Prévia do material em texto

<p>See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/12526496</p><p>Evaluation of the LIAISON® thyroid chemiluminescence immunoassays</p><p>Article  in  Clinical Laboratory · January 2000</p><p>Source: PubMed</p><p>CITATIONS</p><p>7</p><p>READS</p><p>2,133</p><p>7 authors, including:</p><p>Thomas Demant</p><p>Hospital Dresden-Friedrichstadt</p><p>54 PUBLICATIONS   2,530 CITATIONS</p><p>SEE PROFILE</p><p>All content following this page was uploaded by Rudolf Hoermann on 21 November 2017.</p><p>The user has requested enhancement of the downloaded file.</p><p>https://www.researchgate.net/publication/12526496_Evaluation_of_the_LIAISONR_thyroid_chemiluminescence_immunoassays?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_2&_esc=publicationCoverPdf</p><p>https://www.researchgate.net/publication/12526496_Evaluation_of_the_LIAISONR_thyroid_chemiluminescence_immunoassays?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_3&_esc=publicationCoverPdf</p><p>https://www.researchgate.net/?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_1&_esc=publicationCoverPdf</p><p>https://www.researchgate.net/profile/Thomas-Demant-2?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_4&_esc=publicationCoverPdf</p><p>https://www.researchgate.net/profile/Thomas-Demant-2?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_5&_esc=publicationCoverPdf</p><p>https://www.researchgate.net/institution/Hospital-Dresden-Friedrichstadt?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_6&_esc=publicationCoverPdf</p><p>https://www.researchgate.net/profile/Thomas-Demant-2?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_7&_esc=publicationCoverPdf</p><p>https://www.researchgate.net/profile/Rudolf_Hoermann?enrichId=rgreq-56a4ea254e5860772be80aade906d814-XXX&enrichSource=Y292ZXJQYWdlOzEyNTI2NDk2O0FTOjU2MzAxMjUwMDIzNDI0MEAxNTExMjQ0MDQ4MTcy&el=1_x_10&_esc=publicationCoverPdf</p><p>Clin. Lab. 2000;46: I 81- I 89</p><p>OCopyright</p><p>ORIGINAL ARTICLE</p><p>Evaluation of the LIAISON@ Thyroid Chemiluminescence</p><p>Immunoassays</p><p>WALTER HUBLI, DIETER MEIBNER', THOMAS DEMANTI, WOLFGANG BECKER2,</p><p>RUDOLPH HORMANN3, MARION BACH4 AND MICHAEL MACK4</p><p>ttttstitut</p><p>fiir Kti'ist'he Chenie wtd</p><p>f::;i::,::,";;;i:::i*i;':;;ii,:li:,"1:'t;,x:;:x:';::,':::;ichstu'Itt'</p><p>Dresden' Gernru^</p><p>r Klirtische Endokrinologie IJnitersitiitsklinikLun Essen, Essen, Gernnnt'</p><p>t Blk-Sangtec Diugnostica, Diet:enbach, Gennutt'</p><p>SUMMARY</p><p>The LIAISON@ thyroid hormone assays TSH, FT4, FT3, T4 and T3 were evaluated by determining the impre-</p><p>cision, the reference ranges, the functional sensitivity (TSH), the dilution characteristics (accuracy) (FT4, FT3),</p><p>and the recovery after spiking (TSH, T4, T3). Furthermore, inter-method comparisons were perfomed with</p><p>following methods: Elecsys (Roche Diagnostics; TSH), AxSYM (Abbott Diagnostics; TSH, FT4, FT3, T4 ),</p><p>ACS:180 (Bayer Diagnostics; all analytes), Amerlex-M (Johnson & Johnson; T4) and LISO-Phase (Techno</p><p>Genetics; FT4). The fully automated LIAISON@ random access analyser is based on microparticle immunoassays</p><p>and chemiluminescence. The coefficients of variation (CV) of intra-assay imprecision lvere between 0.2 - 6.0Vo,</p><p>except for the control sample with extremely low TSH concentrations and low T3 concentrations. Inter-assay</p><p>imprecision was performed by measuring controls covering the