Liver Function Tests Abnormalities in Patients with Inflammatory Bowel Disease Receiving Artificial Nutrition: A Prospective Randomized Study of Total Enteral Nutrition vs Total Parenteral Nutrition* AGUEDA ABAD-LACRUZ,† FERRAN GONZALEZ-HUIX,‡ MARIA ESTEVE,† FERNANDO FERNÁNDEZ-BAÑARES,† EDUARD CABRÉ,† JAUME BOIX,† DOROTEO ACERO,‡ PERE HUMBERT,† AND MIQUEL A. GASSULL† From the
†Department of Gastroenterology, Hospital Universitari Germans Trias i Pujol, Badalona, Barcelona; ‡Gastroenterology Unit, Hospital de Girona, Girona, Catalunya, Spain
ABSTRACT. Liver and biliary abnormalities are well-known complications of inflammatory bowel disease (IBD). It has been
suggested that using total parenteral nutrition (TPN) may further impair liver function in these patients; this seems not to be so with total enteral nutrition (TEN). However, prospective trials comparing the incidence of liver function test (LFT) abnormalities with either TPN or TEN have not been carried out. Twenty-nine IBD inpatients with normal LFT, randomized to receive either TEN with a polymeric diet or isocaloric, isonitrogenous "all-in-one" TPN because of protein-energy malnutrition and/or severe disease, were included in the study. Sixteen patients (five with ulcerative colitis and 11 with Crohn’s disease) received TEN, and 13 patients (eight ulcerative colitis and five Crohn’s disease) were on TPN. All patients were on systemic steroids, and nine of them were on oral metronidazole. Both groups were homogeneous regarding age, sex, diagnosis, disease activity, nutritional status, daily nutrient supply, and days on artificial nutrition. Serum albumin levels significantly increased with TEN (32 ± 1 to 38.2 ± 1.6 g/liter, p < 0.01), but not with TPN (32.1 ± 2.2 to 33.9 ± 1.4 g/liter, NS). Clinical improvement occurred in both groups of patients as shown by the change in the disease activity indexes. In all cases, measurements of serum alkaline phosphatase, serum
-______n______-----Patients with active inflammatory bowel disease (IBD) often malnourished or prone to develop proteinenergy malnutrition,1-4 and artificial nutritional support, namely total parenteral nutrition (TPN), has become part of the treatment in these cases.1-4 Inasmuch as liver and biliary abnormalities are well-known complications of IBD,,5-’ it has been suggested that using TPN may further impair liver function in these patients.1o~11 Previous work by our group has shown that the use of total enteral nutrition (TEN) does not imply any additional risk for liver function in active IBD.12 The aim of this study was to prospectively compare the incidence of liver function test (LFT) derangement
bilirubin, aspartate aminotransferase, alanine aminotransferase, and γ-glutamyltransferase were performed weekly. There were no significant differences in the initial LFT between both groups. Eight of 13 patients (61.5%) in the TPN group developed some LFT abnormalities, whereas this only occurred in one of 16 patients (6.2%) in the TEN group (p 0.002). In most cases, LFT derangement was mild, the most frequent change being an increase of γ-glutamyltransferase. However, TPN had to be withdrawn in one patient because of severe =
LFT
derangement.
The
mean
period
of time
on
TPN between
patients developing or not LFT derangement was not significantly different (16.9 ± 1.25 us 18.2 ± 2.97 days). In the TPN group, all patients with Crohn’s disease developed LFT abnormalities, whereas this only occurred in three of eight patients with ulcerative colitis ( p 0.043). Possible causes—other than IBD or nutritional support—for LFT derangement did not significantly differ between TEN and TPN groups. These results confirm that LFT abnormalities are significantly more frequent in patients with IBD on TPN than in those on TEN. The maintenance of the integrity of the intestinal mucosa, promoted by the presence of nutrients in the gut lumen, may be necessary in preventing the development of liver damage. :618-621, 1990) (Journal of Parenteral and Enteral Nutrition 14 =
PATIENTS AND METHODS
are
in
patients with active IBD randomly receiving either
TEN
or
TPN.
