Joel D Kopple,
Nancy
ABSTRACT
E Vinton,
Thirty-four
adults
Stewart
A Laidlaw,
undergoing
and
long-term
Marvin
par-
enteral nutrition (TPN) were treated either with or without intravenous taurine for 24 mo. Statistical comparisons were carried out in eight patients randomly assigned to receive intravenous taurine, usually 10 mg. kg . d ‘, and 10 patients not receiving taurine. Compared with normal adults, baseline plasma taunne and urine taurine-creatinine ratios were decreased in both groups and platelet taurine was reduced in the taurine-treated group. During taurine treatment the mean of the mean values for taurine became normal in plasma and platelets and remained normal in erythrocytes, granulocytes, and lymphocytes; urine taurine-creatinine ratios rose to approximately five times normal. During follow-up, patients not given taurine had plasma, erythrocyte, and granulocyte taurine and urine taurine-creatinine ratios below normal values and the concentrations of taurine-treated patients. Their platelet taurine was also subnormal. Thus, 10 mg taurine kg ‘ d ‘ intravenously normalizes plasma and blood cell taunne concentrations in long-term TPN patients. Am J Clin Nutr .
E Ament
rime as well as other sulfur amino acids were not present in the solutions. The present study was part of a clinical trial designed to ascertain the effects of adding taurine to the intravenous solutions on body taurine concentrations and eye function in patients undergoing long-term TPN. Taurine is not well metabolized in the body (1, 9) and the major mechanism for the regulation of body pools is urinary excretion (10, 1 1). Hence, it was not clear whether the intravenous infusion of taurine, which could raise plasma taurine concentrations, might not lead to excessive urinary taurine losses and prevent the attainment of normal taurine concentrations in plasma or blood cells. The present report describes the response of plasma, blood cell, and urine taurine concentrations to daily infusions of taurine.
-
Methods Subjects
and study
design
1990;52:846-53.
clinically stable adult patients followed at the UCLA for the Health Sciences who had been undergoing parnutrition at home for 6 mo and who gave informed were included in this study. Patients first entered a baseline period during which blood and urine tests were performed. Subsequently, they remained on their usual parenteral-nutrition regimen but were randomly assigned to receive cither a daily infusion of taurine, 10 mg. kg body wt, which the patients added each day to their parenteral-nutrition solutions, or to receive no taurine supplement. Patients were maintained on these treatment regimens for a 24-mo follow-up period. Twenty-nine patients were randomly assigned, 15 to receive All
KEY WORDS urine,
blood
cells,
Taurine, methionine,
total parenteral cysteine
nutrition,
plasma,
Introduction Historically,
taunne
(2-aminoethane
sulfonic
acid)
has
not
been considered an essential amino acid for children or adults (I). For this reason it has not been recommended for enteral formulas or parenteral-nutrition solutions except possibly for low-birth-weight infants or neonates (2-4). However, we (5, 6) previously showed that adult patients undergoing long-term TPN have decreased taurine concentrations in plasma, erythrocytes,
lymphocytes,
and
platelets.
Similarly,
children
under-
going long-term TPN have low taurine concentrations in plasma, platelets, and urine (5, 7). The taurine concentrations tended to be particularly low in patients who ingested little or no food. These findings suggested that there may be a generalized depletion of body taurine in patients undergoing longterm TPN (8). The parenteral-nutntion solutions used in these studies provided abundant methionine (1 and range for adult patients: 2. 12, 0.64-3.40 g/d)(6, 7). However, cysteine and tau-
846
Am
J C/in Nuir
l990;52:846-53.
Center enteral consent
I
From
the Departments
of Medicine
and
Pediatrics,
Schools
of
Medicine and Public Health, UCLA and Harbor-UCLA Medical Center, and the UCLA Center for the Health Sciences, Torrance and Los
Angeles, 2
CA.
Supported
by NIH
grant
ROl DK
331 12 and
a grant
Healthcare,
from
Inc. NEV was the recipient ofan NIH Individual Research Service Award. 3 Address reprint requests to JD Kopple, Harbor-UCLA Center, 1000 West Carson Street, Torrance, CA 90509. Received August 7, 1989. Accepted Printed
for publication in USA.
November
© 1990 American
Baxter
National Medical
20, 1989. Society
for Clinical
Nutrition
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
Effect of intravenous taurine supplementation on plasma, blood cell, and urine taurine concentrations in adults undergoing long-term parenteral nutrition13
INTRAVENOUS
did
not
give
protocol,
informed
but
they
receive no taunne. or during the study four other patients long-term parenteral
consent
agreed
to the
to have
Seven pabecause of died. Nine nutrition
granulocytes,
occasional
measurements
of
lymphocytes,
and
platelets
during
the baseline period and at 2, 4, 6, 9, 12, 15, 18, 21, and 24 mo in the follow-up period. Plasma methionine and cystine and other serum chemistry variables were also measured at these times.
For
technical
and
logistical
reasons
blood
specimens
could not be obtained from individuals at every time interval during the follow-up period. To reduce the heterogeneity of each group for the statistical examination, these analyses were performed only on patients for whom plasma and blood-cell taurine concentrations were available at baseline and at three or more visits during the follow-up period. Eighteen patients satisfied these criteria, eight of whom received supplemental taurine, including the women given 2 mg taurine . kg’ . d ‘. Urine specimens were obtained less frequently for taunne measurements. Because not all patients were able to collect 24h urine specimens, part ofa 24-h urine collection or occasionally
single-void
viduals.
