Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
CLIN. CHEM. 36/5, 800-804 (1990) New Ultrafiltration Method for Free Thyroxin Compared with Thyroid Dysfunction and Nonthyroidal Illness SarI Dialysis This new ultrafiltration method for free thyroxin in serum (FT4(U)] is based on radioimmunoassay of the free hormone fraction in ultrafiltrates obtained by centrifuging serum sampIes in Unisep’ Ultracent-1 0 ultrafiltration devices. We compared the results obtained with those by an equilibrium dialysis method [FT4(D)]. In 36 euthyroid healthy subjects, the mean FT4(U) concentration was 24.2 pmol/L and the mean FT4(D) concentration 14.8 pmol/L. In hyperthyroid and hypothyroid patients, results by the ultrafiltration method were also approximately twice as high as those obtained by the dialysis method. In 23 patients with various nonthyroidal illnesses, mean FT4(U) was 41.2 pmoVL and mean FT4(D) 19.8 pmol/L. The mean FT4(U)/FT4(D) ratio in patients with nonthyroidal illnesses (1.97) was not significantly higher than in control subjects (1.68), making it unlikely that the increase in serum FT4 is caused by weakly protein-bound and therefore dialyzable inhibitors of thyroxin binding to carrier proteins. However, two nonthyroidally ill patients with a clearly increased FT4(U) but a normal FT4(D) concentration might have had such inhibitors, whereas for two other nonthyroidaily ill patients a high molar ratio of free fatty acids to albumin is a more likely explanation for increased FT4(U) and FT4(D) concentrations. On theoretical grounds, we consider the FT4(U) concentrations analytically more nearly accurate than FT4(D) values for all patient groups studied. Keyphrases: variation, source radioimmunoassay of When one measures the free (unbound) concentration of an analyte in serum, the analytical conditions must not alter the balance between the bound and the unbound fraction of the analyte. In measuring free thyroxin (Fr4) in serum, methods in which the free fraction is first separated from the serum carrier proteins by equilibrium dialysis or ultrafiltration and thereafter quantified are therefore theoretically sound.5 In assays of FF4 in serum where physiological T4-carrier proteins or anti-P4 autoantibodies are present along with reagent anti-T4 antibody, as in one-step thyroxin-analog tracer assays, accurate determination is possible only if the radioactive tracer is not significantly associated with the endogenous binders. To date, no such analog tracer has been synthesized; therefore, the present generation of analog-based assays of FF,5 give spurious results, to a variable extent, in subjects with quantitati or qualitative abnormalities of T4-binding proteins (1-6) Two-step rapid radioiinmunoassays of VF4 are based o immunoextraction of the free fraction by immobilized antiantibody in the first step and back-titration of unoccupi hormone-binding sites in the second step (2, 6, 7). Th assays, although less prone than the one-step analog assays errors induced by altered T4-protein binding in serum, still not fulfill the requirement that the free hormone equilibri must not be changed, because the extracted hormone fractio much exceeds the true free hormone fraction. F’1 assa involving the use of T4-immunoglobulin-acridinium ester T4-enzyme conjugates (8-11) offer several advantages ov radioimmunoassays, but are also potentially prone to erro caused by protein-binding of conjugates, particularly bin to T4-autoantibodies (10). Our objective in the present study was to measure by new ultrafiltration method the Fr4 concentration [VF4(U in sera from patients with various forms of thyroid dysftm tion and nonthyroidal illness (NT!), and to compare th results with those obtained by a well-validated equilibrium dialysis method [FF4(D)] (2). We also pa formed a correlation analysis between the concentrations o VF4 and free fatty acids (FFA) in