Reprint requests: Dr. M. A. Gassull, Department of GastroenterolHospital Germans Trias i Pujol, Carretera del Canyet s/n, 08916 Badalona, Spain. * This paper was presented as a scientific poster at the 13th Clinical Congress of the American Society for Parenteral and Enteral Nutrition held in Miami Beach. FL, February 5-8, 1989.
ogy,
Patients
Twenty-nine of 36 IBD consecutive inpatients
ran-
random number generator, to receive either TEN or TPN because of protein-energy malnutrition and/or severe disease were included in the study. The remaining seven patients were excluded because of LFT abnormalities prior to the beginning of artificial nutritional support. In all cases the appropriate pharmacological and/or surgical therapy was carried out. The diagnosis of IBD was used based upon clinical.
domized, by
means
of
a
endoscopic, radiologic, isotopic (llln-oxine-leukocyte scan), and histologic findings. Sixteen patients (five ulcerative colitis, UC; 11 Crohn’s disease, CD) received TEN, and 13 patients (eight UC: five CD) were on TPN. All patients were on systemic steroids and none of them received Sulphasalazine or 5ASA. Metronidazole (250 mg t.i.d. per os) was administered to five CD patients of the TEN group and to four patients (three CD, one UC) of the TPN group because of perianal disease of suspicion of small intestine bacte618
Downloaded from pen.sagepub.com at MCMASTER UNIV LIBRARY on March 11, 2015
619
rial overgrowth. IBD activity indexes were calculated at the beginning of the study. The indexes of Truelove and Witts13 and Van Hees et aF4 were used to assess the activity of ulcerative colitis and Crohn’s disease, respectively. In order to perform the statistical analysis, the disease activity indexes were scored as follows: 1 inactive, 2 mild, 3 moderate, 4 severe. Protein-energy nutritional status was assessed by means of triceps skin-fold thickness, midarm muscle circumference, and serum albumin. A patient was considered to be malnourished when at least one of the above-mentioned nutritional parameters was below the 5th percentile of the age- and sex-matched healthy pop=
=
=
=
ulation.&dquo;5 Both groups of
patients proved to be homogeneous regarding age, sex, diagnosis, activity of the disease, and nutritional status (Table I). Liver Function Tests Serum total bilirubin, alkaline phosphatase, y-glutamyltransferase (GGT), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were measuredas part of the routine biochemical assessment (Thecnicon SMA-20 Autoanalyzer)-at the beginning and weekly until the end of the study. Both groups of patients were homogeneous regarding baseline LFT (Table I). Whole blood (or its derivatives) transfusions, surgical procedures, and septic episodes, previous to or during the period of study, were carefully recorded. All these factors were individually considered in each patient in order to evaluate their possible etiologic role in LFT derangement.
Nutritional
Support A high nitrogen polymeric diet (UNIASA, Granada, Spain) containing maltodextrin, whole protein (lactoalbumin and casein), and long-chain triglycerides (milk and vegetable origin) was used for TEN. Enteral diet was continuously infused through a fine bore nasogastric feeding tube with the aid of a peristaltic pump. The composition of the TPN formula included dextrose (10-30%), amino acids (Aminoplasmal L-10 Braun Characteristics and
mean
TABLE I LFT values (±SEM) at admission
of the patients studied
SE, Palex SA, Barcelona. Spain). and long-chain triglycerides (Intralipid 10-20 ~, Kabi Fides. Barcelona. Spain). TPN was infused through a silastic catheter into a central vein. The daily energy requirements were calculated by means of the Harris-Benedict equation modified by Long and for ideal weight.16 Patients in both groups were given no oral food or fluids. Both groups were homogeneous regarding the dailv amount of macronutrients, energy, and nitrogen administered ; the energy/nitrogen ratio; percent of nonprotein calories as lipid; and the mean duration of the nutritional support (Table II). Statistical Analysis Results are expressed as mean ± SEM. Chi-square test with Yates’ correction and Fisher’s exact test were used to test significant differences between categoric variables. Student’s t-test for unpaired data and MannWhitney U-test were used to test significant differences in mean values between TEN and TPN groups. Wilcoxon signed-ranks test was used to assess differences between both initial and final disease activity scores in both groups. RESULTS
TEN was nutritionally judged by the significant
