Does glutamine-enriched parenteral nutrition really affect intestinal morphology and gut permeability?

www.elsevier.com/locate/clnu

ORIGINAL ARTICLE

Does glutamine-enriched parenteral nutritionreally affect intestinal morphology and gutpermeability?

Karel W.E. Hulsew !ea,*, Bernadette A.C. van Ackera, Wim Hameetemanb,Rene R.W.J. van der Hulstc, Tryfon Vainasa, Jan-Willem Arendsd, BernardK. van Kreele, Maarten F. von Meyenfeldta, Peter B. Soetersa

aDepartment of General Surgery, University Hospital Maastricht, The NetherlandsbDepartment of Gastroenterology, University Hospital Maastricht, The NetherlandscDepartment of Plastic and Reconstructive Surgery, University Hospital Maastricht, The NetherlandsdDepartment of Pathology, University Hospital Maastricht, The NetherlandseDepartment of Clinical Chemistry, University Hospital Maastricht, The Netherlands

Received 17 January 2003; accepted 2 April 2004

Summary Background: Nutritional depletion has been related to low glutaminelevels in plasma and gut mucosa. This study was set up to investigate the effects ofglutamine-enriched total parenteral nutrition on intestinal morphology andpermeability.Methods: Twenty-three depleted patients were randomized and after stabilization

baseline measurements were performed. Plasma glutamine concentrations, gutmorphology (including proliferation and lymphocyte markers) and intestinalpermeability were measured. After administration during 8–10 days of a glutamineenriched total parenteral nutrition or an isonitrogenous control solution themeasurements were repeated.Results: No significant changes in glutamine concentrations, intestinal perme-

ability, mucosal morphology or gut mucosal inflammation were observed betweengroups.Conclusions: Glutamine enriched total parenteral nutrition in a depleted patient

population does not result in improvements in gut morphology and gut barrierfunction.& 2004 Elsevier Ltd. All rights reserved.

Introduction

Already in the 1970s it was known that glutamineserves important functions for the gastrointesti-nal tract, which consumes glutamine in large

ARTICLE IN PRESS

KEYWORDS

Glutamine;

Depletion;

Gut;

Permeability;

TPN;

Amino acid metabolism

*Corresponding author. Department of Surgery, MaaslandZiekenhuis, PO Box 5500, 6130 AZ Sittard, The Netherlands.Tel.: þ 31-464597774.E-mail address: [email protected] (K.W.E. Hulsew!e).

0261-5614/$ - see front matter & 2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.clnu.2004.04.002

Clinical Nutrition (2004) 23, 1217–1225

quantities.1 The adverse effects of parenteralnutrition on the gut, which received much atten-tion in the last decade, have been attributed, inpart, to the absence of glutamine in total par-enteral nutrition (TPN).2,3 Glutamine has beenadministered intravenously, but the low solubilityand instability of free glutamine has precluded itswide usage. These practical problems led to thedevelopment of glutamine-dipeptides, such asglycyl-glutamine or alanyl-glutamine, which weremore soluble and stable than free glutamine,enabling addition of glutamine to total parenteralnutrition (TPN).4 Since then many clinical studieshave appeared investigating the clinical effects ofglutamine.

Studies investigating the effects of glutamine-enriched TPN on the intestinal mucosa have yieldedconflicting results. While multiple studies havedemonstrated beneficial effects of glutamine on gutmorphology and barrier function in animal models, inhealthy volunteers and in a patient population,5–9

others were not able to repeat these findings.10

Moreover, studies in rodents have shown considerablylarger effects of TPN and glutamine supplementationin comparison to clinical studies.11,12

There are a few explanations for these conflict-ing results: not all patients included in the studiesmay be lacking adequate endogenous glutaminestores, i.e. glutamine may not be consideredessential in these subjects, thereby reducing theeffects of its supplementation. Furthermore, dif-ferences in baseline nutritional status may existbetween groups especially if baseline samples aretaken without allowing for a standardized stabiliza-tion period: some may be nutritionally depletedwith or without micronutrient deficiencies,whereas others may have been eating until inclu-sion in the study.

Because of these confounding factors we de-signed a randomized trial in order to include ahomogeneous patient population, and to allow for ashort stabilization period during which all subjectswere fed a hypocaloric, glutamine-free parenteralnutrition solution. This was done in order to makethe conditions during baseline measurements morecomparable between subjects and more compar-able to the conditions during the measurements atthe end of the study.

