Optimization and Evaluation of Cardiac Enzym es and Isoenzym es M easured on a Random Access Analyzer

Transcription

1 ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 15, No. 5 Copyright 1985, Institute for Clinical Science, Inc. Optimization and Evaluation of Cardiac Enzym es and Isoenzym es M easured on a Random Access Analyzer JOHN SAVORY, P h.d., ROBERT G. STALLINGS, M.D., DAVID E. BRUNS, M.D., M. GERALDINE SAVORY, M.T.(CLST), MARILYN MARGREY, M.T.(ASCP) and JAMES C. BOYD, M.D. Department of Pathology, University of Virginia Medical Center, Charlottesville, VA ABSTRACT Four serum enzymes and isoenzymes used in the diagnosis of acute m yocardial infarction (AMI), lactate dehydrogenase L D and LD-1, creatine kinase (CK), and CK-MB have been adapted to the Technicon RA autom ated clinical chem istry analyzer. Analytical param eters have been adjusted to provide clinically acceptable precision for all four assays. C orrelations w ith centrifugal analyzer procedures gave correlation coefficients ranging from to A lim ited clinical study of the CK-MB assay indicated that a discrim inant value of 13 U per L could separate AM I from non-am I patients. Introduction M easurem ents of the MB isoenzyme of creatine kinase (CK-MB) and lactate dehydrogenase isoenzym e 1 (LD-1) in serum provide reliable inform ation for the diagnosis of acute m yocardial infarction (AMI). The developm ent of immunoassisted m ethods for th ese two isoenzym es for th e centrifugal analyzer has b e e n re p o rte d by th e p re se n t authors. 2,3,4 T hese reports included clinical as w ell as m ethodological evaluations. T he re c e n t in tro d u ctio n of random access analyzers into the clinical laboratory has sim plified work-flow problem s associated w ith the older batch analyzers. These random access analyzers can be operated to perform m easurem ents of a variety of equilibrium and kinetic reactions and are w ell su ite d to enzym e assays. H ow ever, th e clinical value of CK-MB and LD-1 depends on good analytical precision at low activity values. The present study was initiated to adapt lactate d eh y d ro g en ase L D, creatin e kinase (CK), and th e im m unoassisted LD-1 and CK-MB assays to one of the m ore popular random access analyzers* and to attem pt to make modifications to achieve adequate precision for diagnostic purposes. * Technicon RA-1000, Tarrytown, NY /85/ $00.90 Institute for Clinical Science, Inc.

2 CARDIAC ENZYMES USING A RANDOM ACCESS ANALYZER 401 M aterials and M ethods T he im m unochem ical and enzym e reagents and the immunochemical m ethods for initial sample treatm ent w ere as d e scrib e d p reviously by th e p re se n t authors. 2,3 Briefly, the reagent kit for the im m unochem ical separation of LD -1 from the other four isoenzymes of LD was obtained from a commercial source, f Total LD activity was m easured using the pyruvate to lactate reaction w ith com m ercially available reagents. $ The reagents for the immunochemical separation and selective inhibition of the isoenzymes of CK w ere obtained from a com m ercial source, f T he activities of CK-MB and of total CK w ere m easured using an optim ized assay reagent (BMC CK NAC). This enzym e reag e n t was m odified by the addition of 5M m agnesium a cetate to overcom e th e ED TA p re s e n t in th e im m unochem ical r e agents. The random access clinical analyzer (Technicon RA-1000)* was o p e ra ted according to the m anufacturer s instructions except for the modification of the sample syringe used in the CK-MB activity m easurem ents. The m odification of the syringe was simply a substitution of a 100 (xl for the original 50 xl syringe. This exchange of syringes can be accomplished in approximately 30 seconds. The Technicon RA-1000 is capable of using seven or nine absorbance readings taken 15 or 30 seconds apart. Automatic regression analysis of these absorbances is used to com pute enzym e activity. For the CK- MB m easurem ents the computation program was forced to use nine absorbance readings w ith a 30 second interval. In ad d itio n an absorbance w indow was selected to reject analyses dem onstrating + Roche Diagnostics, Nutley, NJ $ BMC UV System LDH-P; Bio-Dynamics bmc, Indianapolis, IN Hamilton Co., Reno, NV sp e ctro p h o to m etric im precision. This w indow was ± 1. 2 m illiabsorbance around the regression line. Any absorbance read in g o u tsid e of this w indow caused th e analysis to be rejected. In stru m e n t settin g s for all of th e enzyme m easurem ents are listed in table I. For LD-1 and CK-MB, the volumes listed are those of the solution following th e im m unochem ical sep aratio n and inhibition, and not of original sample. All m easurem ents w ere carried out at 37 C and 340 nm. C o m p a r i s o n P r o c e d u r e s C orrelation studies of CK, LD, and LD-1 w ere carried out with the RA-1000 and a centrifugal analyzerj The centrifugal analyzer m ethods w ere described in our earlier studies. 2,3,4 All of the activity m easu rem en ts on th e cen trifu g al analyzer were made at 30 C. Electrophoresis is the technique used by many laboratories for semi-quantitation of CK -M B. O ne of our previous studies2 com pared CK-MB quantitation by centrifugal analyzer w ith electrophoretic estimation by fluorom etric staining using th e C orning AC1 system. 1f This com parison w ith ele c tro p h o resis was repeated in the current study using the optim ized RA-1000 random access analyzer m ethod. Fifty-eight p atien t sam ples w ere selected for this com parison from among specim ens subm itted to the laboratory for CK -M B q u an titatio n. Since we w ere m ost concerned with the performance of the assay at low CK-MB values, only samples with CK-MB activities less than 38 U per L w ere chosen. Patient records w ere reviewed to determine clinical diagnoses. Electrophoresis resu lts w ere g rad ed as n e g a tiv e, trace, or positive. Rotochem lla /3 6, American Instrum ent Company, Silver Spring, MD Corning Medical, Medfield, MA

