Diagnostic accuracy of serum ceruloplasmin in Wilson disease: determination of sensitivity and specificity by ROC curve analysis among ATP7B-genotyped subjects.
Conventionally, diagnosis of WD is based on at least 2 of the following findings: typical neurological symptoms, presence of Kayser-Fleischer (KF) rings by slit lamp examination, and low serum ceruloplasmin concentration. Interestingly, hepatic dysfunction is not one of the criteria. A diagnostic scoring system has been developed that includes 7 parameters (neurological symptoms, KF rings, Coombs-negative hemolytic anemia, serum ceruloplasmin concentration, urinary copper excretion, liver copper measurement, and genetic mutations) (10). Other biochemical markers have been proposed such as serum free copper concentration (11) and copper:ceruloplasmin ratio (12), but these appear to be less reliable. False-positive and -negative rates are high among the traditional markers. For examples, up to 50% patients presenting with hepatic dysfunction or in presymptomatic stages do not have KF rings (13, 14).
An excretion of urinary copper >25 [micro]mol/day after penicillamine challenge was found to be useful in confirming symptomatic WD patients with active liver disease (sensitivity 92%) but unreliable for excluding WD in asymptomatic siblings (sensitivity 46%) (15). Liver copper in quantities >250 [micro]g/g dry weight without cholestasis is considered to be the diagnostic gold standard, with a reported sensitivity of 86.3% and specificity of 98.3% (16), but liver copper less than this threshold does not allow WD to be confidently ruled out. Moreover, the distribution of copper within the liver is heterogeneous, and hepatic copper content may vary up to 500-fold (17). Genetic analysis of ATP7B is the most decisive tool and is considered the only reliable test for family screening of asymptomatic siblings. Comprehensive diagnostic approaches with limited mutation screening have been proposed for populations with common mutations (18-22). More than 340 mutations have been reported so far (http://www. medicalgenetics.med.ualberta.ca/wilson/index/php), and the spectrum is population specific (22).
Serum ceruloplasmin is one of the biochemical markers for the diagnosis of WD. A concentration below the lower reference limit (conventionally taken as 0.20 g/L) is considered diagnostic of WD (1, 10, 23). A lower cutoff of <0.10 g/L has been proposed in the WD scoring system (10); however, such values have not been fully characterized in terms of sensitivity and specificity. In fact, the lower reference limit varies with many factors, e.g., different analytical methods, standardization, and the age of testing subjects. Our current study investigated the diagnostic value of serum ceruloplasmin concentration for WD patients and family screening among genotypically confirmed subjects.
Materials and Methods
We recruited 57 WD patients (ages 4-50 years) from 42 unrelated families. The diagnosis of WD was established by a WD score of 4 or more (10). Serum ceruloplasmin concentrations were measured at the time of diagnosis and before the initiation of chelation therapy. We also recruited 71 family members from these families. The validation group comprised 25 patients suspected of having WD who presented with hepatic dysfunction and/or neurological deficits. The reference interval of ceruloplasmin was established using data from 690 healthy Chinese residents of Hong Kong (ages 18-60 years) in whom there was no clinical evidence of WD. Informed consent was obtained from all the participants, and the local ethics committee approved the study.
MEASUREMENT OF SERUM CERULOPLASMIN
We measured serum ceruloplasmin concentration by use of the IMMAGE Immunochemistry System (Beckman Coulter); specimens collected before May 2, 2002, were measured by use of the Array instrument (Beckman Coulter). Both analyzers employed a nephelometric assay calibrated against the same primary standard CRM470 (RPPHS 91/0619) for measurement of ceruloplasmin. Correlation data between the 2 instruments provided by the manufacturer showed a correlation coefficient of 0.995, slope 0.996, and intercept of -0.02 g/L (Beckman Coulter IMMAGE Immunochemistry chemistry information sheet for CER, Ceruloplasmin, 988627 AE, November 2007). All the samples were handled as routine specimens according to the manufacturer's recommendations.
