Bacterial infections are frequently observed complications during the early postoperative course after liver transplantation (1) leading to increased morbidity and mortality, prolonged hospital stay, and additional costs. There are several reasons for this extraordinary high morbidity: the extensive operation and the immunosuppression given, but also the fact that most of the liver patients are more or less severely malnourished and protein-deficient (2–4). Clinical studies have verified a correlation between preoperative malnutrition and postoperative infections, morbidity, and mortality in liver transplant recipients (2,5). Total parenteral nutrition has been shown to be effective in achieving a positive nitrogen balance and to shorten stay in the intensive care unit when compared with no nutritional support (4,6).
It is obvious that a large proportion of the immune system is localized in the gut. Furthermore, most of pathogens isolated in patients with bacterial infections after liver transplantation are gut-derived (7). Translocation seems to be the major pathogenic factor for these infections (8). Translocation is especially facilitated in jaundiced or cirrhotic patients by the inadequate enterohepatic circulation of bile salts with subsequent bacterial overgrowth and leakage of endotoxin from the gut into the general circulation (9,10). The endotoxins activate gut and liver macrophages to release inflammatory mediators in abundance, leading to an exaggerated acute phase response and further malnutrition and immune suppression. Progressive endotoxemia during the anhepatic phase of a liver transplantation is also reported to correlate with allograft injury (11). Mucosal atrophy caused by preoperative malnutrition and postoperative fasting or total parenteral nutrition increases the rate of translocation and infections (12). The production of immunoglobulin A by the gut, known to prevent translocation, is drastically reduced during mucosal atrophy.
It has been increasingly observed in recent years that the protective gut flora is most often reduced or eliminated as a result of antibiotic treatment or malnutrition (13). During the last decade, two approaches have been tried to reduce translocation and infections after major abdominal surgery: enteral nutrition and selective bowel decontamination (SBD). Enteral nutrition has shown to stimulate bile flow and gut and liver blood flow, to maintain gut structure and function (10), and to be efficient in reducing postoperative morbidity (14,15). Unfortunately so far, mainly fiber-free formulas, which are resorbed in the small bowel and do not reach the colon, have been analyzed in clinical studies.
SBD with oral antibiotics has been extensively used after orthotopic liver transplantation (OLT), and most authors report not only a decrease in the incidence of Gram-negative infections but also an increase in Gram-positive infections (16–18). No overall benefit seems to be achieved by this form of treatment (19). Instead of destroying the gut flora with antibiotics, application of living Lactobacillus appears an attractive alternative. Some Lactobacillus species have been shown to initiate immunoglobulin production (20), to restore macrophage function (21), to stimulate apoptosis (22), and to modulate lymphocyte function (23). In addition, they have been shown to influence cytokine release (24), to increase mucin production (25), to eliminate toxins (26), and to stimulate mucosal growth (27). Lactobacillus has also reported, at least in experimental studies, to reduce permeability to mannitol in the colon, in sharp contrast to Escherichia coli, Klebsiella species, and Streptococcus viridans, which increase lumen-to-blood clearance of mannitol (28).
The Lactobacillus in the colon needs a regular supply of fibers, which they break down to short-chain fatty acids and other important nutrients, for their survival and function (29).
Here is reported a prospective randomized trial comparing the effects of three different treatment strategies on the incidence of early postoperative infections after liver transplantation; early enteral nutrition with SBD and fiber-free solutions, fiber-enriched formula plus living Lactobacillus plantarum 299, and heat-inactivated L plantarum 299.
MATERIAL AND METHODS
One hundred five patients scheduled for liver transplantation entered this prospective trial between October 1997 and October 1999. Those eligible for the study were adult patients undergoing OLT with side-to-side anastomosis of the bile duct. Excluded were patients with severe renal insufficiency (creatinine clearance <50 ml/min), intestinal obstruction (ileus), and cerebral disorders with danger of aspiration. In addition, Roux-en-Y reconstruction of the bile duct was an exclusion criterion.
The study was approved by the local ethics committee, and a written informed consent was obtained before operation from all patients. Ten patients (four of group 1 and 2, two of group 3) did not complete the study because of severe early complications, which made continuous enteral feeding impossible (six patients with relaparotomies within the first two postoperative days, two patients with early ileus, two patients with acute renal failure).