measuring range over a period of 9 to 20 days,</p><p>with CVs ranging from 2.3 - 16.0%o. The suitability of the sample material was determined by analysing serum</p><p>and samples treated with EDTA, citrate or heparin in parallel. The results showed good correlations of the</p><p>thyroid hormone concentrations between serum and plasma samples except for LIAISON@ FT3, for which lower</p><p>results were observed with EDTA-plasma. The regression analysis of correlation studies gave slopes from 0.849 to</p><p>0.957 for TSH, from 1.023 to 1.375 for FT4, from 0.670 to 0.911 for FT3, from 0,917 to 1.166 for T4 and 1.00 for</p><p>T3 depending on the concentration range and the method of comparison. The LIAISON@ FT4 assay showed a</p><p>trend towards higher values in the high concentration range when compared with the ACS:180. The ranges of</p><p>thyroid hormone concentrations determined in serum taken from apparently healthy subjects were found to be in</p><p>accordance with published data. The clinical sample study confirmed that the LIAISON@ thyroid hormone assays</p><p>are sensitive methods for the differentiation of euthyroid subjects and patients with hyper- and hypothyroidism.</p><p>In conclusion, the automated thyroid hormone immunoassays on the random-access LIAISON@ immunoassay</p><p>analyser proved to be very satisfactory, both from the analytical and the clinical point of view. (Clin. Lab.</p><p>2000;46:181-189)</p><p>KEY WORDS</p><p>Chemiluminescence; immunoassay; thyroid; diagnostic</p><p>tests;LIAISONo.</p><p>the LIAISON@ immunoassay analyser. The essential</p><p>objectives of the study were imprecision, functional</p><p>sensitivity of the LIAISON@ TSH assay, dilution cha-</p><p>racteristics (accuracy), recovery after spiking, reference</p><p>ranges and inter-method comparisons with different</p><p>kinds of well established commercial immunoassays for</p><p>thyroid hormones. Clinically relevant samples were</p><p>chosen to compare the concordance between diagnosis</p><p>INTRODUCTION</p><p>The aim of the study was to evaluate the new LIAI-</p><p>SON@ thyroid hormone assays designed to be used on</p><p>Clin. Lab. 3+4/2000 l8l</p><p>Illanuscript accepted December 29, I 999</p><p>W. HUBL et al.</p><p>and measured thyroid hormone concentrations in treated</p><p>and untreated hypo- and hyperthyroid patients, as well</p><p>as in patients with no evidence of thyroid disease who</p><p>acted as euthyroid controls.</p><p>MATERIAL AND METHODS</p><p>Material</p><p>Instrument</p><p>The LIAISONo immunoassay analyser (Byk-Sangtec</p><p>Diagnostica, Dietzenbach, Germany) is a fully automat-</p><p>ed random access analyser. The instrument is based on</p><p>the magnetic particle technology (Dynabeads!; for sep-</p><p>aration and isoluminol chemiluminescence for deter-</p><p>mination of the immunocomplex. Serum samples are</p><p>incubated with coated Dynabeads'u' fbr 10 min priol to</p><p>washing steps. The isoluminol derivative shows flash-</p><p>light kinetics after injection of two ready-to-use trigger</p><p>solutions. The light emission is measured for 3 s. Sam-</p><p>ple concentrations are calculated using a stored master</p><p>curve. First results can be obtained in approx. 15 min</p><p>for one-step assays and in 30 min for two-step assays.