Nutritional depletion has been associated withlow plasma and mucosal glutamine concentrations,suggesting a state of glutamine deficiency.13 Theendogenous glutamine production rate, however,has not been measured in such a population.Furthermore, depletion has been associated withmucosal atrophy and increased intestinal perme-ability.13,14

On the basis of these considerations we hypothe-sized that nutritionally depleted patients are likelyto benefit from glutamine enrichment of TPN.15

The aim of this study was therefore to investigatethe effects of glutamine enrichment of TPN on theplasma glutamine concentrations as well as itseffects on intestinal morphology and permeabilityin depleted patients.

Patients and methods

Patients

Patients admitted for gastrointestinal surgery wereeligible for the study if they were nutritionallydepleted and required preoperative parenteralnutrition. Nutritional depletion was defined as aninvoluntary loss of more than 10% of body weight(BW) within the previous 6 months or a BW of lessthan 95% of the ideal body weight calculated usingthe Metropolitan Life Insurance Tables 16 Further-more, they had to have an intact proximal gut. Ifthere were any clinical signs of severe metabolicstress, such as fever (438.51C), hypothermia(o361C), evidence of sepsis or intra-abdominalabscesses, patients were excluded. Other exclusioncriteria were the presence of renal or hepaticinsufficiency, pregnancy, insulin dependent dia-betes mellitus, ileus or administration of parenteralnutrition within 14 days prior to inclusion in thestudy. The demographic characteristics and dis-eases of the patients are summarized in Table 1.

All participants gave informed consent prior toinclusion in the study. The study was approved bythe Medical Ethics Committee of the AcademicHospital Maastricht.

Study design

After inclusion in the study, the patients wererandomized to receive either a glycyl-glutaminesupplemented TPN solution (GT) based on Glamins

(Fresenius Kabi) or an isonitrogenous control solu-tion (CT) based on Vamin 18EFs (Fresenius Kabi).Only the pharmacist was not blinded with regard tothe TPN solution given. The composition of theamino acid solutions is described in Table 2. Theamino acid solution was mixed with a glucose 50%solution and a lipid solution (Intralipid 20%,Pharmacia & Upjohn, Sweden), each providingapproximately 50% of the calories administered.The total energy expenditure was estimated usingthe Harris and Benedict formula for basal energyexpenditure based on the actual weight and

ARTICLE IN PRESS

1218 K.W.E. Hulsew !e et al.

increased with 40% for total energy expenditure.17

The nitrogen/kcal ratio averaged 1:150. During thestudy the patients were not allowed to eat or drinkexcept for 500ml of water per day. Body composi-tion was determined at baseline with the tracerdilution method using bromide and deuteratedwater for extracellular and total water spacerespectively.18 Ideal body weight was determinedby measuring body length and wrist circumferenceand using the Metropolitan Life Insurance Tables.16

Beginning at 18.00 h, an amino acid solutionlacking glutamine with glucose was infused con-tinuously, providing 50% of the calculated energyneeds. Since these patients were nutritionallydepleted, we gradually increased the TPN adminis-tered in order to prevent the occurrence of therefeeding syndrome. After 38 h baseline measure-ments were performed: an intravenous line forstable isotope infusion and an arterial catheter inthe radial artery for blood sampling was inserted tomeasure whole body protein turnover, arterialplasma glutamine concentration and glutamineproduction. The results of these measurementshave been published elsewhere.19 Blood for gluta-mine concentration was sampled in heparinised

tubes chilled on ice. These were centrifuged at2200 g at 41C during 5min. The plasma wasdeproteinized with 5% sulfosalicylic acid, mixedand frozen in liquid nitrogen and stored at �801Cuntil analysis. In addition, after voiding, a gastro-duodenoscopy was performed and mucosal biopsieswere taken of the descending part of the duode-num. Patients were offered sedation with midazo-lam 5mg i.v. during this procedure. Three biopsieswere fixated in 4% formaldehyde. One biopsy wassnap frozen in thiopental and stored at �801C untilfurther analysis. Following this procedure a smallcatheter was passed under endoscopic guidanceinto the small bowel distal to the site of thebiopsies. A sugar tracer solution containing 10 glactulose, 0.5 g mannitol and 1.0 g rhamnosedissolved in 65ml of water was infused throughthis catheter. During the next 6 h all urine wascollected and preserved with thymol, in order toprevent bacterial breakdown of the sugars. Afterbaseline measurements, the patients received theglutamine containing or the control solution inaccordance with the randomization. The third daythe amount infused equaled 75% of the total energyexpenditure (TEE), and from the 4th day until the

ARTICLE IN PRESS

Table 1 Demographic data of patients.