Hence,
urine
specimens
the urine
data
were
obtained
are expressed
from
some
as the ratio
mdiof tau-
rime to creatinine (taurine:creatinine). Characteristics ofthe two treatment groups of 18 patients are shown in Table 1 There were no significant differences between the groups; there was a trend (NS) for age to be lower in the group not receiving taunne supplement (no-taurine group). There were also no significant differences in the entire group of patients between the 25 no-taurine patients and the 9 taurinetreated patients with regard to the baseline characteristics shown in Table 1 . The ages of the entire group of patients in the no-taurine group (n 25) and the taurine group (n 9) were 50. 1 ± 3.5 and 60.2 ± 4.2 y (A± SEM), respectively (NS). Similarly, within each treatment group there were no significant differences in the baseline characteristics in the entire group ofpatients studied as compared with those who had three or more measurements in the follow-up period. Of the 18 patients with three or more follow-up visits, I 7 patients had short-bowel syndrome caused by Crohn’s disease (2 patients), radiation enteritis (7 patients), ischemic infarction (3 patients), volvulus (1 patient), pseudoobstruction (associated with scleroderma in 2 patients and idiopathic in 1 patient), or peptic ulcer disease (I patient). The patient with peptic ulcer disease had fistulas, adhesions with obstruction, and resection ofthe small intestine. The one patient without short-bowel syndrome had an esophageal resection for carcinoma of that organ. Radiation enteritis was caused by radiotherapy for carci.
=
TPN
TABLE
1
847
PATIENTS
Characteristics
at baseline5 Notaurine (n=l0)
Taurine
(n=8)
randomized-treatment
blood and urine taken while they remained on their usual therapeutic regimen. A total of25 randomly assigned and the above mentioned 9 nonrandomly assigned patients had one baseline and at least one follow-up measurement. Nine ofthese individuals received supplemental taurine: seven for 24 mo and two for 5 mo. Among these nine patients, one woman refused to take the full dose of taurine and was given 2 mg taunne . kg. d ‘ throughout the study. The data from this patient were included in the analyses ofthe taurine group. This study was approved by the UCLA and Harbor-UCLA Medical Center institutional review boards. Blood was drawn for measurement of taurine in plasma, erythrocytes,
IN
=
Age (y) DurationofTPN(mo) Height(cm)
49.0
Weight (kg)
59. 1 ±
Relative Triceps
84.9 ± I .9
3/5 63.9 ± 2.3 56.0± 14.2 162 ±2.3 63.8 ± 4.2 93.2 ± 5.8
13±6.0
12 ±0.50
19
19
M/F
4/6 55.9±
165±2.9
body weight (%) skinfold thickness
skinfold
1 .6
(mm)
Men Women Subscapular
5.9t 11.1
±
3.0
±
(mm) Men Women Arm muscle area (mm2) Men Women
44±5.3 39±4.2
Arm
95
muscle
area (% ofnormal)
Men
5
;:
SEM.
given
4.2 [4]
±
16±2.8 1 3 ± 1.5 [4]
58±4.1 52±8.3
10
102 12.4 8.2 71.0 4.36
1 30 1 1.9
± 20 [4] ±
1 .4
[3]
[7] 15.6 [6]
6.6 80.2
±
1 .9
[6]
±
16.5 [6]
0.84
4.96
±
1.26 [7]
±
I I
±
1.0 [8]
±
1 .0
± ±
[4]
1 14 ± 13 [7] 94±6.7[3]
[7]
Data were obtained immediately before patients began (treatment) phase ofthe study. n less than total sample is
in brackets.
t Normal 15.6
10±3.2 12 ± 2.3
85±21
Women Serum zinc (zmol/L)t Serum selenium (Mmol/L)t Serum vitamin B-6 (nmol/L)t Serum vitamin A (zmol/L)t
the follow-up
±
thickness
mol/L;
serum values: zinc, 1 1.5-18.5 zmol/L; selenium, 10.6vitamin B-6, 19-88 nmol/L; vitamin A, 0.35-1.75
zmol/L.
noma neal
ofthe ovaries, uterine corpus, or cervix or for retropentocarcinoma. Most patients had undergone intestinal resection. No patient had had evidence for active malignancy for 5 y. Twenty adults (9 men, 1 1 women) served as normal control subjects. Their mean age was 54.6 ± 3. 1 y, which was not significantly different from either patient group [analysis ofvariance (ANOVA), Brown-Forsythe test for equality of means (12)]. The intake and intestinal absorption of food, which varied from patient to patient, could affect the contribution of dietary taurine to the patient’s total daily taurine intake. We therefore classified patients by whether their enteral nutrient absorption was estimated to provide < or > 25% of their daily energy requirements. The quantity ofcalories in the nutrients absorbed from the intestinal tract was estimated from dietary histories, from knowledge ofwhether the remaininglength ofsmall intestine was too short to allow absorption of substantial energyyielding nutrients, or occasionally from 72-h fecal fat excretion or serum carotene concentrations. In previous studies these methods of assessment identified parenteral-nutrition patients who tended to have more severely reduced taurine concentrations in plasma or platelets (6, 7). It is estimated that 6 of the 10 no-taurine patients and 5 of the 8 taurine-treated patients absorbed an amount of nutrients from the intestinal tract that provided < 25% of their daily estimated calorie needs. Five of the no-taunne patients and one ofthe taurine patients received parenteral nutrition < 7 d/wk.
