Use of l -glutamine in total parenteral nutrition

JOURNAL OF SURGICAL RESEARCH 44,

506-513(1988)

Use of L-Glutamine in Total Parenteral Nutrition J. P. GRANT, M.D., AND P.J. SNYDER,B.A. Department of Surgery, Duke University Medical Center, Durham, North Carolina 27710 Presented at the Annual Meeting of the Association for Academic Surgery, Orlando, Florida, November l-4, 1987 Gut atrophy develops during prolonged total parenteral nutrition (TPN). TPN solutions do not contain glutamine, an energy substrate of the intestinal tract. This study evaluated the effect of addition of L-glutamine to TPN on gut nitrogen content, histology, and disaccharidase enzyme activity. Five groups of six Fisher 344 rats received rat chow, D5W, TPN (23% calories as lipid), or TPN with 1 or 2% L-glutamine. Animals given TPN received 30 kcal and 0.22 g nitrogen/100 g/day. Metabolic cages allowed nitrogen balance for each group. After 6 days infusion, stomach, small bowel, and colon were assayed for total nitrogen and sucrase,lactase, and maltase activity. Mucosal height and fatty infiltration of the liver were determined from histologic sections. Adding either 1 or 2% L-glutamine resulted in no toxic clinical effects. Glutamine preserved intestinal nitrogen content of the stomach and colon compared to standard TPN and increased nitrogen content of small bowel to greater than that in chow-fed animals. Glutamine maintained mucosaf height of the stomach and colon, but was no better than TPN alone in maintenance of small bowel mucosal height. One percent glutamine increased and standard TPN depressed maltase activity compared to chow. Standard TPN and 1% glutamine both stimulated sucraseand lactase activity compared to chow. Addition of 1 or 2% glutamine protected the liver from fatty infiltration seen with standard TPN. These studies would suggest the addition of glutamine might be beneficial during provision of standard total parenteral nutrition. Q 1988 Academic Fvess.Inc.

coid administration the gut switches from an organ of glucose uptake to one of slight reProvision of nutrients solely by total par- lease [ 1I]. Circulating ketones appear to be enter-al nutrition leads to progressive gut at- metabolized by enterocytes but the imporrophy with loss of mucosal thickness and vil- tance of these fuels is speciesspecific [9]. The lous height [2, 51. Transition to an oral diet dog enterocyte useslittle ketone bodies while after a prolonged period of total parenteral the rat enterocyte appears to readily use nutrition is often associated with diarrhea them. Colonocytes can utilize short-chain due to malabsorption. The progressive atro- fatty acids (n-butyrate) as well as ketone phy is reversible with institution of an enteral bodies [6]. Nonetheless, the large bowel, like feeding program or an oral diet and has not the small intestine, has a relatively high rate been observed with long-term enteral feed- of glycolosis with more than 50% of the gluing. Some of the atrophic changes can be re- cose utilized appearing as lactate. The major duced when the parenteral nutrition solution metabolic fuel of both the rat enterocyte and is infused intragastrically instead of intrave- the rat colonocyte appears, therefore, to be nously but only to a limited degree [8]. Re- the amino acid glutamine. Up to 20 to 30% cent studies of gut metabolism in the experi- of plasma glutamine is extracted with each mental animal have suggested that the gut circulation through the mesenteric system. prefers as its major fuel the amino acid glu- Windmueller demonstrated that approxitamine [ 10, 131. Although substantial mately 64% of the glutamine carbon was oxamounts of glucose are taken up by mucosal idized in the gut to CO2 with the rest appearcells, most appears to be metabolized only to ing as lactate, citrate, and several different pyruvate. Indeed during stressor glucocorti- amino acids [ 121. The glutamine nitrogen INTRODUCTION

0022-4804/88$1.50 Copyright 8 1988 by Academic Press,Inc. All rights of reprodwtion in any form reserved.

506

GRANT AND SNYDER: L-GLUTAMINE TABLE 1

Glucose

Fat

Protein

507

function as reflected by brush border disaccharidase enzyme activity.