Subject TPN Sex Age PIW Loss of BW (%) Diagnosis

103 GT M 36 88 12 Crohn’s disease106 GT f 46 71 18 Gastrocolic fistula108 GT m 65 102 17 Sigmoid stenosis111 GT m 32 95 25 Chronic pancreatitis112 GT m 79 97 13 Colonic stenosis114 GT f 40 93 16 Tuboovarian abscess201 GT f 75 92 18 Rectosigmoid carcinoma202 GT f 65 68 10 Rectosigmoid carcinoma206 GT f 69 94 14 Gastric carcinoma207 GT m 68 82 10 Gastric carcinoma210 GT m 70 115 34 Colonic carcinoma101 CT m 53 111 4 Chronic pancreatitis104 CT m 33 91 14 Stenosis distal ileum105 CT m 34 76 8 Crohn’s disease107 CT m 45 83 26 Chronic pancreatitis109 CT f 61 100 17 Crohn’s disease110 CT m 60 101 13 Colonic stenosis113 CT f 65 86 11 Crohn’s disease203 CT f 70 90 10 Colonic carcinoma204 CT m 64 83 15 Colonic carcinoma205 CT m 47 101 14 Gastric carcinoma208 CT f 77 104 10 Colonic carcinoma209 CT m 64 108 22 Rectosigmoid carcinoma211 CT m 73 82 21 Rectosigmoid carcinoma

PIW¼percentage ideal body weight; BW¼body weight; GT¼ glutamine enriched TPN; CT¼ glutamine free TPN; m¼male;f¼ female.

Glutamine enriched TPN in depleted patients 1219

end of the study 100% of the TEE was administered.After 8–10 days of infusion of the study solutions,the measurements described above were repeated.

Outcome measurements

The primary outcome measure of this study was thelactulose/mannitol ratio as a measure for gutbarrier integrity. Since mannitol is not an idealsugar probe because it can be found in blank urinesamples, we included the rhamnose/lactulose ratioas an alternative measure for intestinal perme-ability. Other secondary outcomes were arterialglutamine concentration, gut mucosal morphology(i.e. crypt depth, villus height and total mucosalthickness), intraepithelial lymphocytes and theproportion of proliferating enterocytes.

Analyses

Plasma glutamine concentrationAminoacid concentrations were determined inarterial plasma using a fully automated highpressure liquid chromatography method as de-scribed previously.20

Morphology/immunohistochemistryAfter fixation in 4% formaldehyde 3 mucosal biop-sies were dehydrated in increasing concentrationsof ethanol and embedded in paraffin. Slices of 4 mMwere cut and stained with haematoxylin andcounterstained with eosin. Measurements of villusheight and crypt depth were performed by one ofthe authors (KH) in blinded sections, without anyknowledge of the corresponding subject or time inrelation to the intervention. For the morphometricmeasurements, an automatic interactive video-analysis system (Jandel video analysis system,Erkath, Germany) was used. The measurementswere performed in at least 10 well oriented villisince this number has been described to result inreliable values.14

Additional slices of 3 mM were cut for immuno-histochemical staining with anti MIB-1, and anti CD-3 antibodies. In short, the slices were deparafinizedand put in methanol with hydrogenperoxide toeliminate endogenous peroxidase activity. Here-after the material was boiled in a citrate bufferusing a microwave oven. After incubation with theprimary antiserum, rat-anti-mouse biotin wasadded. The biotin was bound to horse radishperoxidase which was finally coloured by addingdiaminobenzidine. Labeling of MIB-1 occurs inreplicating enterocytes within the crypt and isexpressed as a percentage of all crypt cells. Atleast 10 well oriented crypts were counted. Anti-CD3 antibodies identify intraepithelial lympho-cytes. At least 1000 enterocytes were countedand the figures found are expressed per 100enterocytes. Positive and negative controls wereincluded in all of these series.

