ARTICLE IN PRESS

Acta histochemica 109 (2007) 122—129

0065-1281/$ - sdoi:10.1016/j.

�CorrespondE-mail addr

www.elsevier.de/acthis

Early glutamine-enriched enteral feedingfacilitates colonic anastomosis healing: Lightmicroscopic and immunohistochemical evaluation

Aysel Guvena,�, Mevlut Pehlivanb, Ibrahim Gokpınarb,Emin Gurleyikb, Meryem C-ama

aDepartment of Histology and Embryology, Duzce University, School of Medicine, 81620 Konuralp, Duzce, TurkeybDepartment of Surgery, Duzce University, School of Medicine, 81620 Konuralp, Duzce, Turkey

Received 24 October 2006; received in revised form 6 November 2006; accepted 7 November 2006

KEYWORDSColon anastomosis;Enteral nutrition;Glutamine;IGF-I;Wound healing

ee front matter & 2007acthis.2006.11.004

ing author. Tel.: +90 380ess: drayselguven@yaho

SummaryProblems related to colonic anastomosis healing constitute the major morbidity incolorectal surgery. Patients without appropriate nutritional support are at higher riskof postsurgical complications, mainly due to reduced wound healing. Therefore, weinvestigated the effect of early and late postoperative total enteral nutrition (TEN)and glutamine addition on colon anastomosis healing using light microscopy andimmunohistochemistry (IGF-I immunolabelling). In this study, 40 Wistar-albino ratsunderwent distal left colonic transection and anastomosis. The rats were thendivided into four groups given different diets: delayed total enteral nutrition (dTEN;beginning 3 days postoperatively), delayed TEN with added glutamine (dTEN+Glutamine), early TEN (eTEN; beginning within 6 h postoperatively), and early TENwith added glutamine (eTEN+Glutamine). Colon segments, including the anasto-mosis, were excised 7 days postoperatively and evaluated histopathologically forinflammation, mucosal healing, submucosal-muscular layer repair, the amounts ofnecrosis and vascularisation and immunohistochemically for IGF-I labelling. Theinflammation and necrosis scores in the dTEN and dTEN+Glutamine groups weresignificantly greater than in the eTEN and eTEN+Glutamine groups. The IGF-Iimmunoreactivity increased in the eTEN, eTEN+Glutamine, and dTEN+Glutaminegroups compared to dTEN (po0.05). We concluded that early TEN and glutamineenrichment in the postoperative period improve anastomosis healing via IGF-I.& 2007 Elsevier GmbH. All rights reserved.

Elsevier GmbH. All rights reserved.

5414485; fax: +90 380 5414487.o.com (A. Guven).

ARTICLE IN PRESS

Colonic anastomosis healing 123

Introduction

Problems with anastomosis healing may lead toserious postoperative complications. Many factorsaffect the healing process, including pathologicalconditions, drugs and patient health. The nutri-tional status of the patient affects wound healingduring the early postoperative period directly, andnutritional support facilitates healing, especially inmalnourished patients. Parenteral or enteral routesmay be used to provide appropriate nutrition.Previous studies have shown that enteral nutritionis superior to the parenteral route in terms ofcalorie and protein intake, progressive immuno-logic status, and in reducing gastrointestinalcomplications (Moore et al., 1989, 1992; Lowry,1990; Kudsk et al., 1992; Feilhauer et al., 2000;Wildhaber et al., 2005).

During wound healing, the epithelium continuesto grow, fibroblasts increase in loose connectivetissue, and newly formed capillaries continue tocontract due to the action of myofibroblasts, withthe involvement of fibroblasts and macrophages atapproximately 1 week after injury (Govan et al.,1995). Myofibroblasts play an important role inwound healing and the fibrosis causes contraction(McKaig et al., 1999). In vitro studies of primaryintestinal subepithelial myofibroblast (ISEMF) cul-tures suggest that the important factors inducingmyofibroblast activation and proliferation aretumour necrosis factor alpha (TNF-a), interleukin-6 (IL-6), epidermal growth factor (EGF), basicfibroblast growth factor (bFGF), insulin like growthfactor-I (IGF-I) and IGF-II, platelet derived growthfactor-BB (PDGF-BB), and interleukin-1b (IL-1b)(the last two act synergistically) (Jobson et al.,1998; Powell et al., 1999).