ANALYSIS OF ATP7B
We extracted genomic DNA from peripheral blood samples using a QIAamp Blood Kit (Qiagen). In all WD patients, the coding exons 1-21 and the flanking introns of the ATP7B gene were amplified by PCR. Family members were screened only for the mutations found in their respective probands. Exons 2, 3, 8, 10, 11, 12, 13, 16, and 18 covered more than 80% of mutations in our study population (24). We screened healthy subjects with ceruloplasmin concentrations <0.20 g/L and all subjects in the validation group for the aforementioned exons. Once a heterozygous mutation was found, all the remaining coding exons were sequenced. We carefully designed the PCR primers to avoid allele dropout, and thermal conditions were as described (25). PCR products were purified by Microspin S-300 HR columns (GE Healthcare) and sequenced using their amplification primers and Big-Dye-Deoxy terminator cycle sequencing reagents, according to the manufacturer's instructions (Applied Biosystems). We purified products of the sequencing reactions using AutoSeq G-50 columns (GE Healthcare), separated purified sequencing fragments by capillary electrophoresis, and detected fragments by laser-induced fluorescence on an ABI Prism 3100 genetic analyzer.
ROC CURVE ANALYSIS
We performed all statistical analyses using Analyze-it Clinical Laboratory version 2.07 software (26). We determined the sensitivity, specificity, and predictive values of serum ceruloplasmin concentration in predicting WD at different cutoff values by the construction of a ROC curve (sensitivity plotted against 1--specificity at different concentrations of ceruloplasmin) including the WD patients, their family members, the validation group, and 40 subjects with serum ceruloplasmin concentrations <0.20 g/L from the reference interval group.
We found disease-causing mutations in all WD patients except for 2 patients in whom only 1 mutation was found after sequencing all the coding exons of ATP7B and about 1 kb of each 5' and 3' untranslated region including the promoter region. Among 71 family members, 49 were heterozygous for their family mutations and 22 were wild-type homozygotes. The ranges of serum ceruloplasmin concentrations for the 49 heterozygotes and 22 wild-type homozygotes were 0.15-0.48 g/L (mean 0.24; SD 0.06) and 0.16-0.40 g/L (mean 0.29; SD 0.06), respectively.
We performed ROC curve analysis with use of data from 57 WD patients, 71 family members, 25 patients of the validation group, and 40 subjects with serum ceruloplasmin concentrations <0.20 g/L from the reference interval group. The ROC curve analysis (Fig. 1) suggested that the most useful cutoff value of serum ceruloplasmin concentration was 0.14 g/L, where the sum of sensitivity (93%) and specificity (100%) was the highest, and the numbers of false results were minimized among all the calculated values with 0 false positives and 4 false negatives. The positive and negative predictive values were 100% and 97.1%, respectively. The conventional cutoff of 0.20 g/L gave a sensitivity of 98.2% and a specificity of 55.9%, with 60 false positives and 1 false negative. The positive and negative predictive values were 48.3% and 98.7%, respectively. The sensitivity and specificity of value <0.10 g/L were 78.9% and 100%, respectively, with 0 false positives and 12 false negatives. The positive and negative predictive values were 100% and 91.9%, respectively. The area under the ROC curve was 0.99 (SE 0.008; 95% CI 0.97-1.01; P < 0.0001).
Four false-negative diagnoses resulted when using a ceruloplasmin threshold of 0.14 g/L; all 4 of these false-negative patients had WD scores [greater than or equal to] 4 despite having serum ceruloplasmin concentrations [greater than or equal to] 0.14 g/L (Table 1). All had KF rings and other abnormal parameters suggestive of impaired copper metabolism (increased 24-h urinary copper excretion > 1.0 [micro]mol/day and/or hepatic copper content >250 [micro]/g dry weight). All presented with acute liver failure; of note, their serum ceruloplasmin concentrations were measured during acute-phase reactions.