Lactulose/mannitol/rhamnoseUrinary lactulose, mannitol and rhamnose weredetermined with gas-liquid chromatography asdescribed by Shippee et al. 21

The excretion percentages of the different sugarprobes were determined and from these theexcretion ratios of lactulose/mannitol and lactu-lose /rhamnose were calculated.

Statistics

Sample size calculationIn order to minimize the number of patientsrequired, we used a sequential design analysis.Unlike the traditional fixed sample design, thesample size in this study was a randomvariable. The confirmatory comparison of thetreatment groups was performed by means of thesequential double triangular procedure proposed by

ARTICLE IN PRESS

Table 2 Composition of aminoacid solutions.

Aminoacids GT CT

Ile 1.6 2.0Leu 2.2 2.8Lys 2.6 3.2Met 1.6 2.0Phe 1.7 2.8Tyr 0.7 0.1Thr 1.6 2.0Trp 0.5 0.7Val 2.1 2.6Ala 4.6 5.7Arg 3.2 4.0Asp 1.0 1.2Glu 1.6 2.0His 2.0 2.4Pro 2.0 2.4Ser 1.3 1.6Gly 3.2 2.8Gln 5.7 0.0

TotalAminoacids 40.2 41.0Nitrogen 6.6 6.5

Expressed as grams of amino acids per 1000ml totalparenteral nutrition solution. Nitrogen content is ex-pressed in g/l. GT is the glutamine enriched TPN solutionand CT is the control TPN solution.


Whitehead.22 This test procedure will always leadto a confirmatory result stating either significantdifferences or significant equivalence between thecompared treatments.

Based on a previous study from our group5 acoefficient of variation of the change in lactulose/mannitol ratios of 100% was assumed. With a type 1error rate of 0.05 and a type 2 error rate of 0.2, anaverage sample number of 11 patients per groupwas calculated.

All patients completing the study except onewere included in the statistical analysis as de-scribed in the results section. All data are pre-sented as mean7standard error of the mean. Forcalculation of differences between groups andbetween baseline and last measurements, the t-test or the Mann–Whitney-U-test was used asappropriate.

Results

Twenty-four patients were included in the studyand randomized to receive either the GT or the CTsolution. The patients were fed parenterallybecause of intolerance to full enteral nutrition. Inthe cases of inflammatory bowel disease andgastrointestinal malignancy this was due to (par-tial) obstruction and in the case of the patient withchronic pancreatitis, this was due to pain. Thestudy was terminated prematurely in three patients(2 of the GT and 1 of the CT group). Two patientsdeveloped symptoms requiring emergency surgery(profuse rectal bleeding in the presence of a rectalmalignancy and spiking fever with abdominal pain,respectively) and one patient developed a general-ized erythema with pruritus in the absence of aclear etiologic factor apart from the parenteralnutrition. This last patient was receiving thestandard glutamine free CT solution. One otherpatient refused to undergo intravascular catheter-ization and the gastroduodenoscopy on the laststudy day but gave consent for the measurement ofintestinal permeability at this time point. Patient

206 withdrew her consent prior to baseline mea-surements. Finally, one patient was excluded fromanalysis because of significant disturbances inglutamine and ammonia metabolism due to portalhypertension which was unknown at the time ofinclusion in the study.

The percentage of ideal body weight of allincluded patients was 93711% (mean7SD)whereas they had lost 15.476.7% (mean7SD) oftheir body weight in the 6 months precedinghospital admission. Data of the body compositionat baseline are presented in Fig. 1. Total bodywater (TBW) was 36.9 l in the GT group and 35.0 l inthe CT group. Intracellular water made up 57% ofTBW in GT and 49% in the CT group. None of thesedifferences were statistically significant.

Baseline plasma glutamine concentrations were609.97736.74 and 562.45723.12 mmol/l for the GTand CT group respectively (NS). After 8–10 days ofadministration of the study solution, these figuresdid not change significantly: 641.41739.15 (GT)and 591.79735.91 (CT). Lactulose, mannitol andrhamnose recoveries are summarized in Table 3.Neither the values at baseline nor the values at theend of the study were significantly differentbetween both groups. The lactulose/mannitol ratioand the lactulose/rhamnose ratios are graphically

ARTICLE IN PRESS

Body composition

0

5

10

15

20

25

30

35

40

GT CT

Vo

lum

e (L

)

ECWICW

Figure 1 Body composition at baseline in liters.ICW¼ intracellular water, ECW¼extracellular water. GTis the glutamine enriched TPN group and CT is the controlgroup.