IGF-I is an important anabolic hormone that isessential for normal growth and development andthe maintenance of lean body mass. IGF-I controlscell proliferation and differentiation by regulatingspecific events in the G1 phase of the cell cycle. Inaddition, IGF-I stimulates myoblast differentiationand myotubule formation and has insulin-likeeffects, such as stimulating glucose consumptionin adipose tissue. IGF-I acts via the IGF-I receptor.IGF-I and IGF-II are expressed in many tissues andcell types. IGF-I is mitogenic for a variety of cells,including fibroblasts, osteoblasts, smooth muscle,foetal brain, neuroglial and erythroid progenitorcells (Froesch et al., 1985; Zumstein and Stiles,1987; Dieguez et al., 1988). The expression of IGF-ImRNA or protein is altered in a tissue-specificmanner under certain pathophysiological condi-tions (Lang et al., 1996). Infection leads to markedalterations in the hepatic, intestinal and extra-

splanchnic IGF-I balance, all of which mightmodulate tissue anabolic responses via both classi-cal endocrine and autocrine–paracrine mechanisms(Jones and Clemmons, 1995; Lang et al., 1998).Several methods and drugs may reduce postsurgicalrisk by improving colon anastomosis healing (Thorn-ton and Barbul, 1997; Kiyama et al., 1999, 2000;Gurleyik et al., 2002). Nutritional support also haspositive effects on the healing process. We inves-tigated the effect of early and delayed enteralnutritional support and glutamine enrichmenton IGF-I immunolabelling and histopathologicalchanges in the healing of colon anastomoses.

Material and methods

Experimental animals

The study was approved by the Animal CareEthics Committee of Duzce University School ofMedicine, Duzce, Turkey. 40 female Wistar-albinorats weighing 232–264 g were obtained from theuniversity animal facility. All the rats were fed acommercial pellet diet and allowed water adlibitum.

Surgical technique

No specific bowel preparation was performed.After an 8-h fast, the rats were anaesthetisedwith 50mg/kg IM ketamine (Ketalar; Parke-Davis,Eczacıbas-ı, Turkey). The skin was shaved andcleaned with povidone–iodine. The left colon wastransacted 3 cm proximal to the peritoneal fold viaa midline laparotomy. An end-to-end anastomosiswas performed using separate 5-0 polypropylenesutures (Prolene; Ethicon, Somerville, NJ, USA).The abdomen was then closed with continuous 3-0polypropylene sutures.

Nutritional support

The study protocol for nutritional support wasbased on early or delayed feeding and glutaminesupplementation (Gokpınar et al., 2006). The ratswere randomly assigned to four equal-sized groupsfor nutritional support beginning 6 h after recover-ing from the surgical procedure.

Delayed enteral nutrition (dTEN): The rats werefed 30% dextrose and water for 3 days postopera-tively and then with a commercial pellet dietand allowed water ad libitum on postoperativedays 4–7.

ARTICLE IN PRESS

A. Guven et al.124

Delayed enteral nutrition+glutamine (dTEN+Glutamine): In addition to the nutritional supportgiven the dTEN group, glutamine was added to thepostoperative diet on days 4–7 via an orogastrictube as 12.5% glutamine solution (L-glutamine,Skip) prepared daily at a dose of 50mg/kg/day.

Early enteral nutrition (eTEN): The animalswere fed commercial pellets and allowed waterad libitum on postoperative days 1–7.

Early enteral nutrition+glutamine (eTEN+Gluta-mine): In addition to the nutritional support giventhe eTEN group, glutamine was added to the dieton postoperative days 1–7 in the same manner as inthe dTEN+Glutamine group.

Sample collection

Five rats died during the study period. Theremaining 35 animals were prepared for surgery asdescribed above and re-laparotomised through thesame incision. A 3 cm colon segment, including theanastomosis, was excised for histological examina-tion. The rats were killed by ether overdose.

