This article was downloaded by: [New York University] On: 08 February 2015, At: 16:33 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of the American College of Nutrition Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uacn20
Abnormal liver function in malnourished patients receiving total parenteral nutrition: a prospective randomized study. a
a
a
a
a
J A Tayek , B Bistrian , N F Sheard , S H Zeisel & G L Blackburn a
Harbor-UCLA Medical Center, Torrance. Published online: 02 Sep 2013.
To cite this article: J A Tayek, B Bistrian, N F Sheard, S H Zeisel & G L Blackburn (1990) Abnormal liver function in malnourished patients receiving total parenteral nutrition: a prospective randomized study., Journal of the American College of Nutrition, 9:1, 76-83, DOI: 10.1080/07315724.1990.10720353 To link to this article: http://dx.doi.org/10.1080/07315724.1990.10720353
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Abnormal Liver Function in Malnourished Patients Receiving Total Parenteral Nutrition: A Prospective Randomized Study John A. Tayek, MD, FACN, Bruce Bistrian, MD, PhD, FACN, Nancy F. Sheard, DSc, RD, Steven H. Zeisel, MD, PhD, FACN, and George L. Blackburn, MD, PhD, FACN
Downloaded by [New York University] at 16:33 08 February 2015
Harbor-UCLA Medical Center, Terranee, California (JA.T.), New England Deaconess Hospital, Boston (B.B., G.L.B.), and Department of Pathology and Pediatrics, Boston University, School of Medicine, Boston (N.F.S., S.H.Z.) Key words: liver function tests, malnutrition, choline, intralipid total parenteral nutrition A prospective study was performed in clinically malnourished patients in which liver function was tested during a 4-week period of total parenteral nutrition (TPN). The purpose was to determine if concomitant intravenous lipid administration would reduce liver function abnormalities noted to occur frequently in patients receiving TPN. Twen ty-five patients were randomly assigned to receive either daily infusions of 200 cc of a 20% lipid emulsion with TPN or TPN without lipid for the first week. In the subsequent 3 weeks all patients received daily intravenous lipid. The early lipid treatment group received 0.7 g lipid/kg BW/day and approximately 280 mg of choline/day from the lecithin emulsifier throughout the entire study period. Liver function tests were performed twice in thefirstweek, then weekly thereafter. There were significant (p < 0.05) elevations in liver function tests in the early lipid treatment group (for aspartate aminotransferase in weeks 1, 2, and 3, and lactic acid dehydrogenase in weeks 2 and 3). Alkaline phosphatase activity was elevated at weeks 2, 3, and 4 for the lipid-treatment group and at week 1 for the lipid-restricted group. The two groups had a similar elevation in γ-glutamyltransferase activity. Analysis of covariance demonstrated that the overall duration of TPN, and not the presence or absence of intravenous lipid, was significantly related to the elevations in both alkaline phosphatase and γ-glutamyltransferase (GGT) levels. In contrast, the early intravenous administration of lipid was significantly related to the increase in aspartate aminotransferase levels. The peak increase in AST was noted at day 7 in the lipid-administration group. The characteristic liver function test abnormalities associated with TPN do occur and, if anything, are more prominent with daily provision of fat calories. Neither the absence of fat, essential fatty acid, nor choline is likely to be primarily responsible for the usual abnormalities in liver function associated with TPN seen after 7 days of TPN. We conclude that early intravenous administration of lipid emulsion with TPN does not reduce the overall occurrence of abnormal liver function.
INTRODUCTION Liver function (LFT) abnormalities are commonly seen during total parenteral nutrition (TPN). Significant increases in aspartate aminotransferase (AST), alkaline phosphatase activity and bilirubin levels occur in 87, 64, and 21% of patients, respectively, when treated with TPN [1,2]. LFT abnormalities occur as early as 5 days after TPN with AST rising first, followed by alkaline phosphatase, and then bilirubin [2,3]. Many hypotheses have been suggested as to the etiology of LFT abnor malities in humans, including excessive dextrose calories [4], camitine deficiency [5], absence of fat calories [6], essential fatty acid deficiencies [7], choline deficiency [8-10], hepatotoxins such as sodium bisulfite [3], abnormalities of amino acid concentrations [11], car-
nitine deficiency [5], and excessive lipid calories [12]. Since fat emulsions provide choline and linoleic acid and allow a reduction in glucose calories, we were inter ested in determining whether daily intravenous lipid ad ministration would limit the development of LFT abnor malities during TPN. Our hypothesis was that early ad ministration of lipids and a reduction in dextrose calories by the intravenous route would reduce LFT abnor malities noted to occur to patients receiving TPN.
MATERIALS AND METHODS Subjects Patients were randomly assigned to a lipid-restricted or daily lipid group as part of a study to determine the
Address reprint requests to Bruce R. Bistrian, M.D., Ph.D., Cancer Research Institute, 194 Pilgrim Road, Boston, Massachusetts 02215.
Journal of the American College of Nutrition, Vol. 9, No. 1, 76-83 (1990) © 1990 John Wiley & Sons, Inc.
CCC 0731-5724/90/010076-08$04.00
LFT Abnormalities on Total Parenteral Nutrition Table 1. Clinical Characteristics by Diet Classification Primary diagnosis
Lipid group (no.)
Inflammatory bowel disease
3
3
Chronic pancreatitis
2
2
Gastric outlet obstruction
2
1
Cancer
3
2
Appendectomy with wound dehiscence
1
0
Atherosclerotic vascular disease
1
2
s/p Mitral valve replacement
0
1
Chronic diarrhea
0
1
Diverticulitis Total
Downloaded by [New York University] at 16:33 08 February 2015
Lipid-re group
effect of TPN on plasma choline levels reported pre viously [13]. Twenty-five hospitalized patients, who met clinical criteria for requiring TPN, were studied (Table 1). There were 12 females and 13 males, ranging from 19 to 90 years of age (61.8 ± 17.4; mean ± SD). In dividuals with a diagnosis of acute hepatocellular disease (AST > 100 IU/L) or patients receiving hepatotoxic medications were not included (Table 1). All 25 patients had the secondary diagnosis of protein-calorie malnutri tion, defined by a low serum albumin and/or recent loss of 10% or more of usual weight. The protocol was ap proved by the Committee on Clinical Investigation of the New England Deaconess Hospital and all study par ticipants gave informed consent. Study Design Patients were randomly assigned to one of two groups. Group I (lipid) had 12 patients who received a combination of amino acids (10% Aminosyn, Abbott Laboratories), 70% dextrose, and daily lipid (20% Liposyn, Abbott Laboratories) during the first week of TPN therapy. Group Π had 13 patients who were lipidrestricted and received only amino acids and dextrose during week 1. After the first week, all subjects received daily intravenous lipid to prevent the development of essential fatty acid deficiency. The patients' oral or enterai diet (Osmolyte, Mead Johnson) was advanced as tolerated with no choline ingested by mouth. Patients were dropped from the study when their daily enterai intake exceeded 1000 calories over a 1-week period. The patients' basal energy expenditures (BEE) were deter mined by the Harris-Benedict equation [14], and energy
0
1
12
13
requirements were estimated to be approximately 40% above BEE. Protein requirements were calculated at 1.5 g protein/kg ideal body weight. All patients received MVI-12 vitamin mix (one vial per day) as well as a minimum of 2 mg of iron per day. Trace minerals (3 mg zinc, 1 mg copper, 0.4 mg manganese, 56 μg iodine, 10 μg chromium, and 50 μg selenium) were administered daily. Parenteral solutions were administered over 24 hours with the exception of lipid, which was ad ministered during the 8-hr afternoon shift. After 1 week, both groups of patients received 200 cc of 20% lipid infusions daily. Complete nutritional assessments including history of weight loss, actual weight, triceps skinfold, and arm muscle circumference were obtained by trained diet technicians. Serum albumin, transferrin, and total lym phocyte counts were also performed as part of the nutri tional assessment. Twenty-four hour urine collections were used to measure urinary creatinine excretion and creatinine—height indices were recorded. Nutritional in dices including the Prognostic Nutritional Index (PNI) [15], Hospital Prognostic Index (HPI) [16], and Catabolic Index [17] were calculated. Prognostic Nutri tion Index (PNI) = 158% - 16.6 · Albumin (g) - 0.78 · Triceps Skinfold (mm) - 0.2 · Transferrin (mg) - 5.8 · Delayed Hypersensitivity (Anergy = 1; Immunocompetence = 2). The PNI predicts the percent probability of subsequent death [15]. Hospital Prognostic Index = 0.91 · Albumin (g) - Delayed Hypersensitivity - 1.44 · Sepsis + 0.98 · Diagnosis - 1.09 (Diagnosis Cancer = 1, Noncancer = 2, Anergy = 1, Immunocompetence = 2). A number less than zero indicates a 50% probability of dying during hospital stay. The units are based on stand-
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
77
LFT Abnormalities on Total Parenteral Nutrition
Downloaded by [New York University] at 16:33 08 February 2015
ard deviations from the mean. Catabolic Index = g of urinary urea nitrogen excretion - [0.5 · nitrogen intake (g) + 3], where -5-0 = no metabolic stress, 0-5 = mild stress, and 5-10 = moderate to severe stress [17]. Blood samples were drawn from each patient both prior to the initiation of parenteral nutrition and throughout their course of therapy (1—4 weeks). Serum chemistries [albumin, transferrin, liver function tests (LFT), serum triglycérides], total lymphocyte counts, and 24-hr urine creatinine analyses were performed at the New England Deaconess Hospital's clinical laboratory. LFTs were performed on a Kodak Ektachem 700 spectrophotometer using established methods [18]. γ-Glutamyltranspeptidase was measured at baseline in only seven patients in each group and in four patients in each group at week 2.
