Letters - Clinical Chemistry

4 downloads 0 Views 215KB Size Report
www.aacc.org/standards/cholesterolinfo.html and http://www.aacc.org/standards/cdc/ cholesterolinfo.stm. 6. Kimberly MM, Leary ET, Cole TG, Waymack PP,.
Letters Proficiency Testing of Therapeutic Drug Monitoring Techniques

To the Editor: In this issue of the Journal, Jenny and Jackson-Tarentino (1 ) have carefully analyzed data from the New York State Department of Health therapeutic drug monitoring proficiency testing program to determine some causes of unsatisfactory performance. In regard to the Beckman Coulter SYNCHRON® Systems, it was noted that 9 results (out of 270) that were flagged “Out of Instrument Range” (OIR) were incorrectly reported as less than the reportable range when in fact they exceeded the range. A software change introduced in 1997 made the OIR flags more explicit by indicating OIR LO or OIR HI. These authors note that the product precision claim is usually greater than that which a user would typically encounter. Most laboratories do not routinely perform precision runs, but use ongoing laboratory qualitycontrol data to determine whether the instrument is performing satisfactorily. I’m not sure what benefit the customer would derive from a tighter advertised precision claim.

tastasize to the leptomeninges (1 ). The prognosis for patients with LMM is poor: most individuals survive a median of only ⬃4 months. Early diagnosis may improve the clinical response to radiotherapy and (intrathecal) chemotherapy, and may lead to more effective palliation and prolonged survival (1 ). Traditionally, a definitive diagnosis of LMM requires cytological detection of malignant cells in the cerebrospinal fluid (CSF). Interpretation is often aided by immunocytochemical techniques. Unfortunately, these CSF samples often contain very few morphologically identifiable malig-

nant cells. In these cases, no definitive diagnosis can be established, leading to “suspicious” or “atypical” diagnoses (1 ). Molecular detection of tumor-derived DNA in CSF can potentially improve early and sensitive detection of LMM because no intact cells are required for diagnosis (2, 3 ). Here we report the detection of K-ras gene mutations in the CSF of two patients with clinical features of LMM and negative cytology. Both patients (A and B) suffered from lung adenocarcinoma, which had metastasized to the cerebellum. Treatment consisted of resection of

Reference 1. Jenny RW, Jackson-Tarentino KY. Causes of unsatisfactory performance in proficiency testing. Clin Chem 1999;45:89 –99.

Nathan Gochman Beckman Coulter, Inc. 200 S. Kraemer Blvd. Brea, CA 92822-8000

Early Detection of Leptomeningeal Metastasis by PCR Examination of Tumor-derived K-ras DNA in Cerebrospinal Fluid

To the Editor: Leptomeningeal metastasis (LMM) occurs in 3– 8% of all cancer patients. Of the solid tumors, breast cancer, lung cancer, and malignant melanoma are the most common to me-

132

Fig. 1. Detection by MASA of mutant K-ras in the lung adenocarcinoma (A) and the cytologically negative CSF (B) of a patient with clinical features of LMM. (A), MASA performed on DNA isolated from the tumor of patient B. Lane 1, amplification of wild-type K-ras; lanes 2 and 3, initial K-ras mutation screening using two PCR mixtures, each containing three MASA primers to detect all codon 12 mutations at nucleotide positions 2 and 1, respectively; lanes 4 – 6, split up of the three MASA primers specific for codon 12, nucleotide 1. (B), MASA performed on DNA isolated from the CSF of patient B (lanes 1–3) and a cell line containing K-ras wild-type alleles (lanes 4 – 6). Three different PCR forward primers were used for specific amplification of: wild-type K-ras (GGT, positive control; lanes 1 and 4), the AGT mutation (MASA specificity control; lanes 2 and 5), and the TGT mutation (lanes 3 and 6) as found in the tumor (Fig. 1A, lane 6). MASA forward primers were described by Hasegawa et al. (6 ). The reverse primer was always 5⬘-ACTCATGAAAATGGTCAGAGAAACCTTTAT-3⬘. PCR products were separated on agarose (2%) and stained with ethidium bromide. M, 50-bp DNA ladder.