[FIGURE 1 OMITTED]
There were no false-positive diagnoses using a ceruloplasmin concentration threshold of 0.14 g/L, whereas a threshold of 0.20 g/L led to 60 false positives. Fourteen of the 60 false positives were heterozygotes (with serum ceruloplasmin concentration range 0.15-0.19 g/L, median 0.18 g/L), and 46 were wild-type homozygotes (ceruloplasmin concentration range 0.14-0.19 g/L, median 0.18 g/L). In 10, we repeated testing of ceruloplasmin, slit lamp examination, 24-h urinary collection for copper excretion, and/or liver ultrasonography. Four had liver biopsy for copper staining and/or copper measurement, but the results of all of these examinations were unremarkable. Five had been treated with chelation therapy (penicillamine and/or zinc) for >3 years.
REFERENCE INTERVAL GROUP
Forty of 690 healthy adults showed serum ceruloplasmin concentrations <0.20 g/L (range 0.16-0.19 g/L). None of these individuals had concentrations <0.14 g/L. Molecular analysis of the 40 subjects revealed normal DNA sequences. The central 95% reference interval for ceruloplasmin was 0.18-0.39 g/L (mean 0.27; SD 0.06). Fig. 2 depicts the serum ceruloplasmin concentrations of all studied subjects.
[FIGURE 2 OMITTED]
We derived an ROC curve for serum ceruloplasmin concentration in the diagnosis of WD from 193 genotypically confirmed subjects (57 WD patients, 71 family members, 25 patients in the validation group, and 40 subjects from the reference interval group). Using 0.20 g/L as the diagnostic threshold, the positive and negative predictive values were 48.3% and 98.7%, respectively. Ceruloplasmin concentrations <0.14 g/L gave the best positive and negative predictive values--100% and 97.1%, respectively--and maximal sensitivity (93%) and specificity (100%). The negative predictive values when using thresholds of 0.20 or 0.14 g/L were similar. Use of 0.20 g/L, however, led to unacceptably high numbers of false positives. An abnormal serum ceruloplasmin result may trigger further investigations, patient anxiety, and even chelation treatment (as was used in some of our patients).
Interpretation of ceruloplasmin using a threshold of 0.14 g/L gave higher specificity and positive predictive value than the postpenicillamine-challenge urinary copper excretion (15) and liver copper quantification (16) tests. If genetic analysis is available, it will provide the highest diagnostic accuracy for family screening.
Application of the WD scoring system includes a full clinical workup, the invasive procedure of liver biopsy, and labor-intensive mutational screening. The scoring system is more reliable when all 7 parameters have been fully assessed. Nonetheless, such information may be difficult to obtain in asymptomatic subjects and patients with atypical presentations. It also is not suited for screening purposes.
None of the 690 healthy subjects we examined had serum ceruloplasmin concentrations <0.14 g/L. This may have implications for population screening of WD to minimize false positives. Although 4 of our WD patients had ceruloplasmin [greater than or equal to] 0.14 g/L, they all presented with acute liver failure, and their serum ceruloplasmin concentrations were measured during the acute-phase reaction. Ceruloplasmin is a positive acute-phase reactant, and this can account for an apparently normal ceruloplasmin concentration in WD patients with fulminant hepatic failure (27).
Ceruloplasmin can be measured by immunochemical and enzymatic activity methods, expressed in mass units and in activity units, respectively. Because the former measures both active holoceruloplasmin and inactive apoceruloplasmin (which has a half-life in the circulation of only a few hours), immunochemical methods may give apparently normal results in individuals with WD. Therefore, it has been suggested that assay of ceruloplasmin oxidase activity is a better indicator of WD than mass immunoassay (28-30). We measured the ceruloplasmin oxidase activities of 2 of our WD patients whose ceruloplasmin concentrations were 0.15 and 0.21 g/L and found that the ceruloplasmin oxidase activities were, respectively, 24.5 and 36.7 [micro]mol/min/L (reference interval 62-108), thus, confirming the potential value of ceruloplasmin oxidase activity measurements.