</p><p>The reagents are stored in a temperature-controlled area</p><p>and can be continuously reloaded. The avelage through-</p><p>put is 100-140 tests/hour. The user intertace is a Win-</p><p>dows-based software with both touch-screen and key-</p><p>board operation. The data reduction is based on a master</p><p>curve with a two-point re-calibration method.</p><p>Reagents</p><p>ZSII: TSH is measured fiom a 200 prl serum santple us-</p><p>ing a one-step anti-TSH B-chain immunoassay. The as-</p><p>say covers a concentration range from 0.004 to 100</p><p>mIU/I.</p><p>FTl: The competitive FT4 assay works according to the</p><p>SPALT principle using an isoluminol labelled mono-</p><p>clonal antibody as tracer. Immune extraction of labelled</p><p>antibodies not bound by endogenous antigen from sam-</p><p>ples is performed using an excess of covalently immo-</p><p>bilised T4 antigen on the solid phase. The assay is per-</p><p>formed by sirnultaneous incubation of a 50 prl sample</p><p>and 200 prl tracer solution and 20 pl magnetic particle</p><p>suspension. The assay has a measuring range fi'orr 0. I</p><p>ng/dl to l0 ng/dl.</p><p>FT3: The competitive FT3 assay works according to the</p><p>SPALT principle using labelled antibodies as tracer and</p><p>covalently bound antigen. High affinity polyclonal anti-</p><p>body</p><p>against T3 is labelled with an activated isoluminol</p><p>derivative. The assay protocol follows a sequential in-</p><p>cubation of reagents. First. the 100 pl sample is incubat-</p><p>ed with labelled antibody solution and then with anti-</p><p>gen-coated magnetic particles. Each incubation step</p><p>takes l0 min. The time to first results is less than 25</p><p>min. The official working ran-qe of the assay is l-20</p><p>pg/ml. In the scope of this evaluation the assay was per-</p><p>formed with an extended workin_e range of 0.001-25</p><p>pg/ml, to investigate the imprecision at very low FT3</p><p>concentrations.</p><p>71, T3: Fol the T4 and T3 cornpetitive assays hi-ch affi-</p><p>nity polyclonal antibodies are covalently linked to r"rni-</p><p>form parama-enetic particles. The antigens are linked via</p><p>a spacer to an activated isoluminol derivative. Antigen</p><p>bound to endo_eenous bindin-s proteins in samples is</p><p>released by a blocker substance. The assays are per-</p><p>fbrmed by simultaneous incubation of a 50 pl (T4) or</p><p>100 pl (T3) sample with 200 prl tracer solution and 20</p><p>pl antibody-coated ma,enetic particles. Reagents are in-</p><p>cubated for l0 min (T4) and 20 min (T3), respectively.</p><p>First results can be expected in less than 15 rnin (T4)</p><p>and 25 min (T3), respectively. The T4 assay covers a</p><p>measuring range from 0.25 to 40 pg/dl and the T3 assay</p><p>from l5 to 800 ng/dl.</p><p>All rea-eents are provided ready-to-use. including the</p><p>two calibratols. The light emission is generated by the</p><p>injection of tr.vo ready-to-use tri-eger solutions with a</p><p>measurin-t tirne of 3 s.</p><p>Specimens</p><p>For the clinical evaluation. rvell characterised speci-</p><p>rnens from different proband ,qrollps were collected. in-</p><p>cludin,s euthyroid subjects. euthyroid pre-snant women</p><p>(3'.'r trirnester). patients with hyper'- and with hypo-</p><p>thyroidism as well as patients suffering frorn Graves'</p><p>disease.</p><p>Methods</p><p>hnprecision was determined by lO-fold replicates of 9</p><p>coutrols covering the whole measuring range. Measure-</p><p>r.nents were perfbrmed at least on 3 different days with</p><p>two different lots.