effective than TPN, as increment in serum albumin concentration (32 ± 1 to 38.2 ± 1.6 g/liter in TEN group vs 32.1 ± 2.2 to 33.9 ± 1.4 g/liter in TPN group; p < 0.01). There were no changes in the remaining nutritional parameters studied. Clinical improvement occurred in both groups of patients as shown by the change in the disease activity scores (3.31 ± 0.15 to 2.31 ± 0.24, p < 0.05 in the TEN group; and 3.38 ± 0.21 to 2.61 ± 0.27, p < 0.05 in the TPN group). Eight of 13 patients (61.5%) in the TPN group developed some LFT abnormalities, whereas this only occurred in one of 16 patients (6.2%) in the TEN group (p 0.002). In the majority of cases, LFT derangement was mild, the most frequent change being an increase of GGT (Table III). However, in one patient TPN had to be withdrawn because of severe derangement of the LFT. There were no significant differences in the mean period more
=
TABLE II Mean nutrient supply
_
¿1.L...oo.L
~VL
1..l.-1.’-..)I1.1.1..I.-}
1Z.VI
1
4,V
1V.V
1
~
1.V
PE~1, protein-energy malnutrition. Normal LFT values parentheses.
1·iJ
are
in
~
~
Vitamins, minerals, and trace elements at the upper limit of the RDA. ANS. artificial nutritional support: BW. body weight. * Nonproteic kilocalories per gram of nitrogen.
Downloaded from pen.sagepub.com at MCMASTER UNIV LIBRARY on March 11, 2015
620
of parenteral and enteral diets in high-risk patients, such as those with IBD.3° The present study meets this statement. Both TPN and TEN groups were homogeneous regarding the activity of the disease, nutritional status, corticosteroid therapy, septic episodes, surgical procedures, transfusions, and other potential causes of liver derangement. LFT abnormalities were significantly more frequent in patients on TPN than in those on TEN. As
TABLE III Patients showing LFT abnormalities
has often been observed in short-term intravenous
Normal values: bilirubin, mol/liter; GGT,
†Department of Gastroenterology, Hospital Universitari Germans Trias i Pujol, Badalona, Barcelona; ‡Gastroenterology Unit, Hospital de Girona, Girona, Catalunya, Spain
ABSTRACT. Liver and biliary abnormalities are well-known complications of inflammatory bowel disease (IBD). It has been
suggested that using total parenteral nutrition (TPN) may further impair liver function in these patients; this seems not to be so with total enteral nutrition (TEN). However, prospective trials comparing the incidence of liver function test (LFT) abnormalities with either TPN or TEN have not been carried out. Twenty-nine IBD inpatients with normal LFT, randomized to receive either TEN with a polymeric diet or isocaloric, isonitrogenous "all-in-one" TPN because of protein-energy malnutrition and/or severe disease, were included in the study. Sixteen patients (five with ulcerative colitis and 11 with Crohn’s disease) received TEN, and 13 patients (eight ulcerative colitis and five Crohn’s disease) were on TPN. All patients were on systemic steroids, and nine of them were on oral metronidazole. Both groups were homogeneous regarding age, sex, diagnosis, disease activity, nutritional status, daily nutrient supply, and days on artificial nutrition. Serum albumin levels significantly increased with TEN (32 ± 1 to 38.2 ± 1.6 g/liter, p < 0.01), but not with TPN (32.1 ± 2.2 to 33.9 ± 1.4 g/liter, NS). Clinical improvement occurred in both groups of patients as shown by the change in the disease activity indexes. In all cases, measurements of serum alkaline phosphatase, serum
-______n______-----Patients with active inflammatory bowel disease (IBD) often malnourished or prone to develop proteinenergy malnutrition,1-4 and artificial nutritional support, namely total parenteral nutrition (TPN), has become part of the treatment in these cases.1-4 Inasmuch as liver and biliary abnormalities are well-known complications of IBD,,5-’ it has been suggested that using TPN may further impair liver function in these patients.1o~11 Previous work by our group has shown that the use of total enteral nutrition (TEN) does not imply any additional risk for liver function in active IBD.12 The aim of this study was to prospectively compare the incidence of liver function test (LFT) derangement
bilirubin, aspartate aminotransferase, alanine aminotransferase, and γ-glutamyltransferase were performed weekly. There were no significant differences in the initial LFT between both groups. Eight of 13 patients (61.5%) in the TPN group developed some LFT abnormalities, whereas this only occurred in one of 16 patients (6.2%) in the TEN group (p 0.002). In most cases, LFT derangement was mild, the most frequent change being an increase of γ-glutamyltransferase. However, TPN had to be withdrawn in one patient because of severe =
LFT
derangement.