Statistical Analysis The results were analyzed by analysis of covariance (time) with diet (lipid versus lipid-restricted) as the de pendent variable. Paired t-tests were then used with each individual acting as his own control. Individual values were compared to baseline values prior to TPN. All data are represented as mean ± SEM, unless otherwise noted, and significance is defined by the 95% confidence inter val.
RESULTS Nutritional Profile The 25 patients were successfully randomized into comparable groups with regard to diagnosis (Table 1), age, amount of weight loss, percent of ideal body weight, and other nutritional assessment data (Table 2). In addition to an average 5-kg weight loss, there was a marked reduction in the creatinine-height index in both the lipid-restricted (73 + 21% of standard) and non-lipidrestricted (67 ± 19% of standard) groups. Triceps skinfold were also markedly reduced (72 ± 9% and 63 ± 9% of standard, respectively. Serum albumin, transferrin, and total lymphocyte counts in the lipid and lipidrestricted groups, respectively, were all below normal at the outset of the study. The reductions in anthropométrie measurements and visceral protein levels indicated the presence of moderate to severe protein-calorie malnutri tion. Both groups were also comparably randomized by similar HPI (0.5 vs 0.4), and PNI (63 vs 59%) in the lipid vs lipid-restricted groups, respectively (Table 2), the latter index consistent with a group at moderate risk of complications. The catabolic index suggests minimal
78
metabolic stress [17] (Table 2). The BEE was 1280 + 50 kcal for the lipid-restricted group and 1240 ± 80 kcal for the nonrestricted group. There was no significant difference between the two groups for protein or caloric intake. Average protein in take was 80 g and average dextrose intake was 350 g/day. Total caloric intake over the 4-week study period averaged 1870 cai or 150% of the BEE (33 + 5 kcal/kg/day; mean ± SD). Oral intake was minimal during the first 3 weeks in both groups and averaged less than 400 kcal/day. There was no significant difference for oral or intravenous caloric intake between the groups (Table 3). Ten of 13 patients randomized to the lipidrestricted group received no intravenous lipid. The other three patients each received one dose of 200 cc of 20% Liposyn during the first week, or V7 of the dose given to patient in the lipid group. The quantity of intravenous lipid administered did not differ between groups after week 1. The amino acid solutions, vitamin mix, iron, and selenium solutions contain no choline. Liposyn (20%) contains trace amounts of free choline (4 nmol/ml) and 13.2 μπιοΐ/ml of phosphatidylcholine [13]. Thus, the daily administration of lipid provides an average of 0.7 μπιοΐ of free choline and 2310 μπιοΐ of phosphatidyl choline. The average amount of lipid administered each day was approximately 32 g or 320 cal, excluding the lipid-restricted group in the first week of parenteral nutrition. This amount of fat given daily represents roughly 20% of total calories or 0.7 g/kg/day. Most (19) patients received 2 weeks of parenteral therapy with a range of 9-28 days. Two patients who expired within the second week and three who had over 1000 calories enterally were dropped from the study. Nineteen patients completed a full 2 weeks of TPN therapy. During the third week, an additional eight patients ate well and were dropped, leaving only 12 patients for the third week. The final week an additional three patients were eating over 1000 cal/day and were dropped from the LFT analysis. Liver Function Abnormalities There was a significant effect of TPN on AST eleva tions in the early lipid group by analysis of covariance (p < 0.05). The mean AST activities for the early lipid group increased from the baseline of 30 ± 7 IU/L (mean ± SEM) to 58 ± 11 IU/L, 52 ± 18 IU/L, and 46+16 IU/L (mean + SEM) for weeks 1, 2, and 3, respectively (p < 0.05; Fig. 1). The peak increase was seen after 7 days of TPN. The lipid-restricted group, in contrast, had a mean AST value of 26 ± 6 IU/L (mean ± SEM), which in creased to 38 ± 9 IU/L at week 1 (p = 0.10) and 41 ± 10 IU/L at week 2 (p = 0.09), both values failing to reach
VOL. 9, NO. 1
LFT Abnormalities on Total Parenteral Nutrition Table 2. Nutritional Assessment Lipid group (mean ± SD)
Lipid-restncted group (mean ± SD)
Age (years)
59.7 ± 16.2
63.8 ± 18.1
Weight (kg)
55.3 ± 13.8
59.6 ±11.4
% Ideal body weight
91 ± 16
98 ± 1 9
Weight loss (kg)
5.0 ± 5.7
5.5 ± 6.2
% Standard arm muscle circumference
92 ± 14
92 ± 1 2
% Standard triceps skinfold
63 ± 3 1
72 ± 3 4
10.7 ± 4.6
11.1 ±6.1
Albumin (g/dl) (NL 3.2)
2.9 ± 0.7
2.9 ± 0.8
Transferrin (mg/dl) (NL 170)
138 ± 55
147 ± 71
14801940
1400 ± 880
-2.6 ± 3.5
0.1 ±4.4
0.5 ± 1.4
0.4 ±1.2
63 ± 2 2
59 ± 2 3
Downloaded by [New York University] at 16:33 08 February 2015
Triceps skinfold
Total lymphocyte count (NL 1500) Catabolic index
a
Hospital prognostic index
b
Prognostic nutritional indexc a
CataboIic index = g urinary urea nitrogen excretion - [0.5 nitrogen intake (g) + 3], where -5-0 = no metabolic stress, 0-5 = mild stress, and 5-10 = moderate to severe stress [17]. Hospital prognostic index obtained from [16]. A number less than zero indicates a greater probability of dying during hospital stay and units are based on standard deviations from the mean. •"Prognostic nutrition index = 158% - 16.6 albumin (g) - 0.78 triceps skinfold (mm) - 0.2 transferrin (mg) - 5.8 delayed hypersensitivity. PNI predicts the percent probability of subsequent death (anergy = 1; immunocompetence = 2) [15].