Table 3 Recovery rates of lactulose, mannitol and rhamnose (mean7SEM).

CT GT

Baseline End of study Baseline End of study

Lactulose recovery rate (%) 1.0170.6 0.7570.2 0.7970.3 1.8370.8Mannitol recovery rate (%) 7.5070.8 6.4070.7 9.9671.8 10.9971.6Rhamnose recovery rate (%) 3.8070.4 2.9370.3 4.0570.6 4.8370.6

CT is the control TPN group and GT is the glutamine enriched TPN group.


displayed in Fig. 2, respectively. The differencesobserved between baseline and the last study daywere also not significantly different betweengroups.

The results of measurements of villus height,crypt depth, total mucosal thickness are describedin Table 4. In addition, the results of the measure-ment of MIB-1 and CD-3 positive intramucosal cellscan be found in the same table. Statistical analysisof these variables showed no differences betweenor within groups.

Discussion

In this study we found no effects of addition ofglutamine to total parenteral nutrition on gutmucosal morphology, mucosal inflammatory para-meters and intestinal permeability to a sugar probesolution. In addition no effects on plasma glutamineconcentrations could be detected. This is in

contrast to results from our previous studies.5,23

There are a few possible factors which may help toexplain these differences.

The crucial issue is whether glutamine is anessential amino acid in nutritional depletion or not.In contrast to the previous studies where depletedpatients were found to have lowered glutamineplasma concentrations, normal values were foundin our current study despite the fact that in thecurrent study all subjects included were nutrition-ally depleted. The degree of nutritional depletionexpressed as percentage ideal body weight (PIW) orrecent weight loss was comparable between thecurrent and previous studies, suggesting thatanother important variable may play a role. More-over, although 1 patient was unjustly included inthe study because he did not fulfill the criteria ofdepletion (patient 101), we included his data in theanalysis. However, recalculation of the datawithout this particular patient gave similar results.Two important differences in patient populationbetween this and our previous study were the

ARTICLE IN PRESS

Figure 2 Lactulose/mannitol and lactulose/rhamnose ratios. Individual data are shown. GT is the glutamine enrichedTPN group and CT is the control TPN group.

Table 4 Histological and immunohistochemical results. Data are presented as mean7SEM.

CT GT

Baseline End of study Baseline End of study P

Villus height 488.9728.7 477.5722.4 462.1722.0 463.8728.2 n.s.Crypt depth 116.973.3 114.379.1 113.174.5 113.378.2 n.s.Mucosal thickness 605.9725.3 591.7723.5 575.2718.9 577.2735.6 n.s.MIB-1 38.573.6 35.772.0 35.472.1 32.571.7 n.s.IEL 20.972.2 18.271.5 21.272.1 20.172.7 n.s.

CT refers to the control group and GT to the glutamine enriched group, MIB-1 is a marker for enterocyte proliferation and isexpressed as a percentage of positive cells in the crypths. IEL are intraepithelial leucocytes and are expressed per 100enterocytes.


inclusion of more cancer patients in the presentand more patients with active Crohn’s disease andulcerative colitis in the previous study. Further-more, the current study excluded specificallypatients with signs of clinically significant inflam-mation, whereas these were included in the formerstudy. Since glutamine levels are also influenced byinflammatory activity,24,25 this may have been aconfounding variable in the correlation of depletionwith low plasma glutamine concentrations.Although the glutamine plasma concentrations didnot rise after glutamine administration, as has beenobserved before,26 we did find increases in totalglutamine rate of appearance.19 However, theconcept of glutamine as a conditionally essentialamino acid remains to be supported or refuted bydemonstrating differences in net glutamine pro-duction rates by peripheral tissues like muscle, andsubsequently in differences in net utilization bycentral tissues like the liver and the immunesystem. We have described the absence of steady-state conditions during infusions of isotopicallylabeled glutamine for up to 11 h.27 The effect ofdepletion on net glutamine production and utiliza-tion rates remain therefore to be determined.27 Asecond important difference between this andother studies is the fact that in former studiesbaseline samples were taken after an overnight fastwhereas in the present study baseline samples weretaken after approximately 38 h of infusion of ahypocaloric amino acid glucose solution. Our aimwas to study the effects of addition of glutamine toTPN. Therefore we tried to make baseline mea-surements as comparable as possible with the finalmeasurement conditions. Furthermore, this stabi-lization period also allowed for a better equaliza-tion of conditions within and between both groups.On the basis of our data we cannot exclude thatinfusion of amino acids (and especially thebranched chain amino acids) led to increases inglutamine synthesis in our patients in the first 112day resulting in normal plasma glutamine concen-trations. If this would be the case, then therationale for glutamine supplementation must bequestioned.