The
mean
period
of time
on
TPN between
patients developing or not LFT derangement was not significantly different (16.9 ± 1.25 us 18.2 ± 2.97 days). In the TPN group, all patients with Crohn’s disease developed LFT abnormalities, whereas this only occurred in three of eight patients with ulcerative colitis ( p 0.043). Possible causes—other than IBD or nutritional support—for LFT derangement did not significantly differ between TEN and TPN groups. These results confirm that LFT abnormalities are significantly more frequent in patients with IBD on TPN than in those on TEN. The maintenance of the integrity of the intestinal mucosa, promoted by the presence of nutrients in the gut lumen, may be necessary in preventing the development of liver damage. :618-621, 1990) (Journal of Parenteral and Enteral Nutrition 14 =
PATIENTS AND METHODS
are
in
patients with active IBD randomly receiving either
TEN
or
TPN.
Reprint requests: Dr. M. A. Gassull, Department of GastroenterolHospital Germans Trias i Pujol, Carretera del Canyet s/n, 08916 Badalona, Spain. * This paper was presented as a scientific poster at the 13th Clinical Congress of the American Society for Parenteral and Enteral Nutrition held in Miami Beach. FL, February 5-8, 1989.
ogy,
Patients
Twenty-nine of 36 IBD consecutive inpatients
ran-
random number generator, to receive either TEN or TPN because of protein-energy malnutrition and/or severe disease were included in the study. The remaining seven patients were excluded because of LFT abnormalities prior to the beginning of artificial nutritional support. In all cases the appropriate pharmacological and/or surgical therapy was carried out. The diagnosis of IBD was used based upon clinical.
domized, by
means
of
a
endoscopic, radiologic, isotopic (llln-oxine-leukocyte scan), and histologic findings. Sixteen patients (five ulcerative colitis, UC; 11 Crohn’s disease, CD) received TEN, and 13 patients (eight UC: five CD) were on TPN. All patients were on systemic steroids and none of them received Sulphasalazine or 5ASA. Metronidazole (250 mg t.i.d. per os) was administered to five CD patients of the TEN group and to four patients (three CD, one UC) of the TPN group because of perianal disease of suspicion of small intestine bacte618
Downloaded from pen.sagepub.com at MCMASTER UNIV LIBRARY on March 11, 2015
619
rial overgrowth. IBD activity indexes were calculated at the beginning of the study. The indexes of Truelove and Witts13 and Van Hees et aF4 were used to assess the activity of ulcerative colitis and Crohn’s disease, respectively. In order to perform the statistical analysis, the disease activity indexes were scored as follows: 1 inactive, 2 mild, 3 moderate, 4 severe. Protein-energy nutritional status was assessed by means of triceps skin-fold thickness, midarm muscle circumference, and serum albumin. A patient was considered to be malnourished when at least one of the above-mentioned nutritional parameters was below the 5th percentile of the age- and sex-matched healthy pop=
=
=
=
ulation.&dquo;5 Both groups of
patients proved to be homogeneous regarding age, sex, diagnosis, activity of the disease, and nutritional status (Table I). Liver Function Tests Serum total bilirubin, alkaline phosphatase, y-glutamyltransferase (GGT), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were measuredas part of the routine biochemical assessment (Thecnicon SMA-20 Autoanalyzer)-at the beginning and weekly until the end of the study. Both groups of patients were homogeneous regarding baseline LFT (Table I). Whole blood (or its derivatives) transfusions, surgical procedures, and septic episodes, previous to or during the period of study, were carefully recorded. All these factors were individually considered in each patient in order to evaluate their possible etiologic role in LFT derangement.