Microscopic examination

All tissue specimens were obtained from theanastomosis area and fixed in 10% bufferedformalin, processed by routine protocols for em-bedding in paraffin wax, and cut into serial 5-mm-thick sections by microtome. The sections werestained with haematoxylin and eosin and Gomoritrichrome using standard protocols and examinedusing a photomicroscope (BX50; Olympus, Tokyo,Japan). The accumulation of polymorphonuclearcells (PMNs), lymphocytes and macrophages(inflammation); thickness of the wall at theanastomosis relative to the thickness of the normalintestinal wall; submucosal-muscular layer repair;and amounts of necrosis and vascularisation onhistopathological examination under light micro-

Table 1. Colon anastomosis healing scores

Score Inflammation Mucosal epithelium

0 Absent Normal prismatic glandularepithelium

1 Slight Mucosa cubic epithelium, glandabsent

2 Moderate Mucosa partially damaged,surrounded by cubic epithelium

3 Severe Mucosa damaged completely,epithelium absent

scopy were scored from 0 to 3, as summarised inTable 1.

Immunohistochemical examination

Insulin like growth factor (IGF-I)Sections from all animals were processed simul-

taneously in order to avoid day-to-day variation inlabelling efficiency. All tissue samples were fixed in10% formalin, processed by routine protocols forembedding in paraffin wax and 5-mm-thick sectionsof anastomotic regions were cut by microtome.Sections were mounted on poly-L-lysine-coatedmicroscope slides (Surgipath, Richmond, USA).Sections were labelled for the binding of a mousemonoclonal anti-IGF-I Ab-1 antibody (Insulin likegrowth factor mouse monoclonal antibody: cloneM23, Ready-to-use Ab-1, NeoMarkers, Fremont, CA,USA), as follows. The sections were deparaffinisedand rehydrated (twice in xylol, twice in 100%, 96%,and 70% ethyl alcohol, twice in redistillated sterilewater, for 15min each). Sections were treated with3% hydrogen peroxide and methanol for 30min toblock endogenous peroxidase activity. Sectionswere washed in phosphate-buffered saline (PBS),pH 7.4. They were incubated overnight at 4 1C withIGF-I Ab-1 antibody. After thorough washing withPBS (3� 5min changes), the sections were floodedwith a solution of 5% hydrogen peroxide, rinsedwith PBS (2� 5min) and incubated with biotiny-lated polyvalent IgG (ready-to-use, SkyTek Labora-tories, Logan, UT, USA) for 15min. Sections wererinsed with PBS (2� 5min) and incubated with anavidin–biotin peroxidase complex (ready-to-use,SkyTek Laboratories, Logan, UT, USA) for 15min.Following rinsing with PBS (2� 5min), sectionswere incubated with substrate chromogen solutionfreshly prepared by dissolving 1mg of 3,30-diami-nobenzidine (DAB) (SkyTek Laboratories, Logan, UT,USA) in 1ml of 0.05M Tris–HCI buffer (pH 7.4)containing 1 ml of H2O2, for 15min. After rinsing indistilled water, the sections were counterstained

Submucosal-muscular healing

Necrosis Vascularisation

Good Absent Absent

s Moderate bridge Slight Slight

Weak bridge Moderate Moderate

Absent Severe Severe

ARTICLE IN PRESS

Colonic anastomosis healing 125

with Harris’s haematoxylin. For dehydration, eachslide was soaked in absolute ethyl alcohol(2� 2min) then cleared with xylene and mountedwith resinous medium. For negative controls, theprimary antisera was omitted and replacedwith PBS.

The intensity of labelling for IGF-I was scored asnone (0), weak (1), moderate (2), or strong (3).‘Strong’ was assessed as intense labelling easilyvisible at low power involving 450% of cells;‘moderate’ as focal intensely labelled areas invol-ving o50% of cells or moderate staining of 450% ofcells; ‘weak’ as focal moderately intense labellingin o50% of cells or weak labelling in any proportionof cells not easily seen under low power. A singleinvestigator who was unaware of the group assign-ments performed the light microscopy and immu-nohistochemical evaluations.