On the other hand, low concentrations of ceruloplasmin can be observed in non-WD patients in decompensated liver failure whose synthetic liver function is greatly reduced. Ceruloplasmin deficiency also has been observed in other pathological conditions, including Menkes disease, protein calorie malnutrition, nephrotic syndrome, protein-losing enteropathy, acquired copper deficiency, and hereditary aceruloplasminemia (31). Patients with the last condition can resemble those with WD in having an undetectable serum ceruloplasmin concentration and similar neurological symptoms.
ATP7B protein has dual roles: copper transport, which is reflected by serum ceruloplasmin concentrations, and cellular trafficking. Some mutations interfering with the trafficking of AT7B within the cell do not lead to low serum ceruloplasmin concentrations. Thus, our results might not be applicable to WD patients with such mutations who present with normal ceruloplasmin concentrations. The correlation between serum ceruloplasmin concentration and ATP7B genotypes is controversial. Merle et al. (32) found no significant correlation, whereas Kumar et al. (33) reported that serum ceruloplasmin concentrations were significantly lower in homozygous WD patients than in compound heterozygotes. Patients in the same family tend to have similar concentrations of serum ceruloplasmin.
In general, the reference interval of serum ceruloplasmin concentration is between 0.20 and 0.40 g/L in normal adults. A concentration less than the lower ref erence limit (0.20 g/L) has been considered as the conventional diagnostic cutoff for WD (1, 10, 23). However, the lower reference limit can vary with different measurement methods, standardization, and the age of subjects. Serum ceruloplasmin concentration is age dependent (34). It is lower in normal neonates and rises to adult concentrations by the age of 6 months. It increases further to a maximum concentration between 2 and 3 years old, and then falls slowly until the teenage years, when it reaches the adult concentrations. In addition, females who are pregnant or on estrogens have higher concentrations. Therefore, a single cutoff value may not be applicable for all age groups (35). The youngest age suitable for use of serum ceruloplasmin measurement in the diagnosis of WD is 3 years (4).
Despite the use of primary protein reference material CRM 470/RPPHS, method-dependent variation of ceruloplasmin measurement is still significant, and the use of CRM 470 alone cannot achieve a universal reference interval for ceruloplasmin (36-38). If the concept of a lower reference limit is applied, only the method-dependent lower reference limit should be used. It cannot be overemphasized that 0.20 g/L is not a universal clinical decision value. For example, a concentration of 0.22 g/L is recommended by the manufacturer instead of 0.20 g/L when serum ceruloplasmin concentration is measured by Beckman IMMAGE (manufacturer insert, 2007). However, we recommend that each laboratory establish the reference interval based on its population. For Hong Kong Chinese, we found that the lower reference limit of serum ceruloplasmin concentration measured by Beckman IMMAGE was 0.18 g/L. The positive and negative predictive values were 70.9% and 99.1 %, respectively. This still gave 23 false positives and 1 false negative (Table 2). In our opinion, the use of the lower reference limit for the diagnosis of WD should be discouraged, and each laboratory should develop cutoff values of serum ceruloplasmin for the diagnosis of WD based on the local population.
Grant/Funding Support: The work described in this paper was partially supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (project no. CUHK4084/ 02M).
Financial Disclosures: None declared.
Acknowledgments: We thank patients, their families, and all the participant subjects without whose cooperation this work would not have been possible. We are indebted to our clinical colleagues at the different hospitals who collected patient blood samples and clinical information.
(1.) Scheinberg I, Stemlieb I. Wilson's disease. Major Probl Intern Med 1984;23:1-24.
(2.) Endo F, Taketa K, Nakamura K, Awata H, Tanoue A, Eda Y, Matsuda I. Measurement of blood holoceruloplasmin by EIA using a mouse monoclonal antibody directed to holoceruloplasmin: implication for mass screening of Wilson disease. J Inherit Metab Dis 1994;17:616-20.
