511
Clinical Chemistry 49, No. 3, 2003
Table 1. PPT1 and TPP1 activities in dried blood samples from patients with different forms of neuronal ceroid lipofuscinosis and from healthy controls. Patient
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Controls (n ⫽ 70)
Diagnosis
PPT1 activity, nmol/spot
TPP1 activity, nmol/spot
CLN1 CLN1 CLN1 CLN1 CLN1 CLN1 CLN2 CLN2 CLN2 CLN2 CLN2 CLN2 carrier CLN2 carrier CLN3 CLN3
0.02 0 0.02 0.02 0.03 0.01 1.13 0.8 0.49 0.41 0.88 0.42 0.46 0.54 0.49
0.22 0.34 0.39 0.27 0.31 0.3 0 0 0 0 0 0.05 0.04 0.21 0.11
0.4–1.52
0.1–0.67
Additionally, because the enzyme activities remain stable over several days, mailing to specialized centers is easier and less expensive. Furthermore, the assay requires only a few drops of blood in contrast to the 2–5 mL of EDTA blood needed for leukocyte assays. We consider the dried blood tests for PPT1 and TPP1 a very useful approach to the diagnosis of CLN1 and CLN2. However, the diagnosis should be confirmed by DNA tests, electron microscopy, and enzyme measurements in skin fibroblasts if only very low or no enzyme activities are detectable.
References 1. Goebel HH, Mole SE, Lake BD, eds. The neuronal ceroid lipofuscinoses (Batten disease). Amsterdam: IOS Press, 1999:197 pp. 2. Van Diggelen OP, Keulemans JL, Winchester B, Hofman IL, Vanhanen SL, Santavuori P, et al. A rapid fluorogenic palmitoyl-protein thioesterase assay: pre- and postnatal diagnosis of INCL. Mol Genet Metab 1999;66:240 – 4. 3. Ezaki J, Takeda-Ezaki M, Oda K, Kominami E. Characterization of endopeptidase activity of tripeptidyl peptidase-I/CLN2 protein which is deficient in classical late infantile neuronal ceroid lipofuscinosis. Biochem Biophys Res Commun 2000;268:904 – 8. 4. Chamoles NA, Blanco MB, Gaggioli D, Casentini C. Hurler-like phenotype: enzymatic diagnosis in dried blood spots on filter paper. Clin Chem 2001; 47:2098 –102. 5. Vanhanen SL, Raininko R, Autti T, Santavuori P. MRI evaluation of the brain in infantile neuronal ceroid-lipofuscinosis. Part 2. MRI findings in 21 patients. J Child Neurol 1995;10:444 –50. 6. Salonen T, Jarvela I, Peltonen L, Jalanko A. Detection of eight novel palmitoyl protein thioesterase (PPT) mutations underlying infantile neuronal ceroid lipofuscinosis (INCL; CLN1). Hum Mutat 2000;15:273–9. 7. Steinfeld R, Heim P, von Gregory H, Meyer K, Ullrich K, Goebel HH, et al. Late infantile neuronal ceroid lipofuscinosis: quantitative description of the clinical course in patients with CLN2 mutations. Am J Med Genet 2002;112:347–54. 8. Kohlschu¨ tter A, Laabs R, Albani M. Juvenile neuronal ceroid lipofuscinosis: quantitative description of its clinical variability. Acta Paediatr Scand 1988; 77:867–72.
Evaluation of a Turbidimetric Denka Seiken C-Reactive Protein Assay for Cardiovascular Risk Estimation and Conventional Inflammation Diagnosis, Thomas Christian Vukovich,* Stefan Mustafa, Helmut Rumpold, and Oswald Wagner (Institute of Medical and Chemical Laboratory Diagnostics, University Hospital of Vienna, AKH Leitstelle 5H, Waehringerguertel 18, A-1090 Vienna, Austria; * author for correspondence: fax 43-1-40400-5389, e-mail
[email protected]) Measurement of C-reactive protein (CRP) is used for conventional inflammation diagnosis (1 ) and diagnosis of low-grade inflammation for risk estimation of cardiovascular events (2, 3 ). Because diagnostic measurement ranges for those two indications differ by two orders of magnitude, different methods or different applications of one method must be used at present to cover both diagnostic measurement ranges (4, 5 ). The aim of this study was to evaluate the analytical performance of the Denka Seiken turbidimetric CRP assay compared with the Dade Behring nephelometric assay across a concentration range of 0.2–300 mg/L. For this evaluation, leftover material was used, which is in concordance with the European Law for Medical and Diagnostic Products. For precision and linearity studies, we prepared serum pools from blood samples with previously measured CRP (BN II nephelometer; Dade Behring). The low and high pools were prepared by combining samples with CRP ⬍1 and 200 –300 mg/L, respectively. The high pool was diluted with the low pool to the following final percentages of high pool: 100%, 33%, 11%, 3.7%, 1.2%, 0.41%, 0.14%, and 0%. The dilutions were aliquoted and stored at ⫺20 °C for a maximum period of 4 weeks until use. Both CRP methods were used according the manufacturers’ instructions. The turbidimetric wide-range CRP assay provided by Denka Seiken [CRP-latex (II)X2 assay, calibrated against reference preparation CRM 470] was performed on a Hitachi 911 (Roche), and the nephelometric assay provided by Dade Behring was performed on a BN II nephelometer (Dade Behring). To examine the precision of the Denka Seiken method compared with the established method, aliquots of serum pool dilutions were measured in duplicate on 10 different days (Table 1). CVs were ⱕ6.8% for the Denka Seiken method and ⱕ4.4% for the Dade Behring method. For all Table 1. Summary of precision and linearity data. Denka Seiken (Hitachi 911)
Pool
Mean, mg/L
1 2 3 4 5 6 7 8
254 84.7 29.7 10.1 3.62 1.57 0.81 0.38
CV, Target, Deviation, % mg/L %
1.3 1.1 84.2 1.6 28.3 2.1 9.78 2.8 3.5 2.0 1.42 3.7 0.74 6.8
0.6 5 3 3 11 9
Dade Behring (BN II) Mean, mg/L
243 82.9 28.9 8.61 3.15 1.27 0.68 0.32
CV, Target, Deviation, % mg/L %
2.1 2.5 80.2 2.3 27 2.9 9.28 2.5 3.3 3.1 1.31 2.6 0.66 4.4
3 7 ⫺7 ⫺5 ⫺3 3
512
Technical Briefs
Fig. 1. Method comparison of selected serum or plasma samples with results ⬎5 mg/L (A; n ⫽ 102) and ⬍5 mg/L (B; n ⫽ 108).