Statistical analysis

The Kruskal–Wallis nonparametric test for multi-ple comparisons was applied to all data and po0.05was considered statistically significant.

Results

Histopathological evaluation

Although the mucosal epithelium integrity wasbetter in the early TEN groups (eTEN, eTEN+Gluta-mine), illustrated in Figs. 1C and D, the differencewas not significant. Of the histopathological criter-ia, only the inflammation and necrosis scoresdiffered significantly between the groups(po0.05). The anastomosis healing, inflammationand necrosis scores were significantly lower(po0.05) in the early TEN groups, Figs. 1C and D,than in the delayed TEN groups, Figs. 1A and B.Although the inflammation score of the eTEN+Glutamine group was lower than that of the dTEN+Glutamine group, the difference was not signifi-cant, whereas the necrosis score was significantlylower (po0.005) with added glutamine in the eTENgroup, Fig. 1D.

Immunohistochemical evaluation

The immunolabelling intensity for IGF-I wasgreater in the eTEN, eTEN+Glutamine and dTEN+Glutamine groups, illustrated in Figs. 2B–D, com-pared to the dTEN group, illustrated in Fig. 2A. TheIGF-I immunolabelling was similar in the dTEN+Glutamine, eTEN and eTEN+Glutamine groups,

Figs. 2B–D. The dTEN group samples labelledweakly for IGF-I, Fig. 2A. The addition of glutamineto the dTEN group, Fig. 2B, resulted in a significantincrease (po0.05) in IGF-I immunolabelling com-pared to dTEN, Fig. 2A. A significant increase(po0.05) occurred in IGF-I immunoreactivity inthe eTEN, eTEN+Glutamine and dTEN+Glutaminegroups compared to the dTEN group.

Discussion

Problems related to colon anastomosis healingare the major source of morbidity in colorectalsurgery. Patients with inadequate nutritional sup-port are at high risk in terms of surgical complica-tions, especially problems related to woundhealing. The main concern in parenteral andenteral nutritional support is to ensure appropriatecalorie and protein intake and, to provide nitrogenbalance for satisfactory wound healing in thesurgical patient. Partial or TEN is the procedureof choice as it has several advantages over theparenteral route if not otherwise contraindicated(Moore et al., 1989, 1992; Lowry, 1990; Kudsket al., 1992; Feilhauer et al., 2000; Wildhaberet al., 2005).

Diets rich in glutamine and arginine give the bestresults when it comes to maintaining body weightand mucosal mass. Glutamine-supplemented totalparenteral nutrition (TPN) protects against thebacterial translocation from the gut seen withstandard formulas. This effect might be mediatedby the immune system through secretory IgA. Thesynthesis and appearance of IgA in secretionsappear to be sensitive to diet and might beimpaired after surgical stress. Data suggest thatthe maintenance of gut mass and barrier functionagainst bacteria via dietary manipulation is essen-tial for ensuring host survival during critical illness(Burke et al., 1989; Alverdy, 1990; Vazquez et al.,1996). The choice of the enteral route for nutri-tional support has many other advantages in termsof wound healing (Zaloga et al., 1992; Kiyamaet al., 1998; Fernandez de Bustos et al., 2006;Ng and Neill, 2006).

Zumstein and Stiles (1987) showed that growthfactors and IGF-I make significant contributions towound healing via mitogenic effects in a variety offibroblasts and smooth muscle cells and that theyare essential for normal growth and development.Growth hormones and IGF-I play important roles inwound healing during intestinal injury and inflam-mation, although indirect evidence suggests thatlocally expressed IGF-I induces excessive collagendeposition, which can lead to intestinal fibrosis

ARTICLE IN PRESS

Figure 1. (A) dTEN group, obvious necrosis (*) and inflammation, (B) dTEN+Glutamine group, inflammation (*),(C) eTEN group, good anastomosis healing, (D) eTEN+Glutamine group, good anastomosis healing. H&E staining;Scale bars: 1A and C: 50 mm; 1B, C-inset and D: 100 mm.

