CLIN. CHEM.23/4, 490-492 (1977)
Evaluationof Serum Triiodothyronineand AdjustedTriiodothyronine
(Free TriiodothyronineIndex) in Pregnancy
Margaret E. Parslow,1 Thomas H. Oddie,2 and Delbert A. Fisher2
We measured serum thyroxine (free and total), triiodothyronine (free and total), thyroxine-binding globulin, and
triiodothyronine
uptake by talc in 97 normal men and 50
pregnant women. Mean serum thyroxine and triiodothyronine concentrations were higher in the pregnant subjects
(104 vs. 78 ag/liter and 1.69 vs. 1.30 zg/liter) because of
a higher mean thyroxine-binding globulin concentration
(70 vs. 38 mg/liter). Mean triiodothyronine uptake by talc
was lower in the pregnant subjects (0.82 vs. 1.03). Mean
free thyroxine concentrations
were similar in the two
were
10% lower in the pregnant subjects. Triiodothyronine
groups, but mean free triiodothyronine concentrations
uptake by talc and the dialyzable thyroxine and triiodothyronine fractions were highly correlated (r = 0.85 and
r = 0.82, P < 0.00 1). Calculated free thyroxine index and
free triiodothyronine index values (thyroxine and triiodothyronine indirectly adjusted, using triiodothyronine talc
uptake to compensate for differences in thyroxine-binding
globulin concentration), were statistically similar (84 vs.
82 and 1.38 vs. 1.34) in pregnant and male subjects. The
results indicate that the total triiodothyronlne concentration
can be normalized on the basis of the triiodothyronine
uptake by talc to correct for variations in thyroxine-binding
globulin concentration.
AdditIonal
Keyphrases:
ine-binding globulin
status
free and total thyroxine
triiodothyronine uptake by talc
.
thyroxthyroid
Increased concentrations
of serum thyroxine-binding
globulin (TBG) are known (1) to account for increases
in the total concentrations
of serum thyroxine (T4) and
triiodothyronine
(T3) during pregnancy.3 Serum T3
resin-uptake test values are decreased; those for free T3
and free T4 remain more or less unchanged (1). Stein
and Price (2) have shown that T4 values may be normalized for variations of TBG by calculating an “adjusted T4,” which is the product of T4 and the T3 resin
uptake. Extension of this adjustment
to T3 results has
been suggested by Bermudez et al. (3).
‘Nichols Institute for Endocrinology, San Pedro, Calif. 90731.
2 UCLA-Harbor
General Hospital, Torrance, Calif. 90509.
Nonstandard
abbreviations
used;
TBG, thyroxine-binding
globulin;
T4, thyroxine;
T3, triiodothyronine;
Here, we describe measurements
intended to investigate this possibility further, on serum from normal
healthy men and from pregnant women.
Materials and Methods
The subjects were 97 apparently normal men, age 20
to 50 years (mean age, 33) and 50 pregnant women, age
17 to 43 years (mean age, 28), who were in the third
trimester of gestation. Men and nonpregnant
women
have similar mean T4 and T3 concentrations,
although
TBG concentrations
are about 8% lower in men.4 Use
of samples from males avoids the effect of oral contraceptive medication.
Pre-breakfast blood samples were obtained from each
subject for measurement of serum T4, T3, TBG, free T4,
and free T3 concentrations,
and T3 uptake by talc
(T3U). T4 and T3 were measured by whole serum radioimmunoassay,
with use of 8-anilino-1-naphthalene
sulfonic acid to block T4 binding to TBG and barbital
buffer to block binding to prealbumin
(4,5). TBG was
measured by radioimmunoassay
(6). Dialyzable T4 and
T3 were measured by a modification of the dialysis and
magnesium
precipitation
method
of Sterling
and
Brenner (7), with correction for sample dilution (8).
Free hormone concentrations
were calculated as the
product of the total concentrations
and the dialyzable
fractions.
T3U was measured by using talc as a binder. The
procedure is similar in principle to T3 resin uptake
methods. Under standardized
conditions labeled T3
distributes
between unbound TBG-binding
sites and
an added T3 adsorbent, in this case talc. All results are
expressed relative to those for a pool of serum from
normal adults; this pool was comprised of equal aliquots
from 100 healthy young men, to minimize effects of
illness.
Reagents
This is prepared by adding talc
tablets (25 mg of talc and other silicates; Ormont Drug
and Chemical Company, Inc., Englewood, N.J. 07631)
T3/talc
mixture.
T3U, triiodothyronine
talc uptake.
Received
490
Aug. 23, 1976; accepted
Dec. 14, 1976.
CLINICALCHEMISTRY.Vol. 23, No. 3, 1977
Our unpublished
observations.
Table 1. Summary of Mean Test Results
M.n
Pregnant
Combined data
wOmen
No.
Variable
No.
Resuft8
No.
