Major surgery induces the production of reactive oxygen species as part of an immuno-inflammatory response that may cause cellular injury by damaging lipids, proteins and DNA. These events may contribute to post-operative complications and organ failure(Reference Park, Brody and Wallace1–Reference Nathens, Neff and Jurkovich9). It is known that after major surgery, levels of antioxidants such as glutathione, vitamin C, vitamin E, β-carotene, Zn and Se are reduced(Reference Heyland, Dhaliwal and Suchner7, Reference Manhart, Vierlinger and Spittler10–Reference Berger and Shenkin12). In addition, the lack of antioxidant intake in surgical patients with upper gastrointestinal cancer increases susceptibility for oxidative stress-related complications(Reference Gurski, Schirmer and Rosa13–Reference Xu, Zhong and Jing15). Supplementation of antioxidants may support recovery in these patients.
In the present study we report on effects of antioxidant-enriched enteral nutrition on several circulating and cellular markers of the immuno-inflammatory response in patients after upper gastrointestinal surgery.
Patients and methods
From February 2002 until May 2003, twenty-one patients who were scheduled for cancer surgery of the oesophagus, stomach, or pancreas, in the VU University Medical Centre (Amsterdam, The Netherlands), were included in the study. Inclusion criteria were: eligibility for jejunostomy feeding, aged 18–75 years, BMI < 35 kg/m2, written informed consent, and a surgical procedure of at least 3 h. Exclusion criteria were: history of cardiovascular or kidney disease, weight loss>10 % in the previous 6 months, steroids or investigational drugs used in the previous 6 weeks and HIV infection. The study was approved by the Medical Ethics Committee of the VU University Medical Centre Amsterdam and conducted according to the Declaration of Helsinki of 1975 (as revised in 1983).
The study was a prospective, open-label, randomised clinical trial of two balanced groups in parallel design in one medical centre. On the first post-operative day patients were randomised to receive either a standard enteral feeding (Sondalis ISO®; Nestlé, Switzerland), or the same enteral formula in combination with Module AOX (Nestlé, Switzerland). Module AOX provides enrichments in glutamine, cysteine, vitamins C and E, β-carotene, Zn and Se (for precise contents, see Table 1 of van Stijn et al. (Reference van Stijn, Ligthart-Melis and Boelens16)). The dosage of each component of Module AOX was established with regards to safety aspects, relying on outcomes of studies on the separate components. Daily intake was calculated by weighing the feeding bags before and after administration.
A jejunostomy was placed preoperatively and continuous feeding was started on the first post-operative day. Patients received two Modules AOX per d when feeding could be increased beyond 500 ml/d. Module AOX was administered for a minimum of 5 d and a maximum of 7 d. The feeding schedule was adjusted according to the energy requirements with the goal of reaching 1500–2000 ml from the third post-operative day of the patient. Patients were not allowed to receive additional vitamins, amino acids or lipid solutions during the study period. Daily fluid requirements were given intravenously, starting directly after surgery, until oral intake was sufficient. This fluid also contained glucose.
We measured C-reactive protein (CRP), peripheral leucocyte count, human leucocyte antigen-DR (HLA-DR) expression on the monocytes, soluble IL-1 receptor II, IL-6, IL-8, elastase, bactericidal/permeability-increasing protein (BPI), soluble TNF receptor 55/60 (TNF-R55) and soluble TNF receptor 75/80 (TNF-R75). Blood samples were taken on the day before surgery (day − 1) and on days 1, 3, 5 and 7 after surgery. The samples on day 1 were taken before the start of the enteral nutrition. All samples were taken between 08.00 and 10.00 hours.
Preparation, storage and analysis of blood samples
After sampling, several sensitive samples were immediately put on melting ice, such as for IL-1 receptor II, IL-6, IL-8, BPI, TNF-R55 and TNF-R75, to avoid degradation. Plasma was prepared by a two-step centrifugation procedure: 1400 g, 4°C, for 10 min, followed by 2700 g, 4°C, for 10 min, and stored at − 80°C.
IL-1 receptor II
The human soluble IL-1 receptor II ELISA kit (Hycult Biotechnology B.V., Uden, The Netherlands) was used to determine the concentration of IL-1 receptor II, representing IL-1 in plasma.
IL-6 was measured with a commercially available automated solid-phase, two-site two-step chemiluminescent immunometric assay (Immulite®; DPC, Los Angeles, CA, USA).
IL-8 was determined using specific sandwich ELISA as described previously(Reference Dentener, Bazil and Von Asmuth17).