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
intravenous taurine and 14 to tients dropped out either before illness or for other reasons and additional patients undergoing
TAURINE
848
KOPPLE
Method
ofparentera/
nutrition
3.50%,
or
4.25%
amino
acids
(Travasol,
Baxter
Healthcare
Laboratories, Deerfield, IL). Lipids were provided from 1 to 7 d/wk as 10% or 20% Intralipid (Cutter Laboratories, Emeryville, CA) or 10% Liposyn (Abbott Laboratories, North Chicago, IL). Vitamins (10 mL/d) were given as MVI-l2 (USVRevlon, Tarrytown, NY). This preparation contained 12 vitamins,
including
4.0 mg pyridoxine
HC1(82.6%
pyridoxine)
and
1.0 mg vitamin A. Trace mineral supplements included 2 mg Zn/L and 1 mg Cu/L. Iron dextran (Imferon, Merrell Dow Pharmaceuticals, Inc, Cincinatti, OH, or Proferdex, Fisons Corp. Rochester, NY) was infused as needed. About 25% of the patients received 100 zg Sel/d and 10 zg Cr/d in their parenteral solutions. During the last 12 mo of the study, the patients received only the 3.50% or 4.25% amino acids and all patients were given selenium and chromium in their parenteral solutions. Analyses
ofblood
and urine
samples
treatment group, differences between baseline and the mean values for each patient during follow-up were evaluated by Student’s I test for paired data. For all tests for statistical significance, data were first subjected to log transformation. ANOVA was used to test for differences in values among the no-taurine group, the taurine group, and the normal control subjects. The Levene test for equality of variances was used to test for differences in the variances among the three groups. If the Levene test was positive (ie, statistically significant), the significance of differences between means was determined by the BrownForsythe test. Ifthe ANOVA indicated a statistically significant difference between means, then comparisons between the two patient groups and between each patient group and the normal control subjects were carried out using Tukey’s test. The main hypothesis proposed before the study was that taurime values were lower than normal in the patients who did not receive
methods of collection, preparation, and analyses of blood-cell and urine specimens for taurine are described in greater detail elsewhere (6, 7). Blood was collected in heparin-treated tubes. The plasma and blood cells were separated on Percoll (Pharmacia, Piscataway, NJ) gradients and prepared for amino acid analysis (6, 7). Blood was collected between 1 300 and 1 500 during the postabsorptive state, 6-8 h after completion ofthe nightly intravenous infusion and 6-8 h after the last meal, ifthe patient ate. Blood was obtained from normal control subjects at the same time, -6-8 h after their last meal. Specimens were deproteinized with sulfosalicylic acid.
Methionine
and
cystine
were
measured
only
with
a cell counter(Coulter
Electronics,
Hialeah,
FL).
Amino
acids were measured with an amino acid analyzer (121 MB, Beckman Instruments, Fullerton, CA) using a single-column three-buffer, lithium citrate elution system (14). Urine creatinine was measured with a creatinine analyzer (Creatinine Analyzer 2, Beckman Instruments), which uses a modification ofthe Jaffe method. Plasma vitamin B-6 was measured by the activation oftyrosine decarboxylase by pyridoxal-5-phosphate at American Biosciences Laboratories (Van Nuys, CA). The other laboratory measurements indicated in Table 1 were performed in the UCLA Hospital clinical laboratories. Blood for these clinical tests was usually obtained from subjects in the postabsorptive state between 1 300 and 1 500. Anthropometry was performed as previously described (13). Statisticalanali’ses Statistical
evaluations
were
with
the BMDP
carried
out
by ANOVA
and
lin-
statistical software package (12). For statistical analyses, in each treatment group the mean of the mean value for each patient during follow-up was compared with the baseline value, with the values for normal control subjects, and with the other treatment group. Within a ear regression,
were
normal
in those
patients
who
received
and
the
follow-up
period
was
available.
For
the
statistical
analyses shown in the text and tables and for the values shown in the tables, only the data from patients who had one taurine measurement
in baseline
and
at least
three
measurements
in
the follow-up period were used. Almost all of these latter patients had been randomly assigned to their respective treatment groups. An exception is the urine taurine data. Because of the more limited number of samples available, the data obtained from all patients were analyzed statistically.
Results
in plasma
because the yield from blood cells was too low to accurately measure intracellular methionine or cystine concentrations. The numbers and volumes ofdifferent blood cells were determined
and
To test this hypothesis, linear-contrast tests were also carried out between the normal control subjects and each of the two patient groups. Variance is expressed as SEM. To show all of the plasma and blood-cell taunne data obtamed in this study, taurine concentrations are shown for the 34 patients in whom at least one measurement in both the baseline
The plasma
taurine
taurine.
The mean taurine concentrations in plasma, blood cells, and urine at each time interval during baseline and follow-up in the entire group of34 patients in the no-taurine and taurine groups who had at least one baseline and one follow-up measurement (Fig I) were very similar to the values from the 18 patients who had three or more follow-up analyses (Table 2). Plasma taurine concentrations at baseline were significantly below normal in both groups of patients (Table 2, Fig 1). During follow-up the mean of the average plasma taurine concentrations increased only in the patients receiving taurine (P < 0.05), and in this group, plasma taurine was no longer different from normal values. In contrast, during follow-up in the no-taurine group, the plasma concentrations remained significantly below normal and also were significantly lower than the plasma concentrations ofthe taurine group. The changes in plasma taurine from the baseline value to the mean concentration during follow-up was significantly greater in the taurine group than in the notaurine group (P < 0.03). Erythrocyte taurine concentrations were not different from control values during baseline (Table 2, Fig 1). However, the mean values during follow-up increased significantly in the taurime group (P < 0.01) and tended to fall in the no-taurine group. The changes in the taurine concentrations from baseline in the taurine group were significantly different from the changes in the no-taurine group (P < 0.02). Moreover, the mean of the
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
Parenteral nutrition was administered over an 8-12-h period, starting at ‘--2000-2200 and ending at --0700-0900, as previously described (6, 13). Patients were infused with 1-3 L/d of solutions that contained 10-25% dextrose and 2.25%,
ET AL
INTRAVENOUS
TAURINE
110
90 80
3,70
jects
30
baseline
change
10
rime concentrations
0
a
2
4
10
6
Durotio”
12 14
16
ences
20
0
14 12 10
I
mean
a
6
10
8
Duration -;.