The absence of changes in villus height is in linewith the absence of changes in gut permeability.Previous papers have described the simultaneousoccurrence of increased permeability and de-creases in villus height during administration ofTPN,5,6 although these phenomena were statisti-cally not related. Buchman et al. determined thechange in villus height to be the result ofhypoplasia.6 Another factor involved may be theeffect of the infusion of hypocaloric TPN prior tobaseline measurements. In a different study we

measured the short-term effects of enteral feedingon villus morphology in healthy volunteers. Wefound a significant increase in villus height within2 h.28 Therefore changes in mucosal architectureoccur rapidly and most likely significant decreasesin villus height will have occurred within 36 h ofTPN. This would explain the absence of any changesin villus morphology in this study. However, it isunlikely that this alone would affect gut barrierfunction since permeability is probably determinedby multiple factors rather than by changes inmucosal architecture alone.29

The same could apply to the absence of changesin IELs: increases in intestinal permeability andintestinal atrophy could have resulted in anincreased inflammatory activity within the gut.Since this is not the case, no changes in IEL are tobe expected. Similarly Buchman et al. found nosignificant changes in mucosal lymphocytes afteradministration of TPN in healthy volunteers.30 Incontrast, van der Hulst et al. found a decrease inIEL after administration of glutamine enriched TPN,associated with decreased gut permeability.23

Finally it is questionable, whether gut barrierfunction can be tested reliably with a sugar probesolution. Although intestinal barrier function isfrequently tested with sugar probe solutions, itsclinical relevance remains disputed.32 These sugarprobes are proposed to pass transcellularly orparacellularly respectively and their excretionratio is a measure of the paracellular transport(which for larger molecules such as lactulose isnormally low). However, the gut barrier is com-posed of several other parts, including the normalgut flora, mucus with several enzymes (e.g.lysozyme), secretory IgA, the gut associated lym-phoid tissue etc. In addition, functional aspects ofthe bowel such as peristalsis also play a role inresisting invasion of microbial products. All thesefactors are not measured with sugar probes. More-over, the results of the sugar absorption tests areinfluenced by other factors such as hydration.33 Butmore importantly, although we know that bacterialtranslocation across the gut occurs, its relation toseptic morbidity in humans is disputed.34,35. De-spite these arguments we included the lactulosemannitol ratio as a primary outcome measure,because both parenteral nutrition and nutritionaldepletion have been correlated with functionaldeterioration of the gut barrier function as mea-sured with these tests, and because currently thereare no better non-invasive tests to assess intestinalpermeability.36 Our data showed no changes in thepermeability of sugar probes within or between thegroups. The extremely abnormal values in onecancer patient in the control group are unex-

ARTICLE IN PRESS


plained, as this patient had no co-morbidity,medication, fluid infusion or symptoms that mightexplain this increased permeability. On average,the values found were in the same range as thosefound in a previous study.5 The absence of an effecton the excretion ratio of lactulose and mannitol isin contrast to studies in animals as well as humansreporting an increased permeability after TPN.6

Possibly the ‘‘damage’’ already occurs within thefirst 36 h, and glutamine cannot repair the damagedbarrier anymore. On the other hand, the values wefound are on average not above normal values for apostabsorptive patient population,5 rendering thisassumption less likely. In the study by Buchman andco-workers, however, the conditions during mea-surement of intestinal permeability were differentbetween baseline and the last measurement:whereas the first lactulose and mannitol solutionswere administered after an overnight fast, thesecond determination of the lactulose/mannitolratio took place during administration of totalparenteral nutrition.6 As referred to before, hydra-tion alone can already account for changes inintestinal permeability as measured with sugarprobes, thus making the value of this observationuncertain.33

In conclusion, this study demonstrates that in anutritionally depleted patient population withoutmajor inflammatory stress, addition of glutamine toTPN does not result in improvements of intestinalpermeability, mucosal morphology and gut mucosalinflammation. Systemic inflammatory reactions,which have profound effects on glutamine meta-bolism, are proposed to be of great importance inthe final effect of glutamine on the gut. The effectsof short term parenteral nutrition on intestinalmorphology and function deserve further study.