(3.) Ohura T, Abukawa D, Shiraishi H, Yamaguchi A, Arashima S, Hiyamuta S, et al. Pilot study of screening for Wilson disease using dried blood spots obtained from children seen at outpatient clinics. J Inherit Metab Dis 1999;22:74-80.
(4.) Yamaguchi Y, Aoki T, Arashima S, Ooura T, Takada G, Kitagawa T, et al. Mass screening for Wilson's disease: results and recommendations. Pediatr Int 1999;41:405-8.
(5.) Garcia-Villarreal L, Daniels S, Shaw SH, Cotton D, Galvin M, Geskes J, et al. High prevalence of the very rare Wilson disease gene mutation Leu708Pro in the Island of Gran Canaria (Canary Islands, Spain): a genetic and clinical study. Hepatology 2000;32:1329-36.
(6.) Olivarez L, Caggana M, Pass KA, Ferguson P, Brewer GJ. Estimate of the frequency of Wilson's disease in the US Caucasian population: a mutation analysis approach. Ann Hum Genet 2001;65: 459-63.
(7.) Hahn SH, Lee SY, Jang YJ, Kim SN, Shin HC, Park SY, et al. Pilot study of mass screening for Wilson's disease in Korea. Mol Genet Metab 2002; 76:133-6.
(8.) Dedoussis GV, Genschel J, Sialvera TE, Bochow B, Manolaki N, Manios Y, et al. Wilson disease: high prevalence in a mountainous area of Crete. Ann Hum Genet 2005;69:268-74.
(9.) Kroll CA, Ferber MJ, Dawson BD, Jacobson RM, Mensink KA, Lorey F, et al. Retrospective determination of ceruloplasmin in newborn screening blood spots of patients with Wilson disease. Mol Genet Metab 2006;89:134-8.
(10.) Ferenci P, Caca K, Loudianos G, Mieli-Vergani G, Tanner S, Stemlieb I, et al. Diagnosis and phenotypic classification of Wilson disease. Liver Int 2003;23:139-42.
(11.) Twomey PJ, Viljoen A, House IM, Reynolds TM, Wierzbicki AS. Relationship between serum copper, ceruloplasmin, and non-ceruloplasmin-bound copper in routine clinical practice. Clin Chem 2005; 51:1558-9.
(12.) Twomey PJ, Viljoen A, House IM, Reynolds TM, Wierzbicki AS. Copper:caeruloplasmin ratio. J Clin Pathol 2007;60:441-2.
(13.) Steindl P, Ferenci P, Dienes HP. Wilson's disease in patients presenting with liver disease: a diagnostic challenge. Gastroenterology 1997;113: 212-8.
(14.) Gow PJ, Smallwood RA, Angus PW, Smith AL, Wall AJ, Sewell RB. Diagnosis of Wilson's disease: an experience over three decades. Gut 2000;46: 415-9.
(15.) Muller T, Koppikar S, Taylor RM, Carragher F, Schlenck B, Heinz-Erian P, et al. Re-evaluation of the penicillamine challenge test in the diagnosis of Wilson's disease in children. J Hepatol 2007; 47:270-6.
(16.) Ferenci P, Steindl-Munda P, Vogel W, Jessner W, Gschwantler M, Stauber R, et al. Diagnostic value of quantitative hepatic copper determination in patients with Wilson's disease. Clin Gastroenterol Hepatol 2005;3:811-8.
(17.) McDonald JA, Snitch P, Painter D, Hensley W, Gallagher ND, McCaughan GW. Striking variability of hepatic copper levels in fulminant hepatic failure. J Gastroenterol Hepatol 1992;7:396-8.
(18.) Wu ZY, Wang N, Murong S, Lin M. Identification and analysis of mutations of the Wilson disease gene in Chinese population. Chin Med J 2000; 113:40-3.