A. Guven et al.126

(Powell et al., 1999; Fruchtman et al., 2005).Egger et al. (2001) demonstrated that the growthfactors tKGF and IGF-I both markedly acceleratethe healing of colon anastomoses in rats. There-fore, the effects of nutritional support and gluta-mine enrichment on the healing of colonanastomoses were investigated based on thehistological changes and IGF-I expression in intest-inal tissue.

In our study, the inflammation scores weresignificantly lower in the eTEN, eTEN+Glutamine,and dTEN+Glutamine groups compared to the dTENgroup because early nutrition and added glutaminesupport anastomosis healing. In addition, the

necrosis scores illustrated the importance of earlynutrition.

Several studies have reported that caloric re-striction and bacterial infection decrease plasmaIGF-I levels. Infection markedly alters the hepatic,intestinal and extra-splanchnic IGF-I balance, all ofwhich are expected to modulate tissue anabolicresponses via classical endocrine mechanisms(Thissen et al., 1994; Lang et al., 1996, 1998). Inour study, dTEN resulted in more inflammation andnecrosis. Therefore, our results parallel those ofthese previous studies; the increased inflammationand necrosis in the dTEN groups induced thereduction in IGF-I immunpositivity.

ARTICLE IN PRESS

Figure 2. (A) dTEN group, weak IGF-I immunolabelling, (B) dTEN+Glutamine group, strong IGF-I immunolabelling,(C) eTEN group, strong IGF-I immunolabelling, (D) eTEN+Glutamine group, strong IGF-I immunolabelling. Scalebars: 2A, B, C and D: 100 mm; A-inset: 50 mm.

Colonic anastomosis healing 127

Carroll et al. (2004) showed that TPN andglutamine (GLN) combined with growth hormone/IGF-I treatment increased plasma IGF-I level com-pared to TPN alone, and resulted in a net proteingain and a net positive protein balance in a group ofcritically ill patients. The addition of glutamineconferred no major benefits on protein metabolismover TPN alone, although glutamine-supplementedTPN resulted in net whole body protein anabolismin severely ill individuals (Carroll et al., 2004). Thisstudy demonstrated the positive effects of IGF-I onprotein metabolism and healing. Petersen et al.(1996) showed that colon-operated rats treatedwith IGF-I increased their postoperative body

weight and collagen deposition was stimulated inleft colon anastomoses; the treatment resulted in athreefold increase in the serum IGF-I of IGF-I-treated rats compared to controls. We did notadminister growth hormone or IGF, but simplyinvestigated the effect of early and delayed enteralnutrition and glutamine enrichment on IGF-I im-munolabelling in support of the healing process incolon anastomosis. Our immunohistochemical ana-lysis revealed that early nutrition via the enteralroute significantly increased IGF-I immunolabellingin colon tissue adjacent to the anastomosis.Conversely, delayed nutrition had a weaker effecton IGF-I immunopositivity compared to the early

ARTICLE IN PRESS

A. Guven et al.128

enteral route. In addition, IGF-I immunolabellingintensitiy increased in the dTEN+Glutamine groupcompared to the dTEN group. We feel that earlyenteral nutrition facilitates protein metabolismand the healing process via growth hormoneexpression. We found that added glutamine in thedelayed nutrition group reversed the disadvantagesof delayed nutrition on the healing process byincreasing IGF-I.

Early TEN supports collagen synthesis, strength-ening the anastomosis during the postoperativeperiod. Glutamine addition to late enteral nutritionreverses the disadvantages of late nutrition. Similareffects of glutamine on collagen synthesis havebeen reported (Gokpınar et al., 2006). In our study,eTEN and eTEN+Glutamine supplementation signif-icantly supported anastomosis healing. Based onthe pathology and IGF-I results, glutamine supple-mentation also reverses the disadvantages ofdelayed nutrition. Therefore, the mechanism ofthe positive effects of early nutrition and glutamineon the healing process can be partly explained byincreased IGF-I immunlocalisation in colon tissue atthe anastomosis. Alexander and Carey (2002)reported that the oral administration of IGF-Istimulates Na+-dependent glutamine absorption inthe piglet small intestine, an effect that isindependent of the changes in intestinal mucosalmass. In this study, we administered glutamine, butnot exogenous IGF-I, and found an increase in IGF-Ilabelling with glutamine addition in the delayednutrition group. We speculate that the addition ofglutamine increases intestinal IGF-I expression bymyofibroblasts.