Plasma was spun twice before storage to prevent leucocyte contamination of plasma. Microtitre plates were coated with human BPI-specific monoclonal antibody 4E3. Washing and dilution buffers for BPI determination contained 80 mm-MgCl2. Mg2+ ions were added to prevent any influence of endotoxin on the BPI measurement. Human recombinant BPI (provided by M. Marra, InCyte, Palo Alto, CA, USA) was used for the standard curve. Diluted samples (1:2 for BPI) were assayed. Biotinylated polyclonal rabbit anti-human BPI IgG was used as the secondary antibody, followed by visualisation using peroxidase-conjugated streptavidin (Dakopatts, Glostrup, Denmark). The level of detection was 200 pg/ml(Reference Hiki, Berger and Mimura18).
TNF receptor 55/60 and TNF receptor 75/80
Monoclonal antibodies specifically directed against TNF-R55 and TNF-R75 were obtained as described elsewhere(Reference van Dielen, van't Veer and Schols19). Polyclonal rabbit antisera, anti-TNF-R55 and anti-TNF-R75 were obtained by immunising rabbits with TNF-R55 and TNF-R75, respectively. We used the monoclonal antibodies that were kindly provided by Dr R. Devos (Hoffmann La-Roche, Welwyn Garden City, Herts, UK).
Plasma levels of CRP were measured by an immunoturbidimetric method, using a Hitachi 747 analyser (Roche Diagnostics, Mannheim, Germany).
Leucocytes were measured using a Sysmex SE9000 analyser (Sysmex Corporation, Kobe, Japan).
Human leucocyte antigen-DR expression
HLA-DR antigen expression was measured in fresh heparinised venous blood after lysis of erythrocytes within 1 h after blood sampling (Q prep; Coulter Corp., Miami, FL, USA). The absolute numbers of leucocytes and the percentage of monocytes (CD14+) were determined. The expression of the HLA-DR antigen on CD14+ cells was evaluated by FACS analysis (FACStar Plus; Becton Dickinson, San Jose, CA, USA) and was expressed as mean channel fluorescence intensity as described previously(Reference Boelens, Houdijk and Fonk20).
Serum levels of elastase were determined by immunoassay (Merck, Darmstadt, Germany) according to the manufacturer's instructions.
The interval and ratio variables are expressed as means and standard deviations. The Mann–Whitney U test was performed to analyse patient characteristics, tumour characteristics and the difference between the treatment group and the control group in change over time. Fischer's exact test was performed to analyse the nominal variables of the patient and tumour characteristics. Differences between the control and treatment group in the course of the post-operative immuno-inflammatory response were analysed using general estimating equations (GEE). GEE-analysis is a linear regression technique and is comparable with the mixed-model factorial ANOVA. However, none of the disadvantages of the mixed-model factorial ANOVA (for example, no missing data allowed, equal time intervals assumed, no ‘real’ effect measures) is present in GEE-analysis(Reference Twisk21).
The Mann–Whitney U test and Fischer's exact test were performed with SPSS 14.0 for Windows® (SPSS, Inc., Chicago, IL, USA). GEE-analysis was performed with STATA® (version 7.0; StataCorp LP, College Station, TX, USA)(Reference Twisk21). For all analyses, P < 0·05 was considered statistically significant.
In total, twenty-seven patients were considered eligible for enrolment in the study, of which eleven patients were included in the control group and ten patients in the treatment group. Six patients did not meet the inclusion criteria and were therefore excluded from the study. In the control group one patient died on the second day after surgery, before receiving any enteral feeding. Another patient in the control group refused further blood sampling. According to the principle of ‘intention-to-treat analysis’, results of all twenty-one patients were analysed. The control and treatment group were comparable with respect to anthropometrics, surgery and tumour characteristics (see Tables 2 and 3 in van Stijn et al. (Reference van Stijn, Ligthart-Melis and Boelens16)).
Patients reached an average 60 % of their daily energy expenditure, and there were no differences between the two groups (control group: 1944 (sd 133) kcal (8134 (sd 556) kJ); treatment group: 1750 (sd 113) kcal (7322 (sd 473) kJ)) (see Fig. 1 in van Stijn et al. (Reference van Stijn, Ligthart-Melis and Boelens16)).
Most immuno-inflammatory markers showed no differences between the groups. Therefore only CRP, IL-6, TNF-R75 and HLA-DR are shown in Fig. 1. In both groups an expected rise in CRP up to 3 d was followed by a decrease. The course of CRP over time was significantly lower in the treatment group (P = 0·04; Fig. 1(a)). The levels of IL-6 (Fig. 1(b)), IL-1, IL-8 and TNF-R55 showed similar courses but no differences were found between the control and treatment groups. A steady increase in TNF-R75 levels was found in the control group during the first 7 d after surgery (P = 0·001; Fig. 1(c)). During the first three post-operative days, TNF-R75 levels behaved significantly different in the treatment group in change over time compared with the control group (P = 0·04); levels continued to rise in the control group, whereas in the treatment group they decreased until the third post-operative day and then stabilised. HLA-DR expression was reduced in both groups (Fig. 1(d)). No significant differences between the groups were found for leucocyte, BPI and elastase levels.