Patients were evaluated by complete history and physical examination, admission laboratory tests, and current weight. The cirrhosis was classified following the Child-Pugh classification. They were stratified using the classification of the American Society of Anesthesiologists (ASA) as follows:
- ASA 1: healthy patient
- ASA 2: patient with mild, controlled, functionally nonlimiting systemic disease
- ASA 3: patient with severe or poorly controlled systemic disease that is functionally limiting
- ASA 4: patient with severe systemic disease that is a constant threat to life
- ASA 5: moribund patient not expected to survive 24 hr with or without surgery
After stratification, patients were randomized by sealed envelope into one of the following three study groups.
Group 1 (SBD group).
Shortly before operation, patients received an enema. On the first postoperative day, a nasojejunal feeding tube was placed with the tip behind the ligament of Treitz. From the second postoperative day on, enteral nutrition (Fresubin, Fresenius, Homburg, Germany, 330 mOsm/L) was given via feeding tube. This formula provides 1000 kcal/L, 38 g protein/L, 138 g carbohydrate/L, and 34 g lipid/L. Feeding began with 25 ml/hr for 24 hr and when tolerated by the patient was advanced to 75 ml/hr on postoperative day 3 (2000 kcal/d) and continued until postoperative day 12. Crystalloid solutions were infused when clinically necessary. Oral intake started with clear liquids on postoperative day 1 and increased thereafter slowly. In addition, 5 ml of SBD containing 80 mg of tobramycin, 500 mg of amphotericin B, and 100 mg colistin sulfate was given orally four times a day for 6 weeks postoperatively.
Group 2 (enteral group with Lactobacillus and fibers).
Enteral nutrition using a supplemented enteral formula (Nutrison L.EN Fibre, Pfrimmer Nutricia, 210 mOsm/L) started within 24 hr after operation. This formula provides 1000 kcal/L, 40 g protein/L, 123 g carbohydrate/L, and 29 g lipid/L. Supplemental nutrients include 15 g/L fiber, divided into 0.6 g/L soluble and 14.4 g/L nonsoluble fibers. During the first 12 days, L plantarum 299 (AB Probi, Lund, Sweden) in a dose of 109 and oat fiber were added twice daily via the feeding tube. Oral intake was started as described for group 1.
Group 3 (enteral group with placebo).
Patients were treated as described for group 2. Instead of living lactobacilli, heat-killed L plantarum 299 (AB Probi) and oat fiber were supplied twice daily.
Postoperative immunosuppression consisted of cyclosporine or tacrolimus, together with prednisolone. Some patients also received an induction therapy with an interleukin-2-receptor antibody. Rapamycin or mycophenolate mofetil were added, if clinically indicated.
Antibiotic Prophylaxis and Treatment
All patients received an intravenous antibiotic prophylaxis with ceftriaxone (2 g twice daily) and metronidazole (500 mg twice daily) 30 min before surgery and until 2 days after surgery. After that, antibiotics were only continued in case of suspected or proven infection or in case of severe complications.
Infections, when occurring, were treated with ciprofloxacin (2×400 mg i.v.) or in accordance with recommendations after cultivation and testing of antibiotic resistance of the isolated bacteria.
Length of surgical procedure and amounts of transfused blood and blood products during and after operation were recorded. Length of stay in the intensive care unit, the first day of bowel movement, side effects of enteral and parenteral nutrition, and kind and amount of antibiotic treatment were also evaluated. Complications including infections, rejections, bile leak, postoperative hemorrhage, primary nonfunction of the transplanted liver, retransplantation, impaired liver and kidney function, and reoperations were recorded. Length of hospital stay was defined as time from the day of operation to the day of discharge.
Blood samples were taken preoperatively and on postoperative days 1, 5, and 10, and the following variables were studied: leukocyte count and differential count, hemoglobin, hematocrit, platelet count, partial thromboplastin time, fibrinogen, prothrombin time, bilirubin, aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transpeptidase, alkaline phosphatase, protein, albumin, sodium, potassium, phosphate, glucose, calcium, blood urea nitrogen, creatinine, cholesterol, triglycerides, amylase, lipase, C-reactive protein, prealbumin, and IgA. In addition, the cellular immune status was measured using a FACScan (CD3+, CD4+, CD8+, CD19+, CD45RA/RO cells, natural killer cells, and the CD4/CD8 ratio).