FEEDINGFORMULATIONS (GRAMSPER 100 RESOLUTION) Group

IN TPN

Glutamine

METHODS

Fisher 334 rats weighing between 180 and 200 g were acclimatized for 1 week in metabolic cagesand then divided into five groups lC of six animals each. All animals were allowed TPN2% 2’ ad Zibitum water. The National Research Council’s guide for the care and use of laboa Rodent Laboratory Chow No. 500 1, Ralston Purina ratory animals was carefully followed. Group Co., St. Louis, MO. 1 was given free access to standard rodent b Liposyn 20% and Aminosyn 8.5%, Abbott Laboratories, North Chicago, IL. laboratory chow (Purina No. 500 1). Group 2 ’ Sigma Chemical Co., St. Louis, MO. received 5% dextrose with added electrolytes and vitamins. Group 3 received intravenous total parenteral nutrition via an indwelling appears in ammonia, alanine, citrulline, and internal jugular catheter which provided 23% proline. The gastrointestinal tract is ideally of the nonprotein caloric load as lipid. suited to utilize glutamine as an energy Groups 4 and 5 received the same parenteral source as it contains the highest glutaminase nutrition solution except Group 4 had 1% of activity of all tissues and the by-product of the amino acid content replaced with L-gluglutamine metabolism, ammonia, passes tamine and Group 5 had 2% of the amino into the portal vein where it can be rapidly acid content replaced. The feeding formulas and amount of nutrients each group received detoxified by the liver. Current parenteral nutrition solutions do are given in Tables l-3. All animals receiving intravenous solunot contain glutamine or glutamate contrary to enteral feeding products and it is proposed tions underwent superior vena caval cannuthat this might in part explain the substantial difference in gut histology when the two TABLE 2 routes of nutritional support are utilized. In starvation and during stress, even with the VITAMINSANDELECTROLYTESINI LITER infusion of total parenteral nutrition, glutaOFIVSOLUTIONS mine is releasedby skeletal muscle and liver. Component TPN, TPN 1 and 2% D5W Uptake by the gastrointestinal tract and kidneys, however, is greatly increased and Na (mes) 55 38 serum concentrations of glutamine are typi- K (meq) 72.5 22 cally depressed. The increased dependency Cl (me@ 104 60 0.8 0 of the gastrointestinal tract on glutamine as Mg @es) Ca @es) 4.6 0 an energy substrate during starvation and Phosphate (g) 0.5 0 stressand the relatively low serum glutamine MVI-12 (ml) 10 10 concentrations have led to the proposition Vitamin K (mg) O.OZ/week O.OZ/week that gut atrophy is the result of glutamine a Armour Pharmaceutical Co., Blue Bell, PA, 10 ml deficiency. contains ascorbic acid 100 mg, vitamin A 3300 IU, vitaThis study was undertaken to evaluate the min D 204 IU, thiamine 3.0 mg, riboflavin 3.6 mg, pyriimpact of addition of L-glutamine to total doxine HCI 4.0 mg, niacinamide 40.0 mg, pantothenic parenteral nutrition (TPN) solutions in acid 15.0 mg, vitamin E 10 IU, biotin 60 mcg, folic acid . Bi2 5 mcg. maintenance of gut histology as well as gut 400 mcg, and vitamin Choti D5W TPN TPN 1%

52 5 34 34 34

4.5 0 5.66 5.Q 5.6’

22 0 4.25b 3.25’ 2.25b

5.3 0 0

508

JOURNAL OF SURGICAL RESEARCH: VOL. 44, NO. 5, MAY 1988

lation under sterile conditions as per the technique of Popp [7]. The catheter was tunneled subcutaneously to the back of the neck and exited through a coil spring which was attached to a swivel allowing free mobility of the animal inside the metabolic cage. Metabolic cages allowed collection of all excrement for determination of nitrogen balances. The environment was maintained at 37°C with a 12-hr day-night cycle of artificial light provided. L-Glutamine was obtained from Sigma Chemical Co., St. Louis, Missouri, and was sterilized by passage through a 0.22~pm Swinnex-GS filter just prior to addition to the nutrient solutions. The parenteral nutrition solutions were formulated 12 hr in advance of usageas a 3-in- 1 mixture and stored at 4°C until used. They were then hung at room temperature and infused over 48 hr. Differences in weight at the time hung and time taken down was used to determine total solution infused. Twenty-four-hour urines were collected on Day 4 and Day 5 of infusion for determination of nitrogen balance. Total urinary nitrogen was measured by a modified Kjeldahl technique. After 6 days, all animals were anesthetized with intraperitoneal injection of phenobarbital and weighed, and their abdominal cavities were opened. Each animal was exsanguinated by drawing arterial blood from the aorta with the blood pooled for each group for determination of blood ammonia. Blood ammonia was determined calorimetrically by the DuPont ACA procedure. After exsanguination, the stomach, small bowel, and

