Download Comparison of TSH Sensitivity: Conventional vs. Solid-Phase IRMA in Thyroid Patients and more Schemes and Mind Maps Radiology in PDF only on Docsity!
onventional competitive binding radioimmunoas
says (RIAs) for h-TSH (1) in clinical use today have a @ maximal analytical sensitivity of 1 MU/mi serum, close
to the mean serum thyrotropin (TSH) concentration en
countered in healthy individuals. They differentiate well
between euthyroid patients and those with primary hypo
thyroidism and are unquestionably of value in the diag
ReceivedMar.25, 1985,revisionacceptedJuly 2, 1985.
For reprintscontact:MonikaE Bayer,MD, Div. of Nuclear
Medicine, Dept. of Radiology, Stanford University Medical
Center,Stanford,CA 94305
nosis of hypothyroidism, but they lack the sensitivity to
measure TSH concentrations at the lower end of the nor mal range precisely enough to distinguish them from the suppressed TSH levels expected in hyperthyroidism.
Wehmann et al. (2) and Spencer and Nicoloff (3) have
reported “highlysensitive, competitive binding, research
RIAs―for TSH which could delineate the entire normal
range and identify subnormal TSH concentrations. Unfor
tunately, both these assays have features that make them
impractical for routine clinical laboratories. For instance,
both assays employ the same antiserum, raised by Parlow
@ in 1975 (4), which has a uniquely high avidity (K@10' 1/
1248 Bayer,Krmss,andMcDougall The Journalof NuclearMedicine
Clinical Experience with Sensitive
Thyrotropin Measurements:
Diagnostic and TherapeuticImplications
Monika F. Bayer, Joseph P Kriss, and I. Ross McDougall Departments ofMedicine and Radiology, Division ofNuclear Medicine, Stanford University School ofMedicine,Stanford,California
A two-siteimmunoradiometricassayfor serumthyrotropin(TSH)wasmodifiedto Improve
the analyticalsensitivity.The sensitivity achieved(detection limft, 0.1 MU/mI;lower Ilmft
@ of quantitative measurement, 0.4 pU/mI) was comparable to that of the best competftlve binding research assays, yet this assay can be performed routlneIy@Serum TSH was 1.
±0.69 (mean ±s.d.) (range 0.4-3.4 MU/mI)in healthy IndividualS and 1.83 ±0.90 @tU/ml
(range 0.7-3.7 @U/ml)In patients with nonthyroldal disorders. By contrast, 97% of
clinically hyperthyroid patients (Graves'disease,toxic nodular goiter) wfth high serum
free T4 (Fr4) and T3 had suppressed serum TSH values, I.e., <0.3 @U/ml.Among patients wfth euthyrOldGraves' ophthalmopathyor nontoxic goiter those clinicallysuspected of
mild hyperthyroldismhad TSHvalues < 0.3 @U/rnl,while those judged euthyrold had
normal values. A large proportion of thyroid patients on antithyrold drugs (poorly to well
controlled)had suppressedTSH.Of Graves' patients in remission(normal FT and T3),
75% had normal TSH, but IndMdual levels changed significantly over time, suggesting
that a progressivedecline In TSHmay be useful In predicting recurrences.in hypothyroid
patients taking L-T4, serum TSH was subnormal In patients wfth elevated FT, but TSH
was also low In six patIentsclinically suspectedto be thyrotoxic despite normal FT@and
T3andIn32%of asymptomaticpatientswithnormalthyroidhormonelevels.Conversely,
23% of thyroid cancer patients who had undergone thyroldectomy were taking
Insufficient L-T4 to completely suppress TSH secretion. In 25 IndIviduals who underwent
thyrotropin releasinghormone (TRH)stimulation tests, the baselineserum TSHvalue
correlated well wfth the peak serum TSH value post-TRH (r = 0.85). We conclude that
sensitiveTSHmeasurementscould establish or confirm the diagnosisof hyperthyroidism
In equivocal cases, replace most TRH-stimulation tests and be of value in optimizing L-T suppression therapy for thyroid cancer patients post-thyroldectomy.