Table 3. Caloric Intakes by Vein and by Moutha Group
n
Week 1
12 13
1649 ± 7 7 1383 ± 9 8 (0.06)
Week 2
Week 3
Week 4
Intravenous calories Lipid group Lipid-restricted group (p value)
1340 ±124 1543 ±185 (0.39)
1540 ±150 1790 ±398 (0.48)
1369 ± 220 1887 ±144 (0.10)
Oral intake Lipid group Lipid-restricted group (p value)
12 13
b
156±69[4] 199 + 86 [4] (0.70)
234± 121 [2] 311 ±146 [4] (0.69)
390±113[6] 48 ± 3 1 [0] (0.07)
868 ± 221 [5] 191 ± 184 [1] (0.06)
a
Mean ± SEM, p value determined by nonpaired t-test not corrected for multiple comparisons. Number of patients whose oral intake averaged > 200 calories/day in each group.
statistical significance from baseline. The only sig nificant increases for the lipid-restricted group were the mean alkaline phosphatase activities, seen only at week
1 (127 ± 19 to 162 ± 24 IU/L; mean ± SEM; p < 0.05) and GGT seen only at week 2 (94 + 24 to 216 + 27 IU/L).
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
79
Downloaded by [New York University] at 16:33 08 February 2015
LFT Abnormalities on Total Parenteral Nutrition By analysis of covariance, the length of TPN ad ministration was an important variable for elevations in alkaline phosphatase and GGT activities (p < 0.01). A significant increase in alkaline phosphatase activity was observed at weeks 2, 3, and 4 in the lipid group by paired t-test (134 ± 33 to 176 ± 39, 223 + 53, 264 + 77; U/L; mean ± SEM). The lipid-restricted group did not show a significant rise in alkaline phosphatase activity after 1 week. LDH levels were also significantly in creased (p < 0.05) for the lipid group from a baseline of 179 + 25 IU/L to 238 ± 25 IU/L and 226 + 26 IU/L (mean ± SEM) for weeks 2 and 3, respectively. There were no significant increases in the lipid-restricted group. Both groups demonstrated significant elevations in γ-glutamyltransferase (GGT) activity at week 2. There was no significant elevation in mean bilirubin seen in either group (Table 4). There were no significant in creases in fasting plasma triglycéride levels noted be tween the two groups (Table 5).
duced liver dysfunction. Elevations of AST and alkaline phosphatase were seen in a study by Lowry et al [4] in which increases in these two enzymes occurred at 14 days of TPN. Another investigation found similar in creases in AST and alkaline phosphatase which occurred earlier, that is, after 10 days of TPN [20]. Grant and others described elevations in AST as early as the fifth day of TPN [3,21]. A recent study has described a stepwise increase in liver function abnormalities beginning with AST, alanine aminotransferase, alkaline phos phatase, and then bilirubin [2]. In the present study, we have observed that liver function abnormalities as measured by alkaline phosphatase and GGT activities were related to the duration of TPN, whereas abnor malities in AST activities were related only to diet com position, i.e., the early provision of lipid in the TPN. The peak elevation in AST levels occurred at day 7 for the early lipid group. The lipid-restricted group failed to demonstrate a significant change in AST levels.
LFT elevations observed in this study were far lower than those seen in hepatocellular disease and are rarely associated with a significant clinical problem. However, these changes that we observed are caused by the abnor mal liver function. Linder et al [1] have described a "meaningful increase" in liver function tests as a value equal to or greater than a 50% increase above baseline values, which results in a value within the abnormal range. A meaningful increase in AST was seen in 58% of the lipid group and in only 38% of the lipid-restricted group. Two patients in the lipid group expired during the second week of TPN. Both patients had meaningful in creases in AST and bilirubin levels but not alkaline phosphatase while on TPN. GGT levels were not measured. At autopsy, the patient with lymphoma, who died from disseminated intravascular coagulation, had focal fatty changes, intrahepatic and extracellular bile stasis, and cholestasis. The other patient, who died from sepsis related to Crohn's disease, demonstrated periportal cellular inflammation and moderate fatty changes in the liver. Both of these changes may be due to the under lying disease and not solely attributable to TPN.
From a theoretical point of view, providing parenteral fat with TPN would be expected to ameliorate changes in liver function by reducing glucose calories, and hepatic lipogenesis [22] and the parenteral fat should provide a source of linoleic acid and choline, the lack of which has been associated with fatty infiltration of liver [7,8,10]. However, contrary to our expectations, liver function was more adversely affected by the daily provision of fat at the outset of TPN. Hall et al have recently reported that both lipid-free and lipid-containing TPN treatments in rats are associated with a threefold increase in hepatic triglycéride content [22]. A recent study in the rat model has demonstrated that the administration of safflower oil calories (20% w/w) to choline-deficient rats produces a greater fatty infiltration of the liver when compared to starch calories [8]. How ever, we have no evidence of clinical choline deficiency in the lipid group when choline levels were maintained although not restored to levels seen in well-nourished individuals [13]. The early concomitant administration of lipid and TPN contributed to an increase in AST ac tivities, but liver biopsies were not available for the en tire group to confirm fatty liver infiltration.
DISCUSSION Among the 25 hospitalized patients who required in travenous hyperalimentation, those who received in travenous lipid at the onset of TPN developed more liver function abnormalities than those not given lipids during the first week (Table 4). The most common elevations were seen in AST, GGT, and alkaline phosphatase ac tivities. Elevations in GGT activities have been noted by Pallares et al [19] to be a sensitive parameter of TPN-in-
80
Liver function tests are abnormal when there is hepatic dysfunction, such as fatty infiltration and cellular damage; both have occurred in patients receiving parenteral nutrition [11]. Preventing LFT abnormalities requires identification of the etiological factors. We found that a significant increase in the liver function tests occurred in both groups of subjects during the course of this study which was not greater in magnitude or more frequent than has been previously described. Elevations of bilirubin and GGT activity occurred in equal percentages and to equal degrees in each group.
VOL. 9, NO. 1
LFT Abnormalities on Total Parenteral Nutrition
LIVER FUNCTION TEST ON TPN
IU/L
Downloaded by [New York University] at 16:33 08 February 2015
• ° * ** I
Lipid Group Lipid Restricted p " °,'°>?] vs. boseline volues p « 0.01 J mean * SEM
IU/L
350r- ALKALINE PHOSPHATASE IU/L
WEEKS Fig. 1. Liver function test results from patients receiving intravenous hyperalimentation (TPN) over a 4-week period. Mean values are compared to baseline values (paired t-testing) prior to TPN for AST (aspartate aminotransferase), LDH (lactic acid dehydrogenase), GGT (γ-glutamyltransferase), and alkaline phosphatase ac tivity. Analysis of covariance demonstrated diet (lipid group) to be significantly different for AST activity (p < 0.05), and time to be an important variable for abnor malities in GGT and alkaline phosphatase activity (p < 0.05; p < 0.001), respective ly. ( · ) Lipid group; (O) lipid-restricted group; *p < 0.05; **p < 0.01 vs baseline values.