TABLE 3

Group Chow

DSW TPN TPNl% TPN296

Calories

Nit(g)

29.0

0.30

7.7 21.3 28.5 27.8

0 0.22 0.23 0.22

Glu (g) Fat(g) 4.5 2.2 11.5 11.9 11.7

3.9 0 1.9 1.9 1.9

AA (g) GLN (9) 1.9 0 1.4 1.4 1.4

0.5 0 0 0.35 0.7

TABLE 4

Group

Blood ammonia k&W

Chow D5W TPN TPN 1% TPN 2%

144 383 119 110 92

Nitrogen balance Weight OWday) W-W 170 -120* 100 30* 30*

1

-49* 12* 3 5

* P < 0.05 compared to chow fed.

colon were dissected free and each was assayed for total nitrogen content. From one animal in each group a small segment of the mid-stomach along the greater curvature, the proximal jejunum 3 cm from the tail of the pancreas, and the mid-transverse colon were taken and placed in formaldehyde. After fixation, histologic slides were prepared with H and E stains for determination of mucosal height. Another segment of small intestine 3.5 cm from the tail of the pancreas was taken for determination of mucosal sucrase, lactase, and maltase activity. Each segment was placed in 5 ml of normal saline and frozen until disaccharidase analysis could be performed. Each segment was weighed and homogenated just prior to analysis. The disaccharidases(sucrase, lactase, and maltase) were analyzed and reported in units per milligram nitrogen of the small intestine using the calorimetric method described by Dahlqvist [I]. Finally, liver biopsies were taken and placed in formaldehyde for histologic evaluation, preparing slides with H and E staining. All statistical analyses were by two-tailed Student’s t test. RESULTS

Nitrogen Balance and Gut HL~olqqy Table 4 shows changesin blood ammon* nitrogen balance, and animal weight for the five groups over the 6-day study period.

GRANT

AND

SNYDER: L-GLUTAMINE IN TPN

TABLES

509

TABLE6 MUCOSALHEIGHTOFINTESTINALSEGMENTS (mm X 100)

Group

Stomach

Small bowel

Colon

Group

Stomach

Small bowel

Colon

Chow D5W TPN TPN 1% TPN 2%

24 16* 17* 24 22

6.5 49 51 s5* 85*

35 25* 28* 33 34

Chow D5W TPN TPN 1% TPN 2%

48 5* 36* 41 57

80 35* 62* 58* 56*

36 31 28* 36 36

* P < 0.05 compared to chow fed.

Blood ammonia concentrations were highest in the intravenous 5% dextrose group and lowest in animals receiving intravenous nutrition with glutamine. Glutamine-supplemented animals had lower blood ammonia than non-glutamine supplemented although not statistically significant. Chow-fed animals had the best average nitrogen balance while animals receiving 5% dextrose intravenously had a dramatic negative nitrogen balance. Animals receiving 1 and 2% supplemented glutamine intravenous nutrition maintained a positive nitrogen balance but significantly less than that of either TPN without glutamine supplementation or the chow-fed group. The positive nitrogen balance was reflected in change in animal weight. Standard intravenous-fed rats had significant increase in weight compared to chow-fed animals averaging 12 g per 6 days. Table 5 depicts the nitrogen content of the stomach, small bowel, and colon averaged for the six animals in each of the five groups. Significant decreased nitrogen content was observed in all three segments of the small bowel in animals receiving 5% dextrose intravenously and in animals receiving standard total parenteral nutrition when compared to the chow-fed animals. Animals receiving 1 or 2% glutamine-supplemented TPN however showed no significant decrease in nitrogen content and indeed a significant increase in the small bowel nitrogen content compared to chow-fed animals.

* P < 0.05 compared to chow fed.