J NuciMed26:1248-1256, 1985
mole), but which is distributed only for research purposes
by the National Pituitary Agency. Also, these assays re
quire either extraction of TSH with concanavalin A-
Sepharose from a large sample volume (5) or sophisti
cated purification procedures for the iodine-l25 (1251)
TSH before each assay, as well as extremely long incuba
@ tion times ( 10 days).
Recently, several immunoradiometric assays (IRMA)
have been developed for TSH in which TSH is bound by
twoormorespecificmonoclonalantibodiesinasand
wich-like fashion. The significance of this new approach
is that higher assay sensitivities can be attained, because
the antibody affinities are a less limiting factor in this type
of assay as compared to competitive binding assays. After preliminary testing and comparison of six com
mercial procedures for TSH a two-site IRMA* was opti
mized for the low concentration range and met best our two criteria, high sensitivity plus suitability for routine
clinical performance.
The following report deals with serum TSH determina
tions by this method in over 200 thyroid patients, their
clinical relevance, and the potential utility of such TSH
measurements.
SUBJECTS AND METHODS
Patients
Nearly all patients were seen and evaluated as outpa
tients in our institution's thyroid clinic by two ofus (J.P.K.
and I.R.M). Only in a few overtly hyperthyroid patients
had the diagnosis been made by other physicians at the
institution's medical center, and in these cases all pertinent data were retrieved from the patients' charts. All patients
had a serum specimen sent to our laboratory for routine
thyroid function tests. The selection of patients for the
“sensitiveTSH test―was random, except for perhaps
preferentially selecting borderline hyperthyroid and diffi cult-to-diagnose patients, or an attempt to keep the size of various patient groups similar. Based on the standard din
ical criteria, history, physical examination, serum VF4, T
(Thble 2), and TSH levels (by a conventional assay, data not given and used only for classification), patients were
assigned to one of eight groups, four groups of hyperthy
mid patients and four groups of hypothyroid patients and/
or patients on L-T4 therapy:
Control 1. Forty-one healthy laboratory volunteers. Control2. 1\venty-oneambulatorypatientswith exclu
sively nonthyroidal disorders including cancer, arterio
sclerosis or hypertension. Group 1. 1\wenty-nine hyperthyroid patients (Graves'
disease, toxic multinodular goiter, toxic nodule, or un
known etiology) who, with a few exceptions of prior ra
dioiodine treatment, were untreated. Group 2. Ten clinically and biochemically euthyroid (or
at most borderline hyperthyroid) patients with various di
agnoses (Graves' ophthalmopathy, simple or multinodular
nontoxic goiter, autonomous nodule).
Group 3. Thirteen hyperthyroid patients under treatment with propylthiouracil; their clinical status at the time of
the TSH measurement ranged from euthyroid to hyper
thyroid.
Group 4. Seven hyperthyroid patients in clinical remis
sion after therapy, all judged clinically euthyroid, taking
no medications for at least several months (five patients
had received antithyroid drugs, one had had thyroid sur
gery and one had received radioiodine treatment).
Group 5. Sixteen overtly hypothyroid patients with sub normal thyroid hormone levels. Group 6. Forty-six patients with a previous diagnosis of hypothyroidism (status post-treatment for Graves' dis ease, Hashimoto's thyroiditis or unknown etiology) who
were taking L-T4 (levothyroxine sodium) at the time of
evaluation. This group was further subdivided, depending
on the patients' clinical assessment and their thyroid hor
more test results; &z: 15 patients whose clinical status
ranged from borderline to clearly hyperthyroid and who
also had abnormally high serum free T4 (FT@)and T
levels, confirming the clinical impression; @:25 patients
who were asymptomatic, had no clinical findings that
could be related to overtreatment and had normal FT@and
T3 levels; &: six patients who had subtle clinical signs of
hyperthyroidism, such as onycholysis, quadriceps weak ness or rest tachycardia, suggesting that the dose of L-T
was excessive, but who had serum FF4 and T3 levels
repeatedly within the normal range. Group 7. Thirty-five patients who had undergone thy
roidectomy and/or iodine-13l (‘@‘I)ablation therapy for
papillary and/or follicular thyroid cancer, all taking Syn
throid at the time of this study.