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
LFT Abnormalities on Total Parenteral Nutrition Table 4. Lipid Treatment Group vs Lipid-Restricted Group: Liver Function Test during TPNa
AST (IU/L)b Lipid (n) Lipid-restricted (NL 10-35)
Week 1/2
30 ± 7 (12) 36 ± 6 (13)
47 ± 15(12) 31 ± 5 (13)
Week 1 58 + 11 (12)** 38 ± 5 (13)
Week 2 52 ± 18(9)* 41 ±10(10)
Week 3 46 ± 1 6 (8)* 29 ± 5 (4)
Week 4 42 ± 20 (6) 22 ± 2 (4)
LDH (IU/L) Lipid Lipid-restricted (NL 80-180)
177 + 23 214 + 26
266 ± 7 3 206 ± 2 6
302 ± 109 211 ± 2 6
238 ± 25** 204 ± 1 9
226 ± 26** 220 ± 32
198 ± 3 6 154 + 24
Alk Phos (IU/L)C Lipid Lipid-restricted (NL 16-106)
134±31 127 ± 19
116 ± 2 0 148 ± 24
147 ± 19 162 ± 24*
176 ±39** 186 ± 2 7
223 ± 53* 182 ± 2 6
264 ±77 186 ± 7
152 ± 5 9 152 ± 126
152 ± 6 4 243 ± 16
140 + 31* 216 ±27*
188 ± 6 2 482
0.9 ± 0.4 0.8 ± 0.2
1.4 ±0.5 0.7 ±0.1
0.6 ±0.1 0.9 ± 0.2
0.6 ±0.1 1.2 ±0.8
GGT (IU/L) Lipid Lipid-restricted (NL 5-40) Bilirubin (mg/dl) Lipid Lipid-restricted (NL 0.2-1.2)
Downloaded by [New York University] at 16:33 08 February 2015
Baseline 0
40±15d 94 ± 2 4
0.8 ± 0.2 0.8 ±0.1
NA NA
0.4 ±0.1 0.9 ± 0.4
a
Mean ± SEM. *p < 0.05; significant elevations above baseline by paired t-test. **p < 0.01; significant elevations above baseline by paired t-test. NA = data not available. Analysis of covariance (time); p < 0.05 diet. c Analysis of covariance (time); p < 0.01 time. GGT difference at baseline is not statistically significant (T = 1.46).
Table 5. Triglycéride Levels during TPNa
n
Baseline
Week 1
Week 2 (mg/dl)
Week 3
Week 4
Lipid group
12
133 ±23
152 ± 4 3
93 ±11
132 ±41
109 ± 19
Lipid-restricted
13
135 ± 1 5
130 ± 1 4
89 ± 6
102 ± 1 6
147 ± 41
a
Mean ± SEM; normal 50-180 mg/dl.
AST value was the only LFT measurement that was sig nificantly influenced by diet (lipid-restricted and earlylipid treated) by analysis of covariance (p < 0.05). The etiology for the abnormal AST activities seen in this prospective study was due to intravenous fat administra tion, unless one might argue that the slight but not sig nificant increase in intravenous calories provided to the lipid group was of greater importance (Table 3). In con trast to this, the duration of TPN and not diet was found to be significant in the development of abnormalities of alkaline phosphatase and GGT activities. For either reason, the elevated AST activity is transient and returns to normal during TPN. Transient elevation in AST ac
82
tivity may be related to intravenous composition but the duration of TPN is associated with an increase in alkaline phosphatase activity which may predispose to early infiltration of the liver. Both groups received at most 1.5 x BEE (31-34 kcal/kg/day), which would not implicate excessive dextrose or total calories as the principal contributing cause of LFT abnormalities. It is unlikely that the ab sence of essential fatty acids, absence of fat calories, or choline deficiency was responsible for the LFT abnor malities. Providing 400 cal of intravenous lipid calories does not seem excessive, but its administration was re lated to the abnormalities in AST. An alternative pos-
VOL. 9, NO. 1
LFT Abnormalities on Total Parenteral Nutrition sibility was that the lipid group was clinically more prone to develop LFT abnormalities. However, this pos sibility is unlikely, since the two groups appear equally randomized with patients predisposed to LFT abnor malities (i.e., pancreatitis and inflammatory bowel dis ease). Despite the numerous conditions known to produce fatty infiltration of the liver, the usual mild al teration in liver function seen during TPN is not easily attributable to any one of these defined abnormalities.
Downloaded by [New York University] at 16:33 08 February 2015
ACKNOWLEDGMENTS This work was supported in part by Grant Nos CA09450 and HD 16727 from the National Institutes of Health, the American McGaw Corporation, and the Clinical Nutrition Research Unit Young Investigator Award CA42710 of the General Clinical Research Cen ter Program RR00425.
REFERENCES 1. Linder KD, Fleming CR, Abrams A, Hirschkorn MA: Liver function values in adults receiving total parenteral nutrition. JAMA 241:2398-2400, 1979. 2. Wagner WH, Lowry AC, Silberman H: Similar liver func tion abnormalities occur in patients receiving glucosebased and lipid-based parenteral nutrition. Am J Gastroenterology 78:199-202, 1983. 3. Grant JP, Cox CZH, Kleinman LM, Maher MM, Pittman MA, Tangrea JA, Brown JH, Gross E, Beazley RM, Jones RS: Serum hepatic enzyme and bilirubin elevations during parenteral nutrition. Surg Gynecol Obstet 145:573-580, 1977. 4. Lowry SF, Brennan MF: Abnormal liver function during parenteral nutrition: relation to infusion excess. J Surg Res 26:300-307, 1979. 5. Worthley LIG, Fishlock RC, Snoswell AM: Carnitine deficiency with hyperbilirubinemia, a generalized skeletal muscle weakness and reactive hypoglycemia in a patient on long-term total parenteral nutrition: treatment with in travenous L-carnitine. J Parenter Enterol Nutr 7:176-180, 1983. 6. McDonald ATJ, Phillips MJ, Jeejeebhoy KN: Reversal of fatty liver by intralipid by patients on total parenteral nutrition. Gastroenterology 64:885, 1973.
7. Zohrab WJ, McHattie JD, Jeejeebhoy KN: Total parenteral nutrition with lipid. Gastroenterology 64:583-592, 1973. 8. Carroll C, Williams L: Choline deficiency in rats as in fluenced by dietary energy sources. Nutr Rep Int 25:773782, 1982. 9. Burt ME, Hanin I, Brennan MF: Choline deficiency as sociated with total parenteral nutrition. Lancet 11:638-639, 1980. 10. Kaminski DL, Adams A, Jellinke M: The effect of hyperalimentation on hepatic lipid content and lipogenic en zyme activity in rats and man. Surgery 88:93-100, 1980. 11. Sheldon GF, Petersen SR, Sanders R: Hepatic dysfunction during hyperalimentation. Arch Surg 113:504-508, 1978. 12. Allardyce DB: Cholestasis caused by lipid emulsions. Surg Gynecol Obstet 154:641-647, 1982. 13. Sheard NF, Tayek JA, Bistrian BR, Blackburn GL, Zeisel SH: Plasma choline concentration in humans fed parenterally. Am J Clin Nutr 43:219-224, 1986. 14. Harris JA, Benedict FG: "A Biometrie Study of Basal Me tabolism in Man," Publication #279 Carnegie Institute of Washington. Philadelphia: JB Lippincott, pp 223-250, 1919. 15. Buzby GP, Mullen JL, Matthews DC, Hobbs CL, Rosato EF: Prognostic nutritional index in gastrointestinal surgery. Am J Surg 139:160-167, 1980. 16. Harvey KB, Moldawer LL, Bistrian BR, Blackburn GL: Biological measures for the formulation of a hospital prognostic index. Am J Clin Nutr 34:2013-2022, 1981. 17. Bistrian BR: A simple way to estimate the severity of stress. Surg Gynecol Obstet 148:675-678, 1979. 18. Test Methods for in-vitro Diagnostic Use: Ektachem 700. Rochester, NY: Kodak, 1985. 19. Pallares R, Sitges-Serra A, Jaurriete E, Fuentes J, Guardia J, Sitges-Creus A: Gamma-glutamyltransferase (GammaGT): the most sensitive parameter for the detection of TPN-induced liver dysfunction. Clin Nutr l(Suppl):F92, 1982. 20. Tulikoura I, Huikuri K: Morphological fatty changes and function of the liver, serum free fatty acids, and triglycérides during parenteral nutrition. Scand J Gastroenterol 17:177-185, 1982. 21. Host WR, Serlin O, Rush BF Jr: Hyperalimentation in cirrhotic patients. Am J Surg 123:57-62, 1971. 22. Hall RI, Grant JP, Ross LH, Coleman RA, Bozovic MG, Quarfordt SH: Pathogenesis of hepatic steatosis in the parenterally fed rat. J Clin Invest 74:1658-1668, 1984.