Nutritional
Support A high nitrogen polymeric diet (UNIASA, Granada, Spain) containing maltodextrin, whole protein (lactoalbumin and casein), and long-chain triglycerides (milk and vegetable origin) was used for TEN. Enteral diet was continuously infused through a fine bore nasogastric feeding tube with the aid of a peristaltic pump. The composition of the TPN formula included dextrose (10-30%), amino acids (Aminoplasmal L-10 Braun Characteristics and
mean
TABLE I LFT values (±SEM) at admission
of the patients studied
SE, Palex SA, Barcelona. Spain). and long-chain triglycerides (Intralipid 10-20 ~, Kabi Fides. Barcelona. Spain). TPN was infused through a silastic catheter into a central vein. The daily energy requirements were calculated by means of the Harris-Benedict equation modified by Long and for ideal weight.16 Patients in both groups were given no oral food or fluids. Both groups were homogeneous regarding the dailv amount of macronutrients, energy, and nitrogen administered ; the energy/nitrogen ratio; percent of nonprotein calories as lipid; and the mean duration of the nutritional support (Table II). Statistical Analysis Results are expressed as mean ± SEM. Chi-square test with Yates’ correction and Fisher’s exact test were used to test significant differences between categoric variables. Student’s t-test for unpaired data and MannWhitney U-test were used to test significant differences in mean values between TEN and TPN groups. Wilcoxon signed-ranks test was used to assess differences between both initial and final disease activity scores in both groups. RESULTS
TEN was nutritionally judged by the significant
effective than TPN, as increment in serum albumin concentration (32 ± 1 to 38.2 ± 1.6 g/liter in TEN group vs 32.1 ± 2.2 to 33.9 ± 1.4 g/liter in TPN group; p < 0.01). There were no changes in the remaining nutritional parameters studied. Clinical improvement occurred in both groups of patients as shown by the change in the disease activity scores (3.31 ± 0.15 to 2.31 ± 0.24, p < 0.05 in the TEN group; and 3.38 ± 0.21 to 2.61 ± 0.27, p < 0.05 in the TPN group). Eight of 13 patients (61.5%) in the TPN group developed some LFT abnormalities, whereas this only occurred in one of 16 patients (6.2%) in the TEN group (p 0.002). In the majority of cases, LFT derangement was mild, the most frequent change being an increase of GGT (Table III). However, in one patient TPN had to be withdrawn because of severe derangement of the LFT. There were no significant differences in the mean period more
=
TABLE II Mean nutrient supply
_
¿1.L...oo.L
~VL
1..l.-1.’-..)I1.1.1..I.-}
1Z.VI
1
4,V
1V.V
1
~
1.V
PE~1, protein-energy malnutrition. Normal LFT values parentheses.
1·iJ
are
in
~
~
Vitamins, minerals, and trace elements at the upper limit of the RDA. ANS. artificial nutritional support: BW. body weight. * Nonproteic kilocalories per gram of nitrogen.
Downloaded from pen.sagepub.com at MCMASTER UNIV LIBRARY on March 11, 2015
620
of parenteral and enteral diets in high-risk patients, such as those with IBD.3° The present study meets this statement. Both TPN and TEN groups were homogeneous regarding the activity of the disease, nutritional status, corticosteroid therapy, septic episodes, surgical procedures, transfusions, and other potential causes of liver derangement. LFT abnormalities were significantly more frequent in patients on TPN than in those on TEN. As
TABLE III Patients showing LFT abnormalities
has often been observed in short-term intravenous
Normal values: bilirubin, mol/liter; GGT,