sera from NT! patien FFA have been suggested by some investigators to functio as inhibitors of T4-protein binding in NT! (4, 12, 13 although others have not been able to provide evidence fo such an effect (14, 15). If, however, the increase in Fr observed in some NT! patients is attributable mainly dialyzable inhibitors of T4-protein binding, one can expe a discrepancy between the Fr4 concentrations determine by equilibrium dialysis and ultrafiltration. I - MaterIals and Methods Subjects. The control group consisted of 36 euthyroic healthy subjects, 18 men and 18 women (ages 19-87 years mean 44.5). We also studied 20 women with hyperthyroidisir (ages 19-79 years, mean 43.7) and 18 patients with hypothyroidism, three men and 15 women (ages 26-73 years, mear 53.8). Their diagnoses were based on clinical findings and standard laboratory tests as reported previously in detail (16) concentrations ofT4 (RIA), triiodothyronine (RIA), and thy. rotropin (immunoradiometric assay) in serum, and a trilodot hyronine-uptake test for calculation of an Fr4 index. Jr addition we studied 23 patients, 12 men and 11 women (ages 17-90 years, mean 53.6), with various severe NTIs: diabeth ketoacidosis (n = 8), cancer (n 8), heart disease (n = 4). bacterial infection (n = 2), or anorexia nervosa (n = 1). Nc patient was given heparin or other drugs known to influenc the concentration of VF4 in serum. Ultrafiltration method. The essential features of thE method are as follows: because of the known importance ol pH for the Fr4 concentration, even within the physiological pH range, one must adjust the serum pH to 7.4 by adding 50 L of 1 molJL 4-(2-hydroxyethyl)-1-piperazineethane sulfonic acid (HEPES) buffer to 1 mL of serum, then incubatE the sample for 20 mm at 37 #{176}C to achieve equilibrium ol = 1 Kuopio University 2Minerva Foundation Central Hospital, Institute, Kuopio; Helsinki; 3Department of Clinical Chemistry, University of Helsinki, Helsinki, Finland. 4Address correspondence to this author at: Medix Biochemica Oy, Asematie 13, 02700 Kauniainen, Finland. 6Nonstandard abbreviations: T4, thyroxin; FF4, free thyroxin; FF4(U), free thyroxin by ultrafiltration assay; Fr4(D), free thyroxin by equilibrium dialysis assay; FFA, free fatty acids; and NT!, nonthyroidal illness. Received November 27, 1989; accepted March 1, 1990. 800 CLINICAL in Patients and B. Krlstlan LiewendahIaS H. Tlkanoja”24 AddItIonal with Equilibrium CHEMISTRY, Vol. 36, No. 5, 1990 inding at this temperature. Wash the Unisep” UltracentLO ultrafiltration devices (Bio-Rad Laboratories, Richnond, CA) by passing 1.5 mL (maximal volume allowed) of 50 mmoIJL phosphate buffer, pH 7.4, and 1.5 mL of listilled water through the membrane. [We selected this Liltrafiltration device for use in our assay after testing ilbumin leakage in four different devices, as reported lsewhere (17).] Apply 1-1.5 mL of serum for ultrafiltration i.r 30 mm at 37#{176}Cand 2000 x g (fixed-angle rotor), liscarding the ultrafiltrate formed during the first 5 mm. Afterwards, without delay, analyze the ultrafiltrate for T4. To quantify the Fr4, we use 100 pL of ultraflltrate in a adioimmunoassay involving sheep anti-T4 antiserum (Inernational Laboratory Services, London, U.K.), in a final lilution of 1:1.5 x iO’, and [1I]T4 (specific activity 4000 i/g; Cambridge Medical Diagnostics, Billerica, MA). An;iserum and tracer are diluted in 50 mmoIJL phosphate )uffer (pH 7.4) containing 2 g of gelatin and 200 mg of odium azide per liter. The tubes are incubated at 4#{176}C )vermght before separation of free and bound radioactivity y adding 1 mL of 250 g/L polyethylene glycol reagent :Carbowax 6000; Fluka, Buchs, Switzerland) and 100 pL of 15 g/L bovine gamma-globulin solution (Sigma Chemical jo., St. Louis, MO). The lowest concentration of standard ve include in the assay is 6.6 pmol/L; the highest, 211 molJL. We correct the