Patient 2 164 a a. a AA / mm of controi - AA / mm ... - Clinical Chemistry

13 downloads 42 Views 388KB Size Report
trometry (GCMS) (1) to measure estradiol in three of these. Table 1. Plasma ... mainly sodium estrone sulfate and sodium equilin sulfate)- or their metabolites ...
reagents, and to measure simultaneously the enzyme activity. Centrifugal analyzers are well suited for such manipulations, all the samples being added simultaneously at the beginning of the centrifugation and measured simultaneously during the centrifugation. We report here our protocol developed for using EMIT-st’i reagents on the Cobas-Bio#{174} centrifugal analyzer (Roche Diagnostics, Etobicoke, Canada M9C 5J4). The main advantage of our protocol is economy of reagents and time. Use the following settings: units M/min; calculation factor 100; standards 1, 2, 3: 0; limit 0; temp. (#{176}C) 30.0; type of analysis 4; wavelength (nm) 340; sample volume (zL) 05; diluent volume (FL) 20; reagent volume (ML) 300; incubation time (s) 0; start reagent volume (ML) 0; time of first reading (s) 30; time interval (s) 60; number of readings 02; blanking mode 1; printout mode 1. Begin the procedure by loading the samples. When ready to start the Cobas-Bio, reconstitute one vial of reagents (for use with as many as nine samples). Compare the results (iA/min x 100) for the samples with that for the calibrator. Figure 1, illustrating the results obtained with the EMITst’rM serum barbiturate assay for positive and negative controls (obtained from Syva), indicates that the method performs in a manner similar to the original application without any overlap. We also use the same protocol for other EMIT-St kits.

c+

Cna2O

n=20

8 7. 6#{149} >0

U

C

a a. aL.

11.

3 2

-O7

.06

.04

.02

AA / mm of controi

0 -

.02

.04

.06 +

AA / mm of caiibrator

of the results obtained for positive (C*) and negative (C) controls by using EMrr-st serum barbiturateassay with the Cobas-Bio Fig. 1. Frequency distributions

Discrepancies in with Commercial A. Jomain (Dept. and Med. School,

Piasma Estradioi Values Obtained Kits, Jennifer A. Nisbet and Patricia of Chem. Pathol., St. George’s Hosp. London, SW17 OQT, U.K.)

We compared the Sorin and “Coat-A-Count” Diagnostic Corp. (DPC) methods for estradiol in plasma. Both kits were used according to the manufacturers’ instructions. For 77 samples, all from female patients, the correlation (Deming’s regression) was DPC = 1.07 Sorin - 0.5, r = 0.95. For a further five samples, large discrepancies were observed (Table 1). We used gas chromatography/mass spectrometry (GCMS) (1) to measure estradiol in three of these

Table 1. Plasma Estradiol Vaiues in Five Patients by Sorin, DPC, and GCMS Methods Compared Estradiol, pmol/L

Patient

Therapy

Sorin

DPC

GCMS

Harmogen,

55 42 25

473 429 457

103

Harmogen, Duphaston Harmogen

111

541

253

Premarin, Duphaston

277

Duphaston 2

Premarin,

88

Duphaston 3 4

164

1399

1040

5

675 1065

and found results at variance with that expected from the correlation of these two methods with GCMS in the UK External Quality Assurance Scheme evaluations [Sorin = 1.1 GCMS + 85 and DPC = 0.97 GCMS + 25]. We cannot rule out operator error as the cause of the discrepancies. However, there was all the specimens that were re-analyzed and consistency in the results: all Sorin values were lower and all DPC values were higher than GCMS values. All patients showing this discrepancy were found to be menopausal or post-menopausal and receiving hormone replacementtherapy. It is possible that the hormones being administered-as Harmogen (piperazine estrone sulfate), Duphaston (dehydroprogesterone), and Premarin (conjugated estrogens, mainly sodium estrone sulfate and sodium equilin sulfate)or their metabolites might cross-react to a greater extent in the DPC method than in the Sorin method. However, DPC lists a 0.1% cross reactivity for Premarin, which would probably not account for the higher estradiol values by the DPC method. No data for either Harmogen or Duphaston are given by either manufacturer. Whatever the explanation, further investigation evidently is required to establish the validity of plasma estradiol results determined by these methods in patients receiving estrogen-replacement therapy. samples

agreementin

We are grateful to Dr. Elizabeth Finlay, Welsh National Schoolof Medicine, Cardiff for quantiiring the plasma estradiol by GCMS.

Reference 1. Gaskell SJ, Brownsey BG. Immunoadsorption to improve gas chromatography/high-resolution mass spectrometry of estradiol17/3 in plasma. Clin Chem 1983;29:677-80.

Analytical Performance of the Vision System Evaluated, Ana Stavijenic, Nada Vrki4 Carol Herak, K. Kummar, and E. Topic (Inst. of Clin. Lab. Diagnostics, School of Med., University of Zagreb, Clin. Hosp. Center, Kipati#{233}eva 12, 41000 Zagreb, Yugoslavia)

Products

1672

CLINICAL CHEMISTRY, Vol. 33, No. 9, 1987

The Vision System (Abbott Laboratories) has recently been introduced (1) as a new clinical chemistry desk-top analyzer for emergency laboratory, physician’s office, or satellite laboratory testing. It is based on the concept of twodimensional centrifugation and the use of a multi-chamber plastic test pack containing liquid reagents.