Acknowledgements

The authors acknowledge the help of Fresenius Kabiwho provided the amino acid solutions and whosupported the realization of this study. Mr. H. vanEijk and Mr. D. Rooyakkers are acknowledged fortheir analytical help in the plasma amino acidconcentrations. Furthermore we thank Mrs. M. vande Wijngaarden for her help with the lactulose,mannitol and rhamnose determinations and theanalists of the immunohistochemistry section of thedepartment of Pathology for their help with theimmunohistochemical stainings. Finally, we like toexpress our gratitude to all gastroenterologists andnurses of the endoscopy department for their helpin this study.

References

1. Windmueller HG, Spaeth AE. Uptake and metabolism ofplasma glutamine by the small intestine. J Biol Chem1974;249:5070–9.

2. Wilmore DW, Smith RJ, O’Dwyer ST, Jacobs DO, Ziegler TR,Wang XD. The gut: a central organ after surgical stress.Surgery 1988;104:917–23.

3. Souba WW. Intestinal glutamine metabolism and nutrition. JNutr Biochem 1993;4:2–9.

4. Furst P, Albers S, Stehle P. Glutamine-containing dipeptidesin parenteral nutrition. J Parenter Enteral Nutr 1990;14:118S–24S.

5. van der Hulst RRWJ, van Kreel BK, von Meyenfeldt MF,Brummer RJ, Arends JW, Deutz NEP, et al. Glutamine and thepreservation of gut integrity. Lancet 1993;341:1363–5.

6. Buchman AL, Moukarzel AA, Bhuta S, Belle M, Ament ME,Eckhert CD, et al. Parenteral nutrition is associated withintestinal morphologic and functional changes in humans.J Parenter Enteral Nutr 1995;19:453–60.

7. Li JL-H, Suzuki K, Stahlgren L. Glutamine prevents par-enteral nutrition-induced increases in intestinal permeabil-ity. J Parenter Enteral Nutr 1994;18:303–7.

8. Tremel H, Kienle B, Weiulemann LS, Stehle P, Furst P.Glutamine dipeptide - supplemented parenteral nutritionmaintains intestinal function in the critically ill. Gastro-enterology 1994;107:1595–601.

9. De Blaauw I, Deutz NEP, van der Hulst RRWJ, RvonMeyenfeldt MF. Glutamine depletion and increased gutpermeability in nonanorectic, non- weight-losing tumor-bearing rats. Gastroenterology 1997;112:118–26.

10. Spaeth G, Gottwald T, Haas W, Holmer M. Glutamine peptidedoes not improve gut barrier function and mucosal immunityin total parenteral nutrition. J Parenter Enteral Nutr1993;17:317–23.

11. Lipman TO. Bacterial translocation and enteral nutrition inhumans: an outsider looks in. J Parenter Enteral Nutr1995;19:156–65.

12. Buchman AL. Glutamine: is it a conditionally requirednutrient for the human gastrointestinal system? J Am CollNutr 1996;15:199–205.

13. Hulst van der RRWJ, Deutz NEP, Meyenfeldt von MF, ElbersJMH, Stockbr.ugger RW, Soeters PB. Nutritional depletion andmucosal glutamine concentration. Clin nutr 1994;13:228–33.

14. van der Hulst RR, von Meyenfeldt MF, van Kreel BK,Thunnissen FB, Brummer RJ, Arends JW, et al. Gutpermeability, intestinal morphology, and nutritional deple-tion. Nutrition 1998;14:1–6.

15. Soeters PB. Glutamine: the link between depletion anddiminished gut function? J Am Coll Nutr 1996;15:195–6.

16. Metropolitan Life Insurance Company. New weight standardsfor men, women. Bull Metropolitan Life Insurance Foundat.1983;64: 1–4.