measured concentration of FT4 for a 10.3% (average) adsorption of T4 to the ultrafiltration nembrane. We determined this adsorption value by adding lIT4 tracer to pooled ultrafiltrates. The within-assay coefficient of variation (CV) for assay of ‘F4 was 10.9% and the between-assay CV 9.7%, as calcu.ated from duplicate determinations of an Fr4 concentraion of 27 pmoLIL. Other methods. We measured serum Fr4 by an equilibium dialysis-based method as described earlier (2). The 220 concentration of total nonesterified FFA was determined enzymatically (NEFA C-Test; Wako Chemicals, Tokyo, Japan) in sera from the NT! patients and controls. The bromcresol purple method (Orion Ltd., Espoo, Finland) was used to quantify albumin in sera from NT! patients. Statistics. Group differences were tested for significance by the nonparametric Wilcoxon’s rank sum test. Correlation coefficients were calculated by Spearman’s method, and linear-regression analysis was performed by the leastsquares method. Results Concentrations of Fr4 in serum measured by ultrafiltration and equilibrium dialysis methods in patients with thyroidal dysfunction or NT! are summarized in Table 1. The distribution of Fr4(U) concentrations in the various groups of patients is presented in Figure 1. Figure 2 shows the correlation between the FT4(U) and FT4(D) concentrations for the subjects studied. Values measured by the ultrafiltration method were higher than by equilibrium dialysis (Table 1). The mean FF4(U) in the euthyroid control group was 24.2 (SD 6.9) pmol/L, whereas in NT! patients the mean value was 41.2 (SD 31.5) pmoIJL. The corresponding values obtained by equilibrium dialysis were 14.8 (SD 3.5) and 19.8 (SD 5.3) pmol/L, respectively. The mean ratio of FT4(U) to Fr4(D) in the group of control subjects was 1.68 (SD 0.41) and in the NT! group 1.97 (SD 1.00). The diagnoses [and FF4(U)/ Fr4(D) ratios] in the four NT! patients with the highest Fr4(U) values were as follows: bacterial endocarditis and heart failure (2.3), severe heart failure caused by persistent truncus arteriosus (5.1), metastatic melanoma (3.3), and metastatic ovarian cancer (4.1). The FT4(U)IF’I’4(D) ratios for hypothyroid and hyperthyroid patients were 2.16 (SD 0.83) and 1.91 (SD 0.50), respectively. There was no overlapping of Fr4(U) or Fr4(D) results in the euthyroid healthy and hyperthyroid groups, whereas several of the hypothyroid patients with mild disease had concentrations in the euthyroid range. The mean FFA concentrations were 0.94 and 0.59 mmol/ 200#{149} 300 180 0 , #{163} leo 250 200 140 f 6. 120 6 150 100 2#{149} 80 too V II’. 80 60 40 ‘ 20 A 25 (C) 75 F14101 8 C 0 ig. 1. Concentrations of FT4 in serum as measured by the iltrafiltration technique in euthyroid healthy subjects (A), and in atlents with nonthyroidal illnesses (, hyperthyroidism (C), or ypothyroidism 80 125 100 l5O lp.,,oILl Correlation between FT4 concentrations as measured by ultrafiltration [FT4(U)] and by equilibrium dialysis [FT4(D)) in subjects with various thyroid states: euthyroid healthy (A), euthyroid sick (Lx), hyperthyroid (#{149}), or hypothyroid (0) The overall regression equation for all the sublects studied is y 1 .78x + 1.72 Fig. 2. = (n = 97, r = 0.93) CLINICAL CHEMISTRY, Vol. 36, No. 5, 1990 801 Table 1. ConcentratIons of FT,8 In Serum as Measured by Ultrafiltration Several Groups of Subjects M.an (and rang.), P <0.01, 36 20 18 23 8P <0.001, as compared 24.2 (14-48) l38.O (61-265) 14.8a (6.9-31) 41.2a (21-138) groups, respectively. In the NT! group, the correlation coefficient for FFA vs Fr4(D) was not significant (r = 0.27), nor was that for FFA vs FT4UJ) (r = 0.07). When the FF4 concentrations in NT! patients were correlated with the FFA to albumin molar ratio, we obtained the following FFA/albumin vs FF4(D), r = 0.59 (P <0.01); FFA/albumin vs Fr4(U), r = 0.39 (P <0.05). Discussion Comparison of the FT4 concentrations