(19.) Weirich G, Cabras AD, Serra S, Coni PP, Nurchi AM, Faa G, Hofler H. Rapid identification of Wilson's disease carriers by denaturing high-performance liquid chromatography. Prev Med 2002; 35:278-84.
(20.) Lovicu M, Dessi V, Zappu A, De Virgiliis S, Cao A, Loudianos G. Efficient strategy for molecular diagnosis of Wilson disease in the Sardinian population. Clin Chem 2003;49:496-8.
(21.) Huster D, Weizenegger M, Kress S, Mossner J, Caca K. Rapid detection of mutations in Wilson disease gene ATP7B by DNA strip technology. Clin Chem Lab Med 2004;42:507-10.
(22.) Ferenci P. Regional distribution of mutations of the ATP7B gene in patients with Wilson disease: impact on genetic testing. Hum Genet 2006;120: 151-9.
(23.) Sternlieb I. Perspectives on Wilson's disease. Hepatology 1990;12:1234-9.
(24.) Mak CM, Lam CW, Tam S, Lai CL, Chan LY, Fan ST, et al. Mutational analysis of 65 Wilson disease patients in Hong Kong Chinese: identification of 17 novel mutations and its genetic heterogeneity. J Hum Genet 2008;53:55-63.
(25.) Lam CW, Mak CM. Allele dropout in PCR-based diagnosis of Wilson disease: mechanisms and solutions. Clin Chem 2006;52:517-20.
(26.) Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 1993;39:561-77.
(27.) Eisenbach C, Sieg 0, Stremmel W, Encke J, Merle U. Diagnostic criteria for acute liver failure due to Wilson disease. World J Gastroenterol 2007;13: 1711-4.
(28.) Walshe JM. Wilson's disease: the importance of measuring serum caeruloplasmin non-immunologically. Ann Clin Biochem 2003;40:115-21.
(29.) Macintyre G, Gutfreund KS, Martin WR, Camicioli R, Cox DW. Value of an enzymatic assay for the determination of serum ceruloplasmin. J Lab Clin Med 2004;144:294-301.
(30.) Gnanou JV, Thykadavil VG, Thuppil V. Pros and cons of immunochemical and enzymatic method in the diagnosis of Wilson's disease. Indian J Med Sci 2006;60:371-5.
(31.) Hellman NE, Gitlin JD. Ceruloplasmin metabolism and function. Annu Rev Nutr 2002;22:439-58.
(32.) Merle U, Schaefer M, Ferenci P, Stremmel W. Clinical presentation, diagnosis and long-term outcome of Wilson's disease: a cohort study. Gut 2007;56:115-20.
(33.) Kumar S, Thapa B, Kaur G, Prasad R. Analysis of most common mutations R778G, R778L, R778W, 11102T and H1069Q in Indian Wilson disease patients: correlation between genotype/phenotype/copper ATPase activity. Mol Cell Biochem 2007;294:1-10.
(34.) Cox DW. Factors influencing serum ceruloplasmin levels in normal individuals. J Lab Clin Med 1966; 68:893-904.
(35.) Mak CM, Tam S, Fan ST, Liu CL, Lam CW. Wilson's disease: a patient undiagnosed for 18 years. Hong Kong Med J 2006;12:154-8.
(36.) Whicher JT, Ritchie RF, Johnson AM, Baudner S, Bienvenu J, Blirup-Jensen S, et al. New international reference preparation for proteins in human serum (RPPHS). Clin Chem 1994;40:934-8.
(37.) Johnson AM, Whicher JT, Ledue TB, Carlstrom A, Itch Y, Petersen PH. Effect of a new international reference preparation for proteins in human serum (certified reference material 470) on results of the College of American Pathologists Surveys for plasma proteins. Arch Pathol Lab Med 2000; 124:1496-501.
