|
|
||||||||
Institute of Nutrition and Food Technology, University of Chile, Santiago, CHILE
| ABSTRACT |
|---|
|
|
|---|
Aim: To demonstrate that nutritional support in ambulatory alcoholic cirrhotic patients improves host defenses.
Methods: Thirty-one male outpatients with alcoholic cirrhosis CHILD-PUGH B or C were included. Twenty-five subjects completed six months consuming daily a nutritional supplement (Ensure®, 1000 Kcal and 35 g protein), in addition to their regular diet. At entrance and every three months, a clinical assessment, nutritional evaluation and indirect calorimetry were performed. Liver function tests and LPS-induced monocyte production of cytokines, salivary secretory IgA, lactulose/mannitol ratio and breath hydrogen tests were also measured in these intervals. Delayed cutaneous hypersensitivity and IgG and IgM antibody response to endotoxin were assessed at entrance and at the end of the study.
Results: Patients drank 85% of the provided supplement as an average. REE, total body fat and serum albumin increased, basal breath hydrogen decreased and cellular immunity improved significantly during the follow up period (p
0.03). All the other parameters remained unchanged throughout the study. Six patients (16.2%) died during the study, five due to upper gastrointestinal bleeding.
Conclusion: Nutritional support in alcoholic cirrhotic patients improves nutritional status and cell mediated immunity.
Key words: Nutrition support, cirrhosis, immunity
| INTRODUCTION |
|---|
|
|
|---|
Nutrition critically affects the immune system, involving both the antigen-nonspecific and the adaptive or antigen-specific responses. Several investigations have confirmed the adverse effects of protein-energy malnutrition (PEM) over different aspects of host defenses [19] and probably also the gut barrier [20,21,22]. Interestingly, it has been demonstrated that immune alterations recover with nutritional replenishment [23].
As malnutrition adversely affects the outcome of many chronic and acute diseases [24,25,26], numerous trials have attempted to correct nutritional status in ALD, hoping to decrease morbidity and mortality [4,27,28].
We previously demonstrated that long-term nutritional support, using a standard enteral formula, reduced the number of admissions due to infections in Child B and C alcoholiccirrhotic patients [4]. The present study was designed to clarify the mechanisms underlying this result. Thus, our aim was to demonstrate that nutritional support in ambulatory alcoholic cirrhotic patients improves host defenses.
| PATIENTS AND METHODS |
|---|
|
|
|---|
a) Clinical evidence of alcoholic liver disease. Patients were stratified according to CHILD PUGH classification [29] B or C at the time of enrollment.
b) A history of at least five years of heavy alcohol consumption (daily alcohol intake>150 g).
c) Absence of hepatitis B surface antigen or hepatitis C antibody.
d) Absence of significant renal, pulmonary or cardiac disease, clinical diabetes or malignant tumors (including hepatoma).
e) Residence in the city where the study was performed (Santiago-Chile).
The study was approved by our local ethics committee, and all eligible patients signed a written informed consent. Each subject was instructed to consume, in addition to his regular diet, one liter of a commercial enteral formula (Ensure®, Abbott Laboratories), that provided 35 g protein, 1000 Kcal/day and 800 mg sodium. The total amount of fluid consumed per day was aimed at 2500 mL/d.
Patients were seen twice a month by a nurse practitioner at the liver disease clinic. On each visit, they were invited to void in a small container to measure urine alcohol with reactive strips (Alcohol Dipstick). They were asked about any pharmacological treatment received, alcohol ingestion and compliance with the nutritional support and were given a new supply of Ensure®. The compliance was measured by interrogating the patient about the amount of daily supplement ingested, the Ensure® leftover, the acceptability or adverse effects and the number and kind of regular meals.
Subjects were examined by a physician at entrance to the study and monthly or more frequently, if necessary, during a six month period. Every third month, a complete clinical and nutritional assessment was performed, including anthropometric measurements (weight, midarm circumference, triceps skinfold thickness using a Lange caliper at four standard locations and hand grip muscle strength using a hand grip dynamometer (model No. 0032, Therapeutic Instruments). Subjective Global Assessment (SGA) of nutritional status was performed only at entrance to the study [30]. Additionally, body fat was measured by Dual Energy X-ray absortiometry (DEXA) in a LUNAR DPX-L densitometer (LUNAR Corp Madison, Wisconsin USA), and resting energy expenditure was assessed by indirect calorimetry in a canopy system (Sensor Medics 2900). Conventional treatment for encephalopathy (metronidazole and lactulose) and ascites was given if required.
Fasting blood samples were obtained at the beginning of the study and every three months, to measure in vitro lipopolysaccharide (LPS)-stimulated peripheral blood monocyte cell (PBMC) production of Il-1ß, Il-6 and TNF-
, and routine laboratory tests (packed red cell volume, erythrocyte sedimentation rate [ESR], albumin, creatinine, blood urea nitrogen, total bilirubin, alkaline phosphatase, aspartate aminotransferase [AST] and prothrombin time) were performed. At the entrance and at the end of the study, serum IgG and IgM antibody responses to endotoxin were measured. At the same visit, a salivary sample was taken to measure secretory IgA. The lactulose/mannitol urinary excretion and a basal breath hydrogen test were performed to estimate intestinal permeability and microbiological contamination of the small intestine, respectively.