Clinical Chemistry 46, No. 1, 2000

Clinical Chemistry 46, No. 1, 2000

metastatic tumor followed by radiotherapy. Clinically suspected LMM emerged 24 (patient A) and 2 weeks (patient B) after the resection. Patient A experienced radicular pain in both arms. Although MRI scanning revealed LMM, cytology of the first CSF specimen was negative. It was only 4 weeks after this initial investigation, at the second CSF puncture, that malignant cells in the CSF could be detected. Patient B suffered from radicular pain in both legs and diplopia. The clinically suspected LMM initially could not be confirmed by cytological examination of the CSF. After 11 weeks, patient B had developed an organic psychiatric syndrome and multifocal radiculopathy. At that time, a second CSF examination was diagnostic for LMM. CSF of both patients was collected at the initial clinical presentation of LMM and centrifuged. The pellet was used for cytology, the results of which were negative. Free DNA was isolated from 0.5 mL of the supernatant. A K-ras mutation (codon 12; GGT3 TGT) common for lung adenocarcinoma (4 ) was found for both patients by mutant-allele-specific amplification (MASA; Fig. 1B) (5, 6 ). Identical mutations had been found previously in the lung primary (Fig. 1A) and the cerebellum metastasis. Other K-ras mutations were absent. In controls, including two patients with LMM without a K-ras mutation in the tumor and three patients with nonneoplastic disease (multiple sclerosis, meningitis, and Alzheimer disease), no mutant K-ras could be detected in the CSF supernatant. We conclude that detection of Kras mutations in the CSF of clinically suspected LMM patients is a promising tool for early diagnosis, followup, and treatment monitoring of LMM derived from primary adenocarcinoma of the lung. Moreover, with sensitive methods such as MASA, it is relatively easy to screen for K-ras point mutations in the CSF supernatant, which is not used for cytology, even without previous knowledge of the tumor mutations. Because K-ras mutations are found in only 30% of the adenocarcinoma of the lung (4 ) and are rare in other

solid tumors that frequently metastasize to the leptomeninges (7 ), additional molecular markers for the management of LMM are of potential interest (8, 9 ).

References 1. DeAngelis LM. Current diagnosis and treatment of leptomeningeal metastasis. J Neurooncol 1998;38:245–52. 2. Rhodes CH, Glantz MJ, Glantz L, Lekos A, Sorenson GD, Honsinger C, Levy NB. A comparison of polymerase chain reaction examination of cerebrospinal fluid and conventional cytology in the diagnosis of lymphomatous meningitis. Cancer 1996;77:543– 8. 3. Rhodes CH, Honsinger C, Sorenson GD. Detection of tumor-derived p53 DNA in cerebrospinal fluid. Am J Clin Pathol 1995;103:404 – 8. 4. Rodenhuis S, Slebos RJC. Clinical significance of ras oncogene activation in human lung cancer. Cancer Res 1992;52:2665s–9s. 5. de Kok JB, van Solinge WW, Ruers TJM, Roelofs RWHM, van Muijen GNP, Willems JL, Swinkels DW. Detection of tumour DNA in serum of colorectal cancer patients. Scand J Clin Lab Investig 1997;57:601– 4. 6. Hasegawa Y, Takeda S, Ichii S, Koizumi K, Maruyama M, Fujii A, et al. Detection of K-ras mutations in DNA’s isolated from feces of patients with colorectal tumors by mutant-allele-specific amplification (MASA). Oncogene 1995;10:1441–5. 7. Bos JL. ras oncogenes in human cancer: a review. Cancer Res 1989;49:4682–9. 8. Fujita J, Ueda Y, Bandoh S, Namihira H, Ishii T, Takahara J. A case of leptomeningeal metastasis from lung adenocarcinoma diagnosed by reverse transcriptase-polymerase chain reaction for carcinoembryonic antigen. Lung Cancer 1998;22:153– 6. 9. de Kok JB, Zendman AJ, van de Locht LT, Ruers TJM, van Muijen GNP, Mensink EJ, Swinkels DW. Real-time hTERT quantification: a promising telomerase-associated tumor marker. Lab Investig 1999;79:911–2.