3 402 SAVORY, STALLINGS, BRUNS, SAVORY, MARGREY, AND BOYD C h o i c e o f E n z y m e R e a g e n t s The pyruvate to lactate procedure 8 for LD assays was chosen as th e ro u tin e assay in the authors laboratory since it follows the natural equilibrium w ith a rate 2.5 tim e that of the reverse reaction. The CK activity was determ ined using reagents recom m ended by Szasz12 with the im portant additions of 10.2 xm diadenosine-5'-pentaphosphate to inhibit adenylate kinase and 2.0 mm ethylene diamine tetracetic acid (EDTA) as cation ch elato r w hich has b een show n to increase CK activity. 9 The CK assay chosen in th e p re s e n t study was a two reagent system in which the CK in the sam ple is activated w ith sulfhydryl d u r ing the first delay (180 s) and the reaction th en is trig g e re d w ith c re a tin e p h o s phate. P rein cu b atio n of sam ple w ith sulfhydryl activator has been reported to produce m ore com plete reactivation of CK in some sam ples. 1,5,7 An additional advantage of this approach over a single reagent system is that substrate is not utilized during th e tim e of enzym e activation and thus the linear range of the assay is increased. The present m ethod is linear to 2300 U per L whereas with a single reagent m ethod the corresponding linearity is approxim ately 1150 U/L. As m en tio n e d earlier, both LD and CK assays are available in com m ercial kits. There are several im m unoinhibition kits available for CK-MB. However, the only one which is free from interferences TA BLE I Parameters fo r RA-1000 Enzyme Measurements LD LD -1 CK CK-MB Sample v o l. (pi) s t reagent v o l. (yl) s t delay (sec) nd reagent v o l. (yl) nd d elay (sec) Temperature ( C) W avelength (nm) TABLE I I S e n s it iv i t ie s o f th e Enzyme Assays Ab s /m in a t t h e A b s / m i n / u n i t / L D e c i s i o n V a lu e (m A /M in /U /L ) (m A /M in ) LD (515) LD (195) CK (210) CK-MB (30 y l sample) ( 13) CK-MB (60 y l sample) ( 13) Decision values are upper limits of the reference range and are given in parentheses. from CK-BB, adenylate kinase and atypical CK, is that chosen for the present investigation.* Reagents from the same company w ere used for the im m unoprécipitation assay for LD-1 since, to the author s knowledge, these are the only ones available com m ercially. T he p rin ciples of th ese m ethods have been described clearly by th eir originators. 13,14 Results an d Discussion S e n s it i v i t y a n d P r e c i s i o n The sen sitiv ity and precisio n of the analyses are closely linked. O ne factor affecting the sensitivity of the Technicon RA-1000 is that the optical light path is 7 m m ra th e r th an th e u su al 10 m m. Thus, these absorbance readings are less than w ith a m ore conventional optical system. Another limiting feature of the RA-1000 is that an unm odified in stru m ent allows a maximum sample volume of only 30 xl. Respecting these lim itations, assays for LD, LD-1, CK, and CK- MB w ere developed and gave sensitivities shown in table II. These sensitivities are expressed in absorbance change per m inute per unit and absorbance change p er m inute at the decision value. The param eters for all assays w ere as shown in table I except for CK-M B w here an * Roche Diagnostics, Inc.