Definition of Bacterial Infection
A bacterial infection was diagnosed in case of fever (body temperature >38°C), leukocytosis with a left shift of the differential, elevation of C-reactive protein, and specific signs of bacterial infection that resolve under antibiotic treatment such as the following:
- pneumonia: pulmonary infiltrate, dyspnea, and reduced arterial oxygen
- cholangitis: elevation of alkaline phosphatase and γ-glutamyl transpeptidase, a positive culture from the T-tube, or histologic-proven cholangitis
- sepsis: sepsis syndrome with positive blood cultures
- urinary tract infection: dysuria and bacteriuria >10,000 colony-forming units/ml
- wound infection: pus
Surveillance of Infection
Body temperature was measured three times daily. In case of suspected infection, chest x-rays and ultrasound sonography of the abdomen were performed. Cultures were taken twice weekly from urine, blood, abdominal drainages, feeding tubes, sputum, and wounds. If pneumonia was suspected, a bronchoalveolar lavage was done.
Statistical analysis was performed using the SPSS 10.0 program (SPSS Inc., Chicago, Illinois). To compare discrete variables, the extended chi-square test was used. For nonparametric analysis of continuous distributed variables, the Mann-Whitney U test and the Kruskal-Wallis test were used. A P <0.05 was regarded as statistically significant with a power of 80%.
Ninety-five patients in total completed the study. Patient demographics and general state of health for the groups are presented in Table 1. Regarding the ASA classification, most patients were classified as ASA 3 (19 patients in group 1; 24 patients in group 2; 25 patients in group 3). 10, 6, and 7 patients were classified as ASA 2 in groups 1, 2, and 3, respectively, and 3, 1, and 0 patients as ASA 4 in groups 1, 2, and 3, respectively. There were no significant differences among the groups for mean age, gender, Child-Pugh classification of cirrhosis, and preoperative ASA classification.
Operative data are also listed in Table 1. The mean length of surgical procedure was comparable. The amount of intraoperative blood transfusions was higher in the living Lactobacillus group, and the amount of postoperative blood transfusions was higher in the SBD group, but these differences were not statistically significant.
Postoperative immunosuppression consisted of cyclosporine in 11 patients of group 1, 16 patients of group 2, and 13 patients of group 3. Twenty-one group 1 patients, 15 group 2 patients, and 19 group 3 patients received tacrolimus. In all patients prednisolone was administered. Sixty patients received additional immunosuppression with either an interleukin-2-receptor antibody (n=44), rapamycin (12) or mycophenolate mofetil (n=4). The differences regarding immunosuppression were not statistically significant.
Preoperatively, serum albumin was significantly lower in group 2 (3.42±0.12 g/dl) than in group 1 (3.84±0.12 g/dl). The other variables were equally distributed. On postoperative days 5 and 10, serum creatinine was significantly higher in group 1 (1.7±0.25 mg/dl and 1.9±0.3 mg/dl) compared with group 2 (1.1±0.09 mg/dl and 1.3±0.1 mg/dl). In addition, blood urea nitrogen was also higher in group 1 on postoperative day 10 (111±10 mg/dl versus 85±8 mg/dl). The other variables did not differ.
The course of changes in leukocyte count is depicted in Figure 1. Leukocytes were lower in group 2 than in the other groups, but the difference was not statistically significant.
Cellular immune variables were comparable within the groups throughout the whole study period. Figure 2 shows the course of the CD4/CD8 ratio. Although it was higher in group 2 than in groups 1 and 3, the difference was not statistically significant (P =0.06).
Postoperative infections and other complications.