Dorine W. Swinkels1* Jacques B. de Kok1 Anton Hanselaar2 Karel Lamers1 Rudolf H. Boerman3 Departments of 1 Clinical Chemistry, 2 Pathology, and 3 Neurology University Hospital Nijmegen P.O. Box 9101 6500 HB Nijmegen, The Netherlands *Address correspondence to this author at: Department of Clinical Chemistry (CKCL 564), University Hospital Nijmegen, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. Fax 31-243541743; email [email protected].

133

Total Error Evaluation of Roche Direct HDL-Cholesterol Reagent and Calibrator across 31 Lot Combinations: A 2-Year Experience

To the Editor: Professionally set quality specifications are needed as major considerations in development of new reagents (1 ). Two years ago, the Dutch Lipid Reference Laboratory, a permanent international member of the CDC Cholesterol Reference Method Laboratory Network (2, 3 ), in collaboration with Roche Diagnostics Nederland B.V. (formerly Boehringer Almere, The Netherlands), began evaluating lot-to-lot differences of the new direct HDLcholesterol (HDL-C) reagent from Kyowa Medex (cat. no. 1731157) and of the HDL-C/LDL-cholesterol cfas calibrator (cat. no. 1778501) before distribution on the Dutch market. To this end, a scaled down split-sample comparison protocol was used, essentially according to the design described in the HDL Cholesterol Method Evaluation Protocol for Manufacturers (4, 5 ). The accuracy platform was the CDC Designated Comparison Method (DCM) for HDL-C (2, 4 ); the test method was run on an Hitachi 911 analyzer (Boehringer Mannheim/Roche). A split-sample comparison was done with six specimens covering the HDL-C concentration range. All specimens were from individual donors. The matrix types investigated with the test method were serum and heparin plasma, whereas serum was used in combination with the DCM. Fresh specimens, intermittently stored at 4 °C, were analyzed with the test method, whereas frozen split sera were analyzed with the HDL-C DCM (5, 6 ). With the test method, all specimens were run in duplicate during 3 consecutive days for a total of 36 measurements (6 ⫻ 3 ⫻ 2 measurements); with the DCM, duplicate analyses were performed in one analytical run for a total of 12 measurements (6 ⫻ 2 measurements). Both test and reference data were produced in the

134

Lipid Reference Laboratory in Rotterdam. Analytical performance was checked against the National Cholesterol Education Program guidelines (4 ). New lot combinations were acceptable when the total error criterion of 13% was met. Total error can be considered as an error budget that can be divided between imprecision and bias (4, 7 ). During a 2-year period, 31 reagent/calibrator lot combinations for direct HDL-C were evaluated. For serum, the mean (⫾ SD) results were as follows: mean bias (%), ⫺1.6% ⫾ 3.0%; mean absolute bias (%), 3.0% ⫾ 1.9%; overall imprecision, 1.5% ⫾ 0.7%; and total error, 5.5% ⫾ 2.5%. For heparin plasma, the mean (⫾ SD) results were as follows: mean bias (%), ⫺2.7% ⫾ 3.1%; mean absolute bias (%), 3.8% ⫾ 2.1%; overall imprecision, 1.6% ⫾ 0.7%; and total error, 6.2% ⫾ 2.7%. In four reagent/calibrator lot combinations, the mean bias exceeded ⫾ 5%, i.e., the excessive bias was related to one specific lot of calibrator in three combinations and to a specific reagent lot only once. In none of the combinations did the total error exceed 13%, the maximum total error being 11.1% for one serum matrix and 12.1% for the corresponding heparin plasma. Hence, all tested combinations were sold on the Dutch market. In the HDL-C reagent kit, a “Document of Comparison” issued by the Lipid Reference Laboratory Rotterdam is inserted, stating the overall analytical CV (%), the percentage of bias, and the percentage of total error for both matrices and for the tested reagent/calibrator lot numbers on a Hitachi 911 system. It is concluded that the lot-to-lot differences of the direct HDL-C reagent and calibrator during the past 2 years were acceptable and in agreement with the 1998 National Cholesterol Education Program recommendations for HDL-C method performance (4 ). We have been assured that Roche will make further efforts to decrease the method bias to less than ⫾ 5% in all combinations. The inserted Document of Comparison with the stated method bias and total error is objective evidence that should aid clinical chemists in the