Received March 1989; revision accepted July 1989.
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
83
Journal of the American College of Nutrition Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uacn20
Abnormal liver function in malnourished patients receiving total parenteral nutrition: a prospective randomized study. a
a
a
a
a
J A Tayek , B Bistrian , N F Sheard , S H Zeisel & G L Blackburn a
Harbor-UCLA Medical Center, Torrance. Published online: 02 Sep 2013.
To cite this article: J A Tayek, B Bistrian, N F Sheard, S H Zeisel & G L Blackburn (1990) Abnormal liver function in malnourished patients receiving total parenteral nutrition: a prospective randomized study., Journal of the American College of Nutrition, 9:1, 76-83, DOI: 10.1080/07315724.1990.10720353 To link to this article: http://dx.doi.org/10.1080/07315724.1990.10720353
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Abnormal Liver Function in Malnourished Patients Receiving Total Parenteral Nutrition: A Prospective Randomized Study John A. Tayek, MD, FACN, Bruce Bistrian, MD, PhD, FACN, Nancy F. Sheard, DSc, RD, Steven H. Zeisel, MD, PhD, FACN, and George L. Blackburn, MD, PhD, FACN
Downloaded by [New York University] at 16:33 08 February 2015
Harbor-UCLA Medical Center, Terranee, California (JA.T.), New England Deaconess Hospital, Boston (B.B., G.L.B.), and Department of Pathology and Pediatrics, Boston University, School of Medicine, Boston (N.F.S., S.H.Z.) Key words: liver function tests, malnutrition, choline, intralipid total parenteral nutrition A prospective study was performed in clinically malnourished patients in which liver function was tested during a 4-week period of total parenteral nutrition (TPN). The purpose was to determine if concomitant intravenous lipid administration would reduce liver function abnormalities noted to occur frequently in patients receiving TPN. Twen ty-five patients were randomly assigned to receive either daily infusions of 200 cc of a 20% lipid emulsion with TPN or TPN without lipid for the first week. In the subsequent 3 weeks all patients received daily intravenous lipid. The early lipid treatment group received 0.7 g lipid/kg BW/day and approximately 280 mg of choline/day from the lecithin emulsifier throughout the entire study period. Liver function tests were performed twice in thefirstweek, then weekly thereafter. There were significant (p < 0.05) elevations in liver function tests in the early lipid treatment group (for aspartate aminotransferase in weeks 1, 2, and 3, and lactic acid dehydrogenase in weeks 2 and 3). Alkaline phosphatase activity was elevated at weeks 2, 3, and 4 for the lipid-treatment group and at week 1 for the lipid-restricted group. The two groups had a similar elevation in γ-glutamyltransferase activity. Analysis of covariance demonstrated that the overall duration of TPN, and not the presence or absence of intravenous lipid, was significantly related to the elevations in both alkaline phosphatase and γ-glutamyltransferase (GGT) levels. In contrast, the early intravenous administration of lipid was significantly related to the increase in aspartate aminotransferase levels. The peak increase in AST was noted at day 7 in the lipid-administration group. The characteristic liver function test abnormalities associated with TPN do occur and, if anything, are more prominent with daily provision of fat calories. Neither the absence of fat, essential fatty acid, nor choline is likely to be primarily responsible for the usual abnormalities in liver function associated with TPN seen after 7 days of TPN. We conclude that early intravenous administration of lipid emulsion with TPN does not reduce the overall occurrence of abnormal liver function.
INTRODUCTION Liver function (LFT) abnormalities are commonly seen during total parenteral nutrition (TPN). Significant increases in aspartate aminotransferase (AST), alkaline phosphatase activity and bilirubin levels occur in 87, 64, and 21% of patients, respectively, when treated with TPN [1,2]. LFT abnormalities occur as early as 5 days after TPN with AST rising first, followed by alkaline phosphatase, and then bilirubin [2,3]. Many hypotheses have been suggested as to the etiology of LFT abnor malities in humans, including excessive dextrose calories [4], camitine deficiency [5], absence of fat calories [6], essential fatty acid deficiencies [7], choline deficiency [8-10], hepatotoxins such as sodium bisulfite [3], abnormalities of amino acid concentrations [11], car-
nitine deficiency [5], and excessive lipid calories [12]. Since fat emulsions provide choline and linoleic acid and allow a reduction in glucose calories, we were inter ested in determining whether daily intravenous lipid ad ministration would limit the development of LFT abnor malities during TPN. Our hypothesis was that early ad ministration of lipids and a reduction in dextrose calories by the intravenous route would reduce LFT abnor malities noted to occur to patients receiving TPN.
MATERIALS AND METHODS Subjects Patients were randomly assigned to a lipid-restricted or daily lipid group as part of a study to determine the
Address reprint requests to Bruce R. Bistrian, M.D., Ph.D., Cancer Research Institute, 194 Pilgrim Road, Boston, Massachusetts 02215.
Journal of the American College of Nutrition, Vol. 9, No. 1, 76-83 (1990) © 1990 John Wiley & Sons, Inc.
CCC 0731-5724/90/010076-08$04.00
LFT Abnormalities on Total Parenteral Nutrition Table 1. Clinical Characteristics by Diet Classification Primary diagnosis
Lipid group (no.)
Inflammatory bowel disease
3
3
Chronic pancreatitis
2
2
Gastric outlet obstruction
2
1
Cancer
3
2
Appendectomy with wound dehiscence
1
0
Atherosclerotic vascular disease
1
2
s/p Mitral valve replacement
0
1
Chronic diarrhea
0
1
Diverticulitis Total
Downloaded by [New York University] at 16:33 08 February 2015
Lipid-re group
effect of TPN on plasma choline levels reported pre viously [13]. Twenty-five hospitalized patients, who met clinical criteria for requiring TPN, were studied (Table 1). There were 12 females and 13 males, ranging from 19 to 90 years of age (61.8 ± 17.4; mean ± SD). In dividuals with a diagnosis of acute hepatocellular disease (AST > 100 IU/L) or patients receiving hepatotoxic medications were not included (Table 1). All 25 patients had the secondary diagnosis of protein-calorie malnutri tion, defined by a low serum albumin and/or recent loss of 10% or more of usual weight. The protocol was ap proved by the Committee on Clinical Investigation of the New England Deaconess Hospital and all study par ticipants gave informed consent. Study Design Patients were randomly assigned to one of two groups. Group I (lipid) had 12 patients who received a combination of amino acids (10% Aminosyn, Abbott Laboratories), 70% dextrose, and daily lipid (20% Liposyn, Abbott Laboratories) during the first week of TPN therapy. Group Π had 13 patients who were lipidrestricted and received only amino acids and dextrose during week 1. After the first week, all subjects received daily intravenous lipid to prevent the development of essential fatty acid deficiency. The patients' oral or enterai diet (Osmolyte, Mead Johnson) was advanced as tolerated with no choline ingested by mouth. Patients were dropped from the study when their daily enterai intake exceeded 1000 calories over a 1-week period. The patients' basal energy expenditures (BEE) were deter mined by the Harris-Benedict equation [14], and energy
0
1
12
13
requirements were estimated to be approximately 40% above BEE. Protein requirements were calculated at 1.5 g protein/kg ideal body weight. All patients received MVI-12 vitamin mix (one vial per day) as well as a minimum of 2 mg of iron per day. Trace minerals (3 mg zinc, 1 mg copper, 0.4 mg manganese, 56 μg iodine, 10 μg chromium, and 50 μg selenium) were administered daily. Parenteral solutions were administered over 24 hours with the exception of lipid, which was ad ministered during the 8-hr afternoon shift. After 1 week, both groups of patients received 200 cc of 20% lipid infusions daily. Complete nutritional assessments including history of weight loss, actual weight, triceps skinfold, and arm muscle circumference were obtained by trained diet technicians. Serum albumin, transferrin, and total lym phocyte counts were also performed as part of the nutri tional assessment. Twenty-four hour urine collections were used to measure urinary creatinine excretion and creatinine—height indices were recorded. Nutritional in dices including the Prognostic Nutritional Index (PNI) [15], Hospital Prognostic Index (HPI) [16], and Catabolic Index [17] were calculated. Prognostic Nutri tion Index (PNI) = 158% - 16.6 · Albumin (g) - 0.78 · Triceps Skinfold (mm) - 0.2 · Transferrin (mg) - 5.8 · Delayed Hypersensitivity (Anergy = 1; Immunocompetence = 2). The PNI predicts the percent probability of subsequent death [15]. Hospital Prognostic Index = 0.91 · Albumin (g) - Delayed Hypersensitivity - 1.44 · Sepsis + 0.98 · Diagnosis - 1.09 (Diagnosis Cancer = 1, Noncancer = 2, Anergy = 1, Immunocompetence = 2). A number less than zero indicates a 50% probability of dying during hospital stay. The units are based on stand-
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
77
LFT Abnormalities on Total Parenteral Nutrition
Downloaded by [New York University] at 16:33 08 February 2015
ard deviations from the mean. Catabolic Index = g of urinary urea nitrogen excretion - [0.5 · nitrogen intake (g) + 3], where -5-0 = no metabolic stress, 0-5 = mild stress, and 5-10 = moderate to severe stress [17]. Blood samples were drawn from each patient both prior to the initiation of parenteral nutrition and throughout their course of therapy (1—4 weeks). Serum chemistries [albumin, transferrin, liver function tests (LFT), serum triglycérides], total lymphocyte counts, and 24-hr urine creatinine analyses were performed at the New England Deaconess Hospital's clinical laboratory. LFTs were performed on a Kodak Ektachem 700 spectrophotometer using established methods [18]. γ-Glutamyltranspeptidase was measured at baseline in only seven patients in each group and in four patients in each group at week 2.