The rate and kind of postoperative infections and the isolated bacteria in the three groups are shown in Table 2. Significantly (P =0.017) more infections occurred in group 1 (48%) compared with group 2 (13%). The incidence of infections in group 3 was also lower than in group 1 (34%), but this difference did not reach statistical significance. Cholangitis and pneumonia were the most frequent infections. Enterococci were the predominant bacteria isolated in all groups (n=17), but most pronounced in group 1. Other frequent pathogens seen were staphylococci and E coli. Enterococci and staphylococci were seen only in one patient each in the group supplied living Lactobacillus. The mean time of onset of infection after transplantation was 11.7 days in group 1, 12 days in group 2, and 12.9 days in group 3.
The mean cumulative length of antibiotic therapy is listed in Table 2. Patients in groups 1 and 3 received antibiotic therapy longer than patients in group 2, but this difference was not statistically significant.
Noninfectious complications occurred in 15 patients of group 1, 16 patients of group 2, and 19 patients of group 3. The incidence of acute rejections was 10 of 32 in group 1 (four with OKT3 therapy), 10 of 31 in group 2 (two with OKT3 therapy), and 15 of 32 in group 3 (three with OKT3 therapy). The rate of renal insufficiencies requiring hemodialysis was 8 of 32 in group 1, 2 of 31 in group 2, and 4 of 32 in group 3. Six patients of group 1 had a relaparotomy performed because of hemorrhage, bile leakage, and primary nonfunction of the graft compared with four patients in group 2 (hemorrhage, biliary leak) and two patients in group 3 (hemorrhage, arterial stenosis). No perioperative deaths occurred.
Bowel movement, weight, and length of hospital stay.
The first bowel movement occurred on average after 2.7 days in group 1 compared with 2.2 days in group 2 and 2.4 days in group 3. This difference was not statistically significant.
The length of hospital stay in the intensive care unit was longest in group 1, followed by groups 3 and 2 (Table 2). The total length of hospital stay was also longer in group 1 (39 days) compared with group 3 (36 days) and group 2 (35 days). However, none of these differences were statistically significant.
Side effects of the nutrition formulas.
Both enteral formula and Lactobacillus were well tolerated. Abdominal side effects (distension, cramps, or diarrhea) were seen in 8 of 32 patients in the SBD group (group 1) compared with 6 of 31 patients in the live Lactobacillus group 2 and 11 of 32 patients in the inactivated Lactobacillus group (group 3).
In this prospective, randomized study, a new concept with early enteral nutrition using a fiber-enriched formula and live Lactobacillus achieved a significant reduction of bacterial infections after liver transplantation compared with supply of no fibers or SBD. Living Lactobacillus was more effective than inactivated Lactobacillus.
The Baylor Medical Center liver transplant group reported a decrease of bacterial infections from 29% in nonfed patients to 14% in patients fed enterally (30). However, Wicks et al. (3) found no difference in bacterial infections between patients on parenteral nutrition or enteral nutrition, but the study was probably too small as only 24 patients were enrolled. Significant positive effect of early enteral nutrition after general abdominal surgery in comparison with parenteral nutrition has in recent years been demonstrated both in experimental and clinical studies (31). The fact that enteral nutrition decreases the incidence of postoperative infections and morbidity does not only seem to relate to improvement in nutritional factors. Shirabe et al. (32) compared parenteral and enteral nutrition after major hepatic resection, and found no significant differences in levels of prealbumin, transferrin, or retinol binding protein. In contrast, immunologic factors such as natural killer activity, phytohemagglutinin response, and lymphocyte numbers were enhanced in the enteral group, and the incidence of infections was significantly lower (31% vs. 8%).
Although preoperative albumin was significantly lower in the Lactobacillus group in our study compared with the SBD group, creatinine and blood urea nitrogen were significantly higher on postoperative days 5 and 10 in the patients receiving SBD. In addition, more patients required hemodialysis. Whether this effect is really caused by more effective nutrition and prevention of a catabolic state cannot be answered by the present study. Most of the cellular immune variables did not differ in our study. Only the CD4/CD8 ratio was higher in the live Lactobacillus group than in the SBD group, and this difference did not reach statistical significance. Immune variables are difficult to compare in the transplant setting because they are mainly influenced by immunosuppressants. For bacterial infections, B cells are the most important marker, but B cells were not different in the three groups. As there were enormous clinical differences, the variables measured in this study do not seem to be good indicators for infections.