Details of the Vision System operation have been previously described (2). Each test pack is provided with a code for investigation and intended for single use. It can accommodate 40-120 L of whole blood, plasma, or serum. Clinical chemistry tests performed on the Vision System are optimized and based on the known-solution chemistry methods (3-5). We evaluated the precision and accuracy of five assays: glucose, urea nitrogen, triglycerides, cholesterol, and alkaline phosphatase (AP; EC 3.1.3.1). The within-day and between-day precision of the five tests, performed with control sera and reference sera, was as follows:

Assay, units

Glucose, mmol/L

5.74

1.99

13.92

1.48

Cholesterol,mmol/L

4.32 4.38

1.99

Triglycerides, mmol/L

2.57

WIthin-day

2,50

17.30 17.28 140

Urea N, mmol/L AP, U/L =

10, replicates 14, 14 days

of

2.42 3.11 2.10

1.70

10.

1965:221-2.

Table 1. Correlation between Vision Results and Those by Other Methods Analyte

Urea N, mmol/L Cholesterol, mmol/L

Triglycerides, mmol/L AP, U/L n

=

35 each.

2. Holden JT, Schultz SG, Wong ST. The application of twodimensionalcentrifugation to clinical chemistry testing. Uppsala J Med Sci 1986;91:151-4.

4. Tietz NW. Fundamentals of clinical chemistry. PhiladelphiaLondon-Toronto: WB Saunders Co., 1987; 427, 677. 5. Wieland 0. Glycerol. In: Bergmeyer HU, ed. Methods of enzymatic analysis, 2nd ed., vol. 1. New York: Academic Press,

Between-day imprecision did not exceed 3.5% for the five tests, which is considered satisfactory in conventional techniques. The analytical recovery results were as follows: glucose, 98.6%; urea, N, 99.1%; cholesterol, 97.0%; triglycerides, 97.8%; and AP, 97.3%. To determine what the frequency of recalibration should be, we assayed serum calibrators as i.mknowns for 14 days. The system maintained stable calibration during the two weeks. The results obtained with the Vision System for 35 sera from hospitalized patients, with wide ranges of analyte concentrations, were compared with results by the methods available on the Abbott VP, RA-1000, and ASTRA-8 (Table 1). Reasonable agreement was regularly obtained. The most pronounced difference was observed between the Vision hexokinase method for glucose and the glucose oxidaseperoxidase method on the Technicon RA-1000. Analysis of variance for the five analytes revealed no significant differences between this system and the comparison methods or analyzers. The practical advantages of this technology include the

Glucose, mmol/L

Chem 1985;31:1457-63.

927.

2.03 1.65 2.15 2.20

1 .&.i

References

3. Henry RJ, Cannon DC, Winkelman JW. Clinical chemistry, principles and techniques. New York: Harper and Row, 1974:516,

3.26 3.42

0.89 0.95

62 =

Be sen-day 3,45

1.90 4.24 3.20

of

1. Schultz SG, Holden JT, Donohue JP, Francoeur TA. Twodimensional centi-ifugation for desk-top clinical chemistry. Clin

CV, %

Mean concn

following. Because the chemistry used in this system is the same as in the conventional methods, the reference values are directly comparable with those obtained from the central hospital laboratory for the same patient. The test pack and analyzer have sufficient flexibility to perform not only endpoint, but also rate reactions, enzyme immunoassays, and organic extraction assays (the latter is important for rapid determination drugs). The Vision System is very easy to use; its ability to automatically separate plasma from whole blood simplifies the testing. The time required for a particular test to be completed averages 60 s.

Comparison method

y.lntercept 1.25

Slope

r

0.992

2.02

0.990 0.831

-0.19 0.99

1.079 0.893

1.03

0.842

0.985

RA-1000

0.16

1.011

0.996

RA-1000

0.38 0.67 0.06

0.992 0.894 0.980 1.109

0.942

0.982

0.995

0.884

0.998

VP

RA-1000 ASTRA-8 VP ASTRA VP VP

RA-1000 VP RA-1000

-0.32 6 -8

0.983 0.983 0.996

0.933 0.985 0.973

OptimIzed Kinetic Glutamate Dehydrogenase Assay Adapted to the Cobas Blo Centrifugal Analyzer, David A. Armbruster (Dept. of Pathol., Med. College of Virginia, Richmond, VA 23278; current address: Clin. Chem. Section, Lab. Services Branch, Epidemiology Div., USAF School of Aerospace Med., Human Systems Div. (AFSC), Brooks Air Force Base, TX 78235-5301) Optimized kinetic assays for glutamate dehydrogenase (GDH, i-Glutamate:NAD oxido-reductase, deaminase; EC 1.4.1.2) have been developed by Ellis and Goldberg (1) and by Jung et al. (2). I have adapted both assays to the Cobas Bio centrifugal analyzer (Roche, Nutley, NJ), using the following settings:

Alpha 1. 2. 3. 4. 5. 6.

7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

= 13 (decrease in initial absorbance is measured) Units = 1 (U/L) Calculation factor = 659.6 Standard 1 concn = 0 Standard 2 concn = 0 Standard 3 concn = 0 Limit = 350 U/L Temperature = 37#{176}C Type of analysis = 3 Wavelength = 340 rim Sample vol = 60 jAL Diluentvol = 35L Reagent vol = 225 jL Incubation time = 300 s Start reagent vol = 15 L Time of first reading = 30 s Time Interval = 10 s No. readings = 10 Blanking mode = 1 Printout mode = 1

CLINICALCHEMISTRY, Vol. 33, No. 9, 1987 1673