4 CARDIAC ENZYMES USING A RANDOM ACCESS ANALYZER 403 ad d itio n al non-o p tim ized assay w ith a sam ple volum e of 30 xl is listed. The obvious im provem ent in sensitivity can be seen using the 60 (jli sample size. Also, it should be noted that the com bined volum e of the two reagents for CK-MB is less th an for to tal CK and re p re s e n t approxim ately th e m inim al volum e acceptable by the instrum ent. The precision of the four total enzyme and isoenzyme assays using optimal conditions is shown in table III. These data represent analyses carried out over a 24- day p e rio d using th e two com m erical controls. The sensitivities and precision of the LD, LD-1, and CK assays were satisfactory using param eters on the RA similar to those used for our original clinical stu d ies on th e centrifugal analyzer. 2,3,4 The obvious differences in the two in stru m e n ts such as optical path length (7 mm vs. 10 mm) and the necessity of using a sample diluent on the centrifugal analyzer, did not affect the overall perform ance of the assay. However, there w ere major problem s with the CK- MB assay especially at the decision level, which in our laboratory is 13 U per L. O ur original assay using 30 jxl of supernatant for the im m unoprecipitation reaction provided precision of ± 2 0 percent at this decision level. This level of im precision m ade the assay of lim ited value for the diagnosis of acute myocardial infarction and it becam e essential to provide a means of improving the sensitivity and, th ere fo re, th e precision of th e p ro ced u re. T he m ajor change m ade to im- TABLE IV C o r rela tio n o f RA-1000 w ith C en trifu gal A nalyzer Procedures Assay a b (U /L ) r SE E (U /L ) CK LD LD (Y = ax + b; Y = RA-1000; X - C en trifu gal Analyzer) prove the precision of this assay was to substitute a jjli sample syringe for the original 50 xl syringe which allowed us to increase the maximum sample volume from 30 jjli to 60 xl. The total sample and reagent volume was 345 ul which is only slightly above the minimal am ount which can be safely used in th e RA-1000 cu v ettes. Also, th e authors becam e aware of the fact that the immunoinhibition reagents for CK-MB provided by the m anufacturer* co n tain ed ED TA. The reagents used for the total CK assayf also co n tain ed E D TA and m agnesium su p p lem en tatio n for activation of the CK. W e found it necessary to increase the magnesium concentration in the subtrate to overcome the effects of the additional ED TA in this im m unoinhibition reagent. C o r r e l a t i o n S t u d i e s The new procedures for CK, LD, and LD-1 were correlated with similar m eth- * Roche Diagnostics, Nutley, NJ t BMC UV Systems, Indianapolis, IN TABLE I I I P re c is io n o f Enzyme Assays LD LD -1 CK CK-MB QC1 QC2 QC1 QC2 QC1 QC2 QC1 QC2 n Mean, U/L SD, U/L CV, p e r c e n t

5 404 SAVORY, STALLINGS, BRUNS, SAVORY, MARGREY, AND BOYD ods which we had adapted for use on a centrifugal analyzer and which had been in routine use in our clinical laboratory for several years, and th e resu lts are shown in table IV. As expected these correlation studies showed higher values on the Teehnicon RA-1000 which was operated at 37 C rather than 30 C used on the centrifugal analyzer. The non-zero Y intercepts can be explained by the differential effect of tem p e ra tu re on reaction rate w hich has b een re p o rte d for L D 6 and also a sp artate am in o tran s ferase. 10,11 The results of the comparison of CK- MB determ inations by electrophoresis and Technicon RA-1000 are illustrated in figure 1. This plot is similar to that seen in our previous comparison of CK-MB data from electrophoresis and centrifugal analyzer m ethods. 2 In the current study, sam ples w hich co n tain ed no CK-M B activity d e te c ta b le by electro p h o resis had quantitative CK-MB activities less than or equal to 10 U per L. Samples positive for CK-M B by electrophoresis had CK-M B activities exceeding 10 U per L with one exception. The exception involved a patient who had undergone coronary artery bypass grafting the p re vious day. Samples showing trace CK- MB activity by electrophoresis had in te r m ediate quantitative values. Twelve of the 58 patient samples were from three patients with the clinical diagnosis of acute myocardial infarction. Peak CL-MB values for each patient exceeded 13 U per L. These CK-MB values were higher than those reported in our earlier clinical stu d y 2 since th e p re se n t assay was carried out at 37 C rather than the earlier 30 C. Conclusion The excellent analytical performance of the LD, LD-1, and CK assays coupled w ith th e correlatio n d ata p ro v id ed us with confidence that these assays could provide valid resu lts w hen in ro u tin e clinical use. W e m ade this ju d g m e n t based upon several years of experience using very sim ilar assays on the centrifugal analyzer. However, the borderline sensitivity of the CK-MB assay at a decision level of 13 U per L indicated that a lim ited clinical study and com parison with electrophoresis w ere required. This study was carried out on patients and the d ata are p re se n te d in figure 1. From these data it can be concluded that the assay can provide excellent discrim ination b e tw e en AM I and non-a M I patients. References CK-MB ELECTROPHORESIS F i g u r e 1. Comparison of CK-MB by quantitative and electrophoretic methods (n = 58). 1. A b b o t, L. B. and L o t t, J. A. : Reactivation of serum creatine kinase isoenzym e BB in patients with malignancies. Clin. Chem. 30: , B r u n s, D. E., C h i t w o o d, J., K o l l e r, K., H i l l, K. E., M o s t r o m, J., and S a v o r y, J.: Creatine kinase-mb activity: Clinical and laboratory studies of specific immunochemical technique with optimized enzymatic assay. Ann. Clin. Lab. Sci. 13: , 1983.