</p><p>Tlre functional sensitivity of TSH was determined on</p><p>serum pools with initial TSH concentrations between</p><p>0.08 and 0.82 rnIU/I. These pools were diluted step-</p><p>rvise with a "TSH-free" huruan serum. Dilution series</p><p>were rneasured over five consecutivc asstys.</p><p>Recovery experiments were perforr.ned with the LIAI-</p><p>SON TSH and the LIAISON'' T4 assay by using na-</p><p>tive sin_ele donor samples (analyte conc.: TSH: 0.45 -</p><p>9.6 mIU/I. T4: 1.3 - 7.8 pg/dl.). The analyte concentra-</p><p>tions of the samples were determined before and after</p><p>spiking with ar.r antigen stock solution. Recovery after</p><p>spikin_e was defined by the pelcenta-ee of the measured</p><p>concentration in relation to the theoretically expected</p><p>value.</p><p>The accutacy of the FT3 and FT4 assays was investi-</p><p>gated by exarnining the dilution characteristics of these</p><p>assays. Serum samples rvere diluted step-wise with 0. I</p><p>rnol/l hepes buffer (FT3) or 0.01 mol/l PBS br,rffer</p><p>(FT4). respectively. The highest dilution quotient was</p><p>182 Clin. Lab. 3+4/2000</p><p>30</p><p>o</p><p>120</p><p>?o</p><p>?ls</p><p>G</p><p>*10</p><p>5</p><p>0</p><p>30</p><p>.20</p><p>o</p><p>G</p><p>d lu</p><p>0</p><p>0,1 1.0 10,0</p><p>TSH concentraiion (m ltyl)</p><p>1</p><p>FT4 concentration (n9/dl)</p><p>0 1 2 3 4 5 6 7 I I 101112</p><p>FT3 concentration (pg/m l)</p><p>30</p><p>25</p><p>o</p><p>irs</p><p>o</p><p>*10</p><p>5</p><p>0</p><p>;20o</p><p>5 '10 15 20 25 30</p><p>T4 concentration (Fg/dl)</p><p>0</p><p>EVALUATION OF THE LIAISON] THYROID CHEMILUMINESCENCE IMMUNOASSAYS</p><p>Figure 1: Intra-Assay precision profiles of serum samples determined in duplicates</p><p>0.2 for FT3 and 0.01 for FT4. Analyte concerltrations</p><p>were fileasured in duplicate.</p><p>For the investigation of suitable sample material, paired</p><p>samples of serum and EDTA plasma. citrated plasma or</p><p>heparinised plasma were prepared in parallel from one</p><p>blood sample and measured in direct compadson. The</p><p>concentration values obtained for citrated plasma were</p><p>corrected according to the respective predilution factor</p><p>for plasma preparation.</p><p>RESULTS</p><p>Intra-assay imprecision</p><p>Intra-assay imprecision was evaluated by duplicate de-</p><p>termination of native serum samples. The results are</p><p>summarized in precision profiles shown in Figure l.</p><p>The ratio of sera determined with an intra-assay CV</p><p>4 pg/rrl</p><p>showed slightly increased values upon the first dilution</p><p>step (volume fraction 0.8). Upon further dilution, nearly</p><p>constant concentration levels were observed down to a</p><p>sample volume fraction of 0.4. At a sample volume</p><p>Figure 2:</p><p>Functional sensitivity</p><p>LIAISON@ TSH</p><p>fraction of 0.2 a slight decrease of measured concentra-</p><p>tion was observed, leading to FT3 values close to the</p><p>initial concentration of the respective sample.</p><p>However, these data document that both the LIAISON@</p><p>FT3 assay and the LIAISON@ FT4 assay meet the re-</p><p>quirements for the correct determination of free thyro-</p><p>xine and free triiodothyronine, respectively.</p><p>60</p><p>aa</p><p>A</p><p>LLab2</p><p>.Lab1</p><p>I I</p><p>A</p><p>aa</p><p>A</p><p>aao\</p><p>t</p><p>d40</p><p>ooo</p><p>G</p><p>I</p><p>o</p><p>c</p><p>20</p><p>0,00 0,02 0,04 0,06 0,08 0,10 0,12 0,14 0,16 0,18 0,20 0,22 0,24</p><p>Mean TSH concentration (mlM)</p><p>184 Clin. Lab. 3+412000</p><p>LIAISON FT3</p><p>0,8 0.6 0.4</p><p>Volume fraction of the sample</p><p>LIAISON FT4</p><p>5,0</p><p>e</p><p>E +,0</p><p>E</p><p>oa</p><p>C^</p><p>3Es'o</p><p>oOFE</p><p>llv</p><p>E 2,0.