in serum reported by investigators who used either equilibrium dialysis (2, 18-22) or ultrafiltration (23-29) reveals that ultrafiltration methods usually gave higher estimates, although to a variable extent; with the ultrafiltration and equilibrium dialysis methods used by Surks et al. (30), the dialysis method yielded the higher Fr4 estimates. In our euthyroid control subjects, the Fr4 concentration determined by ultrafiltration was 64% higher than that measured by equilibrium dialysis. In the various groups of patients we studied, the relative differences between the FF4 concentrations determined by ultrafiltration and dialysis were even greater (84-108%), but for no patient group was the mean Fr4(U)/FF4(D) ratio significantly higher than that for the control group. The finding that Fr4(D) concentrations are lower than corresponding Fr4(U) concentrations is explained, at least partly, by our observation that a 10-fold dilution of normal serum by buffer caused a decrease of about 30% in the measured Fr4(U) concentration, whereas further dilution had no additional effect (unpublished data). Interestingly, a similar effect of dilution on Fr4(D) has been reported by some earlier investigators (see references cited in 22). This phenomenon may be due to a decreased affinity of carrier proteins for T4, attributable to the components of the buffers, or may indicate the existence of physiological inhibitors of T4-protein binding in serum. According to a hypothesis by Chopra et al. (12, 31), an increased concentration of FFA, particularly oleic acid, is responsible for the increase in the Fr4 concentration in NT!. Observations in our laboratory provide further evidence for the validity of this hypothesis, because NT! patients with high ratios for unsaturated FFAialbumin had a supranormal Fr4 concentration measured (13, 32). In vitro studies conducted by others also have demonstrated the necessity of a very high FFA/albumin ratio before the concentration of Fr4 serum can be expected to increase (14, 33,34). According to our experience, only in patients with a total FFA/albumin ratio >2 can the Fr4 concentration be expected to exceed the mean + 2 SD limit of the control population (13). Our conclusion that FFA-induced increases of serum FT4 are infrequent in NT! patients is supported somewhat by a recent report in which no NT! patient studied had an FFA/albumin ratio (>1.7) that could CLINICAL CHEMISTRY, (D) Methods ii FT4D) 14.8 (8-23) 75.1#{176}(29.-135) 7.8#{176}(3-14) l9.8t(l534) FT4(U)/FT4(D) 1.68 (0.9-2.5) 1.91 (1.1-3.3) 2.16 (1.0-3.8) 1.97(1.1-5.1) with the euthyroidhealthy group. L in the NT! and control 802 Dialysis pmol/L FT4(U) n Euthyroid Hyperthyroid Hypothyroid NTI (U) and Equilibrium Vol. 36, No. 5, 1990 be expected to increase the Fr4 concentration (35). In the NT! group, we observed a stronger associatioi between serum FFA/albumin ratio and Fr4 as measured b: dialysis than with Fr4 by the ultrafiltration method. I seems unlikely that FFA are significantly diluted durin dialysis, because they are tightly bound to albumin and t thyroxin-binding globulin (36). Elsewhere (37) we hay observed some generation of FFA from triglycerides durin overnight dialysis of serum at 37 #{176}C, the FFA concentratioi being somewhat higher after than before dialysis. For th time being, we conclude that the stronger correlatioi between FFA/albumin ratio and Fr4(D) than betweei FFAIalbumin ratio and Fr4(U) is possibly due to this small in vitro and therefore artifactual increase in the FFA concentration during dialysis. Notably, Wang et al. (38) observed that heparin therapy caused a larger increase in the Fr4 fraction when measured by equilibrium dialysis than when measured by ultraffitration, again probably because during dialysis there is more time for liberation FFA, thereby displacing T4 from proteins, than during ultrafiltration. If, in