Result8
Talc 13 uptake fraction of poo11)
Log10(13)b
13, ag/liter
Log10 (free 13 index)c
95
1.033±0.075
50
50
0.816±0.032
0.229 ± 0.082
1.69 (1.40-2.05)
0.140±0.075
1.38(1.16-1.64)
Free T3 Index, zg/Ilter#{176}
Log1o (14)b
T4, gig/liter
Log10 (free T4 lndex)C
Free 14 index, ig/lIter#{176}
Free 13 b, pg/liter
Free 14 C, ng/liter
TBGb, mg/liter
Mean and SD; logarithmic
Difference between men
C Difference
between men
d x2-test shows log-normal
a
b
97
95
0.113±0.076
1.30 (1.09-1.54)
0.127±0.067
1.34(1.15-1.56)
1.901 ± 0.072
79.6 (67.5-94)
1.913±0.066
93
47
95
4240±710
23.5 ± 4.0
36.5 ± 7.8
95
97
50
50
2.015±0.087
50
1.926 ± 0.079
84 (70-101)
3770 ± 590
24.0 ± 4.9
79.1 ± 11.4
145
Result8
O.l3l±O.O7Od
1.35(1.15-1.59)
104 (85-126)
82 (70-95)
48
28
50
145
1.917±0.071
83 (70-97)
75
23.6 ± 4.4
values translated to arithmetic mean and ± 1 SD range.
and pregnant women significant by f-test at P < 0.001 level.
and pregnant women not significant by f-test.
distribution (P = 0.7).
Adjusted T3 or 14.
to give a concentration
of 27.5 g/liter of barbital buffer
(75 mmol/liter, pH 8.6; Beckman buffer B2). The buffer
is mixed with a magnetic stirrer until the tablets are
dispersed, and [125ljtriiodothyronine
is added to provide
about 60 000 cpm/ml.
similar highly significant correlation
(r = 0.82, P <
0.00001) between the fractional T3U and dialyzable T3
fraction (Figure 2), represented
by the regression
equation:
T3U (fraction
of pool)
Procedure
Pipet 1 ml of T3/talc reagent into 10 X 75 mm glass
test tubes. Add 50 sl of the pooled serum to triplicate
tubes. Add 50 il of each unknown serum to duplicate
tubes. Vortex-mix the contents of the tubes, allow them
to stand at room temperature
for 15 mm or longer,
vortex-mix
a second time, and centrifuge
(2000 rpm, 10
mm). Decant each supernatant
solution into a counting
tube for gamma counting, and count each tube for 1
mm. Triiodothyronine
uptake is calculated:
1251
T3U
Within-assay
=
1251
cpm (standard)
cpm (unknown)
coefficient of variation
2%; across-assay
(CV) is less than
CV does not exceed 3.3%.
Results
Mean values and standard deviations for the several
measured variables are shown in Table 1 for the normal
male subjects and the pregnant women. Combined results are shown for the values of adjusted T3, adjusted
T4, and free T4. The adjusted T4 indicates total T4
concentration
“adjusted”
for TBG variations as suggested by Stein and Price (2):
Adjusted
T4
=
T4 (sg/liter)
X T3U (fraction
of pool)
(1)
This correction is justified because of the high correlation (r = 0.85) between the results for fractional
T3U and the dialyzable T4 fraction. There was also
= 0.509 + [1.546
X (T3 dialyzable fraction
Therefore, the total T3 concentration
follows:
Adjusted
T3
=
in %)].
was adjusted
T3 (j.tg/liter)
X T3U (fraction
of pool)
as
(2)
Both T3 and T4 values were distributed log-normally.
Figure 1 shows the distribution
of the logarithm of adjusted T3 for the men, the pregnant women, and the
combined data, satisfactory by x2-test (P = 0.7). When
adjustment
was made in accordance with equations 1
and 2 the mean adjusted T3 and adjusted T4 results
were very similar for the two groups. Free T4 values also
were similar in the two groups of subjects (23.5 and 24
ng/liter, Table 1). Free T3 values differed significantly
by t-test (P <0.001).
Discussion
Serum T4 and T3 are largely bound to serum proteins, the most important of which is TBG. As a result,
total serum T4 or T3 concentrations
vary directly with
concentrations
of binding protein. The thyroid hormones (T4 and T3) available to tissues are the free or
dialyzable fractions. These free hormone concentrations
can be estimated as the product of measurements
of the
dialyzable
fractions
and total hormone concentrations.
Measurements
of thyroid hormone dialyzable fractions are rather laborious and expensive and thus not
suited to routine thyroid assessment. It has been obCLINICALCHEMISTRY,Vol. 23, No. 3, 1977 491
DISTRIBUTION
OF Log, (ADJUSTED
13) RESULTS FOR
50 PREGNANT FEMALES AND 95 CONTROL MALES
.4
ONSaaiD
0
30
9
.2
CONTNOI.
Nail
0
PNSINANT
S
.5
1.0
I.
0
U20
I-.
0.8
U
di
10
0.2
0.3
0:4
13 DIALYZABLEFRACTIONPERCENT
1
0
SD.
.