An additional immunologic effect of enteral nutrition has been demonstrated on the rate of early rejection (33). In the mentioned study, enteral nutrition reduced the rate of early rejection from 44% in the parenteral group to 7%. The rejection rate was not significantly different in our patients, but all of them received enteral nutrition and immunosuppression was not standardized.
As observed by other groups in the past, Gram-positive bacteria and especially enterococci were the most frequent pathogens isolated in the SBD group. These bacteria are not destroyed by the antibiotic combination used. Enterococci were responsible for 30% of bacteremia episodes in the intensive care unit in a report from Baltimore (34). Seventy percent of Enterococcus faecium were vancomycin-resistant. SBD caused a reduction of nosocomial infections in most of the published studies, but there was no difference in overall infection rate and length of hospital stay, and the survival was not improved. Costs were substantially increased (16–18). In contrast, living Lactobacillus reduced the incidence of infections, the amount of antibiotics administered, and the length of hospital stay. Although no cost-analysis was done in the present study, we assume that costs were lower in the living Lactobacillus group because costs of the two nutrition formulas did not differ and lactobacilli are much cheaper than SBD. As heat-inactivated Lactobacillus did not alter the rate of infections, the effect cannot be caused by fiber alone. Several studies show that it is possible by enteral immunonutrition to enhance immune response and to control infection. Preliminary studies in patients with multiple organ failure demonstrated a significant improvement in Acute Physiology and Chronic Health Evaluation II scores during 5 and 10 days of treatment with Lactobacillus and fiber (36).
An alternative not tried, but possible in elective operations like many liver transplantations, could be to pretreat the recipient before operation with pre- and probiotics. In future, efforts should be made to start enteral nutrition and supply both pre- and probiotics before liver transplantation to the recipient, but also to the donor, to prevent endotoxemia during transplantation. This approach should be especially interesting in living related donor settings.
1. Engelhard D. Bacterial infections. In: Bowden RA, Ljungman P, Paya CV, eds. Transplant infections. Philadelphia: Lippincott-Raven, 1998: 153.
2. Pikul J, Sharpe MD, Lowndes R, Ghent CN. Degree of preoperative malnutrition is predictive of postoperative morbidity and mortality in liver transplant recipients. Transplantation 1994; 57 (3): 469.
3. Wicks C, Somasundaram S, Bjamason I, et al. Comparison of enteral feeding and total parenteral nutrition after liver transplantation. Lancet 1994; 344: 837.
4. Reilly J, Mehta R, Teperman L, et al. Nutritional support after liver transplantation: a randomized prospective study. J Parent Ent Nutr 1990; 14: 386.
5. Figueiredo F, Dickson ER, Pasha T, et al. Impact of nutritional status on outcomes after liver transplantation. Transplantation 2000; 70: 1337.
6. Buzby GP. Veterans Affairs Total Parenteral Nutrition Cooperative Study Group: perioperative total parenteral nutrition in surgical patients. N Engl J Med 1991; 325: 525.
7. Schwartz MN. Hospital-acquired infections: diseases with increasingly limited therapies. Proc Natl Acad Sci USA 1994; 91: 2420.
8. Wiesner RH, Hermans P, Rakela J, et al. Selective bowel decontamination to prevent Gram-negative bacterial and fungal infection following orthotopic liver transplantation. Transplant Proc 1987; 1: 2420.
9. Cunningham-Rundles S, Lin DH. Nutrition and the immune system of the gut. Nutrition 1998; 14: 573.
10. Helton WS. Nutritional issues in hepatobiliary surgery. Semin Liver Dis 1994; 14: 140.
11. Yokoyama I, Todo S, Selby R, et al. Endotoxemia and human liver transplantation. Transplant Proc 1989; 21: 3833.
12. Bengmark S, Anderson R, Mangiante G. Uninterrupted perioperative enteral nutrition. Clin Nutr 2001; 20: 11.
13. Bengmark S. Gut environment and immune function. Curr Opin Clin Nutr Metab Care 1999; 2: 83.
14. Windsor ACJ, Kanwar S, Li AGK et al. Compared with parenteral nutrition, enteral feeding attenuates the acute phase response and improves disease severity in acute pancreatitis. Gut 1998; 42: 431.