NT!, unsaturated FFA displace T4 from its protein binding sites, one would expect total T4 to correlate negatively with total FFA or with the unsaturated FFAJalbumm ratios, in particular. However, neither we (32) nor Haynes et al. (36) have observed such correlations, although Csako et al. (4) did. The absence of an inverse correlation between T4 and FFA does not exclude the possibility that in NT! the increase in serum Fr4 might be induced by the displacement of T4 by unsaturated FFA: the correlation between FFA and total T4 may be obscured by other factors affecting the total T4 concentration, e.g., the marked decrease in the concentrations of the T4-carrier proteins (13, 15). According to observations by Nelson and Weiss (22), dilution of sera from NT! patients results in a progressive fall in the FF4 concentration, a phenomenon not found in sera from healthy subjects. In our equilibrium dialysis system, the final serum dilution is 1:55, which could result in a spuriously low concentration of Fr4 in NT! sera if inhibitors of T4-protein binding are diluted. The risk for a dilution effect on serum Fr4 is particularly apparent in sera from patients taking drugs such as phenytoin (39), salicylate (40), or furosemide (41), which are known to displace T4 from its protein binding sites. Indeed, several of the earlier studies involving NTI patients cannot be considered to have adequately excluded the possibility that the increase in serum Fr4 was attributable to the drugs given the patients. One risks reaching erroneous conclusions concerning the mechanism of the increased concentrations of Fr4 in serum when patients treated in intensive-care units are studied, because these patients often receive drugs that interfere with thyroid-hormone binding to se- ofi um proteins or with the peripheral metabolism of thyroid ormones. NT! sera could also contain unknown endogenous dialyzLble inhibitors. However, our results provide no clear vidence for the existence of as-yet-unidentified dialyzable ithibitors of T4-protein binding in NT! sera, the Fr4(U)/ T4(D) ratio in NT! patients being not significantly higher han in the control subjects. This does not exclude the ossibility that sera from the four NT! patients with the dghest Fr4(U) concentrations might contain such inhibiors, because the three highest Fr4(U)/FI’4(D) ratios were Lctually observed in this group of cases. However, two of hese NT! patients also had high (>3) FFA/albumin ratios, hich could have caused the observed increase in the T4(U) and FT4(D) concentrations. On the other hand, the wo other NT! patients had low (<1.5) FFAialbumin ratios; herefore, their increased Fr4(U) concentrations might tave been caused by dialyzable inhibitors, especially given hat their Fr4(D) concentrations were normal. Obviously, ne must account for the possibility that changes in Fr4 oncentrations in NT! patients can be caused by both FFA nd dialyzable inhibitors. Interestingly, Mendel and Cavaieri (42), according to a preliminary communication, were Lnable to demonstrate the presence of binding inhibitors on he basis of expected changes in FF4 fractions in mixtures f NT! and normal pool sera; their normal poo1 had a low oncentration of triglyceride, to minimize the possible in itro generation of FFA. In conclusion, the ultrafiltration method we developed ives a significantly higher and possibly more nearly accuate estimate of Fr4 in serum than does the equilibrium ialysis method in healthy subjects and in patients with hyroid dysfunction and nonthyroidal illnesses. The ultraItration method, being more practicable and significantly ss time consuming than the equilibrium dialysis method, an therefore be used in clinical service laboratories with adioimmunoassay experience and appropriate equipment. We are indebted to Drs. Margaretha Turula, Matti VAlim#{228}ki, nd Gustav WAgar for providing us with sera