=</p><p>-rgI 3 1.0a=</p><p>a</p><p>a</p><p>a</p><p>a</p><p>a</p><p>a</p><p>t</p><p>a</p><p>t</p><p>t</p><p>I</p><p>a</p><p>a</p><p>a</p><p>a</p><p>1,8</p><p>I,O</p><p>c</p><p>€ 1,4</p><p>6</p><p>E I,Z</p><p>o</p><p>oi = 1,0</p><p>XE</p><p>TPO,8</p><p>E 0,6</p><p>o nLG -' 'o</p><p>= 0,2</p><p>0,0</p><p>a</p><p>a</p><p>o</p><p>0,0</p><p>0,333 0,111 0,037</p><p>Volume fraction otthe sample</p><p>EVALUATION OF THE LIAISON- THYROID CHEMILUMINESCENCE IMMUNOASSAYS</p><p>Figure 3: Dilution characteristics of LIAISON@ FT4 and LIAISON@ FT3</p><p>Figure 4: Recovery after spiking - LIAISON@ TSH, T4 and T3</p><p>Recovery experiments</p><p>Recovery was investigated on the basis of 12 patient</p><p>sera</p><p>with TSH concentrations between 0.45 and 9.6</p><p>mIU/I. For 10 samples, recoveries between 95 and</p><p>1087o were observed. In 2 cases. recoveries were found</p><p>to be 71c/o and797o, respectively. A dependence of ini-</p><p>tial TSH concentrations on the samples and the results</p><p>ofrecovery was not observed.</p><p>The recovery of T4 was investigated on the basis of</p><p>patient sera with T4 concentrations between 1.3 and 7.8</p><p>prg/dl. For 7 samples, recoveries between 90 and lll7o</p><p>were observed. ln 2 cases, recoveries were found to be</p><p>837o and87Vo.</p><p>The recovery of T3 was investigated on the basis of</p><p>patient sera with T3 concentrations between 45 and 168</p><p>ng/dl. The recoveries of T3 were between 97Vo and</p><p>109Vo.</p><p>In a further step, the spiked serum samples were diluted</p><p>with the respective original (i.e. unspiked) sera. The</p><p>TSH, T4 and T3 concentrations were re-calculated by</p><p>multiplication of the obtained values by the respective</p><p>Clin. Lab. 3+4/2000 1 85</p><p>dilution factor and were found to be linear upon dilution</p><p>(Figure 4).</p><p>Inter-method comparisons</p><p>Each laboratory correlated the LIAISONo thyroid re-</p><p>sults with its own routine methods. The data of regres-</p><p>sion analysis are shown in Table 2.</p><p>Regression analysis of correlation studies gave slopes</p><p>from 0.849 to 0.957 for TSH and intercepts from 0.030</p><p>to 0.134.</p><p>The LIAISON@ FT4 assay showed a trend towards</p><p>higher values in the high concentration range, especially</p><p>in comparison with the ACS:180 assay. However, at</p><p>FT4 concentrations 3 pglml the LIAISON@</p><p>FT3 assay showed a trend towards lower values (lead-</p><p>ing to a slope of 0.7). The concordance of the measuring</p><p>levels between the LIAISOND FT3 and the ACS:180</p><p>FT3 assay was found to be better.</p><p>The LIAISON@ T4 assay showed a slight trend towards</p><p>higher values compared with the ACS:180 T4 assay.</p><p>r86 Clin. Lab. 3+412000</p><p>EVALUATION OF THE LIAISON€ THYROID CHEMILUMINESCENCE IMMUNOASSAYS</p><p>100,00</p><p>10,00</p><p>1,00</p><p>0,10</p><p>0,01</p><p>0,00</p><p>o Labl</p><p>. Lab2</p><p>il il $l ; i-------.-----r- . * t.:.)</p><p>*</p><p>a^</p><p>a ',,.l</p><p>12,0</p><p>11,0</p><p>10,0</p><p>9.0</p><p>2.O</p><p>1.0</p><p>E</p><p>E,e</p><p>E6</p><p>c</p><p>o</p><p>o</p><p>FL</p><p>E</p><p>L</p><p>.9</p><p>(lt</p><p>o,oc</p><p>oo</p><p>Ia</p><p>F</p><p>8,0</p><p>7.O</p><p>6.0</p><p>5,0</p><p>4.0</p><p>3.0</p><p>t</p><p>= 6,5</p><p>e 6,0</p><p>the non-parametric method</p><p>for the calculation ofthe median and percentiles.