According to the histopathological findings,which showed that necrosis and inflammation weresignificantly lower in eTEN than in dTEN, eTENdecreases the rate of complications by supportinganastomosis healing. We conclude that postopera-tive nutritional support using the enteral routeimproves the healing process partly via IGF-Iexpression. In certain situations, when enteralnutrition is delayed, glutamine supplementationmay reverse the disadvantages of late nutrition,and improve the healing process by increasing IGF-Iexpression. TEN should be administered as early inthe postoperative period as possible.

References

Alexander AN, Carey HV. Insulin-like growth factor-Istimulates Na+-dependent glutamine absorption inpiglet enterocytes. Dig Dis Sci 2002;47(5):1129–34.

Alverdy JC. Effects of glutamine-supplemented diets onimmunology of the gut. JPEN J Parenter Enteral Nutr1990;14:109–13.

Burke DJ, Alverdy JC, Aoys E, Moss GS. Glutamine-supplemented total parenteral nutrition improves gutimmune function. Arch Surg 1989;124(12):1396–9.

Carroll PV, Jackson NC, Russell-Jones DL, Treacher DF,Sonksen PH, Umpleby AM. Combined growth hormone/insulin-like growth factor I in addition to glutamine-supplemented TPN results in net protein anabolism incritical illness. Am J Physiol Endocrinol Metab 2004;286(1):151–7.

Dieguez C, Page MD, Scanlon MF. Growth hormoneneuroregulation and its alterations in disease states.Clin Endocrinol 1988;28:109–43.

Egger B, Inglin R, Zeeh J, Dirsch O, Huang Y, Buchler MW.Insulin-like growth factor I and truncated keratinocytegrowth factor accelerate healing of left-sided colonicanastomoses. Br J Surg 2001;88(1):90–8.

Feilhauer K, Schall H, Butters M. Early postoperativenutrition in colorectal surgery: jejunal vs. parenteralfeeding: a prospective randomized trial. Clin Nutr2000;19:57.

Fernandez de Bustos A, Creus Costas G, Pujol Gebelli J,Virgili Casas N, Pita Merce AM. Per os early nutritionfor colorectal pathology susceptible of laparoscopy-assisted surgery. Nutr Hosp 2006;21(2):173–8.

Froesch G, Schmid C, Schwander J, Zapf J. Action ofinsulin-like growth factors. Ann Rev Physiol 1985;47:433–67.

Fruchtman S, Simmons JG, Michaylira CZ, Miller ME,Greenhalgh CJ, Ney DM, et al. Suppressor of cytokinesignaling-2 modulates the fibrogenic actions of GH andIGF-I in intestinal mesenchymal cells. Am J PhysiolGastrointest Liver Physiol 2005;289(2):342–50.

Gokpınar I, Gurleyik E, Pehlivan M, Ozcan O, Ozaydin I,Aslaner A, et al. Early enteral and glutamine enrichedenteral feeding ameliorates healing of colonic anasto-mosis: experimental study. Turkish J Trauma EmergSurg 2006;12(1):17–21.

Govan DT, Macfarlane PS, Callander R. Pathology illu-strated. Healing. New York: 1995. p. 52–66.

Gurleyik G, Gurleyik E, Yilmazcan A, Ozcan A, Onaran I,Unalmiser S. Effects of neurotensin on the healing ofexperimental anastomosis of the colon. Acta Chir Belg2002;102:33–6.

Jobson TM, Billington CK, Hall IP. Regulation of prolifera-tion of human colonic subepithelial myofibroblasts bymediators important in intestinal inflammation. J ClinInvest 1998;101:2650–7.

Jones JI, Clemmons DR. Insulin-like growth factors andtheir binding proteins: biological actions. Endocr Rev1995;16:3–34.