At the beginning and at the end of the study, cell mediated immunity was assessed using a delayed hypersensitivity skin multitest with seven antigens (MULTITEST IMC, Pasteur Merieux). Results were expressed as the sum of the major diameter of all positive reactions (mm).
| Analytical Procedures |
|---|
|
|
|---|
were measured as previously described [31] using a specific commercial enzyme-linked immunoabsorbent assay (R&D System, Minneapolis). IgG and IgM antibody responses to endotoxin were quantified using an enzyme-linked immunoabsorbent assay (EndoCAb), as previously described [32]. Salivary secretory IgA levels were measured by a radial immunodiffusion kit (The Binding Site Limited, Birmingham, England).
The lactulose/mannitol test was performed withdrawing antibiotics and lactulose seven days before the test. After an overnight fast, subjects drank 200 mL of a solution containing seven g lactulose and two g mannitol. Subsequently, urine was collected during the following five hours, where both sugars were measured by gas chromatography.
The breath hydrogen test was performed withdrawing antibiotics and/or lactulose one week before the test. After eight hours fast, expired air was collected using Milar bags with a two-way valve. Hydrogen was measured the same day by gas chromatography [33].
Results from the four methods just described were compared with values obtained in twelve age and socioeconomic-status paired healthy subjects.
If necessary, patients were admitted to the hospital and followed by the staff in charge of the protocol. Criteria for admission were the following:
a) Upper gastrointestinal bleeding evidenced by hematemesis, melena or rectal bleeding.
b) Progressive ascites despite the use of diuretics and salt restriction.
c) Progressive encephalopathy despite adequate ambulatory management (lactose and/or neomycin).
d) Clinical evidences of severe infections such as pneumonia or spontaneous bacterial peritonitis.
e) Any other life threatening condition.
In case of admission to another hospital (usually due to emergencies), relatives were instructed to provide information about such events and the outcome data (death or discharge diagnoses). All hospitalizations were registered.
If a patient died, the cause was recorded, and if he failed to attend two or more follow up visits, efforts were made to find him. Those who were not located after one month were considered definitively lost from control and not readmitted to the protocol even if they returned, although their clinical care was maintained.
| Statistical Analysis |
|---|
|
|
|---|
| RESULTS |
|---|
|
|
|---|
|
|
|
0.05) during the first three months and serum albumin at the end of the study compared with the basal assessment (Table 4). Liver function tests and salivary secretory IgA did not change during the study. Packed red cell volume increased, and ESR decreased significantly at the end of the study period compared with the basal assessment (Table 5).
|
|
, or lactulose/mannitol ratio were observed during 6-month (Table 6).
|
|
|
Serum IgG and IgM endotoxin antibodies levels did not change significantly during the study. However, those patients with persisting alcohol intake had higher initial and final levels of IgG endotoxin antibodies p<0.006 (1217±747 initial, 1380±626 MU/mL final) compared with abstinent patients (588±400 initial, 641±574 MU/mL final) (Fig. 3).
|
| DISCUSSION |
|---|
|
|
|---|
Throughout the study, the rate of hospitalizations due to infections was comparable to that of the supplemented group of our previous trial (three cases out of the 31 patients included in this study, versus two out of 26 in the former trial). Mortality was also similar to that of previous studies [4,34] and was determined mostly by upper gastrointestinal bleeding associated with alcohol ingestion, a factor we could not directly control.
Resting energy expenditure increased during the period of nutritional supplementation. This change probably reflects an increase in food intake or lean body mass accretion during this period [35]. Unfortunately, we were not able to demonstrate directly an increase in lean body mass, since body composition was assessed using DEXA, that estimates lean tissues incorrectly in patients with cirrhosis and renal failure, as it depends on the hydration status of soft tissues. Therefore, under the given conditions, this method is only reliable for the estimation of body fat mass, which contains a low proportion of water [36]. In fact, we observed an increase in this in our patients, as assessed by DEXA and skinfold thickness. Thus, we did not attempt to draw conclusions regarding the effects of nutrition support on lean body mass. Other causes of increased REE, such as infections and stress, were ruled out. Our patients were free of these complications at the entry to the study, and resting energy expenditure was always performed when no acute event was present.
In cirrhosis, both the hepatic and the whole body respiratory quotient are markedly reduced. In overnight fasted cirrhotic patients, lipid oxidation increases and glucose oxidation, decreases, compared with normal subjects. These changes in endogenous substrate use is similar to those of subjects adapted to prolonged starvation and persisted even after variable periods of nutritional support [12,13,37]. The behavior of our patients did not differ from that of those in other series.