Letters

field to judge whether the quality of different lots of direct HDL-C reagent and calibrator meets recommended standards (4 ). In general, the results obtained for analytical performance characteristics should be compared to well-documented, objective quality specifications. We believe that assessment of the analytical performance by independent bodies across different sales lots should become part of (inter)national reagent release. References 1. Fraser CG, Hyltoft Petersen P. Analytical performance characteristics should be judged against objective quality specifications. Clin Chem 1999;45:321–3. 2. Myers GL, Cooper GR, Henderson LO, Hassemer DJ, Kimberly M. Standardization of lipid and lipoprotein measurements. In: Rifai N, Warnick GR, Dominiczak MH, eds. Handbook of lipoprotein testing. Washington, DC: AACC Press, 1997:223–50. 3. McNamara JR, Leary ET, Ceriotti F, Cobbaert C, Cole TG, Hassemer DJ, et al. Status of lipid and lipoprotein standardization. Clin Chem 1997; 43:1306 –10. 4. Warnick GR, Wood PD, for the National Cholesterol Education Program Working Group on Lipoprotein Measurement. National Cholesterol Education Program recommendations for measurement of high-density lipoprotein cholesterol: executive summary. Clin Chem 1995;41: 1427–33. 5. National Reference System for Cholesterol and Cholesterol Reference Method Laboratory Network. HDL cholesterol method evaluation protocol for manufacturers. http:// www.aacc.org/standards/cholesterolinfo.html and http://www.aacc.org/standards/cdc/ cholesterolinfo.stm. 6. Kimberly MM, Leary ET, Cole TG, Waymack PP, for the Cholesterol Reference Method Laboratory Network. Selection, validation, standardization, and performance of a designated comparison method for HDL-cholesterol for use in the Cholesterol Reference Method Laboratory Network. Clin Chem 1999;45:1803–12. 7. Cobbaert C, Mulder PGH, Baadenhuijsen H, Zwang L, Weykamp CW, Demacker PNM. Survey of total error of precipitation and homogeneous HDL-cholesterol methods and simultaneous evaluation of lyophilized, saccharosecontaining candidate reference materials for HDL-cholesterol. Clin Chem 1999;45:360 –70.

Christa M. Cobbaert1* Thomas K. J. Luderer2 1

Lipid Reference Laboratory University Hospital Rotterdam Dr. Molewaterplein 40 3015 GD Rotterdam, The Netherlands and Hospital de Baronie Department of Clinical Chemistry Langendijk 75 4819 EV Breda, The Netherlands

2

Roche Diagnostics Nederland B.V. Markerkant 13-10 1314 AN Almere, The Netherlands

*Address correspondence to this author at: De Baronie Hospital, Department of Clinical Chemistry, Langendijk 75, 4819 EV Breda, The Netherlands. Fax 31-76-5277-043; e-mail cobbaert@ worldonline.nl.

High Prevalence of Factor V Mutation (Leiden) in the Eastern Mediterranean

To the Editor: Factor V (FV), a 330-kDa procofactor, is a plasma protein that acts in concert with other plasma factors in regulating the blood coagulation cascade (1 ). Upon its proteolysis by factor Xa and/or thrombin, factor Va is inactivated by proteolysis by activated protein C (1, 2 ). A defect in FV hydrolysis brought about in part by resistance to activated protein C hydrolysis translates into poor anticoagulant activity, leading to thrombosis and other disorders of poor anticoagulation (3–5 ). Factor V mutation-Leiden (FV-Leiden) is a specific point mutation, identified as a G-A substitution in nucleotide 1691 in the factor V gene that leads to Arg506-Gln conversion (2 ) and is associated with hypercoagulability and increased risk of venous thromboembolism. The prevalence of FV-Leiden is high among venous thromboembolism patients (20 – 60%) compared with otherwise healthy individuals (2–13%) (6, 7 ). FV-Leiden has been described as a predisposing as well as an additional risk factor for venous thrombosis (3 ), pulmonary embolism (5 ), and to a lesser extent coronary artery disease (8 ). However, the latter remains to be established in light of reports disputing any link between FV-Leiden and coronary artery disease (3, 9 ). High prevalence of FV-Leiden has been reported in Caucasians (Europeans and Arabs) but not in nonCaucasians (Africans, Asians, and Eskimos) (10 –12 ), thus suggesting a single origin of the mutation (13 ). In view of its role as the most common