Table 6 summarizes mucosal height of the stomach, small bowel, and colon. Again, 5% dextrose and standard total parenteral nutrition resulted in a significant decreasein mucosal height compared to chow-fed animals which was prevented by the addition of glutamine in the stomach and colon. The small intestine, however, demonstrated a significant decreasein mucosal height compared to chow-fed animals whether or not glutamine was added to the parenteral nutrition solution.

DisaccharidaseActivity Table 7 reports the average disaccharidase activities for the one animal in each group, reported as units of activity per milligram nitrogen of small bowel and total units per entire small bowel. Supplementation with 2% glutamine demonstrated a slight decrease in disaccharidase activity compared to TABLE1 DISACCHARIDASEACTIVITYOFSMALLINTESTINE (UNITSPERMILLIGRAMNOFSMALLBOWEL/ UNITSPERENTIRESMALLBOWEL) Group

Sucrase

Lactase

Maltase

Chow D5W TPN TPN 1% TPN 2%

7114615 4112009 1X3/8058 80/6800 53/4505

6 l/3965 2811372 14817548 98/8330 3 l/2635

309/20,085 15711,693 17919,129 510/43,350 190/16,150

510

JOURNAL OF SURGICAL RESEARCH: VOL. 44, NO. 5, MAY 1988

chow-fed animals whereas 1% glutamine demonstrated an increased disaccharidase activity. Animals receiving nonsupplemented standard intravenous nutrition had markedly elevated sucrase and lactase, and normal maltase activities in spite of mucosal atrophy. Five percent dextrose significantly reduced activity of all three disaccharidases.

DISCUSSION

Approximately 21% of the amino acid composition of a standard rat diet consists of glutamine or glutamate. This high concentration is readily utilized by the gastrointestinal tract metabolizing much of it to carbon dioxide and water with the releaseof ammonia and alanine into the portal blood stream. Except in the presence of severe liver injury, the ammonia is readily detoxified. This study Liver Histology confirms findings of previous studies that the intravenous administration of glutamine, Figure 1 depicts normal histology of rat bypassing the gastrointestinal tract, is well liver when animals are fed with standard rat tolerated when a 1 to 2% solution is infused chow. Figure 2 demonstrates fatty infiltra[4]. Blood ammonia levels were comparable tion present following 6 days of standard in- to those when glutamine was not added and travenous nutrition. A dramatic decreasein less than seen with standard rat chow. No steatosis is evident in animals who receive clinical signs of glutamine toxicity were eviglutamine supplementation to the standard dent as all animals appeared healthy during the 6 days of infusion. intravenous nutrition solution (Fig. 3).

FIG. 1. Normal histological picture of the rat liver.

GRANT AND SNYDER: L-GLUTAMINE

IN TPN

511

FIG. 2. Fatty infiltration of the rat liver after 6 days of intravenous feeding with TPN solution containing no glutamine.

The addition of glutamine at either 1 or 2% concentration resulted in reduced nitrogen retention compared to chow feeding or standard TPN. There is no explanation for this finding. Weight gain was similar for the chow-fed and glutamine-supplemented groups but significantly less than for standard TPN. Perhaps glutamine supplementation reduces free water retention and fat deposition known to occur with standard TPN. As discussed later, glutamine supplementation did reduce hepatic steatosisin this study. In spite of the difference in nitrogen balance, addition of glutamine at either 1 or 2% concentration resulted in maintenance of the nitrogen content of the stomach and colon and indeed increased nitrogen content of the small bowel when compared to standard intravenous nutritional support or chow-fed

animals. Indeed standard intravenous nutritional support was no better than intravenous 5% dextrose in maintenance of gut nitrogen content. Mucosal height was likewise preserved in the stomach and colon with the addition of glutamine compared to 5% dextrose or standard intravenous nutrition support. Preservation of mucosal height was however not seen in the small bowel where there was a significant decrease from chowfed animals but not to the level of intravenous dextrose. A most interesting finding was the increased concentration of the disaccharidase enzymes, sucrase, lactase, and maltase, whether or not glutamine was added to the nutritional support solution when compared to chow-fed animals and animals given only intravenous 5% dextrose. Total disacchari-