Kiyama T, Witte MB, Thornton FJ, Barbul A. The route ofnutrition support affects the early phase of woundhealing. JPEN J Parenter Enteral Nutr 1998;22:276–9.

Kiyama T, Efron DT, Tantry U, Barbul A. Effect ofnutritional route on colonic anastomotic healing inthe rat. J Gastrointest Surg 1999;3:441–6.

Kiyama T, Onda M, Tokunaga A, Yoshiyuki T, Barbul A.Effect of early postoperative feeding on the healing ofcolonic anastomoses in the presence of intra-abdom-inal sepsis in rats. Dis Colon Rectum 2000;43:54–8.

ARTICLE IN PRESS

Colonic anastomosis healing 129

Kudsk KA, Croce MA, Fabian TC, Minard G, Tolley EA,Poret HA, et al. Enteral versus parenteral feeding.Effects on septic morbidity after blunt and penetrat-ing abdominal trauma. Ann Surg 1992;215:503–11.

Lang CH, Fan J, Cooney R, Vary TC. Interleukin-1 receptorantagonist attenuates sepsis-induced alterations inthe insulin-like growth factor system and proteinsynthesis. Am J Physio 1996;270:430–7.

Lang CH, Frost RA, Ejiofor J, Lacy DB, McGuinness OP.Hepatic production and intestinal uptake of IGF-I:response to infection. Am J Physiol Gastrointest LiverPhysiol 1998;275:1291–8.

Lowry SF. The route of feeding influences injuryresponses. J Trauma 1990;30:10–5.

McKaig BC, Makh SS, Hawkey CJ, Podolsky DK, Mahida YR.Normal human colonic subepithelial myofibroblastsenhance epithelial migration (restitution) via TGF-beta3. Am J Physiol 1999;276:1087–93.

Moore FA, Moore EE, Jones TN, McCroskey BL, PetersonVM. TEN versus TPN following major abdominal trauma-reduced septic morbidity. J Trauma 1989;29:916–22.

Moore FA, Feliciano DV, Andrassy RJ, McArdle AH, BoothFV, Morgenstein-Wagner TB, et al. Early enteralfeeding, compared with parenteral, reduces post-operative septic complications. The results of a meta-analysis. Ann Surg 1992;216:172–83.

Ng WQ, Neill J. Evidence for early oral feeding of patientsafter elective open colorectal surgery: a literaturereview. J Clin Nurs 2006;15(6):696–709.

Petersen TI, Kissmeyer-Nielsen P, Flyvbjerg A, Laurberg S,Christensen H. Effect of insulin-like growth factor I(IGF-I) administration on the healing of colonicanastomoses in rats. Int J Colorectal Dis 1996;11(1):19–24.

Powell DW, Mıfflın RC, Valentıch JD, Crowe SE, Saada JI,West AB. Myofibroblasts. II. Intestinal subepithelialmyofibroblasts. Am J Physiol Cell Physiol 1999;277:183–201.

Thissen JP, Ketelslegers JM, Underwood LE. Nutritionalregulation of the insulin-like growth factors. EndocrRev 1994;15:80–101.

Thornton FJ, Barbul A. Healing in the gastrointestinaltract. Surg Clin North Am 1997;77:549–73.

Vazquez P, Gomez de Segura IA, Cos A, Candela CG, DeMiguel E. Response of the intestinal mucosa todifferent enteral diets in situations of surgical stressand malnutrition. Nutr Hosp 1996;11(6):321–7.

Wildhaber BE, Yang H, Spencer AU, Drongowski RA,Teitelbaum DH. Lack of enteral nutrition—effects onthe intestinal immune system. J Surg Res 2005;123:8–16.

Zaloga GP, Bortenschlager L, Black KW, PrielippR. Immediate postoperative enteral feeding decreasesweight loss and improves wound healing after abdom-inal surgery in rats. Crit Care Med 1992;20:115–8.

Zumstein P, Stiles CD. Molecular cloning of genesequences that are regulated by insulin-like growthfactor I. J Biol Chem 1987;262(23):11252–60.