from patients with iyroidal and nonthyroidal diseases. This study was supported by rants from the Finnish Medical Society (Finska LakaresAllskaet) and the Finnish Cultural Foundation, to which we are rateful. eferences Nishi K, Nakatani K, et a!. Effect of albumin )ncentration on the assay free thyroxin by equilibrium sdioimmunoassay with labeled thyroxin analog (Amerlex#{174}Free ). Clin Chem 1983;29:321-5. Helenius T, Liewendahi K. Improved dialysis method for free iyroxin in serum compared with five commercial radioimmunoasys in nonthyroidal illness and subjects with abnormal concenations otthyroxin-binding globulin. Clin Chem 1983;29:816-22. Rajatanavin R, Fournier L, DeCosimo D, Abreau C, Braverman 1. Amino E. Elevated N, ofserum serum free thyroxine by thyroxine analog radjoim- iunoassays in euthyroid patients with familial dysalbuminemic yperthyroxinemia. Ann Intern Med 1982;97:865-6. Csako G, Zweig MH, Glickman J, Kestner J, Ruddel M. Direct nd indirect techniques for free thyroxin compared in patients ith nonthyroidal illness I. Effect of free fatty acids. Clin Chem 989;35:102-9. Ekins R. Validity of analog free thyroxin immunoassays. Clin hem 1987;33:2137-52. Ooi DS, Mahadevan MS, Greenway DC, Gertler SZ. Evaluation rfour commercially available assays for free thyroxin. Clin Chem 88;34:2302-7. Rajan MGR, Samuel AM. A two-step ee triiodothyronine in serum. Clin Chem radioimmunoassay 1987;33:372-6. for 8. Weetall HH, Hertl W, Ward FB, Hersh LS. Enzyme immunoassay for free thyroxin. Clin Chem 1982;28:666-71. 9. Ito M, Miyai K, Doi K, Mizuta H, Amino N. Enzyme immunoassay of free thyroxin in serum in clinical samples. Clin Chem 1984;30:1682-5. 10. Beaman J, Woodhead JS, Liewendahi K, M#{224}hOnenH. The evaluation of a chemiluminescent assay for free thyroxine by comparison with equilibrium dialysis in clinical samples. Clin Chim Acts 1989;186:83-90. 11. Bounard J-Y, Bounard M-P, Begon F. One-step chemiluminescent immunoassay of free thyroxin with acridinium-esterlabeled thyroxin evaluated and compared with a two-step radioimmunoassay. Clin Chem 1988;34:2556-60. 12. Chopra U, Chua Teco GN, Mead JF, Huang T-S, Beredo A, Solomon DH. Relationship between serum free fatty acids and thyroid hormone binding inhibitor in nonthyroid illnesses. J Clin Endocrinol Metab 1985;60:980-4. 13. Liewendahl K, Tikanoja S, MahOnen H, Helenius T, VfiliinAki M, Taligren LG. Concentrations of iodothyronines in serum of patients with chronic renal failure and other nonthyroidal ill- nesses: role of free fatty acids. Clin Chem 1987;33:1382-6. 14. Mendel CM, Frost PH, Cavalieri RH. Effect of free fatty acids on the concentration of free thyroxine in human serum: the role of albumin. J Clin Endocrinol Metab 1986;63:1394-9. 15. Konno N, Hirokawa J, Tsuji M, et al. Concentration of free thyroxin in serum during nonthyroidal illness-calculation or measurement? Clin Chem 1989;35:159-63. 16. Liewendahl K, MShOnen H, Tikanoja S, Helenius T, Turula M, Vfilimaki M. Performance of direct equilibrium dialysisand analogue-type free thyroid hormone assays, and an imxnunoradiometnc TSH method in patients with thyroid dysfunction. Scand J Clin Lab Invest 1987;47:421-8. 17 Tikanoja S, Liewendahl K. Protein leakage in ultrafiltration devices [Abstract]. 8th Eur Congr of Clin Chem, Milano. Biochim Cliii 1989(Suppl. 1/8);13:175. 18. Ekins RP, Ellis SM. The radioiinmunoassay of free thyroid hormones in serum. In: Braverman LE, Robbins J, eds., Thyroid research, proc seventh mt thyroid conf Boston, 1975. Amsterdam: Excerpts Medica, 1976:597-600. 19. Weeke J, Orskov H. Ultrasensitive radioimmunoassay for direct determination of free triiodothyronine concentration in serum. Scand J Clin Lab Invest 1975;35:237-.44. 20. Giles AF. An improved method for the radioimmunoassay of free-thyroxine in serum dialysates. Clin Endocrinol 1982;16: 1015. 