12 14 16
18202224
of study (mo)
25
00
2
20
I 10
a
2
4
6
8
10
12
Dumotion of
14
study
16
ie 20
22 24
0
c
I ! io is
0
2
4
6
8
10 12
Duration
14
16 18 20 22
24
of study (mo)
300 250 0
; -
There
and follow-up
in either
patient
in the change with the taurine
granulocyte
taunne
was
no
significant
in the granulocyte group,
nor
was
from baseline group. There concentrations
tau-
there
a sig-
in the no-tauwere no differbetween
the
200
150
100
j50
0
2
4
6
8
10
Ourotion
12 14 of study
ofthe
mean
lymphocyte
taurine
concentrations
16 18202224 (mo)
FIG 1. Plasma, erythrocyte, granulocyte, lymphocyte, and platelet taurine concentrations in 34 patients undergoing long-term TPN who were given intravenous infusions containing no taurine (0, n 25) or taurine (#{149}, n 9). Each patient had one baseline measurement of taurime (at zero time) and one or more follow-up measurements, n 8 for no taurine (range 5-1 1) and n 7 for taurine (range 4-1 1). .± SEM. =
=
=
=
erythrocyte taurine values during follow-up in the nogroup was significantly lower than the values in both the taurine group and the normal control subjects.
rose
sig-
nificantly in both groups of patients during follow-up but remained no different from normal values. The change in lymphocyte taurine concentrations between baseline and followup and the absolute values at either baseline or follow-up were not different between the two patient groups. Baseline platelet taurine concentrations were significantly lower than normal only in the taurine group (Table 2, Fig 1). However, during follow-up the values tended to rise in the taurime group (NS) but fell significantly in the no-taurine group. The changes in taurine concentrations between baseline and follow-up in the taurine group as compared with the no-taurine group was ofborderline significance (P 0.06). During followup the mean of the mean platelet taurine concentrations was significantly below normal only in the no-taurine group. There were no significant differences in the baseline or follow-up platelet taurine values between the taurine group and the notaurine group. Urine taurine:creatinine values at baseline were lower in each of the two patient groups compared with the normal control subjects (Table 2, Fig 2). During follow-up the mean of the mean taurine:creatinine values did not change in the notaurine group, although mean concentrations tended to rise near the end of the follow-up period and the ratios remained lower than in the normal control subjects. During follow-up in the taurine group, there was a dramatic and significant rise in the urine taurine:creatinine (P < 0.001). The follow-up urine taurine:creatinine in the taurine group was approximately five times greater than the normal value (P < 0.0 1 ) and about seven times greater than the ratio for the no-taurine group (P < 0.01). The increase in taurine:creatinine between baseline and followup in the taurine group was significantly greater than in the notaurine group (P < 0.01). The slope of the changes in taurine concentrations between the 2nd and 24th mo offollow-up, determined by linear-regression analysis, was not significant for plasma, urine, or any blood-cell type in the taurine and no-taurine groups. Moreover, there was no significant difference between the two groups during follow-up for the slopes ofthe taurine concentrations in plasma, urine, or any blood cell. The plasma and blood-cell taurine concentrations in the one woman given 2 mg . taurine . kg . d ‘ , in general, rose during the follow-up period. Her taurine concentrations at baseline and during follow-up, respectively, were as follows: plasma, 61 =
(mo)
20
.
in the
or follow-up.
baseline
two groups ofpatients at baseline, but the mean follow-up taurime concentrations were significantly greater in the taunne group than in the no-taurine group. Lymphocyte taunne concentrations at baseline tended to be lower than normal in both groups of patients, although these differences were not statistically significant(Table 2, Fig 1). The
1
!