</p><p>Clinical evaluation</p><p>The clinical application of the LIAISON@ thyroid hor-</p><p>mone assays was investigated by their ability to diffe-</p><p>rentiate between euthyroid patients and patients with</p><p>thyroid disease. Figures 5 to 9 show the distribution</p><p>patterns of serum TSH, FT4, FT3, T4 and T3 levels</p><p>measured with the LIAISON@ thyroid hormone assays</p><p>in patients with different kinds of thyroid diseases.</p><p>Lower TSH values were observed in sera from hyper-</p><p>thyroid patients and in contrast, TSH levels within the</p><p>reference range were found in sera from autonomous</p><p>o</p><p>rl!a</p><p>Y/a</p><p>-- n- - - -6- - - -. - -- - - - - - -. ----'r-ili i+</p><p>11</p><p>o</p><p>.g</p><p>o</p><p>9.</p><p>a8r</p><p>a^.sf.</p><p>.las--------*-----ao .E&oo</p><p>1...o</p><p>foo</p><p>-a--- -------- ------ --- ----- ----^o----</p><p>"*ar,</p><p>:; :l</p><p>$r</p><p>W. HUBL et al.</p><p>=E</p><p>E</p><p>c 15</p><p>oz</p><p>G</p><p>c</p><p>c ,v</p><p>o</p><p>F</p><p>3</p><p>It buu</p><p>5 500</p><p>E</p><p>I aoo</p><p>oe</p><p>5 s00</p><p>e</p><p>P zoo</p><p>I = healthy f'emale adults (n=48); 2 = healthy male adults (n=39); -3 =</p><p>pregnant women 3'd trimester (n=23); 4 = autonomous thyroid nodules</p><p>rvith eu- or hyperthyroidism (n=45): 5 = hyperthyroidism under</p><p>therapy (n=8); 6 = hypothyroidism (n=36)</p><p>Figure 8: LIAISON@ T4 values in healthy adults and</p><p>patients with thyroid diseases</p><p>Table 4: Reference ranges for the LIAISON@ thyroid</p><p>hormones</p><p>Analyte</p><p>TSH (mIU/l)</p><p>FT4 (ng/dl)</p><p>FT3 (pg/ml)</p><p>Ta (pddl)</p><p>T3 (n9dl)</p><p>n Mean 2.S-Voile 97.5-7oile</p><p>87 t.27 0..15 2.46</p><p>87 l.l8 0.82 l.7t</p><p>u7 2.17 t.72 3.-15</p><p>87 9.0 5.s t2.l</p><p>19 132 102 187</p><p>NIin-Max</p><p>0.2 - 3.3</p><p>0.7 - 2.3</p><p>l.+-+.1</p><p>.1.-+ - l.{.0</p><p>97 - t95</p><p>thyroid nodules with euthyroidism. All sera of hypo-</p><p>thyroid patients revealed TSH values significantly</p><p>above the euthyloid reference range.</p><p>FT4 and FT3 values in pregnant women (3'd trimester;</p><p>were found at the low end or slightly lower than the</p><p>values observed for the group of healthy women. Ele-</p><p>vated FT4 and FT3 values were observed in the majol-</p><p>ity of sera from hyperthyroid patients and patients with</p><p>Graves' disease. On the other hand, the majority of sera</p><p>taken from hypothyroid patients revealed FT4 and FT3</p><p>values that were significantly lower than the reference</p><p>range. In some cases, however, the FT4 and FT3 values</p><p>of both groups were within the reference range. FT4</p><p>and FT3 concentrations within the reference range were</p><p>observed in serum samples from patients with hepatic</p><p>diseases and diabetes mellitus. As expected, patients</p><p>with congenital abnormalities of the thyroxine binding</p><p>globulin showed normal FT4 values. Samples from</p><p>women taking oral contraceptives also showed FT4 and</p><p>FT3 values within the reference range. Patients with</p><p>severe nonthyroidal illness (NTI) showed in agreement</p><p>with Meinhold et al. (7) decreased FT4 and FT3 values.