pools, values measured by the Denka Seiken method were somewhat higher than those measured by the Dade Behring method. To study the linearity of each method, we calculated the target concentrations of pools 2–7 from the mean concentrations of pool 1 (100% high pool) and pool 8 (100% low pool) as measured by the respective methods. Table 1 shows the calculated target concentrations of the pool dilutions as well as the percentages of deviation of the measured concentrations from the respective targets. The Denka Seiken method revealed positive deviations, whereas the comparison method revealed positive as well as negative deviations. Deming regression of measuredvs-target pool values revealed that the slopes of the Denka Seiken method were closer to 1 when calculated for the whole range (pools 2–7; slope, 1.01; intercept, 0.09 mg/L; Sy兩x ⫽ 0.43 mg/L; r ⫽ 1) as well as for the low range (pools 4 –7; slope, 1.03; intercept, 0.06 mg/L; Sy兩x ⫽ 0.09 mg/L; r ⫽ 0.999) than were the values obtained with the Dade Behring comparison method (pools 2–7; slope, 1.04; intercept, 0.16 mg/L; Sy兩x ⫽ 0.72 mg/L; r ⫽ 0.999; and pools 4 –7; slope, 0.92; intercept, 0.09 mg/L; Sy兩x ⫽ 0.10; r ⫽ 0.999). To study the concordance of results obtained from plasma and serum samples, we analyzed 30 pairs of heparin and serum samples from the same blood donation with the Denka Seiken method. Deming regression analysis of these measurements revealed optimum correlation: y ⫽ 1.00x ⫺ 0.006 mg/L [r ⫽ 1.000; range, 0.3– 83 mg/L; mean (SD) 24.9 ⫾ 26.6 mg/L in serum vs 24.8 ⫾ 26.5 mg/L in plasma]. To study the correlation of the Denka Seiken method with the comparison method, we collected 191 serum and 19 heparin-plasma samples from routine requests for traditional CRP analysis or high-sensitivity CRP. Of these, 108 samples had concentrations ⬍5 mg/L and 102 had concentrations ⬎5 mg/L. The samples were then analyzed with the Denka Seiken and the comparison method (Fig. 1). Fig. 1 shows that at concentrations ⬍5 mg/L, the slope was higher for the Denka Seiken method compared with the Dade Behring method (slope, 1.15; intercept, 0.09 mg/L; Sy兩x ⫽ 0.14 mg/L; r ⫽ 0.986). At concentrations ⬎5 mg/L, however, the slopes were close to the lines of unity for both methods (slope, 1.03; intercept, 0.11 mg/L; Sy兩x ⫽
3.76 mg/L; r ⫽ 0.998). This bend in the slopes might be attributed to the opposite nonlinearity between the Denka Seiken method and the Dade Behring method, as shown in Table 1. To evaluate the concordance of both method in cardiovascular risk assessment, the recently proposed cutoff values for the Dade Behring method (6 ) were adjusted for the Denka Seiken method by the regression equation calculated from patient samples ⬍5 mg/L (Fig. 1B). At these adjusted cutoff values, 96% of patients were allocated to identical risk groups. No patient was mismatched more than one adjacent risk group. In conclusion, the linearity of the Denka Seiken CRP assay is slightly better than that of the comparison method. Because of the bend in the linearity curve for the Dade Behring method, correlation of patient values revealed different slopes at high and low CRP concentrations. Therefore, sufficient concordance between methods for cardiovascular risk estimation might be obtained only after adjustment of cutoff values by the regression equations. The Denka Seiken CRP assay covers in a single determination the ranges for diagnosis of both conventional and low-grade inflammation. This method therefore improves laboratory throughput by reducing the number of retests with different sample dilutions or a different method. References 1. Colley CM, Fleck A, Goode AW, Muller BR, Myers MA. Early time course of the acute phase response in man. J Clin Pathol 1983;36:203–7. 2. Ridker PM, Buring JE, Shih J, Matias M, Hennekens CH. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. Circulation 1998;98:731–3. 3. Ko¨ nig W, Sund M, Frohlich M, Fischer HG, Lowel H, Doring H, et al. C-Reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men. Results from MONICA (monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation 1999;99:237– 42. 4. Roberts WL, Sedrick R, Moulton L, Spencer A, Rifai N. Evaluation of four automated high-sensitivity C-reactive protein methods: implications for clinical and epidemiological applications. Clin Chem 2000;46:461– 8. 5. Hamwi A, Vukovich T, Wagner O, Rumpold H, Spies R, Stich M, et al. Evaluation of turbidimetric high-sensitivity C-reactive protein assays for cardiovascular risk estimation. Clin Chem 2001;47:2044 – 6. 6. Rifei N, Ridker PM. Proposed cardiovascular risk assessment algorithm using high-sensitivity C-reactive protein and lipid screening. Clin Chem 2001;47: 28 –30.