between
nificant difference rime as compared
18202224
of study (mo)
18
mean
at either
20
-
849
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
s0 40
taurine
PATIENTS
Oranulocyte taurine concentrations in the no-taurine group tended to be lower than in the normal control subjects at baseline (P < 0.06) and were significantly below normal during follow-up (Table 2, Fig 1). The granulocyte taurine concentrations in the taurine group were not different from control sub-
100
6
IN TPN
KOPPLE
850 TABLE
2
Taurine
concentrations
in plasma,
blood
cells, and urine No
I 1.7 6.0
± ±
Taurine
10)
8.l 4.5II
38.2 72.6
43.8 5 1.4
±
73.1
1 .0 2.0ff
± ±
8.5
15 110±3531 [7] 15 751 ± 2 644[7]
3367±585[9] 6696 ± 457 [9]t ± 45
3359±564 9 298 ±
[9]
±
14 [9]59
±
15.4
[l2]
±
1 3.0
[ 15]9
Nancy
ABSTRACT
E Vinton,
Thirty-four
adults
Stewart
A Laidlaw,
undergoing
and
long-term
Marvin
par-
enteral nutrition (TPN) were treated either with or without intravenous taurine for 24 mo. Statistical comparisons were carried out in eight patients randomly assigned to receive intravenous taurine, usually 10 mg. kg . d ‘, and 10 patients not receiving taurine. Compared with normal adults, baseline plasma taunne and urine taurine-creatinine ratios were decreased in both groups and platelet taurine was reduced in the taurine-treated group. During taurine treatment the mean of the mean values for taurine became normal in plasma and platelets and remained normal in erythrocytes, granulocytes, and lymphocytes; urine taurine-creatinine ratios rose to approximately five times normal. During follow-up, patients not given taurine had plasma, erythrocyte, and granulocyte taurine and urine taurine-creatinine ratios below normal values and the concentrations of taurine-treated patients. Their platelet taurine was also subnormal. Thus, 10 mg taurine kg ‘ d ‘ intravenously normalizes plasma and blood cell taunne concentrations in long-term TPN patients. Am J Clin Nutr .
E Ament
rime as well as other sulfur amino acids were not present in the solutions. The present study was part of a clinical trial designed to ascertain the effects of adding taurine to the intravenous solutions on body taurine concentrations and eye function in patients undergoing long-term TPN. Taurine is not well metabolized in the body (1, 9) and the major mechanism for the regulation of body pools is urinary excretion (10, 1 1). Hence, it was not clear whether the intravenous infusion of taurine, which could raise plasma taurine concentrations, might not lead to excessive urinary taurine losses and prevent the attainment of normal taurine concentrations in plasma or blood cells. The present report describes the response of plasma, blood cell, and urine taurine concentrations to daily infusions of taurine.
-
Methods Subjects
and study
design
1990;52:846-53.
clinically stable adult patients followed at the UCLA for the Health Sciences who had been undergoing parnutrition at home for 6 mo and who gave informed were included in this study. Patients first entered a baseline period during which blood and urine tests were performed. Subsequently, they remained on their usual parenteral-nutrition regimen but were randomly assigned to receive cither a daily infusion of taurine, 10 mg. kg body wt, which the patients added each day to their parenteral-nutrition solutions, or to receive no taurine supplement. Patients were maintained on these treatment regimens for a 24-mo follow-up period. Twenty-nine patients were randomly assigned, 15 to receive All
KEY WORDS urine,
blood
cells,
Taurine, methionine,
total parenteral cysteine
nutrition,
plasma,
Introduction Historically,
taunne
(2-aminoethane
sulfonic
acid)
has
not
been considered an essential amino acid for children or adults (I). For this reason it has not been recommended for enteral formulas or parenteral-nutrition solutions except possibly for low-birth-weight infants or neonates (2-4). However, we (5, 6) previously showed that adult patients undergoing long-term TPN have decreased taurine concentrations in plasma, erythrocytes,
lymphocytes,
and
platelets.
Similarly,
children
under-
going long-term TPN have low taurine concentrations in plasma, platelets, and urine (5, 7). The taurine concentrations tended to be particularly low in patients who ingested little or no food. These findings suggested that there may be a generalized depletion of body taurine in patients undergoing longterm TPN (8). The parenteral-nutntion solutions used in these studies provided abundant methionine (1 and range for adult patients: 2. 12, 0.64-3.40 g/d)(6, 7). However, cysteine and tau-
846
Am
J C/in Nuir
l990;52:846-53.
Center enteral consent
I
From
the Departments
of Medicine
and
Pediatrics,
Schools
of
Medicine and Public Health, UCLA and Harbor-UCLA Medical Center, and the UCLA Center for the Health Sciences, Torrance and Los
Angeles, 2
CA.
Supported
by NIH
grant
ROl DK
331 12 and
a grant
Healthcare,
from
Inc. NEV was the recipient ofan NIH Individual Research Service Award. 3 Address reprint requests to JD Kopple, Harbor-UCLA Center, 1000 West Carson Street, Torrance, CA 90509. Received August 7, 1989. Accepted Printed
for publication in USA.
November
© 1990 American
Baxter
National Medical
20, 1989. Society
for Clinical
Nutrition
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
Effect of intravenous taurine supplementation on plasma, blood cell, and urine taurine concentrations in adults undergoing long-term parenteral nutrition13
INTRAVENOUS
did
not
give
protocol,
informed
but
they
receive no taunne. or during the study four other patients long-term parenteral
consent
agreed
to the
to have
Seven pabecause of died. Nine nutrition
granulocytes,
occasional
measurements
of
lymphocytes,
and
platelets
during
the baseline period and at 2, 4, 6, 9, 12, 15, 18, 21, and 24 mo in the follow-up period. Plasma methionine and cystine and other serum chemistry variables were also measured at these times.
For
technical
and
logistical
reasons
blood
specimens
could not be obtained from individuals at every time interval during the follow-up period. To reduce the heterogeneity of each group for the statistical examination, these analyses were performed only on patients for whom plasma and blood-cell taurine concentrations were available at baseline and at three or more visits during the follow-up period. Eighteen patients satisfied these criteria, eight of whom received supplemental taurine, including the women given 2 mg taurine . kg’ . d ‘. Urine specimens were obtained less frequently for taunne measurements. Because not all patients were able to collect 24h urine specimens, part ofa 24-h urine collection or occasionally
single-void
viduals.