21. Jiang N-S, Tue KA. Determination of free thyroxine in serum by radioimmunoassay. Clin Chem 1977;23:1679-83. 22. Nelson JC, Weiss RM. The effect of serum dilution on free thyroxine (T4) concentration in the low T4 syndrome of nonthyroidal illness. J Clin Endocrinol Metab 1985;61:239-46. 23. Pedersen KO. Simultaneous estimation of the free thyroxine and tniiodothyronine fractions in serum. Scand J Clin Lab Invest 1974;34:241-6. 24. Wang Y-S, Hershman JM, Pekary AE. Improved ultrafiltration method for simultaneous measurement of free thyroxin and free triiodothyronine in serum. Clin Chem 1985;31:517-22. 25. Faber J, Rogowski P, Kirkegaard C, Siersbaek-Nielsen K, Friis T. Serum free T4, T3, r’P3, 3,3’-diiodothyronine and 3’,5’diiodothyronine measured by ultrafiltration. Acts Endocrino! (Copenh) 1984;107:357-65. 26. Weeke J, Boye N, Orskov H. Ultrafiltration method for direct radioimmunoassay measurement of free thyroxine and free tniodothyronine in serum. Scand J Clin Lab Invest 1986;46:381-9. 27. Thorson SC, Wilkins GE, Schaffrin M, Morrison RT, McIntosh HW. Estimation of serum-free thyroxine concentration by ultrafiltration. J Lab Cliii Med 1972;80:14&-54. 28. Sophianopoulos J, Jerkunica I, Lee CN, Sgoutas D. An improved ultrafiltration method for free thyroxine and triiodothyronine in serum. Clin Chem 1980;26:159-62. 29. Lee LA, Mooney RA, Woolf PD. Clinical utility of measuring free thyroxin and free triiodothyronine in serum of critically ill patients by ultrafiltration. Clin Chem 1986;32:797-800. 30. Surks MI, Hupart KH, Pan C, Shapiro LE. Normal free thyroxine in critical nonthyroidal illnesses tration of undiluted serum and equilibrium cninol Metab 1988;67:1031-9. CLINICAL CHEMISTRY, measured by ultrafildialysis. J Clin Endo- Vol. 36, No. 5, 1990 803 31. Chopra U, Huang T-S, Solomon DH, Chaudhuri G, Chua Teco ON. The role of thyroxine (T4)-binding serum proteins in oleic acid-induced increase in free T4 in nonthyroidal illnesses. J Clin Endocninol Metab 1986;63:776-9. 32. Tikanoja SH, Joutti A, Liewendahl BK. Association between increased concentrations of free thyroxine and unsaturated free fatty acids in non-thyroidal illnesses: role of albumin. Clin Chim Acts 1989;179:33-44. 33. Bregeng#{226}rd C, Kirkegaard C, Faber J, Poulsen S, SiersbaekNielsen K, Friis T. The influence of free fatty acids on the free fraction of thyroid hormones in serum as estimated by ultrafiltration. Acts Endocrinol (Copenh) 1987;116:102-7. 34. Lim C-F, Bai Y, Topliss DJ, Barlow JW, StockigtJR. Drug and fatty acid effects on serum thyroid hormone binding. J Clin Endocrinol Metab 1988;67:682-8. 35. Nicolson RE, Reilly CP, Pannall PR, Esposito L, Weliby ML. Do nonestenifled fatty acids displace thyroxin from its plasma binding sites in severe nonthyroidal illnesses? Clin Chem Endoermol 1989;31:25-30. 37. Liewendahl K, Tikanoja S, Helenius T, V#{227}lim#{228}ki M. Effect the free fatty acids on concentrations of iodothyronines in plas during nonthyroidal illness [Letter]. Clin Chem 1988;34:1370-1 38. Wang Y-S, Hershman JM, Smith V, Pekary AE. Effect heparin on free thyroxin as measured by equilibrium and ultn tration [Tech Briefi. Clin Chem 1986;32:700. 39. Liewendahl K, Tikanoja 5, Helenius T, Majuri H. Free th roxin and free tniiodothyronine as measured by equilibrium dia ysis and analog radioimmunoassays in serum of patients taki phenytoin and carbamazepine. Clin Chem 1985;31:1993-6. 40. Larsen PR. Salicylate-induced increases in free triiodoth nine in human serum. J Clin Invest 1972;51:1125-34. 41. Stockigt JR, Lim C-F, Barlow JW, et al. Interaction of furo mide with serum thyroxine-binding sites: in vivo and in vii studies and comparison with other inhibitors. J Clin Endocrin 1989;35:931-4. 36. Haynes 42. Mendel CM, Cavalieri RH. Inability to detect an inhibitor T4-serum protein binding in sera from patients with nonthyroi illness [Abstract]. 