6 CARDIAC ENZYMES USING A RANDOM ACCESS ANALYZER B r u n s, D. E., E m e r s o n, J. C., B e r t h o l f, R. L., H il l, K. E., and S a vory, J.: Lactate dehydrogenase isoenzyme 1 with the centrifugal analyzer after immunochemical removal of other isoenzymes. Clin. Chem. 27: , B r u n s, D. E., E m e r s o n, J. C., I n t e m a n n, S., B e r t h o l f, R., H i l l, K. E., J r., and S a v o r y, J. : Lactate dehydrogenase isoenzym e-1: Changes during the first day after acute myocardial infarction. Clin. Chem. 27:1821, B r u n s, D. E., M o r g a n, W. S., D a v is, J. E., and L a d e n s o n, J. H.: L o w apparent creatine kinase activity and prolonged lag phases in serum of patients with metastatic disease: Elimination by treatment of sera with sulfhydryl agents. Clin. Chem. 22: , B u h l, S. N., J a c k s o n, K. Y., and G r a f f u n d e r, B.: Optim al reaction conditions for assaying human lactate dehydrogenase pyruvate to lactate at 25, 30 and 37. Clin. Chem. 24: , C h a n d l e r, W. L., C l a y s o n, K. J., L o n g s t r e t h, W. T., Jr., and F i n e, J. S.: C reatine kinase isoenzym es in human cerebrospinal fluid and brain. Clin. Chem. 30: , D e u t s c h e s G e s e l l s c h a f t F ü r K l i n i s c h e C h e m i e. Standardisierung von M ethoden zur B estim m ing von Enzym aktivitäten in biologischen Flüssigkeiten. Experimentelle Begründung der optimierten Standard-Bedingungen. Z. klin. Chem. klin. Biochem. 10:182, R o l l o, J. L., D a v i s, J. E., L a d e n s o n, J. H., M c D o n a l d, J. M., and B r u n s, D. E.: Effects of b-mercaptoethanol and chelating agents on the stability and activation of creatine kinase in serum. Clin. Chim. Acta $7: , R o s a l k i, S. B., B r o w n, S. S., F l e c k, A., M c C o r m a c k, J. J., P a d m o r e, G. R. A., S m i t h, A. F., a n d W i l k i n s o n, J. H.: I n v e s t i g a t i o n o f t h e v a l i d i t y o f t e m p e r a t u r e c o r r e c t i o n f a c t o r s f o r s e r u m a s p a r a t e a n d a l a n i n e t r a n s a m i n a s e s. Ann. C l i n. B i o c h e m. 12:78, S c h i e l e, F., G a l t e a u, M.-M., and S i e s t, G. : Du danger d utiliser des facteurs de temperature en enzym ologie. Organisation des Laboratories- Biologie Prospective, 3em e Colloque de Pont-a- Mousson. Paris, L Expansion Scientifique Française, 1975, pp S z a s z, G., G r u b e r, W., and B e r n t, E.: Creatine kinase in serum: 1. Determination of optimum reaction conditions. Clin. Chem. 22: , U s a t e g u i - G o m e z, M., W i c k s, R. W., and W a r - s h a w, M.: Im m unochem ical determ ination of heart iso en zym e o f lactate d eh yd rogen ase (LDH,) in human serum. Clin. Chem. 25: , W i c k s, R., U s a t e g u i- G o m e z, M., M i l l e r, M., and W a r s h a w, M. : Im m unochem ical determ i nation of CK-MB isoenzym e in human serum. II. An enzymic approach. Clin. Chem. 28:54 58, 1982.