15. Gianotti L, Braga M, Vignali A, et al. Effect of route of delivery and formulation of postoperative nutritional support in patients undergoing major operations for malignant neoplasm. Arch Surg 1997; 132: 1222.
16. Cerra FB, Maddaus MA, Dunn DL, et al. Selective gut decontamination reduces nosocomial infections and length of stay but not mortality or organ failure in surgical intensive care unit patients. Arch Surg 1992; 127: 163.
17. Gastinne H, Wolff M, Delatour F, et al. A controlled trial in intensive care units of selective decontamination of the digestive tract with nonabsorbable antibiotics. N Engl J Med 1992; 326: 594.
18. Hammond JM, Potgieter PD, Saunders GL, Forder AA. Double-blind study of selective decontamination of the digestive tract in intensive care. Lancet 1992; 340: 5.
19. Condon RE. Selective bowel decontamination. Arch Surg 1990; 125: 1537.
20. Yasui H, Nagaoke N, Mike A, et al. Detection of Bifidobacterium
strains that induce large quantities of IgA. Microb Ecol Health Dis 1992; 5: 155.
21. Hatcher GE, Lamprecht RS. Augmentation of macrophage phagocytotic activity by cell-free extracts of selected lactic acid-producing bacteria. J Dairy Sci 1993; 76: 2485.
22. Heerdt BG, Houston MA, Augenlicht LH. Potentiation by specific short-chain fatty acids of differentiation and apoptosis in human colonic carcinoma cell lines. Cancer Res 1994; 54: 3288.
23. Kirjavainen PV, El-Nezami HS, Salminen SJ, et al. The effect of orally administered viable probiotic and dairy lactobacilli on mouse lymphocyte proliferation. FEMS Immunol Med Microbiol 1999; 26: 131.
24. Henderson B, Poole S, Wilson M. Microbial/host interactions in health and disease: who controls the cytokine network? Immunopharmacology 1996; 35: 1.
25. Mack DR, Michail S, Wei S, et al. Probiotic inhibit enteropathogenic E. coli
adherence in vitro by inducing intestinal mucin gene expression. Am J Physiol 1999; 276: G941.
26. Haskard C, Binnion C, Ahokas J. Factors affecting the sequestration of aflatoxin by Lactobacillus rhamnosus
strain GG. Chem Biol Interact 2000; 128: 39.
27. Ichikawa H, Kuroiwa T, Inagaki A, et al. Probiotic bacteria stimulate epithelial cell proliferation in rats. Dig Dis Sci 1999; 44: 2119.
28. Garcia-Lafuente A, Antolin M, Guarner F, Crespo E, Malagelada JR. Modulation of colonic barrier function by the composition of the commensal flora in the rat. Gut 2001; 48: 503.
29. Bengmark S. Immunonutrition: role of biosurfactants, fiber, and probiotic bacteria. Nutrition 1998; 14: 585.
30. Hasse J, Blue L, Liepa G, et al. Early enteral nutrition support in patients undergoing liver transplantation. JPEN J Parenter Enteral Nutr 1995; 19: 437.
31. Moore FA, Feliciano DV, Andrassy RJ, et al. Total enteral feeding, compared with parenteral, reduces postoperative septic complications: the results of a meta-analysis. Ann Surg 1992; 216: 172.
32. Shirabe K, Matsumata T, Shimada M, et al. A comparison of parenteral hyperalimentation and early enteral feeding regarding systemic immunity after major hepatic resection: the results of a randomized prospective study. Hepatogastroenterology 1997; 44: 205.
33. Sharpe MD, Pikul J, Lowndes R, et al. Early enteral feeding reduces incidence of early rejection following liver transplantation [Abstract]. London: Joint Congress of Liver Transplantation, 1995.
34. Mainous MR, Lipsett PA, O’Brien M, and the Johns Hopkins SICU group. Enterococcal bacteremia in the surgical intensive care unit: does vancomycin resistance affect mortality? Arch Surg 1997; 132: 76.
36. Bengmark S. Econutrition and health maintenance: a new concept to prevent GI inflammation, ulceration and sepsis. Clin Nutr 1996; 15: 1.