Hence,
urine
specimens
the urine
data
were
obtained
are expressed
from
some
as the ratio
mdiof tau-
rime to creatinine (taurine:creatinine). Characteristics ofthe two treatment groups of 18 patients are shown in Table 1 There were no significant differences between the groups; there was a trend (NS) for age to be lower in the group not receiving taunne supplement (no-taurine group). There were also no significant differences in the entire group of patients between the 25 no-taurine patients and the 9 taurinetreated patients with regard to the baseline characteristics shown in Table 1 . The ages of the entire group of patients in the no-taurine group (n 25) and the taurine group (n 9) were 50. 1 ± 3.5 and 60.2 ± 4.2 y (A± SEM), respectively (NS). Similarly, within each treatment group there were no significant differences in the baseline characteristics in the entire group ofpatients studied as compared with those who had three or more measurements in the follow-up period. Of the 18 patients with three or more follow-up visits, I 7 patients had short-bowel syndrome caused by Crohn’s disease (2 patients), radiation enteritis (7 patients), ischemic infarction (3 patients), volvulus (1 patient), pseudoobstruction (associated with scleroderma in 2 patients and idiopathic in 1 patient), or peptic ulcer disease (I patient). The patient with peptic ulcer disease had fistulas, adhesions with obstruction, and resection ofthe small intestine. The one patient without short-bowel syndrome had an esophageal resection for carcinoma of that organ. Radiation enteritis was caused by radiotherapy for carci.
=
TPN
TABLE
1
847
PATIENTS
Characteristics
at baseline5 Notaurine (n=l0)
Taurine
(n=8)
randomized-treatment
blood and urine taken while they remained on their usual therapeutic regimen. A total of25 randomly assigned and the above mentioned 9 nonrandomly assigned patients had one baseline and at least one follow-up measurement. Nine ofthese individuals received supplemental taurine: seven for 24 mo and two for 5 mo. Among these nine patients, one woman refused to take the full dose of taurine and was given 2 mg taunne . kg. d ‘ throughout the study. The data from this patient were included in the analyses ofthe taurine group. This study was approved by the UCLA and Harbor-UCLA Medical Center institutional review boards. Blood was drawn for measurement of taurine in plasma, erythrocytes,
IN
=
Age (y) DurationofTPN(mo) Height(cm)
49.0
Weight (kg)
59. 1 ±
Relative Triceps
84.9 ± I .9
3/5 63.9 ± 2.3 56.0± 14.2 162 ±2.3 63.8 ± 4.2 93.2 ± 5.8
13±6.0
12 ±0.50
19
19
M/F
4/6 55.9±
165±2.9
body weight (%) skinfold thickness
skinfold
1 .6
(mm)
Men Women Subscapular
5.9t 11.1
±
3.0
±
(mm) Men Women Arm muscle area (mm2) Men Women
44±5.3 39±4.2
Arm
95
muscle
area (% ofnormal)
Men
5
;:
SEM.
given
4.2 [4]
±
16±2.8 1 3 ± 1.5 [4]
58±4.1 52±8.3
10
102 12.4 8.2 71.0 4.36
1 30 1 1.9
± 20 [4] ±
1 .4
[3]
[7] 15.6 [6]
6.6 80.2
±
1 .9
[6]
±
16.5 [6]
0.84
4.96
±
1.26 [7]
±
I I
±
1.0 [8]
±
1 .0
± ±
[4]
1 14 ± 13 [7] 94±6.7[3]
[7]
Data were obtained immediately before patients began (treatment) phase ofthe study. n less than total sample is
in brackets.
t Normal 15.6
10±3.2 12 ± 2.3
85±21
Women Serum zinc (zmol/L)t Serum selenium (Mmol/L)t Serum vitamin B-6 (nmol/L)t Serum vitamin A (zmol/L)t
the follow-up
±
thickness
mol/L;
serum values: zinc, 1 1.5-18.5 zmol/L; selenium, 10.6vitamin B-6, 19-88 nmol/L; vitamin A, 0.35-1.75
zmol/L.
noma neal
ofthe ovaries, uterine corpus, or cervix or for retropentocarcinoma. Most patients had undergone intestinal resection. No patient had had evidence for active malignancy for 5 y. Twenty adults (9 men, 1 1 women) served as normal control subjects. Their mean age was 54.6 ± 3. 1 y, which was not significantly different from either patient group [analysis ofvariance (ANOVA), Brown-Forsythe test for equality of means (12)]. The intake and intestinal absorption of food, which varied from patient to patient, could affect the contribution of dietary taurine to the patient’s total daily taurine intake. We therefore classified patients by whether their enteral nutrient absorption was estimated to provide < or > 25% of their daily energy requirements. The quantity ofcalories in the nutrients absorbed from the intestinal tract was estimated from dietary histories, from knowledge ofwhether the remaininglength ofsmall intestine was too short to allow absorption of substantial energyyielding nutrients, or occasionally from 72-h fecal fat excretion or serum carotene concentrations. In previous studies these methods of assessment identified parenteral-nutrition patients who tended to have more severely reduced taurine concentrations in plasma or platelets (6, 7). It is estimated that 6 of the 10 no-taurine patients and 5 of the 8 taurine-treated patients absorbed an amount of nutrients from the intestinal tract that provided < 25% of their daily estimated calorie needs. Five of the no-taunne patients and one ofthe taurine patients received parenteral nutrition < 7 d/wk.