Group 8. Ten patients on L-T4 suppression therapy for benign thyroid nodules.
A TRH-stimulation test was performed on 21 appar
ently healthy individuals and on four patients with equivo
cal thyroid disease in the course of their clinical
evaluation.
Procedures
Sensitive TSH measurements were carried out using a
two-site solid-phase, sequential IRMA*, which we modi
fled for optimal sensitivity. The assay principle is as fol
lows: standards or sera are first incubated with a (3-subu
nit specific monoclonal anti-TSH antibody immobilized
on large polystyrene beads (one bead/tube). After decant
ing the assay mix, the beads are reincubated with a see
ond, iodine-125-labeled, a-subunit specific monoclonal
anti-TSH antibody. The binding of the latter is propor
tional to the amount of TSH bound to the bead during the
first incubation and, in turn, to the TSH concentration in standard or sample. (The specificity of the two antibodies
Volume26 •Number11 •November1985 1249
CONTROL 1 CONTROL 2 GROUP 1 GROUP 2 GROUP 3
HYPERTHYROID EUTHROID OR HYPERTHYROID
BORDERLINE ON ANTI HYPERTHYROID @HYROID DRUG
GROUP 4
HYPERTHYROID IN REMISSION
FT4 FT4 TSH
TSH
...
:R
.. ...
FT 00
0 (^00) 0
j
U
TSH
•(6.0J
.. •A• .. B;
TSH
-I..
TSH FT
FT
RR o@
S
0 0 0
13 220
5.
4.
3 3. 0 3 @ 2.
1. 0.
0.
FIGURE
Distributionof serumTSH (closedsymbols)and FT@(open symbols)in healthyindividuals(Control1), patientswith
nonthyroidaldiseases(Control2) and four groupsof hyperthyroidpatients.Group 2: (I ,O)Graves'ophthalmopathy;
(•,D)Simple/multinodulargoiter; (A,L@)Hot nodule. Group 3: (U,D) Elevated FT; (•,O) Normal FT@.Group 4 (seven
patients,12studies):(•,0)Postprophylthiouracil(SD) Postsurgery;(A,L@)Post 1311
- 2s.d. was determined according to Rodbard (10). The
lower limit of quantitative determination defmed by the
expression s.d. (x) 10% of x (measurement error is
not more than 10%) according to Kalman (11) was esti
mated graphically from the interassay coefficients of van
ation of the four serum pools with very low TSH concen
trations.
RESULTS
The assay performance characteristics are summarized
in1@ible 1.
The distribution of serum TSH and FT@results in con
trols and various groups ofpatients is illustrated in Figs. 1
and 2. Means for serum TSH and serum thyroid hor
mones are shown in ‘Thble2 and individual cases that
might be of particular interest are listed in Table 3. Serum TSH concentrations measured in healthy mdi viduals (Control 1) were 1.82±0.69 @.&U/ml(mean±s.d.) (range 0.4-3.4 MU/mi) and patients with miscellaneous
nonthyroid disorders (Control 2) had similar TSH levels,
1.83±0.90 MU/mi (range 0.7-3.7 @iU/m1).
Clinically hyperthyroid patients with Graves' disease or
toxic multinodular goiter (Group 1) had subnormal serum
TSH concentrations of <0.3 MU/mi, with only one cx
ception. The one patient whose TSH was not suppressed
had a serum FF4 of 4.3 ng/dl, a T3 of 195 ng/dl, and a
TSH of0.6 iU/ml. Anti-I4 autoantibodies, which in rare
instances can cause misleadingly high VF4 or T4 results, were tested for in this case, but not found. Among untreated patients with normal to borderline
high thyroid hormone levels (Group 2), four clinically
euthyroid patients (two with Graves' ophthalmopathy, one
with a goiter and one with a hot nodule) had normal to
slightly elevated serum TSH results of 1.76-6.0 MU/mi, (FT4 1.2-1.9 ng/dl; T3 95—160ng/dl), while six other patients with a similar diagnosis (four Graves' disease,
two goiters), clinically suspected to be mildly hyperthy
roid had significantly lower serum TSH levels of <0.3-
0.4 @uimm(FT4 1.3-2.5 ng/dl; T3 155-200 ng/dl). Fur thermore, one of the later (TSH < 0.3 @UImm;FT@ 1. ng/dl; T3 200 ng/dl) became grossly hyperthyroid within —2mo.
Of thyrotoxic patients who were undergoing treatment
with antithyroid drugs (Group 3), approximately half
were poorly controlled as judged by their elevated serum
thyroid hormone levels (n=7, FT@2.4—5.0ngldl; T
105-350 ng/dl), and they had serum TSH of <0.3-0.
@uimm.Of theotherpatientswithnormalthyroidhor
mone levels (n=6, FF4 1.0-1 .8 ng/dl; T3 100—200ng/di)
four had suppressed and two had normal TSH values.
The majority (75%) ofpreviously thyrotoxic patients in
clinical remission (Group 4) had normal rather than sub
normal serum TSH. Sequential studies in three of these patients indicated that their serum TSH changed signifi candy in all ofthem in the course of 3 mo; TSH increased from subnormal to normal in two patients and fell from a borderline high level to low normal in the third patient [Tkble 3(B)]. Clinically hypothyroid patients had TSH concentrations
25 MU/mi.hi hypothyroidpatientson thyroidreplace
Volume26 •Number11 •November1985 1251
5.
E
@ 3.0 -
E
@ 2. 3. (0) **I I-
0.**
TSH FT
E
V 0 U E -3, I U) I.
TABLE Estimatesfor Serum TSH, FT4 and T3 Concentrationsin VariousPatient Groups No.of patients TSH(@U/ml) Group Clinical condition (male, female) mean (s.d.)* RangeFT@
(ng/dl) mean(s.d.)RangeT
(ng/dl) (s.d.)RangeC-i mean
(18)90—175C-2^ Healthy^ 41 (10 M,3119^ 1.82 (0.69)^ 0.4—3.41.43^ (0.22)1.0—1. .9——2 Euthyroidsick 21 (5 M, 16 F) 1.83 (0.90) 0.7—3.71 .39(0.25)0.9— (164)195-8202^ Hyperthyroid^ 29 (5 M,24 F)^ <0.3^ <0.3t4.79^ (1.08)3.0—>5.0@ Euthyroidorborderline 10(2 M, 8 F) seetext < 0.3—6. hyperthyroid
(34)95—2003 .70 (0.46)1 .2—2.
(87)100—3504Hyperthyroidon therapy 13(2 M, 11 F) seetext < 0.3—1.852.72 (1.65)1 .0-> 5. (47)85—2005 Hyperthyroidin remission 7 (1 M,6 F) seetext <0.3—6.01 .51(0.32)0.9—2. 0.2—0.8)——6Overthypothyroid 16(4 M, 12F) 80.8(26.9) 25—>1000.38 (0.20)< OnT4,previously hypothyroid
a) highFT4,T3 15(1M,14F) <0.3 <0.3@
b) normalFT4.T3clincally euthyroid 25(3 M, 22 F) 0.98(0.93) < 0.3—3. c) normalFT4,T3mildly hyperthyroid 6 (6 F) <0.3 <0.3—0.4)3.
1.62(0.20)2.4—4.5@
0.9—2.
1.3—1.
—
130(23)150-
—
105— Thyroidcancer, (0.67)1.2—3.6——8postsurgery/1311,on T4 35 (8 M, 27 F) 0.45(0.74) <0.3—4.02. .7—3.9——‘AThyroidnodule,on T4 10(1 M,9 F) 0.33(0.32) < 0.3—1.12.28 (0.64) valueof0.15wasusedinsteadof < 0.3forcalculationofmean(s.d.). TOneexception:TSH,0.6RU/mI. @12studies. 1Oneexception:TSH,I .0;FT@,3.8;T3,195. @Excludingone case of T3toxicosis.
GROUP 5 GROUP 6 ON THYROXINE REPLACEMENT
GROUP7 GROUP 8
5.
‘..u ‘1_@ .@ 3
3. .@^0 3 2.
1. 0.
5.
4.
2.
FiGURE 2 Distribution of serum TSH (closed symbols) and FT@(open symbols) in four groups of hypothyroid patients or patients taking L-T4.Group 7 and 8: (•,0) Normal FT@,(U ,D) Elevated FT@
ment therapy (Group 6) serum TSH varied considerably:
Patients taking excessive amounts of levothyroxine so
dium as indicated by their elevated serum FT@(Group 6a)
had, with one exception, subnormal TSH of <0.3 @tU/
ml. All six patients of Group 6c who had repeatedly nor mal serum thyroid hormone concentrations, but who pre
1252 Bayer,Kriss,andMcDougall The Journal of Nuclear MedicIne
with initially suppressed serum TSH showed that the TSH
could be normalized by changing the L-T4 treatment dos
age (Table 3(D), #8—10).The observed changes in the
TSH were similar to those noted in hyperthyroid patients
after successful treatment by surgery or radioiodine [Ta
ble 3(A)].
Among thyroid cancer patients who had undergone thy
roidectomy and/or ‘311-ablationand who were receiving
levothyroxine sodium as replacement and suppression
therapy (Group 7), one-third (33%) had serum TSH levels of <0.3 (or at most 0.4 @iU/mi)paired with a normal serum VF4 (range, 1.2—2.3ng/dl), as is considered opti
mal. 43 % of the patients had an elevated FT@(range 2.4—
3.6 ng/dl) and TSH <0.3 MU/mi. The remaining 23%,
however, had clearly measurable serum TSH ranging
from 0.5-4.0 MU/mi(VF4 1.2-2.5 ng/dl).
In one of the thyroid cancer patients, because of re
peated abortions possibly due to thyrotoxicosis, the L-T
dose was deliberately reduced to allow the TSH to rise
into the normal range (Thble 3(C), No. 11). In other
thyroid cancer patients an attempt to normalize serum FT@
concentrations by adjusting the L-T4 dosage downward
also sometimes resulted in concomitant normal TSH (Th
ble 3(C), No. 12), generally not regarded as optimal sup
pressive therapy.
In patients on suppression therapy for benign thyroid
nodules (Group 8), eight patients (80%) had suppressed
serum TSH of <0.3 MU/mi, while two patients had a normal TSH. Data collected from 25 TRH stimulation tests indicated
a good correlation between baseline serum TSH levels
pre-TRH and peak serum TSH values 20-30 mm after
TRH injection (Fig. 3, r=0.85). A subnormal TSH of
<0.3 was found in a patient with no rise in TSH, or
borderline TSH values of 0.4-0.5 @tU/mlin two patients
with a blunted response. All individuals witha!ormal to
slightly exaggerated TRH test (@ TSH 5 @iU/ml)had a
normal to borderline high baseline TSH.
DISCUSSION
One of the principle findings of this investigation was
that IRMA-type assays which make use of two or more monoclonal anti-TSH antibodies in sequential fashion,
can be optimized to achieve sensitivities comparable to
the best conventional competitive binding research assays
and, in contrast to the latter, can be performed routinely.
The specific TSH assay under investigation* had, after @ minor modifications, a detection limit of 0.1 MU/mi, a
value similar to those reported for the two most sensitive
competitive binding RIM (0.33 or 0.05-0. 1 1@U/mi) (2,3).
Although the detection limit (or detection threshold) is
widely used to characterize assay sensitivity, it merely
states the smallest hormone concentration that can be dis
criminated from “zero―hormone, and as such is probably
E
D 3.
U)^ I LU
z
-J LU U)
MAXIMUMTSHPOSTTRH (pU/mI)
FIGURE 3 Correlation between baseline serum TSH levels measured
before administrationof TRH and peak TSH levels ob
served 20-30 mm post Lv. bolus of 400 @gTRH in 21 individualswithout any known organic disease (•)and four
thyrokipa@entswfthequivocaldiagnosis(U)
n = 25;y = 0.lOx+ 0.21;r= 0.
inadequate when assay sensitivity is the key issue. There
fore, we have also determined the “lowerlimit of quanti
tative determination,―as recently proposed (11), to mdi
cate that the smallest TSH concentration which can be
precisely measured (s.d. 10% of mean) by the assay is
—0.4@iU/ml.Values below 0.3 @iU/mlare not as repro
ducible and are thus stated as <0.3 MU/mi.
Serum TSH concentrations found in healthy individuals
without thyroid, pituitary, or hypothalamic disorders were
consistently above this lower limit of quantitative mea
surement, ranging from 0.4-3.4 @iU/mi.In patients with
miscellaneous nonthyroidal diseases the range was 0.7-
3.7 @UImi.By contrast, 97% of hyperthyroid patients
had a serum TSH of <0.3 MU/mi, i.e., below the limit of
quantitative determination. Thus the method provides
practically a complete separation between normal serum
TSH concentrations and those found in hyperthyroid pa
tients. Moreover, the results for the normal range, or for
patients with hyperthyroidism, respectively, are similar to
those obtained by competitive binding assays (3,12).
While this work was in progress, Pekany and Hershman
(13) reported a higher detection limit of0.6 MU/mi, larger
intra-, as well as interassay coefficients of variation (15.
or 37.7%, respectively, at 1.5 @UTSH/m1) and a consid
erable overlap (20%) between TSH levels in normal and
1254 Bayer,Kriss,andMcDougall The Journal of Nuclear Medicine
hyperthyroid patients, using the same assay reagents as
we, but the protocol supplied by the manufacturer. We
believe these differences in the assay performance can be
ascribed to our modifications of the procedure.
In accord with the negative feedback relationship be
tween serum VF4concentrations and pituitary TSH secre
tion, we also observed subnormal serum TSH levels in
patients of various other groups who had elevated serum
thyroid hormones, e.g. , in thyrotoxic patients not yet
properly controlled by antithyroid drugs, or in patients
taking excessive amounts of exogenous L-T Clinically, and particularly relevant from a diagnostic
point of view, was the observation that some patients
whose serum FF4 and T3 were within the normal range
determined for the general population, but who were din ically suspected of borderline hyperthyroidism, did in deed have suppressed serum TSH concentrations, thus
confirming the clinical impression. In one of the “euthy
roid-borderline hyperthyroid―Graves' patients studied,
the detection of a suppressed TSH preceded overt hyper
thyroidism by a few months. These fmdings provide fur
ther evidence that suppression of pituitary TSH synthesis
and/or secretion could be measured routinely as the first
biochemical change in the development of hyperthyroid
ism and underscore the clinical utility of sufficiently sen
sitive TSH assays in the early detection of hyperthyroid
ism. The good correlation found between (precisely
measured) basal serum TSH and the results of TRH-stim
ulation tests would imply that sensitive TSH measure
ments could replace some, if not all, TRH-tests in the
diagnosis of equivocal cases of hyperthyroidism, since
TRH-tests are time consuming, more stressful to be
patient and much more expensive.
Although our follow-up data on Graves' patients cur
rently in remission [Thble 3(B)] are too few to draw any
definitive conclusions, they suggest that sequential serum
TSH measurements may be useful in predicting recur
rences, or detecting recurrences before they become clii
cally overt. A marked decline in serum TSH may identify
patients who are at greatest risk of suffering a relapse,
while a rise in TSH from subnormal to normal levels
probably reflects a return to the euthyroid state.
With respect to the management ofhypothyroid patients
on thyroid replacement therapy, we found subnormal se
rum TSH in a rather large percentage (45 %) of patients
who appeared treated appropriately with L-T4 based on
their normal serum VF4and T3 (Groups 6b and 6c). Some
ofthese patients (19%, Group 6c) had subtle signs suspi
cious of slight overtreatment, but in the majority of the
cases (26%) there was no clinical evidence of thyrotoxico
sis medicamentosa.
It is difficult to explain all the TSH results, specifically
the two most extreme sets of data listed in Thble 3(C),
i.e. , in Patient 6 a subnormal serum TSH in the presence
of a borderline low serum FT@while another patient (No.
- had a normal TSH associated with a borderline high
VF4.We cannot rule out completely any hypothalamic, or
pituitary abnormalities, and we do not know with cer
tainty whether these patients complied with their pre
scribed treatment plan and whether, at the time of the TSH
measurement, the serum FT@concentrations and pituitary
TSH secretions had indeed reached a steady state, as
would be required for any meaningful interpretation of the
results. However, the fact that it was possible to normalize
TSH levels through changes in the treatment regimen in
most patients where serial measurements are available is
consistent with the concept that TSH synthesis and secre
tion by the pituitary thyrotroph is tightly regulated by relatively small fluctuations in the peripheral serum FT@ concentrations (14) (and intrapituitary FT@and T3). The serum level of VF4 which is associated with a normal
serum TSH level, may vary considerably from patient to
patient. Consequently, when only thyroid hormone levels
are monitored, a very significant number of hypothyroid
patients are actually receiving thyroid suppression therapy
instead of thyroid hormone replacement therapy, and vice
versa, some thyroid cancer patients are merely receiving
replacement therapy when pituitary TSH secretion should
be fully suppressed.
We routinely monitor our patients by serum VF4 mea
surements, while Welunann et al. (12) used serum total T
in the evaluation of their patients. Our fmding of sub
normal serum TSH levels in 45 % of hypothyroid patients
taking Synthnoid,despite normal FT@and T3values, is not
significantly different from “approximately50% “re
ported by Wehmann. Thus, serum FF@does not seem to
be a significantly better marker for the negative feedback
regulation of TSH secretion, unless binding protein ab
normalities are present. Rather, as suggested by
Wehmann, after the serum thyroid hormone levels have
been brought into the normal range, a further “finead
justment―of the treatment dose based on serum TSH
levels appears to be necessary to achieve true physiologic
and biochemical euthyroidism. It remains doubtful
whether such fine adjustments will always be mirrored by
discernable changes in the patients' symptoms or clinical
signs and whether they offer a clear benefit to the patient.
Nonetheless, we conclude that sufficiently sensitive
TSH measurements are of clinical value in the diagnosis
and management of many hyperthyroid patients, espe
cially those with equivocal disease, or those presumably in remission after therapy, and also in optimization of L
14 suppressiontherapyfor thyroid cancerpatients.
FOOTNOTES
*“Tandem..RTSH―,two- step immunoradiaometricassay by
Hybritech,SanDiego, CA.
tTSH Maiaclone―by Serono,Randolph,MA.
1―QuantimunehTSH IRMA―by Bio-RadLabs., Richmond,
CA.
Volume26 •NumberI 1 •November1985 1255