(38.) Beetham R, White P, Riches P, Bullock D, MacKenzie F. Use of CRM 470/RPPHS has not achieved true consensus for ceruloplasmin measurement. Clin Chem 2002;48:2293-4.
Chloe M. Mak, [1,2] Ching-Wan Lam,  * and Sidney Tam 
 Department of Chemical Pathology, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong SAR, China;  Division of Clinical Biochemistry, Queen Mary Hospital, Hong Kong SAR, China.
 Nonstandard abbreviations: WD, Wilson disease; KF, Kayser-Fleischer.
* Address correspondence to this author at: Department of Chemical Pathology, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong. Fax 852-26365090; e-mail firstname.lastname@example.org.
Received January 15, 2008; accepted May 17, 2008.
Previously published online at DOI: 10.1373/clinchem.2008.103432
Table 1. Clinical and laboratory data of 4 false negatives using serum ceruloplasmin concentration <0.14g/L as diagnostic cutoff. (a) Age at onset, Patient Sex years Presentation KF rings 1 M 14 Acute liver failure Yes 2 M 21 Acute liver failure Yes 3 M 50 Acute liver failure Yes 4 F 27 Acute liver failure Yes Basal Serum ceruloplasmin Ceruloplasmin urinary copper, Patient concentration, g/L oxidase activity, [micro]mol/day [micro]mol/min/L 1 0.14 ND ND 2 0.15 24.5 30 3 0.16 ND 13 4 0.21 36.7 97.4 Liver copper, WD Mutation Patient [micro]g/g score analysis 1 1462 5 p.G943D and p.I1148T 2 ND 5 p.P1273Q and 1 unknown 3 ND 5 p.R778L and p.T935 mol/L 4 ND 4 p.V176SfsX28 and c.1708- 5T>G (IVS4-5T>G) (a) Reference interval of ceruloplasmin oxidase activity is 62-108 mol/min/L. ND, not done. Table 2. Diagnostic value of serum ceruloplasmin concentration at different decision cutoffs for the diagnosis of WD. Serum ceruloplasmin Sensitivity, Specificity, concentration, g/L % (95% CI) % (95% CI) 0.22 100 (93.7-100) 52.7 (44.2-61.6) 0.20 98.2 (90.6-100) 55.9 (47.1-64.4) 0.18 98.2 (90.6-100) 83.1 (75.7-89) 0.14 93 (83-98.1) 100 (97.3-100) 0.10 78.9 (66.1-88.6) 100 (97.3-100) Serum ceruloplasmin Positive Negative concentration, g/L predictive value, predictive value, % (95% CI) % (95% CI) 0.22 47.1 (36.8-63.1) 100 (94.1-100) 0.20 48.3 (39.5-58.8) 98.7 (91.8-100) 0.18 70.9 (61.7-76.3) 99.1 (94.5-100) 0.14 100 (96.4-100) 97.1 (88.2-99.4) 0.10 100 (96.4-100) 91.9 (77-93.1) Serum ceruloplasmin FALSE FALSE concentration, g/L positives negatives 0.22 64 0 0.20 60 1 0.18 23 1 0.14 0 4 0.10 0 12 Serum ceruloplasmin concentration, g/L Source 0.22 Lower reference limit stated in Beckman IMMAGE insert 2007 0.20 Conventional cutoff (1, 10, 23) 0.18 Lower reference limit established using Beckman IMMAGE based on Hong Kong Chinese population in this study 0.14 Diagnostic cutoff for WD proposed by this study 0.10 WD scoring system (10)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||General Clinical Chemistry|
|Author:||Mak, Chloe M.; Lam, Ching-Wan; Tam, Sidney|
|Date:||Aug 1, 2008|
|Previous Article:||Commutable calibrator with value assigned by the IFCC reference procedure to harmonize serum lactate dehydrogenase activity results measured by 2...|
|Next Article:||Atherosclerosis Risk in Communities (ARIC) Carotid MRI flow cytometry study of monocyte and platelet markers: intraindividual variability and...|