</p><p>Postoperative measurements during severe sepsis and</p><p>I = healthy female adults (n=19): 2 = hcalthy male adults (n=30): 3 =</p><p>pregnant rvomen 3'd trimester (n=20): 4 = autonomous thyroid nodules</p><p>rvith eu- or hyperthyroidism (n=21): j = hypothyroidism (n=23): 6 =</p><p>Craves' disease (n=20)</p><p>Figure 9: LIAISON@ T3 values in healthy adults and</p><p>patients with thyroid diseases</p><p>recovery revealed FT4 and FT3 values below. within</p><p>and also above the reference range.</p><p>The T4 and T3 values of pregnant women were found to</p><p>be similar to the values observed fbr the group of</p><p>healthy subjects. Elevated T4 and T3 values were obser-</p><p>ved in sera from hyperthyroid patients. As expected.</p><p>patients with latent hyperthyroidism or latent hypo-</p><p>thyroidism showed normal T3 and T4 values with de-</p><p>creased and increased TSH concentrations. respectively.</p><p>DISCUSSION</p><p>Automation of thyroid immunoassays is an important</p><p>goal for the clinical laboratoly. High analytical sensiti-</p><p>vity. speed and flexibility are required (1. 2). A new</p><p>diagnostic approach with an analyte detection based on</p><p>the magnetic particle technology and isoluminol chemi-</p><p>luminescence was recently introduced by Byk-Sangtec</p><p>Diagnostica and has used the technology for develop-</p><p>ment of the LIAISON5 system (3). In the present study,</p><p>we assessed the analytical performance and the clinical</p><p>usefulness of the fully automated LIAISON@ thyroid</p><p>hormone assays.</p><p>The results of intra-assay and inter-assay imprecision on</p><p>the LIAISON@ thyroid assays were highly acceptable</p><p>and in agreement with other thyroid hormone assay sys-</p><p>tems (8). Excellent results were obtained for concentra-</p><p>tions within the reference ranges for TSH, FT3, FT4</p><p>and T4.</p><p>The actual lorver detection limit. as defined by the func-</p><p>tional sensitivity (4. 5, 6) is a more convenient approach</p><p>to the problem than the analytical sensitivity. Analytical</p><p>sensitivity reflects an over-optimistic estimate of sensi-</p><p>tivity since it is based on the intra-assay imprecision of</p><p>t</p><p>a</p><p>a</p><p>!</p><p>a</p><p>o</p><p>a</p><p>.iiili a</p><p>188 Clin. Lab. 3+4/2000</p><p>EVALUATION OF THE LIAISON. THYROID CHEMILUMINESCENCE IMMUNOASSAYS</p><p>the zero calibrator. To date, the most important criterion</p><p>proposed for the assessment of third-generation TSH</p><p>assays demands that the functional sensitivity at</p><p>third--ueneration assay (Amerlite</p><p>TSH-30) of thyroid-stirnulating hormone in serurn or plasma</p><p>assessed. Clin Chem 1993: 39: 2166-7 3.</p><p>Spencer CA. Takeuchi M. Kazarosl,an M: Curcnt status and</p><p>pertbrmance goals for serum thyrotropin (TSH) irssays. Clin</p><p>Chem l996:42: 140-5.</p><p>Spencer CA. LoPresti JS. Patel A. Guttler RB. Eigen A. She n.D:</p><p>Application of a neu, chemiluminometric thyrottopin assa), to</p><p>sutrnornral mcasurement. J Clin Enclocrinol Metab 1990: 70:</p><p>4-53-60.</p><p>Meinhold H. Gramr.n HJ. Finke R: Potcntial of new third-</p><p>generation thl,rotropin assirys tbr routine rneasurement of sub-</p><p>normal serum TSH lcvels in thyroicl diseases and ser,erc non-</p><p>thyroidal illnesses. J Lab Mcd 1999:23: 281-287.</p><p>Grisslcr N: Pertorrnancc of thc Vitros ECI analyze r - cornpari-</p><p>son with other immunoassay syslems for deterrnination of</p><p>thyroid parameters. Clin Lab 19991 45: 151 - 161.</p><p>Bartalena L. Bo_sazzi F. Bro-qioni S. Burelli A. Scarcello G and</p><p>Martino E: Measurenrent of serum fiee thyroid hornrone concen-</p><p>trations: an essential t</p>