin, m a y promote gastrointestinal motifity (24). To our knowledge, the possible influence of lipids in T P N on h o r m o n e s has not b e e n elucidated. However, it has recently b e e n suggested that the addition of lipid to T P N in conscious rats depresses i n s u l i n concentration (25). As i n s u l i n has b e e n found to disrupt migrating motor complexes (26), the possibility that it might exert an influence u p o n G E c a n n o t be discounted. In the present study we observed n o differences in the effects o n G E of e m u l s i o n s c o n t a i n i n g 40% of total n o n - p r o t e i n energy in the form of either L C T or c o m b i n e d L C T / M C T . Results of other studies are conflicting. A c c o r d i n g to M e y e r (27), the effects of enterally-administered M C T are more potent than LCT, particularly at pH7. Siegel et al (28) f o u n d in premature infants that L C T exerted a greater effect u p o n G E than M C T . H u n t and K n o x (29) observed a more potent effect of M C T . Despite the differing results there is agreement that both types of triglyceride possess a slowing effect on G E w h e n given orally. The accelerating effect of high lipid T P N o n G E could facilitate transitional feeding to enteral nutrition and help to avoid associated aspiration and regurgitation. Further research in the clinical setting is required to investigate this.
4. 5. 6. 7. 8. 9.
14. 15. 16. 17. 18.
References 1. Record K E, Kolpeck J H, Rapp R P. Long-chain versus medium-chain triglycerides. Review of their metabolism and clinical use. Nutrition in Clinical Practice 1986; 1 (3): 129-135. 2. Cervera P. Nutrici6n parenteral total. In: Marti E, Romero J A,
eds. Nutrici6n artificial hospitalaria y enfermerfa. IIaJomadas de Nutrici6n Hospitalaria y Enfermerfa 1991; 7: 119-135. Me Fie J, Smith R C, Hill G L. Glucose or fat as a non-protein energy source? A controlled clinical trial in gastroenterological patients requiring intravenous nutrition. Gastroenterology 1981; 80 (1): 103-107. Mc Fie J, Courmey D, Salter M C P, Brennan T G. Continuous versus intermittent infusion of fat emulsions during i. v. nutrition. Br J Surg 1985; 72:398 (Abstract). Johnson R C, Cotter R. Metabolism of medium-chain triglyceride lipid emulsion. Nutr Intemat 1986; 2: 150-158. Bach A C, Frey A, Lutz A. Clinical and experimental effects of medium-chain triglyceride-basai fat emulsions. A review. Clinical Nutrition 1989; 8: 223-235. Keinke O Ehrlein H. J. Effect of oleic acid on canine gastroduodenai motility, pyloric diameter, and gastric emptying. J Exp Physiol 1983; 68: 675-686. Groisman S D, Kharclienko N M. Characteristics of the evacuatory function of the stomach when fat is added to the food. Fiziol Zh SSSR 1989; 75 (2): 239-244. Heddle R, Collins P J, Dent Jet al. Motor mechanism associated with slowing of the gastric erupting of a solid meal by an intraduodenal lipid infusion. J Gastroenterol Hepatol 1989; 4 (5): 437-447. Stacher G, Bergman H, Ganpmann Get al. Fat preload delays gastric emptying: reversal by cisapride. Br J Clin Pharmacol 1990; 30 (6): 839-845. Stacher G, Granser G V, Bergman H, Kugi A, Stacher-Janotta G, Hoebart J. Slow gastric emptying induced by high fat content of meal accelerated by cisapride administered rectally. Dig Dis Sci 1991; 36 (9): 1259-1265. Casanbon P R, Dahlstrom K A, Vargas J, Hawkins R, Mogard M, Ament M E. Intravenous fat emulsion (intralipid) delays gastric emptying, but does not cause gastroesophageal reflux in healthy volunteers. JPEN 1989; 13 (3): 246-248. Kalfarentzos F, Vagenas C, Michall Aet al. Gallbladder contraction after administration of intravenous amino acids and long-chain tfiacylglycerols in humans. Nutrition 1991; 7 (5): 347-349. CAnerM, Meguid M M. Effect of intravenous or intragastric nutrients on food intake in rats. J Surg Res 1991; 51: 259-266. Steiger E, Vats H M, Dudrick S J. A technique for long-term intravenous feeding in unrestrained rats. Arch Surg 1972; 104: 330-332. Faweet J K, Scott J E. A rapid and precise method for the determination of urea. J Clin Pathol 1960; 13: 156-159. Scarpignato C, Coruzzi C, Bertaccini G. Effect of histamine and related compounds on gastric emptying of the conscious rat. Pharmacology 1981; 23: 185. Lawaetz O, Bloom S R, Stimpel H, Siemssen O J. Effect of soya bean oil on symptoms, gastric emptying and gut hormone release in patients with postvagotomy symptoms. Ann Chir Gynaecol 1986; 75 (6): 308-313. Fioramonti J, Fargeas M J, Bueno L. Involvement of endogenous opiates in regulation of gastric emptying of fat meals in mice. Am J Physiol 1988; 255 (2): G158-G161. Mogard M H, Maxwell V, Wong H, Reedy T J, Sytnik B, Walsh J H. Somatostatin may not be a hormonal messenger of fat-induced inhibitionof gastric functions. Gastroenterology 1988; 94 (2): 405-408. Morgan L M, Tredger J A, Hampton S M, French A P, Peake J C, Marks V. The effect of dietary modification and hyperglycaemia on gastric emptying and gastric inhibitory polypeptide (GIP) secretion. Br J Nutr 1988; 60(1): 29-37. Gottschilich M M, Warden M Aet al. Diarrhea in tube-fed
CLINICALNUTRITION 253 burn patients: incidence, etiology, nutritional impact, and prevention. JPEN 1988; 12: 338. 23. Andel M, Nozickova M, Bartos V. The effect of intravenous administration of fat emulsion on the flow rate and composition of thoracic duct lymph in the dog. Clinical Nutrition 1989; 8: 75. 24. Weisbrodt N W. Motility of the small intestine. In: Johnson L R, ed. Physiology of the Gastrointestinal Tract, 2nd ed. New York: Raven Press, 1987: 1: 631-663. 25. Nussbaum M S, Shujun L, Bower R H, McFadden D W, Dayal R, Fischer J E. Addition of lipid to total parenteral nutrition prevents hepatic steatosis in rats by lowering the portal venous insulin/glucagon ratio. JPEN 1992; 16: 106-109.
26. Bueno L, Ruckebusch M. Insulin and jejunal electrical activity in dogs and sheep. Am J Physiol 1977; 230: 1538-1544. 27. Meyer J H. Motility of the stomach and gastroduodenal junction. In: Johnson L R, ed. Physiology of the gastrointestinal tract, 2nd ed. New York: Raven Press, 1987; 1: 613-629. 28. Siegel M, Krantz B, Lebenthal E. Effect of fat and carbohydrate composition on the gastric emptying of isocaloric feedings in premature infants. Gastroenterology 1983; 89 (4): 785-790. 29. Hunt J N, Knox M T. The slowing of gastric emptying by nine acids. J Physiol 1972; 201: 161-179.
Submission date: 20 January 1995; Accepted after revision: 8 May 1995