135

Clinical Chemistry 46, No. 1, 2000

inherited predisposing factor for venous thrombosis and other disorders of poor anticoagulation (14 ), coupled with its preferential distribution in selective areas and in different ethnic groups (6, 12 ), we assessed the prevalence of the FV-Leiden mutation in Lebanon. Our results were compared to those of other countries, in particular Eastern Mediterranean countries. The data obtained support earlier findings regarding the high prevalence of FV-Leiden among Caucasians (12, 15 ) and suggest that this mutation arose in the Eastern Mediterranean. Healthy, unrelated Lebanese individuals (n ⫽ 155) of both sexes and of different age groups from the five provinces of Lebanon, representing the two major religious groups, were recruited into the study after signing a consent form. All institutional ethics requirements were met. Total genomic DNA was extracted from peripheral blood mononuclear cells by incubation of the cells in a proteinase K-containing lysis buffer, followed by extraction with phenolchloroform, and precipitation with 7.5 mol/L ammonium acetate and ice-cold absolute ethanol. FV-Leiden was assessed by PCR, followed by hybridization of PCR products with wild-type (G) and mutant (A) DNA

probes and detection by DNA enzyme immunoassay according to the manufacturer’s instruction (SORIN). The results were interpreted as follows: FV-Leiden non-carriers, positive Gnegative A; FV-Leiden heterozygote, positive G-positive A; and FV-Leiden homozygote, negative G-positive A. The prevalence of FV-Leiden was determined for 155 healthy, unrelated Lebanese subjects (mean age, 32 ⫾ 11 years). The subjects, recruited from different provinces, were of both sexes (81 males and 74 females) and represented the two major religious groups (60 Moslems and 95 Christians). All participants were free of blood coagulation disorders, and none was on anticoagulant therapy. FV-Leiden was present in 22 of 155 participants (14.2%; Table 1). The mean age of FV-Leiden carriers was similar to that of FV-Leidennegative individuals (32.2 ⫾ 11.4 years and 32.3 ⫾ 11.3 years, respectively; P ⫽ 0.96). The frequencies of FV-Leiden were similar in females and males (13 of 74 vs 9 of 81, respectively; P ⫽ 0.25) and in Christians and Moslems (16 of 95 vs 6 of 60, respectively; P ⫽ 0.39). Heterogeneity in FV-Leiden distribution was noted among the different provinces of Lebanon, exemplified by the high prevalence in Mount Lebanon and

the low prevalence in South Lebanon (7 of 43 and 1 of 32, respectively; P ⫽ 0.21; Table 1), thus suggesting differential demographic distribution of FV-Leiden. We found 21 heterozygotes and 1 homozygote. The homozygote, a 61-year-old male, had no signs of thrombosis at the time of specimen collection. The observed homozygote-to-heterozygote ratio was consistent with Hardy-Weinberg equilibrium (p ⫽ 0.858; 2pq ⫽ 0.135, and q ⫽ 0.00645). The high prevalence of FV-Leiden in Lebanon (14.2%) was similar to prevalences in Syria (13.6%; A. Nehme, personal communication), Greece-Cyprus (13.4%) (6 ), Jordan (12.3%) (16 ), and Turkey (7.4%) (17 ), further supporting the concept (13 ) that the FV-Leiden mutation arose in the Eastern Mediterranean basin. FVLeiden is common in countries with predominantly Caucasian populations [e.g., Sweden (11.1%) (18 ) and Germany (7.13%) (19 )] and also in countries associated with significant migrations of Lebanese/Syrians and Greeks [e.g., UK (8.9%) (6 ), Argentina (5.12%) (11 ), Australia (20 ), and Spain (3.33%) (4 )]. In contrast, FVLeiden is nearly absent in Senegalese (6 ), African-Americans (15 ), Koreans (21 ), Japanese (22 ), and among Greenland Inuits (23 ), further con-