Statistical Analysis The results were analyzed by analysis of covariance (time) with diet (lipid versus lipid-restricted) as the de pendent variable. Paired t-tests were then used with each individual acting as his own control. Individual values were compared to baseline values prior to TPN. All data are represented as mean ± SEM, unless otherwise noted, and significance is defined by the 95% confidence inter val.
RESULTS Nutritional Profile The 25 patients were successfully randomized into comparable groups with regard to diagnosis (Table 1), age, amount of weight loss, percent of ideal body weight, and other nutritional assessment data (Table 2). In addition to an average 5-kg weight loss, there was a marked reduction in the creatinine-height index in both the lipid-restricted (73 + 21% of standard) and non-lipidrestricted (67 ± 19% of standard) groups. Triceps skinfold were also markedly reduced (72 ± 9% and 63 ± 9% of standard, respectively. Serum albumin, transferrin, and total lymphocyte counts in the lipid and lipidrestricted groups, respectively, were all below normal at the outset of the study. The reductions in anthropométrie measurements and visceral protein levels indicated the presence of moderate to severe protein-calorie malnutri tion. Both groups were also comparably randomized by similar HPI (0.5 vs 0.4), and PNI (63 vs 59%) in the lipid vs lipid-restricted groups, respectively (Table 2), the latter index consistent with a group at moderate risk of complications. The catabolic index suggests minimal
78
metabolic stress [17] (Table 2). The BEE was 1280 + 50 kcal for the lipid-restricted group and 1240 ± 80 kcal for the nonrestricted group. There was no significant difference between the two groups for protein or caloric intake. Average protein in take was 80 g and average dextrose intake was 350 g/day. Total caloric intake over the 4-week study period averaged 1870 cai or 150% of the BEE (33 + 5 kcal/kg/day; mean ± SD). Oral intake was minimal during the first 3 weeks in both groups and averaged less than 400 kcal/day. There was no significant difference for oral or intravenous caloric intake between the groups (Table 3). Ten of 13 patients randomized to the lipidrestricted group received no intravenous lipid. The other three patients each received one dose of 200 cc of 20% Liposyn during the first week, or V7 of the dose given to patient in the lipid group. The quantity of intravenous lipid administered did not differ between groups after week 1. The amino acid solutions, vitamin mix, iron, and selenium solutions contain no choline. Liposyn (20%) contains trace amounts of free choline (4 nmol/ml) and 13.2 μπιοΐ/ml of phosphatidylcholine [13]. Thus, the daily administration of lipid provides an average of 0.7 μπιοΐ of free choline and 2310 μπιοΐ of phosphatidyl choline. The average amount of lipid administered each day was approximately 32 g or 320 cal, excluding the lipid-restricted group in the first week of parenteral nutrition. This amount of fat given daily represents roughly 20% of total calories or 0.7 g/kg/day. Most (19) patients received 2 weeks of parenteral therapy with a range of 9-28 days. Two patients who expired within the second week and three who had over 1000 calories enterally were dropped from the study. Nineteen patients completed a full 2 weeks of TPN therapy. During the third week, an additional eight patients ate well and were dropped, leaving only 12 patients for the third week. The final week an additional three patients were eating over 1000 cal/day and were dropped from the LFT analysis. Liver Function Abnormalities There was a significant effect of TPN on AST eleva tions in the early lipid group by analysis of covariance (p < 0.05). The mean AST activities for the early lipid group increased from the baseline of 30 ± 7 IU/L (mean ± SEM) to 58 ± 11 IU/L, 52 ± 18 IU/L, and 46+16 IU/L (mean + SEM) for weeks 1, 2, and 3, respectively (p < 0.05; Fig. 1). The peak increase was seen after 7 days of TPN. The lipid-restricted group, in contrast, had a mean AST value of 26 ± 6 IU/L (mean ± SEM), which in creased to 38 ± 9 IU/L at week 1 (p = 0.10) and 41 ± 10 IU/L at week 2 (p = 0.09), both values failing to reach
VOL. 9, NO. 1
LFT Abnormalities on Total Parenteral Nutrition Table 2. Nutritional Assessment Lipid group (mean ± SD)
Lipid-restncted group (mean ± SD)
Age (years)
59.7 ± 16.2
63.8 ± 18.1
Weight (kg)
55.3 ± 13.8
59.6 ±11.4
% Ideal body weight
91 ± 16
98 ± 1 9
Weight loss (kg)
5.0 ± 5.7
5.5 ± 6.2
% Standard arm muscle circumference
92 ± 14
92 ± 1 2
% Standard triceps skinfold
63 ± 3 1
72 ± 3 4
10.7 ± 4.6
11.1 ±6.1
Albumin (g/dl) (NL 3.2)
2.9 ± 0.7
2.9 ± 0.8
Transferrin (mg/dl) (NL 170)
138 ± 55
147 ± 71
14801940
1400 ± 880
-2.6 ± 3.5
0.1 ±4.4
0.5 ± 1.4
0.4 ±1.2
63 ± 2 2
59 ± 2 3
Downloaded by [New York University] at 16:33 08 February 2015
Triceps skinfold
Total lymphocyte count (NL 1500) Catabolic index
a
Hospital prognostic index
b
Prognostic nutritional indexc a
CataboIic index = g urinary urea nitrogen excretion - [0.5 nitrogen intake (g) + 3], where -5-0 = no metabolic stress, 0-5 = mild stress, and 5-10 = moderate to severe stress [17]. Hospital prognostic index obtained from [16]. A number less than zero indicates a greater probability of dying during hospital stay and units are based on standard deviations from the mean. •"Prognostic nutrition index = 158% - 16.6 albumin (g) - 0.78 triceps skinfold (mm) - 0.2 transferrin (mg) - 5.8 delayed hypersensitivity. PNI predicts the percent probability of subsequent death (anergy = 1; immunocompetence = 2) [15].