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
intravenous taurine and 14 to tients dropped out either before illness or for other reasons and additional patients undergoing
TAURINE
848
KOPPLE
Method
ofparentera/
nutrition
3.50%,
or
4.25%
amino
acids
(Travasol,
Baxter
Healthcare
Laboratories, Deerfield, IL). Lipids were provided from 1 to 7 d/wk as 10% or 20% Intralipid (Cutter Laboratories, Emeryville, CA) or 10% Liposyn (Abbott Laboratories, North Chicago, IL). Vitamins (10 mL/d) were given as MVI-l2 (USVRevlon, Tarrytown, NY). This preparation contained 12 vitamins,
including
4.0 mg pyridoxine
HC1(82.6%
pyridoxine)
and
1.0 mg vitamin A. Trace mineral supplements included 2 mg Zn/L and 1 mg Cu/L. Iron dextran (Imferon, Merrell Dow Pharmaceuticals, Inc, Cincinatti, OH, or Proferdex, Fisons Corp. Rochester, NY) was infused as needed. About 25% of the patients received 100 zg Sel/d and 10 zg Cr/d in their parenteral solutions. During the last 12 mo of the study, the patients received only the 3.50% or 4.25% amino acids and all patients were given selenium and chromium in their parenteral solutions. Analyses
ofblood
and urine
samples
treatment group, differences between baseline and the mean values for each patient during follow-up were evaluated by Student’s I test for paired data. For all tests for statistical significance, data were first subjected to log transformation. ANOVA was used to test for differences in values among the no-taurine group, the taurine group, and the normal control subjects. The Levene test for equality of variances was used to test for differences in the variances among the three groups. If the Levene test was positive (ie, statistically significant), the significance of differences between means was determined by the BrownForsythe test. Ifthe ANOVA indicated a statistically significant difference between means, then comparisons between the two patient groups and between each patient group and the normal control subjects were carried out using Tukey’s test. The main hypothesis proposed before the study was that taurime values were lower than normal in the patients who did not receive
methods of collection, preparation, and analyses of blood-cell and urine specimens for taurine are described in greater detail elsewhere (6, 7). Blood was collected in heparin-treated tubes. The plasma and blood cells were separated on Percoll (Pharmacia, Piscataway, NJ) gradients and prepared for amino acid analysis (6, 7). Blood was collected between 1 300 and 1 500 during the postabsorptive state, 6-8 h after completion ofthe nightly intravenous infusion and 6-8 h after the last meal, ifthe patient ate. Blood was obtained from normal control subjects at the same time, -6-8 h after their last meal. Specimens were deproteinized with sulfosalicylic acid.
Methionine
and
cystine
were
measured
only
with
a cell counter(Coulter
Electronics,
Hialeah,
FL).
Amino
acids were measured with an amino acid analyzer (121 MB, Beckman Instruments, Fullerton, CA) using a single-column three-buffer, lithium citrate elution system (14). Urine creatinine was measured with a creatinine analyzer (Creatinine Analyzer 2, Beckman Instruments), which uses a modification ofthe Jaffe method. Plasma vitamin B-6 was measured by the activation oftyrosine decarboxylase by pyridoxal-5-phosphate at American Biosciences Laboratories (Van Nuys, CA). The other laboratory measurements indicated in Table 1 were performed in the UCLA Hospital clinical laboratories. Blood for these clinical tests was usually obtained from subjects in the postabsorptive state between 1 300 and 1 500. Anthropometry was performed as previously described (13). Statisticalanali’ses Statistical
evaluations
were
with
the BMDP
carried
out
by ANOVA
and
lin-
statistical software package (12). For statistical analyses, in each treatment group the mean of the mean value for each patient during follow-up was compared with the baseline value, with the values for normal control subjects, and with the other treatment group. Within a ear regression,
were
normal
in those
patients
who
received
and
the
follow-up
period
was
available.
For
the
statistical
analyses shown in the text and tables and for the values shown in the tables, only the data from patients who had one taurine measurement
in baseline
and
at least
three
measurements
in
the follow-up period were used. Almost all of these latter patients had been randomly assigned to their respective treatment groups. An exception is the urine taurine data. Because of the more limited number of samples available, the data obtained from all patients were analyzed statistically.
Results
in plasma
because the yield from blood cells was too low to accurately measure intracellular methionine or cystine concentrations. The numbers and volumes ofdifferent blood cells were determined
and
To test this hypothesis, linear-contrast tests were also carried out between the normal control subjects and each of the two patient groups. Variance is expressed as SEM. To show all of the plasma and blood-cell taunne data obtamed in this study, taurine concentrations are shown for the 34 patients in whom at least one measurement in both the baseline
The plasma
taurine
taurine.
The mean taurine concentrations in plasma, blood cells, and urine at each time interval during baseline and follow-up in the entire group of34 patients in the no-taurine and taurine groups who had at least one baseline and one follow-up measurement (Fig I) were very similar to the values from the 18 patients who had three or more follow-up analyses (Table 2). Plasma taurine concentrations at baseline were significantly below normal in both groups of patients (Table 2, Fig 1). During follow-up the mean of the average plasma taurine concentrations increased only in the patients receiving taurine (P < 0.05), and in this group, plasma taurine was no longer different from normal values. In contrast, during follow-up in the no-taurine group, the plasma concentrations remained significantly below normal and also were significantly lower than the plasma concentrations ofthe taurine group. The changes in plasma taurine from the baseline value to the mean concentration during follow-up was significantly greater in the taurine group than in the notaurine group (P < 0.03). Erythrocyte taurine concentrations were not different from control values during baseline (Table 2, Fig 1). However, the mean values during follow-up increased significantly in the taurime group (P < 0.01) and tended to fall in the no-taurine group. The changes in the taurine concentrations from baseline in the taurine group were significantly different from the changes in the no-taurine group (P < 0.02). Moreover, the mean of the
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
Parenteral nutrition was administered over an 8-12-h period, starting at ‘--2000-2200 and ending at --0700-0900, as previously described (6, 13). Patients were infused with 1-3 L/d of solutions that contained 10-25% dextrose and 2.25%,
ET AL
INTRAVENOUS
TAURINE
110
90 80
3,70
jects
30
baseline
change
10
rime concentrations
0
a
2
4
10
6
Durotio”
12 14
16
ences
20
0
14 12 10
I
mean
a
6
10
8
Duration -;.