Table 1. Prevalence of FV-Leiden. FV-Leiden positive Group

Total Mean age, years Sex Males Females Religion Moslems Christians Region Greater Beirut Mount Lebanon South Lebanon Others a

Number

%

Number

%

Allele frequencya

22 32.3 ⫾ 11.4

14.2

133 32.2 ⫾ 11.3

85.8

0.074 (0.021)

9 13

11.1 17.6

72 61

88.9 82.4

6 16

10.0 16.8

54 79

8 8 1 5

14.8 18.6 3.1 19.2

46 35 31 21

SE in parentheses. OR, odds ratio; CI, confidence interval. c Two-tailed Student’s t-test. d Pearson ␹2 test. e Compared with Greater Beirut. b

FV-Leiden negative

P

ORb

0.96c

⫺5.3 to 5.1

0.062 (0.019) 0.088 (0.022)

0.25d

1.71

0.68–4.86

90.0 83.2

0.050 (0.018) 0.089 (0.023)

0.39d

1.82

0.67–4.96

85.2 81.4 96.9 80.8

0.074 (0.021) 0.093 (0.023) 0.031 (0.014) 0.115 (0.026)

0.21d

1.31e 0.19e 1.37e

0.45–3.85 0.02–1.56 0.4–4.69

95% CI

136

firming the preferential prevalence of FV-Leiden among Caucasians, as suggested (10, 12, 15 ). FV-Leiden is the largest inherited risk factor of venous thrombosis (14 ). However, the pathogenesis of venous thrombosis is multifactorial, including inherited and non-inherited risk factors (24 ). Whereas our results do not support random and indiscriminate screening for FV-Leiden, they do recommend screening for FV-Leiden in relatives of FV-Leiden carriers. References 1. Kalafatis M, Bertina RM, Rand MD, Mann KG. Characterization of the molecular defect in factor V R506Q*. J Biol Chem 1995;270:4053–7. 2. Bertina RM, Koeleman BPC, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994; 369:64 –7. 3. Ridker PM, Hennekens CH, Lindpainter K, Stampfer MJ, Eisenberg PR, Miletich JP. Mutation in the gene coding for coagulation factor V and the risk of myocardial infarction, stroke, and venous thrombosis in apparently healthy men. N Engl J Med 1994;332:912–7. 4. Garcia-Gala JM, Alvarez V, Pinto CR, Soto I, Urgelles MF, Menendez MJ, et al. Factor V Leiden (R506Q) and risk of venous thromboembolism: a case-control study based on the Spanish population. Clin Genet 1997;52:206 – 10. 5. Hirsch DR, Mikkola KM, Marks PW, Fox EA, Dorfman DM, Ewenstein BM, et al. Pulmonary embolism and deep venous thrombosis during pregnancy or oral contraceptive use: prevalence of factor V Leiden. Am Heart J 1996;131: 1145– 8. 6. Rees DC, Cox M, Clegg JB. World distribution of factor V Leiden. Lancet 1995;346:1133– 4. 7. Dahlback B. Are we ready for factor V Leiden screening? Lancet 1996;347:1346 –7. 8. Marz W, Seydewitz H, Winkelmann B, Chen M, Nauck M, Wit I. Mutation in coagulation factor V associated with resistance to activated protein C in patients with coronary artery disease. Lancet 1995;345:526 –7. 9. Samani NJ, Lodwick D, Martin D, Kimber P. Resistance to activated protein C and the risk of premature myocardial infarction. Lancet 1994;344:1709 –10. 10. Gregg JP, Yamane AJ, Grody WW. Prevalence of factor V-Leiden mutation in four distinct American ethnic populations. Am J Med Genet 1997;73:334 – 6. 11. Herrmann FH, Koesling M, Schoeder W, Latman R, Jimenez-Bonilla R, Lopacciuk S, et al. Prevalence of factor V Leiden mutation in various populations. Genet Epidemiol 1997;14: 403–11. 12. Pepe G, Rickards O, Vanegas OC, Brunelli T, Gori AM, Giusti B, et al. Prevalence of factor V Leiden mutation in non-European populations. Thromb Haemost 1997;77:329 –31. 13. Zivelin A, Griffin JH, Xu X, Pabinger I, Samama M, Conrad J, et al. A single genetic origin for a common Caucasian risk factor for venous thrombosis. Blood 1997;89:397– 402. 14. Rosen SB, Sturk A. Activated protein C resistance–a major risk factor for thrombosis. Eur J Clin Chem Clin Biochem 1997;35:501–16.