Table 3. Caloric Intakes by Vein and by Moutha Group
n
Week 1
12 13
1649 ± 7 7 1383 ± 9 8 (0.06)
Week 2
Week 3
Week 4
Intravenous calories Lipid group Lipid-restricted group (p value)
1340 ±124 1543 ±185 (0.39)
1540 ±150 1790 ±398 (0.48)
1369 ± 220 1887 ±144 (0.10)
Oral intake Lipid group Lipid-restricted group (p value)
12 13
b
156±69[4] 199 + 86 [4] (0.70)
234± 121 [2] 311 ±146 [4] (0.69)
390±113[6] 48 ± 3 1 [0] (0.07)
868 ± 221 [5] 191 ± 184 [1] (0.06)
a
Mean ± SEM, p value determined by nonpaired t-test not corrected for multiple comparisons. Number of patients whose oral intake averaged > 200 calories/day in each group.
statistical significance from baseline. The only sig nificant increases for the lipid-restricted group were the mean alkaline phosphatase activities, seen only at week
1 (127 ± 19 to 162 ± 24 IU/L; mean ± SEM; p < 0.05) and GGT seen only at week 2 (94 + 24 to 216 + 27 IU/L).
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
79
Downloaded by [New York University] at 16:33 08 February 2015
LFT Abnormalities on Total Parenteral Nutrition By analysis of covariance, the length of TPN ad ministration was an important variable for elevations in alkaline phosphatase and GGT activities (p < 0.01). A significant increase in alkaline phosphatase activity was observed at weeks 2, 3, and 4 in the lipid group by paired t-test (134 ± 33 to 176 ± 39, 223 + 53, 264 + 77; U/L; mean ± SEM). The lipid-restricted group did not show a significant rise in alkaline phosphatase activity after 1 week. LDH levels were also significantly in creased (p < 0.05) for the lipid group from a baseline of 179 + 25 IU/L to 238 ± 25 IU/L and 226 + 26 IU/L (mean ± SEM) for weeks 2 and 3, respectively. There were no significant increases in the lipid-restricted group. Both groups demonstrated significant elevations in γ-glutamyltransferase (GGT) activity at week 2. There was no significant elevation in mean bilirubin seen in either group (Table 4). There were no significant in creases in fasting plasma triglycéride levels noted be tween the two groups (Table 5).
duced liver dysfunction. Elevations of AST and alkaline phosphatase were seen in a study by Lowry et al [4] in which increases in these two enzymes occurred at 14 days of TPN. Another investigation found similar in creases in AST and alkaline phosphatase which occurred earlier, that is, after 10 days of TPN [20]. Grant and others described elevations in AST as early as the fifth day of TPN [3,21]. A recent study has described a stepwise increase in liver function abnormalities beginning with AST, alanine aminotransferase, alkaline phos phatase, and then bilirubin [2]. In the present study, we have observed that liver function abnormalities as measured by alkaline phosphatase and GGT activities were related to the duration of TPN, whereas abnor malities in AST activities were related only to diet com position, i.e., the early provision of lipid in the TPN. The peak elevation in AST levels occurred at day 7 for the early lipid group. The lipid-restricted group failed to demonstrate a significant change in AST levels.
LFT elevations observed in this study were far lower than those seen in hepatocellular disease and are rarely associated with a significant clinical problem. However, these changes that we observed are caused by the abnor mal liver function. Linder et al [1] have described a "meaningful increase" in liver function tests as a value equal to or greater than a 50% increase above baseline values, which results in a value within the abnormal range. A meaningful increase in AST was seen in 58% of the lipid group and in only 38% of the lipid-restricted group. Two patients in the lipid group expired during the second week of TPN. Both patients had meaningful in creases in AST and bilirubin levels but not alkaline phosphatase while on TPN. GGT levels were not measured. At autopsy, the patient with lymphoma, who died from disseminated intravascular coagulation, had focal fatty changes, intrahepatic and extracellular bile stasis, and cholestasis. The other patient, who died from sepsis related to Crohn's disease, demonstrated periportal cellular inflammation and moderate fatty changes in the liver. Both of these changes may be due to the under lying disease and not solely attributable to TPN.
From a theoretical point of view, providing parenteral fat with TPN would be expected to ameliorate changes in liver function by reducing glucose calories, and hepatic lipogenesis [22] and the parenteral fat should provide a source of linoleic acid and choline, the lack of which has been associated with fatty infiltration of liver [7,8,10]. However, contrary to our expectations, liver function was more adversely affected by the daily provision of fat at the outset of TPN. Hall et al have recently reported that both lipid-free and lipid-containing TPN treatments in rats are associated with a threefold increase in hepatic triglycéride content [22]. A recent study in the rat model has demonstrated that the administration of safflower oil calories (20% w/w) to choline-deficient rats produces a greater fatty infiltration of the liver when compared to starch calories [8]. How ever, we have no evidence of clinical choline deficiency in the lipid group when choline levels were maintained although not restored to levels seen in well-nourished individuals [13]. The early concomitant administration of lipid and TPN contributed to an increase in AST ac tivities, but liver biopsies were not available for the en tire group to confirm fatty liver infiltration.
DISCUSSION Among the 25 hospitalized patients who required in travenous hyperalimentation, those who received in travenous lipid at the onset of TPN developed more liver function abnormalities than those not given lipids during the first week (Table 4). The most common elevations were seen in AST, GGT, and alkaline phosphatase ac tivities. Elevations in GGT activities have been noted by Pallares et al [19] to be a sensitive parameter of TPN-in-
80
Liver function tests are abnormal when there is hepatic dysfunction, such as fatty infiltration and cellular damage; both have occurred in patients receiving parenteral nutrition [11]. Preventing LFT abnormalities requires identification of the etiological factors. We found that a significant increase in the liver function tests occurred in both groups of subjects during the course of this study which was not greater in magnitude or more frequent than has been previously described. Elevations of bilirubin and GGT activity occurred in equal percentages and to equal degrees in each group.
VOL. 9, NO. 1
LFT Abnormalities on Total Parenteral Nutrition
LIVER FUNCTION TEST ON TPN
IU/L
Downloaded by [New York University] at 16:33 08 February 2015
• ° * ** I
Lipid Group Lipid Restricted p " °,'°>?] vs. boseline volues p « 0.01 J mean * SEM
IU/L
350r- ALKALINE PHOSPHATASE IU/L
WEEKS Fig. 1. Liver function test results from patients receiving intravenous hyperalimentation (TPN) over a 4-week period. Mean values are compared to baseline values (paired t-testing) prior to TPN for AST (aspartate aminotransferase), LDH (lactic acid dehydrogenase), GGT (γ-glutamyltransferase), and alkaline phosphatase ac tivity. Analysis of covariance demonstrated diet (lipid group) to be significantly different for AST activity (p < 0.05), and time to be an important variable for abnor malities in GGT and alkaline phosphatase activity (p < 0.05; p < 0.001), respective ly. ( · ) Lipid group; (O) lipid-restricted group; *p < 0.05; **p < 0.01 vs baseline values.