Although no differences in delayed cutaneous hypersensitivity were observed between normal subjects and our patients in the basal period, an improvement in this parameter was observed after the nutritional support. A boosting effect of repeated skin test measurements is highly unlikely after six months, since it has only been observed when the tests are repeated after a short period, of three weeks or less [38]. Cirrhosis and alcoholism are both associated with immunological changes. Alcohol consumption appears to attenuate the production and migration of polymorphonuclear leukocytes and inhibits cell-mediated immunity [39,40]. Cirrhosis is associated with a lower T lymphocyte count, cutaneous anergy, chemotactic activity disturbances, decreased complement factors and depressed macrophage phagocytic activity [5]. These changes are very similar to those caused by protein energy malnutrition [41,42]. As discussed previously, we had no objective indicator of lean body mass depletion in our patients. However the low SGA of nutritional status at baseline likely reflects the nutritional derangement caused by alcoholism and cirrhosis. The later method has been extensively validated by us and other authors [30,43].
Bacterial translocation and intestinal permeability are increased in animal models of cirrhosis [44,45]. Other causes of elevated intestinal permeability are malnutrition and the metabolic response to injury [22,46]. Bacterial translocation is associated with a higher risk of developing gut derived infections such as spontaneous bacterial peritonitis and bacteraemia. It also activates phagocytic cells to produce lymphokines, whose adverse effects can perpetuate liver damage or induce wasting. Although endotoxemia is common in cirrhotic patients with portal hypertension [47,48], intestinal permeability has been reported to be normal [49], as in our patients. The lack of effect of nutritional support on intestinal permeability is consistent with other studies, performed in acutely ill patients, in whom this parameter did not improve after a successful nutritional replenishment [50]. Other factors such as acute infectious events or burns also influence intestinal permeability [51,52]. A normal intestinal permeability, assessed through the lactulose/mannitol ratio, does not exclude an increased bacterial translocation or a reduced clearance of intestinal bacteria in cirrhosis [43]. The effect of alcohol ingestion on endotoxin antibody levels probably indicates that ethanol increases bacterial translocation, although it was not reflected by changes in cytokine production or lactulose/mannitol ratio. We have not previously found modifications in lactulose mannitol ratio in alcoholics [53].
In this study, the reduction in breath hydrogen presumably reflects less intestinal contamination [54]. This effect could be due to an improvement in nutritional status or a more efficient management of intestinal microflora with oral antibiotics or lactulose. The lack of alcohol effect on breath hydrogen is in contradiction to the reports showing a higher intestinal contamination in recently drinking alcoholics [55]. Noteworthy, as mentioned before, our patients were receiving treatment for encephalopathy.
PBMC production of IL-1ß and TNF-
has been reported normal or elevated in cirrhotic patients [56]. An enhanced production of these cytokines is associated with acute events such as infections and acute alcoholic hepatitis [30]. IL-6 is elevated in patients with chronic liver disease, associated with ascites and end stage cirrhosis [57,58,59]. As expected in this study, performed in stable Child B or C cirrhotic patients PBMC production of IL1-ß and TNF-
at baseline were not different from normal controls, and IL-6 was higher. On the other hand, we did not observe changes in PBMC production of cytokines during the nutritional supplementation, suggesting that the initial low rate of endotoxemia did not worsen throughout the study.
Salivary secretory IgA levels were high in our healthy subjects and in cirrhotic patients, when compared with other reference populations. These differences are possibly explained by a high oral contamination that is common in low socioeconomic levels. Secretory IgA is lower in animals models of stress or starvation and in malnourished children [60,61]. The lack of changes with nutritional supplementation in our patients indicates that the regulation of salivary secretory IgA depends on multiple factors, besides nutrition.
Serum albumin levels increased during the nutritional intervention. However, this parameter is not a sensitive indicator of visceral protein storage in cirrhotic patients. Albumin concentrations levels are a function of its rate of synthesis, volume of distribution and catabolism. The causes of hypoalbuminemia in cirrhotic subjects are an enlarged volume of distribution and increased catabolic rate, without a compensatory increase in albumin synthesis, due to inadequate synthetic reserve, inadequate protein intake and frequent superimposed infections [28]. The observed improvement in serum albumin levels probably reflect an increase in protein intake and a low incidence of infections in these patients, since other liver function tests did not change. Volume depletion due to a more efficient ascites management could also play a role.
In summary, nutritional support in alcoholic cirrhotic patients improves nutritional status and host defenses due to better cellular immunity and less intestinal bacterial overgrowth. In addition, alcohol abstinence in these patients probably exerts an additive effect.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Received February 1, 1999. Accepted July 1, 1999.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. R. DiCecco and N. Francisco-Ziller Nutrition in Alcoholic Liver Disease Nutr Clin Pract, June 1, 2006; 21(3): 245 - 254. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. de la Maza, M. Gotteland, C. Ramirez, M. Araya, T. Yudin, D. Bunout, and S. Hirsch Acute Nutritional and Intestinal Changes after Pelvic Radiation J. Am. Coll. Nutr., December 1, 2001; 20(6): 637 - 642. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |