Jan 23, 1985 - Linpisarn S, Kricka LI, Kennedy JM, Whitehead TP. Sensitive sandwich enzyme ... and Jerome F. Strauss, 11112. We evaluated a new assay ...
liver or spleen (13). In our study, we established four calibration curves with four ferritin standards from liver; the reactivity of the standards is not the same, giving very different results for patients’ serum samples. Consequently, it is necessary to establish the normal range for each method. Generally, there is a good correlation among the different methods (r >0.98), but absolute values can vary greatly (14). To standardize the results, all liver ferritin standards should be calibrated vs the National Institute for Biological Standards and Controls reference standard (15). In conclusion: the method described here is simple, fast, and presents the same precision and accuracy as the preceding (1) two-step method. The non-significant batch to batch variation (due to coating of tubes) does not affect the longterm reproducibility of the assay. We recommend use of the National Institute for Biological Standards and Controls reference standard to establish the standard curve. Furthermore, this assay procedure is less expensive than ferritin commercial kits.
References 1. Revenant MC. “Sandwich” enzyme immunoassay for serum ferritin with polypropylene test tubes as the solid phase. Clin Chem 29, 681-683 (1983). 2. Goudable J, Beaudonnet A, Revenant MC, Monnet L. Mesure de l’activite peroxydasique avec l’orthoph#{233}nyl#{232}ne diarnine comme chromogene dans les m#{233}thodes enzymoimmunologiques. Ann Bid Clin 42, 231-235 (1984). 3. Lee M, Burgett MW. A solid phase enzyme immunoassay for the quantitation of serum ferritin. Clin Chim Acta 112,241-246(1981).
4. Linpisarn S, Kricka LI, Kennedy JM, Whitehead TP. Sensitive
sandwich enzyme immunoassay for serum ferritin on microtitre plates. Ann Clin Biochem 18, 48-53 (1981). 5. Watanabe N, Niitsu Y, Ohtsuka S, et al. Enzyme immunoassay for human ferritin. Clin Chem 25, 80-82 (1979). 6. Page M, Theriault L, Nilsson M. Solid phase ELISA for serum ferritin. Scand J Clin Lab Invest 40, 641-645 (1980). 7. Ishikawa
K, Narita 0, Saito H, Kate K. Determination
of
ferritin in urine and in serum of normal adults with a sensitive enzyme immunoassay. Clin Chim Acta 123, 73-81 (1982). 8. Theriault L, Page M. A solid phase enzyme immunoassay for serum ferritin. Clin Chem 23, 2142-2144 (1977). 9. Miles LEM, Lipschitz DA, Bieber CP, Cook JD. Measurement of serum ferritin by a 2-site immunoradiometric
assay. Anal Biochem
61, 209-224 (1974). 10. Revenant MC, Beaudonnet A. Serum ferritin determination by enzyme immunoassay: Importance of sample dilution (the “hook effect”). Clin Chem 28, 253-254 (1982). Letter. 11. Ng RH, Brown BA, Valdes R. Three commercial methods for serum ferritin compared and the high-dose “hook effect” eliminated. Clin Chem 29, 1109-1113 (1983). 12. Anaokar 5, Garry PJ, Standefer JC. Solid-phase enzyme immunoassay for serum ferritin. Clin Chem 25, 1426-1431 (1979). 13. Van Oo8t BA, Willekens FLA, Van Neerbos BR, Van Den Bald B. Implications of using different tissue ferritins as antigens for ferritin in serum: Four radioimmunoassay kits compared. Clin Chem 28, 2429-2433 (1982). 14. Wood WG. A comparison of eleven commercial kits for the determination of serum ferritin levels. J Clin Chem Clin Biochem 19, 947-952 (1981). 15. lacobello C, Ghiehni 5, Belloli 5,. et al. Use of a reference standard to improve the accuracy and precision of seven kits for determination of ferritin in serum. Clin Chem 30, 298-301 (1984).
CLIN. CHEM. 31/4, 642-644 (1985)
Identificationof Hyperthyroid Patients by Means of a Sensitive Assay for Thyrotropi n James S. Mafter,1Suzanne
E. Manotti,’ Gerald R. Knee,’ David B. P. Goodman,’
We evaluated a new assay (TSH3 MAlAclone) for thyrotropin (TSH) with improved sensitivity, testing a series of hospital inpatientswith increased free thyroxin indices (thyroxin concentration x triiodothyronine uptake on resin). This assay involves use of three monoclonal antibodies and an antibody-magnetic particle conjugate that rapidly and Completely separates bound and free tracer in a magnetic field. The assay turnaround time is 3 h. By the TSH3-MAlAclone assay, 65% of these patients with an increased free thyroxin index were identified on the basis of a TSH value 0.50 milli-int. unit/L. In contrast, another commercially available assay for TSH detected suppressed TSH concentrations in less than 5% of these patients. We conclude that the TSH3 MAlAclone assay markedly improves our ability to discriminate hyperthyroidism from euthyroidism. Addftlonal Keyphrases:
thyroid status
use of magnetizablesolidphase The regulation
immunoradiometry by
cutoffvalue
of pituitary secretion of thyrotropin
(TSH)
Departments of’ Pathology and Laboratory Medicine and#{176} Obstetrics & Gynecology, Hospital of the University of Pennsylvania, 3400 Spruce St., Philadelphia, PA 19104. Received December 10, 1984; accepted January 23, 1985. 642 CLINICALCHEMISTRY, Vol. 31, No. 4, 1985
and Jerome F. Strauss, 11112
by thyroxin (T4) is well described (1 )3 TSH stimulates T4 secretion, whereas circulating T4 modulates additional TSH production via negative feedback to the pituitary gland. Given the nature of this feedback relationship, measurement of TSH in serum could provide critical information regarding the site of thyroid axis dysfunction as well as revealing the efficacy of thyroxin replacement therapy. This would be particularly useful in patients with intercurrent illness, either with suspected thyroid disease or receiving thyroid medication, because intercurrent illness often masks the signs and symptoms of thyroid dysfunction. However, until recently, routine assays for TSH were characterized by poor sensitivity at low concentrations (2). Consequently, it has not been possible to distinguish primary hyperthyroidism or secondary hypothyroidism from the euthyroid state, or to provide a reasonably rapid assessment of treatment by thyroid-hormone replacement. We have used a new, sensitive assay for TSH to study a population of inpatients in a tertiary-care medical center. In this assay, three monoclonal antibodies are directed against 3Nonstandard abbreviations: ThH, thyrotropin (thyroid-stimulating hormone); T4, thyroxin; T3, triiodothyronine; FTI, free thyrosin index; TRH, thyroliberin (thyrotropin-releasing hormone); ECS, Environmental Chemical Specialties.
independent determinants on the TSH molecule, and a unique antibody-magnetic particle conjugate permits rapid, complete separation of bound from free tracer in a magnetic field. One can detect 0.25 milli-int. unit of TSH per liter (3, 4). For comparison, we also evaluated another commercially available assay for TSH.
cal), performed according to the manufacturer’s instructions. “Serono Low” and “Sero-test” controls were obtained from Serono Diagnostics. Environmental Chemical Specialties (ECS) controls were purchased from Bio-Rad Corp., Anaheim, CA.
Results Materials and Methods We studied
65 patients attending the Hospital of the University of Pennsylvania between November 1983 and June 1984, who had free thyroxin indices (FF1) exceeding 11.8. The thyroid-function tests were repeated on more than one occasion, to confirm the increased FF1. We also measured triiodothyronine (T3) in these patients: 46% had an increased T3 as measured by RIA (>2.00 .tgfL), the rest having T3 concentrations within the normal interval. In general, the population under study had significant medical problems that necessitated hospitalization, multi-drug therapy, and in some cases surgery. Complicating medical problems were cardiopulmonary disease, malignancy, diabetes, and neurologic disorders. Medications taken by the patients included drugs known to influence the thyroidpituitary axis, the binding of T4 to serum proteins, and thyroxin metabolism-e.g., i-thyroxin, propylthiouracil, phenytoin, furosemide, and amiodarone (1, 5). The euthyroid group comprised 18 healthy individuals who were participating in a study of thyroliberin (TRH) stimulation testing. None had a history of thyroid disease, and none was using any medications known to affect the thyroid or thyroid-function tests. Each euthyroid control had TSH values determined just before and 30, 60, and 180 mm after administration of 400 pg of TRH as an intravenous bolus. All of these individuals had basal values for FF1 within the normal interval ( ± SD = 7.0 ± 1.85, normal 3.8-11.8) and T3 by RIA in the normal interval (0.85 ± 0.29, normal 0.60-2.00 pg/L), and all responded appropriately to T.RH stimulation with respect to TSH secretion. Blood samples from each participant were collected into evacuated specimen-collection tubes (Vacutainer Tubes, Becton Dickinson) and allowed to clot. The serum, obtained by centrifugation, was stored at 4#{176}C until assayed for T4, T3resin uptake, and T3 by RIA with commercially available methods (Corning Medical, Medfleld, MA). Specimens for TSH determination were stored at -20 #{176}C for no longer than three months before measurement. The TSH3-MAlAclone assay (Serono Diagnostics, Braintree, MA) was provided by the manufacturer and has been described elsewhere (3, 4). Briefly: an aliquot of serum or standard is incubated for 2 h with anti-TSH reagent (a mixture