Major surgical procedures are frequently associated with tissue damage and ischaemia–reperfusion injury, resulting in oxidative stress and deterioration of the immune response. After major surgery, antioxidant levels are low and antioxidant supplementation may help to moderate the inflammatory response in an attempt to improve post-operative course(Reference Park, Brody and Wallace1, Reference Decker, Tolba and Springer2). In the present prospective study, an enteral antioxidant formula (Module AOX) was studied in patients undergoing major upper gastrointestinal surgery.
The surgical trauma was reflected in the rise of the pro-inflammatory mediators CRP, IL-6, and TNF-R55 in both groups (Fig. 1(a)–(c)). After surgery, HLA-DR expression was significantly reduced, indicating an injury-related cellular immune deficiency, which is consistent with other reports (Fig. 1(d))(Reference Park, Brody and Wallace1, Reference Tashiro, Yamamori and Takagi22, Reference Spittler, Razenberger and Kupper23).
CRP is produced rapidly and at high levels, by the liver, during acute-phase reactions(Reference Black, Kushner and Samols24). CRP recognises pathogens by ligand binding which efficiently activates the classical complement cascade(Reference Black, Kushner and Samols24, Reference Marnell, Mold and Du Clos25). Module AOX had a moderating effect on the course of CRP production after surgery. After the initial rise in CRP level up to day 3 the decrease in CRP levels was more pronounced in the Module AOX group.
Major surgery induces the production of TNF-α, a cytokine that activates the innate immune system, by induction of other cytokine production, activation and expression of adhesion molecules, and growth stimulation. This potent biological activity of TNF can be harmful when not adequately controlled and can lead to systemic inflammatory response syndrome(Reference Decker, Tolba and Springer2, Reference Hehlgans and Pfeffer26). TNF-α is a ligand for the soluble TNF receptors TNF-R55 and TNF-R75 that are markers of the pro-inflammatory response and responsible for inactivation and clearance of TNF(Reference el Barbary and Khabar27, Reference Bemelmans, van Tits and Buurman28). In trauma patients who received enteral nutrition enriched with glutamine, reduced TNF-R75 levels were associated with lower infectious morbidity(Reference Houdijk, Rijnsburger and Jansen29). In the present study, from the moment Module AOX was administered at day 1, TNF-R75 levels decreased until the third post-operative day and then stabilised, whereas in the control group, TNF-R75 levels increased during the entire post-operative period. This seems to suggest that hyperactivity of the immune response might be moderated by administration of Module AOX.
Limitations of the study
The present pilot study was part of a feasibility study evaluating the safety and tolerance of Module AOX, which included only a small group of patients. Therefore, our findings on the inflammatory markers need to be interpreted with caution. The more severely ill and malnourished patients were not included, by using the 10 % 6-month weight loss as an exclusion criterion. It is known that better clinical effects in nutritional intervention studies can be expected in patients with a greater severity of illness and malnutrition, using the NRS-2002 (Nutritional Risk Screening) as the nutritional assessment(Reference Kondrup, Rasmussen and Hamberg30). Although the study was not strongly powered, significance was still reached for several immuno-inflammatory markers, despite the fact that more severely ill and malnourished patients were excluded. A more pronounced result can be expected with this kind of nutrition in severely ill and malnourished patients.
In conclusion, the present study shows that enteral nutrition enriched with glutamine and antioxidants is a safe and well-tolerated supplement to standard enteral nutrition. Major surgery induces a systemic immuno-inflammatory response with reduced HLA-DR expression and increased concentrations of CRP, IL-6 and TNF-R75. Enteral nutrition enriched with glutamine and antioxidants possibly moderates the immuno-inflammatory response (CRP, TNF-R75) after surgery. Larger studies are needed to reproduce these results and to investigate the effects of Module AOX on morbidity.
The present study was supported by a grant from Nestlé Nutrition.
P. G. B. and P. A. M. v L. designed and coordinated the study. P. G. B., G. C. L.-M. and J. D. collected the data. P. G. B., G. C. L.-M., M. F. M. v S., M. C. R., A. P. J. H. and P. A. M. v L. drafted the manuscript. G. C. L.-M., M. F. M. v S. and J. W. R. T. performed the statistical analysis. All authors read and approved the final manuscript.
There are no conflicts of interest.