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
LFT Abnormalities on Total Parenteral Nutrition Table 4. Lipid Treatment Group vs Lipid-Restricted Group: Liver Function Test during TPNa
AST (IU/L)b Lipid (n) Lipid-restricted (NL 10-35)
Week 1/2
30 ± 7 (12) 36 ± 6 (13)
47 ± 15(12) 31 ± 5 (13)
Week 1 58 + 11 (12)** 38 ± 5 (13)
Week 2 52 ± 18(9)* 41 ±10(10)
Week 3 46 ± 1 6 (8)* 29 ± 5 (4)
Week 4 42 ± 20 (6) 22 ± 2 (4)
LDH (IU/L) Lipid Lipid-restricted (NL 80-180)
177 + 23 214 + 26
266 ± 7 3 206 ± 2 6
302 ± 109 211 ± 2 6
238 ± 25** 204 ± 1 9
226 ± 26** 220 ± 32
198 ± 3 6 154 + 24
Alk Phos (IU/L)C Lipid Lipid-restricted (NL 16-106)
134±31 127 ± 19
116 ± 2 0 148 ± 24
147 ± 19 162 ± 24*
176 ±39** 186 ± 2 7
223 ± 53* 182 ± 2 6
264 ±77 186 ± 7
152 ± 5 9 152 ± 126
152 ± 6 4 243 ± 16
140 + 31* 216 ±27*
188 ± 6 2 482
0.9 ± 0.4 0.8 ± 0.2
1.4 ±0.5 0.7 ±0.1
0.6 ±0.1 0.9 ± 0.2
0.6 ±0.1 1.2 ±0.8
GGT (IU/L) Lipid Lipid-restricted (NL 5-40) Bilirubin (mg/dl) Lipid Lipid-restricted (NL 0.2-1.2)
Downloaded by [New York University] at 16:33 08 February 2015
Baseline 0
40±15d 94 ± 2 4
0.8 ± 0.2 0.8 ±0.1
NA NA
0.4 ±0.1 0.9 ± 0.4
a
Mean ± SEM. *p < 0.05; significant elevations above baseline by paired t-test. **p < 0.01; significant elevations above baseline by paired t-test. NA = data not available. Analysis of covariance (time); p < 0.05 diet. c Analysis of covariance (time); p < 0.01 time. GGT difference at baseline is not statistically significant (T = 1.46).
Table 5. Triglycéride Levels during TPNa
n
Baseline
Week 1
Week 2 (mg/dl)
Week 3
Week 4
Lipid group
12
133 ±23
152 ± 4 3
93 ±11
132 ±41
109 ± 19
Lipid-restricted
13
135 ± 1 5
130 ± 1 4
89 ± 6
102 ± 1 6
147 ± 41
a
Mean ± SEM; normal 50-180 mg/dl.
AST value was the only LFT measurement that was sig nificantly influenced by diet (lipid-restricted and earlylipid treated) by analysis of covariance (p < 0.05). The etiology for the abnormal AST activities seen in this prospective study was due to intravenous fat administra tion, unless one might argue that the slight but not sig nificant increase in intravenous calories provided to the lipid group was of greater importance (Table 3). In con trast to this, the duration of TPN and not diet was found to be significant in the development of abnormalities of alkaline phosphatase and GGT activities. For either reason, the elevated AST activity is transient and returns to normal during TPN. Transient elevation in AST ac
82
tivity may be related to intravenous composition but the duration of TPN is associated with an increase in alkaline phosphatase activity which may predispose to early infiltration of the liver. Both groups received at most 1.5 x BEE (31-34 kcal/kg/day), which would not implicate excessive dextrose or total calories as the principal contributing cause of LFT abnormalities. It is unlikely that the ab sence of essential fatty acids, absence of fat calories, or choline deficiency was responsible for the LFT abnor malities. Providing 400 cal of intravenous lipid calories does not seem excessive, but its administration was re lated to the abnormalities in AST. An alternative pos-
VOL. 9, NO. 1
LFT Abnormalities on Total Parenteral Nutrition sibility was that the lipid group was clinically more prone to develop LFT abnormalities. However, this pos sibility is unlikely, since the two groups appear equally randomized with patients predisposed to LFT abnor malities (i.e., pancreatitis and inflammatory bowel dis ease). Despite the numerous conditions known to produce fatty infiltration of the liver, the usual mild al teration in liver function seen during TPN is not easily attributable to any one of these defined abnormalities.
Downloaded by [New York University] at 16:33 08 February 2015
ACKNOWLEDGMENTS This work was supported in part by Grant Nos CA09450 and HD 16727 from the National Institutes of Health, the American McGaw Corporation, and the Clinical Nutrition Research Unit Young Investigator Award CA42710 of the General Clinical Research Cen ter Program RR00425.
REFERENCES 1. Linder KD, Fleming CR, Abrams A, Hirschkorn MA: Liver function values in adults receiving total parenteral nutrition. JAMA 241:2398-2400, 1979. 2. Wagner WH, Lowry AC, Silberman H: Similar liver func tion abnormalities occur in patients receiving glucosebased and lipid-based parenteral nutrition. Am J Gastroenterology 78:199-202, 1983. 3. Grant JP, Cox CZH, Kleinman LM, Maher MM, Pittman MA, Tangrea JA, Brown JH, Gross E, Beazley RM, Jones RS: Serum hepatic enzyme and bilirubin elevations during parenteral nutrition. Surg Gynecol Obstet 145:573-580, 1977. 4. Lowry SF, Brennan MF: Abnormal liver function during parenteral nutrition: relation to infusion excess. J Surg Res 26:300-307, 1979. 5. Worthley LIG, Fishlock RC, Snoswell AM: Carnitine deficiency with hyperbilirubinemia, a generalized skeletal muscle weakness and reactive hypoglycemia in a patient on long-term total parenteral nutrition: treatment with in travenous L-carnitine. J Parenter Enterol Nutr 7:176-180, 1983. 6. McDonald ATJ, Phillips MJ, Jeejeebhoy KN: Reversal of fatty liver by intralipid by patients on total parenteral nutrition. Gastroenterology 64:885, 1973.
7. Zohrab WJ, McHattie JD, Jeejeebhoy KN: Total parenteral nutrition with lipid. Gastroenterology 64:583-592, 1973. 8. Carroll C, Williams L: Choline deficiency in rats as in fluenced by dietary energy sources. Nutr Rep Int 25:773782, 1982. 9. Burt ME, Hanin I, Brennan MF: Choline deficiency as sociated with total parenteral nutrition. Lancet 11:638-639, 1980. 10. Kaminski DL, Adams A, Jellinke M: The effect of hyperalimentation on hepatic lipid content and lipogenic en zyme activity in rats and man. Surgery 88:93-100, 1980. 11. Sheldon GF, Petersen SR, Sanders R: Hepatic dysfunction during hyperalimentation. Arch Surg 113:504-508, 1978. 12. Allardyce DB: Cholestasis caused by lipid emulsions. Surg Gynecol Obstet 154:641-647, 1982. 13. Sheard NF, Tayek JA, Bistrian BR, Blackburn GL, Zeisel SH: Plasma choline concentration in humans fed parenterally. Am J Clin Nutr 43:219-224, 1986. 14. Harris JA, Benedict FG: "A Biometrie Study of Basal Me tabolism in Man," Publication #279 Carnegie Institute of Washington. Philadelphia: JB Lippincott, pp 223-250, 1919. 15. Buzby GP, Mullen JL, Matthews DC, Hobbs CL, Rosato EF: Prognostic nutritional index in gastrointestinal surgery. Am J Surg 139:160-167, 1980. 16. Harvey KB, Moldawer LL, Bistrian BR, Blackburn GL: Biological measures for the formulation of a hospital prognostic index. Am J Clin Nutr 34:2013-2022, 1981. 17. Bistrian BR: A simple way to estimate the severity of stress. Surg Gynecol Obstet 148:675-678, 1979. 18. Test Methods for in-vitro Diagnostic Use: Ektachem 700. Rochester, NY: Kodak, 1985. 19. Pallares R, Sitges-Serra A, Jaurriete E, Fuentes J, Guardia J, Sitges-Creus A: Gamma-glutamyltransferase (GammaGT): the most sensitive parameter for the detection of TPN-induced liver dysfunction. Clin Nutr l(Suppl):F92, 1982. 20. Tulikoura I, Huikuri K: Morphological fatty changes and function of the liver, serum free fatty acids, and triglycérides during parenteral nutrition. Scand J Gastroenterol 17:177-185, 1982. 21. Host WR, Serlin O, Rush BF Jr: Hyperalimentation in cirrhotic patients. Am J Surg 123:57-62, 1971. 22. Hall RI, Grant JP, Ross LH, Coleman RA, Bozovic MG, Quarfordt SH: Pathogenesis of hepatic steatosis in the parenterally fed rat. J Clin Invest 74:1658-1668, 1984.
Received March 1989; revision accepted July 1989.
JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION
83