17. Roza AM, Shizgas HM. The Harris–Benedict equation reea-valuated: resting energy requirements and the body cellmass. Am J Clin Nutr 1984;40:168–82.

18. Ellis KJ. Human body composition: in vivo methods. PhysiolRev 2000;80:649–80.

19. van Acker BA, Hulsewe KW, Wagenmakers AJ, von Meyen-feldt MF, Soeters PB. Response of glutamine metabolism toglutamine-supplemented parenteral nutrition. Am J ClinNutr 2000;72:790–5.

20. van Eijk HM, van der Heijden MA, van Berlo CL, Soeters PB.Fully automated liquid-chromatographic determination ofamino acids. Clin Chem 1988;34:2510–3.

ARTICLE IN PRESS


21. Shippee RL, Johnson AA, Cioffi WG, Lasko J, LeVoyer TE,Jordan BS. Simultaneous determination of lactulose andmannitol in urine of burn patients by gas-liquid chromato-graphy. Clin Chem 1992;38:343–5.

22. Whitehead J. The design, analysis of sequential clinicaltrials, 1 edn. Chichester, Ellis Horwood, 1992.

23. Hulst van der RRWJ, Meyenfeldt von MF, Tiebosch A,Buurman WA, Soeters PB. Glutamine and intestinal immunecells in humans. J Parenter Enteral Nutr. 1997;21:310–5.

24. Jackson NC, Carroll PV, Russell-Jones DL, Sonksen PH,Treacher DF, Umpleby AM. The metabolic consequences ofcritical illness: acute effects on glutamine and proteinmetabolism. Am J Physiol 1999;276:E163.

25. Parry Billings M, Baigrie RJ, Lamont PM, Morris PJ, News-holme EA. Effects of major and minor surgery on plasmaglutamine and cytokine levels. Arch Surg 1992;127:1237–40.

26. Powell-Tuck J, Jamieson CP, Bettany GE, Obeid O, FawcettHV, Archer C, et al. , A double blind, randomised, controlledtrial of glutamine supplementation in parenteral nutrition.Gut. 1999;45:82–8.

27. Van Acker BA, Hulsewe KW, Wagenmakers AJ, Deutz NEP, VanKreel BK, Halliday D, et al. Absence of glutamine isotopicsteady state: implications for the assessment of whole-bodyglutamine production rate. Clin Sci (Colch). 1998;95:339–46.

28. Hulsew!e KWE, van Acker BAC, Hameeteman W, vonMeyenfeldt MF, Soeters PB. Effects of a bolus meal onintestinal mucosa in man. Clin Nutr 1997;16:46.

29. Unno N, Fink MP. Intestinal epithelial hyperpermeability.Mechanisms and relevance to disease Gastroenterol ClinNorth Am 1998;27:289–307.

30. Buchman AL, Mestecky J, Moukarzel A, Ament ME. Intestinalimmune function is unaffected by parenteral nutrition inman. J Am Coll Nutr 1995;14:656–61.

32. Uil JJ, van Elburg RM, van Overbeek FM, Mulder CJ,VanBerge-Henegouwen GP, Heymans HS. Clinical implica-tions of the sugar absorption test: intestinal permeabilitytest to assess mucosal barrier function. Scand J Gastro-enterol 1997;223(Suppl):70–8.

33. Hallemeesch MM, Lamers WH, Soeters PB, DeutzNE. Increased lactulose/rhamnose ratio during fluidoad is caused by increased urinary lactulose excretion.Am J Physiol Gastrointest Liver Physiol 2000;278:G83–8.

34. Lemaire LC, van Lanschot JJ, Stoutenbeek CP, van DeventerSJ, Wells CL, Gouma DJ. Bacterial translocation in multipleorgan failure cause or epiphenomenon still unproven. Br JSurg 1997;84:1340–50.

35. MacFie J, O’Boyle C, Mitchell CJ, Buckley PM, Johnstone D,Sudworth P. Gut origin of sepsis: a prospective studyinvestigating associations between bacterial translocation,gastric microflora, and septic morbidity. Gut 1999;45:223–8.

36. Bjarnason I, MacPherson A, Hollander D. Intestinal perme-ability: an overview. Gastroenterology 1995;108:1566–81.

ARTICLE IN PRESS


Documents

Does glutamine-enriched parenteral nutrition really affect intestinal morphology and gut permeability?