12 14 16
18202224
of study (mo)
25
00
2
20
I 10
a
2
4
6
8
10
12
Dumotion of
14
study
16
ie 20
22 24
0
c
I ! io is
0
2
4
6
8
10 12
Duration
14
16 18 20 22
24
of study (mo)
300 250 0
; -
There
and follow-up
in either
patient
in the change with the taurine
granulocyte
taunne
was
no
significant
in the granulocyte group,
nor
was
from baseline group. There concentrations
tau-
there
a sig-
in the no-tauwere no differbetween
the
200
150
100
j50
0
2
4
6
8
10
Ourotion
12 14 of study
ofthe
mean
lymphocyte
taurine
concentrations
16 18202224 (mo)
FIG 1. Plasma, erythrocyte, granulocyte, lymphocyte, and platelet taurine concentrations in 34 patients undergoing long-term TPN who were given intravenous infusions containing no taurine (0, n 25) or taurine (#{149}, n 9). Each patient had one baseline measurement of taurime (at zero time) and one or more follow-up measurements, n 8 for no taurine (range 5-1 1) and n 7 for taurine (range 4-1 1). .± SEM. =
=
=
=
erythrocyte taurine values during follow-up in the nogroup was significantly lower than the values in both the taurine group and the normal control subjects.
rose
sig-
nificantly in both groups of patients during follow-up but remained no different from normal values. The change in lymphocyte taurine concentrations between baseline and followup and the absolute values at either baseline or follow-up were not different between the two patient groups. Baseline platelet taurine concentrations were significantly lower than normal only in the taurine group (Table 2, Fig 1). However, during follow-up the values tended to rise in the taurime group (NS) but fell significantly in the no-taurine group. The changes in taurine concentrations between baseline and follow-up in the taurine group as compared with the no-taurine group was ofborderline significance (P 0.06). During followup the mean of the mean platelet taurine concentrations was significantly below normal only in the no-taurine group. There were no significant differences in the baseline or follow-up platelet taurine values between the taurine group and the notaurine group. Urine taurine:creatinine values at baseline were lower in each of the two patient groups compared with the normal control subjects (Table 2, Fig 2). During follow-up the mean of the mean taurine:creatinine values did not change in the notaurine group, although mean concentrations tended to rise near the end of the follow-up period and the ratios remained lower than in the normal control subjects. During follow-up in the taurine group, there was a dramatic and significant rise in the urine taurine:creatinine (P < 0.001). The follow-up urine taurine:creatinine in the taurine group was approximately five times greater than the normal value (P < 0.0 1 ) and about seven times greater than the ratio for the no-taurine group (P < 0.01). The increase in taurine:creatinine between baseline and followup in the taurine group was significantly greater than in the notaurine group (P < 0.01). The slope of the changes in taurine concentrations between the 2nd and 24th mo offollow-up, determined by linear-regression analysis, was not significant for plasma, urine, or any blood-cell type in the taurine and no-taurine groups. Moreover, there was no significant difference between the two groups during follow-up for the slopes ofthe taurine concentrations in plasma, urine, or any blood cell. The plasma and blood-cell taurine concentrations in the one woman given 2 mg . taurine . kg . d ‘ , in general, rose during the follow-up period. Her taurine concentrations at baseline and during follow-up, respectively, were as follows: plasma, 61 =
(mo)
20
.
in the
or follow-up.
baseline
two groups ofpatients at baseline, but the mean follow-up taurime concentrations were significantly greater in the taunne group than in the no-taurine group. Lymphocyte taunne concentrations at baseline tended to be lower than normal in both groups of patients, although these differences were not statistically significant(Table 2, Fig 1). The
1
!
between
nificant difference rime as compared
18202224
of study (mo)
18
mean
at either
20
-
849
Downloaded from https://academic.oup.com/ajcn/article-abstract/52/5/846/4651059 by Washington University School of Medicine Library user on 24 February 2019
s0 40
taurine
PATIENTS
Oranulocyte taurine concentrations in the no-taurine group tended to be lower than in the normal control subjects at baseline (P < 0.06) and were significantly below normal during follow-up (Table 2, Fig 1). The granulocyte taurine concentrations in the taurine group were not different from control sub-
100
6
IN TPN
KOPPLE
850 TABLE
2
Taurine
concentrations
in plasma,
blood
cells, and urine No
I 1.7 6.0
± ±
Taurine
10)
8.l 4.5II
38.2 72.6
43.8 5 1.4
±
73.1
1 .0 2.0ff
± ±
8.5
15 110±3531 [7] 15 751 ± 2 644[7]
3367±585[9] 6696 ± 457 [9]t ± 45
3359±564 9 298 ±
[9]
±
14 [9]59
±
15.4
[l2]
±
1 3.0
[ 15]9