-2
-1.5
-I
,
-0.5
DEVIATION
,
,
0
+0.5
Fig. 2. Correlation between 13 talc uptake result and dialyzable
13 fraction in 136 subjects
+1
+1.5
+2
S.D.
FROM MEAN VALuE
Fig. 1. DistributIon of adjusted 13 results for 50
and 95 control men
pregnant
women
Values are recorded as log,,, of the adjusted T3 results. Each column shows
pregnant female, control male, and combined results. See text for details
served that there is a high correlation between the dialyzable fraction of T4 and the result for T3U expressed
as a fraction of the normal serum value (2). Thus, the
effective thyroxine concentration
has been estimated
by calculation of the free T4 index.
Free T4 index
=
total T4 (tg/liter)
X
T3U (fraction)
This normalized T4 result has also been referred to as
the “adjusted T4” or “corrected T4” (2).
Bermudez et al. (3) suggested that a free T3 index can
be similarly calculated. However, the degree of correlation between the T3U and dialyzable T3 fraction has
not been investigated,
and the extent to which the adjusted T3 will normalize T3 for variation in TBG concentrations has not been rigorously tested. The present
results provide this information.
The serum T4 and T3 concentrations
in the men and
in the pregnant women differed statistically
(Table 1)
owing to the marked difference in mean TBG concentrations (36.5 vs. 79.1 mg/liter, Table 1). As expected,
the high mean TBG concentration,
resulted in a low
T3U. When we adjusted the total T4 and total T3 results in the pregnant subjects for this variation in TBG,
the results were similar to values in the men (Table 1).
The mean free T4 concentrations,
like the mean adjusted T4 results in the two groups, also were similar
(Table 1). The free T3 concentration
was about 10%
lower (P <0.001) in the pregnant women than in the
men. Whether this is real or artifactual, in the light of
the similar values for free T4, adjusted T4, and adjusted
T3 in the two groups of subjects, is not clear. In any case,
the difference was small.
The results indicate that the total T3 concentration
by radioimmunoassay
in subjects with above-normal
492 CLINICALCHEMISTRY.Vol. 23, No. 3, 1977
The T3 talc uptake result is recorded as a ratio to the control serum. The correlation coefficient was r = 0.82. signIficant at the P < 0.0000 1 level
TBG concentrations,
like the total T4 concentration,
can be normalized on the basis of the T3 resin uptake
or similar test to correct for high TBG concentration.
This correction is carried out exactly as the T4 correction. As suggested by Stein and Price (2) the term “adjusted T3” emphasizes the fact that this product adjusts
for any alterations in total hormone concentration
due
to changes in TBG concentration.
The term “free T3
index” has been recommended
by the committee on
nomenclature
of the American Thyroid Association
(10).
References
I. Erenberg, A., Phelps, D., Lain, R., and Fisher, D. A., Total and free
thyroid
hormone
concentrations
in the neonatal
53, 211 (1974).
2. Stein, R. B., and Price, L., Evaluation
(free thyroxine
index) as a measure
docrinol. Metab. 34, 225 (1972).
period.
of adjusted
of thyroid
Pediatrics
total thyroxine
J. Clin. En-
function.
3. Bermudez, F., Surks, M. I., and Oppenheimer,
of decreased
non-thyroidal
J. H., High incidence
serum triiodothyronine
concentration
in patients
with
disease. J. Clin. Endocrinol. Metab. 41,27 (1975).
4. Chopra, I. J., A radioimmunoassay
for measurement of thyroxine
in unextracted
serum. J. Clin. Endocrinol. Metab. 34,938 (1972).
5. Chopra, I. J., Ho, R. S., and Lam, R. W., An improved radioimmunoassay
of triiodothyronine
in serum: its application
to clinical
and physiological
studies. J. Lab. Clin. Med. 80,729 (1972).
6. Levy, R. P., Marshall, J. S., and Velayo, N. L., Radioimmunoassay
of human thyroxine-binding
globulin
(TBG). J. Clin. Endocrinol.
Metab. 32, 372 (1971).
7. Sterling, K., and Brenner,
M. A., Free thyroxine
simplified measurement
with the aid of magnesium
Gun. Invest.
in human serum:
precipitation.
J.
45, 153 (1966).
8. Chopra, I. J., Sack, J., and Fisher, D. A., 3,3’,5’-Triiodothyronine
(reverse T3) and 3,3’,5’-triiodothyronine
(T3) in fetal and adult sheep:
studies of metabolic clearance rates, production rates, serum binding,
and thyroidal content relative to thyroxine. Endocrinology
97, 1080
(1975).
9. Nie, N. H., Hull, C. H., Jenkins, J. G., et al., Statistical
Package
for the Social Sciences, 2nd ed., McGraw Hill, New York, 1975.
10. Solomon, D. H., Benotti, J., DeGroot, L. H., et al., Revised nomenclature for tests of thyroid hormones in serum: the Committee
on Nomenclature
of The American
Thyroid
Association.
J. Clin.
Endocrinol.
Metab. 42, 595 (1976).