of two radioiodinated monoclonal antibodies to TSH), then mixed with anti-TSH-magnetic particle conjugate. An additional 5-mm incubation is then followed by separation and washing. We counted radioactivity for 1 mm in a Micromedic System 4/200 automated gamma counter equipped with a Hewlett-Packard HP200 microprocessor for data reduction. Standard curves were run for each assay, from TSH standards provided by Serono (0.25, 0.5, 1, 5, 10, and 30 milli-int. units/L). The 0.5 milli-int. unit/L standard was diluted with an equal volume of zero standard to produce the 0.25 milli-int. unitlL standard. Patients’ specimens and controls were at first run in quadruplicate, but later in duplicate. Any serum containing TSH greater than 50 milli-int. units/L was diluted and reassayed. Using linear regression analysis, we compared results from the TSH3-MAIAclone assay with data obtained by another commercially available assay for TSH, based on polyclonal anti-TSH antibodies (Immophase; Corning Medi-
By using log-log data reduction, we found that a linear relationship between boundltotal radioactivity vs TSH concentration was maintained over the clinically relevant range of 0.25 to 50 milli-int. unitsfL. Standard curves and controls were run on eight separate occasions over a 30-day period to evaluate the precision and reproducibility of individual points for the controls run in quadruplicate. The inter-assay imprecision (mean, SD, and CV) was: Serono Low, 0.89 ± 0.10 milli-int. unitlL, 11.7%; Sero-test, 4.54 ± 0.31 milli-int. unitsfL, 4.9%; ECS 2, 11.0 ± 0.24 milli-int. units/L, 2.2%; and ECS 3, 30.2 ± 1.10 milli-int. units/L, 3.6%. We determined the cross reactivity of the antibodies with follitropin, lutropin, and choriogonadotropin by assaying TSH in sera from clinically euthyroid individuals having known increased concentrations of these hormones and normal
values
for the FF1. Cross reactivity
was 4.0
increased FF1 values. Most of these patients were on multidrug regimens and were hospitalized with severe illness. The TSH3-MAlAclone assay identified two-thirds of these patients as being hyperthyroid by detecting suppressed TSH concentrations; by comparison the PSH Immophase assay identified less than 5% by this criterion. Although two-thirds of the patients we investigated had suppressed TSH concentrations as measured with the PSH3MAlAclone assay, a third of them had TSH values within the normal range. T3 as measured by RIA in most of the latter patients (22 of 23) were either normal or within 0.11 pg/L of the upper limit of normal. Normal T3 concentrations associated with increased FTIs and normal TSH concentrations have been described in patients with severe systemic illness. This is thought to be the result of impaired peripheral conversion of T4 to T (8). Hence it is likely that some of the apparent discrepancies between TSH values and FFIs are a consequence of the complex and severe nature of the disease processes (predominantly cardiopulmonary and neurologic) in this subgroup of patients.
TSH
IfJIU/L Fig. 1. Companson of TSH values for patients with increased FTIs (FTI 1 1.8) determined with the TSH3-MAIAclone and TSH Immophase assays Doftedlinedelineateslower limitof normal for the respectiveassays.Abscissa units: milli-int units/L medications had appropriately low serum ‘NH concentra-
tions in conjunction with an increased
FF1.
Discussion Radioimmunoassays for T4, T3, and TSH have greatly simplified and improved the diagnosis of thyroid disorders. However, diagnosis of mild primary hyperthyroidisni is still difficult. Routinely used TSH assays are not sufficiently sensitive to discriminate depressed TSH values from those found in euthyroid individuals (6). The TSH3-MAlAclone assay reported here is sensitive to at least 0.25 milli-int. unitlL, with an acceptable CV, allowing the identification of primary hyperthyroidism (3,4, 7). Linearity of the assay is maintained to 50 milli-int. unita/L, covering the clinically significant range. Additionally, the TSH3 MAlAclone assay correlates highly significantly with results by the Immophase TSH assay for TSH values in the moderately abovenormal range, and the assay requires only 3 h. Of particular interest is the performance of the PSI!3MAlAclone assay with the sera of the 65 inpatients with
644
CLINICAL CHEMISTRY, Vol. 31, No. 4, 1985
We gratefully acknowledge Serono Diagnostics, Braintree, MA, for providing reagents. We thank Ms. Janet Brennar for her help in preparation of this report. References 1. Ingbar SH, Woeber KA. The thyroid gland. In Texibook Endocrinology.
RH Williams,
Ed., WB Saunders
of
and Co., Philadel-
phia, PA, 1981, pp 117-247. 2. Wide L, Dahlberg PA. Quality requirements of basal s-TSH assays in predicting an s-TSH response to TRH. Scand J Clin Lab Invest 40, Suppl 155, 101-110 (1980). 3. Rattle S,J, Purnell DR. Williams PIM, et al. New separation method for monoclonal immunoradiometric assays and its application to assays for thyrotropm and human choriogonadotropin. Cliii Chem 30, 1457-1461 (1984). 4. Cobb WE, Lamberton RP, Jackson IMD. Use of a rapid, sensitive immunoradioactive assay for thyrotropin to distinguish normal from hyperthyroid subjects. Clin Chem 30, 1558-1560 (1984). 5. Martino E, Safran M, Aghini-Lombardi F, et al. Environmental iodine intake and thyroid dysfunction during chronic amiodarone therapy. Ann Intern Med 101, 28-34 (1984). 6. Spencer CA, Nicoloff JT. Improved radioimmunoassay for human TSH. Clin Chim Acts 108, 415-424 (1980). 7. Allen KR, Watson D. Thyrotropin as the initial screening test for thyroid disease. Clin Chem 30, 502-503 (1984). 8. Cavalieri RR, Rapoport B. Impaired peripheral conversion of thyroxine to triiodothyronine. Ann Rev Med 28, 57-65 (1977). Review.