512

JOURNAL OF SURGICAL RESEARCH: VOL. 44, NO. 5, MAY 1988

dase activity was higher for maltase with glutamine supplementation and essentially equal for sucrase and lactase but all superior to chow fed. It would follow that preservation of gut nitrogen content and mucosal height with glutamine infusion would not confer any improvement in sugar absorption. Further studies are necessary to evaluate if there is any improvement in protein or fat absorption. The maintenance of mucosal height and gut nitrogen content may, however, be of great clinical significance if the gut mucosal barrier were maintained by glutamine preventing translocation of bacteria or bacterial toxins during total parenteral nutrition in stressedpatients. Also of interest is the observation that steatosis secondary to total parenteral nutrition is eliminated with the addition of 1 or 2% glutamine to the feeding formulation. As all other

FIG.

substrates were unaltered, one might speculate that with support of glutamine metabolism in the gut by intravenous glutamine infusion, resulting in an increased flow of alanine into the portal vein, more normal hepatic metabolism of nutrients occurred than during glutamine-free nutrition. Hall et al. [3] reported hepatic steatosisduring TPN in the Fisher 344 rat to be due to four abnormalities: increased hepatic synthesis of fat, impaired mobilization of fat out of the liver, increased uptake of fat from the blood, and reduced metabolism of fat by the liver. Which mechanism is corrected by glutamine infusion remains to be identified. The beneficial effects of glutamine infusion on gut nitrogen content and mucosal height may suggestfurther beneficial applications in patients with short bowel syndrome, inflammatory bowel disease,and enterocutaneous fistulas.

3. Elimination of fatty infiltration when 1% glutamine is added to TPN solution.

GRANT AND SNYDER L-GLUTAMINE

REFERENCES 1. Dahlqvist, A. Assay of intestinal disaccharidases. Enzym. Biol. Clin. 11: 52, 1970. 2. Eastwood, G. L. Small bowel morphometry and epithelial proliferation in intravenously alimented rabbits. Surgery 82: 6 13, 1977. 3. Hall, R. I., Grant, J. P., Ross, L. H., Coleman, R. A., Bozovic, M. G., and Quarfordt, S. H. The pathogenesis of hepatic steatosisin the parenterally fed rat. J. Clin. Invest. 74: 1658, 1984. 4. Hwang, T. L., O’Dwyer, S. T., Smith, R. J., and Wilmore, D. W. Preservation of the small bowel mucosa using glutamine-enriched parenteral nutrition. Surg. Forum 37: 56, 1986. 5. Koga, Y., Ikeda, K., Inokuchi, K., Watanabe, H., and Hashimoto, N. The digestive tract in total parenteral nutrition. Arch. Surg. 110: 742, 197.5. 6. Kripke, S. A., Fox, A. D., Berman, J. M., Settle, G., and Rombeau, J. L. Stimulation of mucosal growth with intracolonic butyrate infusion. Surg. Forum 38: 41, 1987. I. Popp, M. B., and Brennan, M. F. Long-term vascu-

IN TPN

513

lar accessin the rat: Importance of asepsis.Amer. J. Physiol. 241: H606, 1981.

8Y. Ram, F., Galluser, M., and Doffoel, M. A comparison of intestinal adaptation to short-term intravenous versus intragastric diet in adult rats. J. Purenter. Enteral Nutr. 11: 389, 1987. 9. Roediger, W. E. W. Utilization of nutrients by isolated epithelial cells of the rat. Gastroenterology 83: 424, 1982. 10. Souba, W. W., Smith, R. J., and Wilmore, D. W. Glutamine metabolism by the intestinal tract. J. Parenter. Enteral Nutr. 9: 608, 1985.

11. Souba, W. W., and Wilmore, D. W. Gut-liver interaction during accelerated gluconeogenesis. Arch. Surg. 120: 66, 1985.

12. Windmueller, H. G. Glutamine utilization by the small intestine. In S. Fleischer and L. Packer (Eds.), Advances in Enzymology. San Diego: Academic Press, 1982. Vol. 53, Pp. 202-237. 13. Windmueller, H. G., and Spaeth, A. E. Uptake and metabolism of plasma glutamine by the small intestine. J. Biol. Chem. 249: 5070, 1914.