64th meeting, Am Thyroid Assoc 1989:T-5. thyroxine IG, Lockett SJ, Farmer binding inhibitor in the CLIN. CHEM. 36/5, 804-807 MJ, serum et al. Is oleic acid the of ill patients? Clin Metab 1985;60:1025-31. (1990) Therapeutic Monitoring of Cyclosporine: Impact of a Change in Standards on 1251-Monoclonal RIA Performance in Comparison with Liquid Chromatography Paul A. Keown,12 Jane Glenn,1 Jorge Denegri,1 Ursula Maclejewska,’ David Seccombe,’ Marilyn Stawecki,4 David Calvin Stiller,4 Christopher Shackleton,3 Eugene Cameron,2 This study examines the measurement of cyclosporine (CsA) by 1251-monoclonal AlA, and describes the impact of the recent change in the standard curve provided. GsA concentrations in serum and whole-blood control samples measured by 1251-RIA were initially 8-18% higher than those by H PLC. During the first two months of 1989, a significant and sustained deviation in the 125i..RIA produced results that exceeded the HPLC results by 21-28% (P <0.001). Introduction of the new standard curve in March 1989 returned the concentration of the whole-blood controls to the previous range (11-12% above HPLC, P <0.001). Measurement of clinical samples from heart, liver, and bone-marrow graft recipients by 1251-RIA by both old and new kit standards produced a close linear correlation (y = 0.89 x 19.02; r 0.99; n = 75, range 40-850 ig/L), with use of the new standards yielding results 82 (SD 8)% of those with the preceding assay. However, even with the new standard curve, CsA concentrations by 1251-RIA in the clinical samples exceeded those by HPLC by a factor of 1.37 (SD 0.18) to 1.52 (SD 0.19). Segregation for transplant type did not affect the RIA’ HPLC ratio. The results suggest cross-reactivity of the 1251-RIA with material present in vivo. - = Additional Keyphrases: British Room Columbia 19, 855 West control materials variation, source of Transplant Society, Heather Pavilion, D-10, 12th Avenue, Vancouver, British Columbia, V5Z 1M9, Canada; and the Departments of ‘Pathology, 2Methcine, and Surgery, Vancouver General Hospital and the University of British Columbia, Vancouver, and the Departments of 4Medicine and 5Pharmacology, University of Western Ontario, London, Ontario, Canada. Received December 11, 1989; accepted March 9, 1990. 804 CLINICAL CHEMISTRY, Vol. 36, No. 5, 1990 Freeman,4’5 and G. Phillips2 One of the most important developments in the thera. peutic monitoring of cyclosporine (CsA) is the recent intro duction of a monoclonal antibody that is selective for thE parent CsA molecule (1).6 This provides the theoretica] advantages of consistency and specificity in a rapid radio. immunoassay (RIA), which correlates reasonably closel3 with measurement by HPLC. Two commercial RIA kits currently use this antibody with either a 3H-labeled tracer and charcoal separation (Sandinmiune-SP; Sandoz Ltd., Basel, Switzerland), or 1251. labeled tracer with double-antibody precipitation (CYCLO. Trac-SP; INCStar Ltd., Stillwater, MN). Because gamma counting offers advantages of availability and speed, ani avoids the need for quench correction with whole-blooc samples, the latter RIA has become the predominani method for therapeutic monitoring of CsA in Canada. Both kits have comparable operating characteristics although the ‘251-RIA appears to record CsA concentra tions greater than those measured by the 3H-RIA or HPLC (2-6). This difference in concentrations has beer attributed to an inaccuracy in the standard curve, and th manufacturers have recently changed the standards sup plied to bring the results for these techniques into close alignment. Here, we chronicle the impact of this change or the concentrations of CsA measured in both control anc clinical samples. 6Nonsdard abbreviations: CsA, cyclosporine; VGH, Vancou General Hospital; SDI, standard deviation index; and UHL University Hospital, London. 7Canadian Working Group on Cyclosponine Monitoring. Unpub lished data. yen