Letters

15. Ridker PM, Miletich JP, Hennekens, CH, Buring JE. Ethnic distribution of factor V Leiden in 4047 men and women. Implications for venous thromboembolism screening. JAMA 1997;277: 1305–7. 16. Awidi A, Shannak M, Bseiso A, Kailani MA, Omar N, et al. High prevalence of factor V Leiden in healthy Jordanian Arabs. Thromb Haemost 1999;81:582– 4. 17. Ozbek U, Tangun Y. Frequency of factor V Leiden (Arg506Gln) in Turkey. Br J Hematol 1997;97:504 –5. 18. Svensson PJ, Zoller B, Mattiasson I, Dahlback B. The factor VR506Q mutation causing APC resistance is highly prevalent amongst unselected outpatients with clinically suspected deep venous thrombosis. J Intern Med 1997; 241:379 – 85. 19. Schroder W, Koesling M, Wulff K, Wehnert M, Herrmann FH. Large-scale screening for factor V Leiden mutation in north-eastern German population. Haemostasis 1996;26:223– 6. 20. Pecheniuk NM, March NA, Walsh NA, Dale JL. Use of first nucleotide change technology to determine the frequency of factor V Leiden in a population of Australian blood donors. Blood Coagul Fibrinol 1997;8:491–5. 21. Kim TW, Kim WK, Lee JH, Kim SB, Kim SW, Suh C, et al. Low prevalence of activated protein C resistance and coagulation factor V Arg506 to Gln mutation among Korean patients with deep venous thrombosis. J Korean Med Sci 1998; 13:587–90. 22. Takamiya O, Ishida F, Kodaira H, Kitano K. APC-resistance and MnlI genotype (Gln 506) of coagulation factor V are rare in Japanese population. Thromb Haemost 1995;74:996. 23. de Maat MPM, Kluft C, Jespersen J, Gram J. World distribution of factor V Leiden mutation. Lancet 1996;347:58. 24. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999;353:1167–73.

Noha Irani-Hakime1 Hala Tamim4 Ghanem Elias2 Ramzi R. Finan3 Jocelyn L. Daccache1 Wassim Y. Almawi1* Departments of 1 Laboratory Medicine, 2 Cardiology, and 3 Obstetrics and Gynecology St. Georges Hospital Beirut, Lebanon 4

Faculty of Health Sciences Balamand University Beirut, Lebanon

*Address correspondence to this author at: Department of Laboratory Medicine, St. Georges Hospital, P.O. Box 166378-6417, Beirut, Lebanon. Fax 961-1582560.

Use of Turbidity-Correction Algorithm Eliminates the Effect of Perflubron Emulsion on CO-Oximeter Results

To the Editor: Shepherd and Steinke (1 ) reported that perflubron emulsion, a potential blood substitute soon to be in clinical use, interferes with measurements by several CO-oximeters. The analyzer affected most was the AVL Omni: at 29 g/L perflubron, oxyhemoglobin was decreased by 12.3%, carboxyhemoglobin was decreased by 3.5%, and methemoglobin was increased by 1.0%. At higher concentrations of perflubron, the AVL Omni gave error messages without results. Although the Radiometer OSM3 Hemoximeter reported results at all concentrations of perflubron studied, it too was affected: at 29 g/L perflubron, oxyhemoglobin was decreased by 5.3%, carboxyhemoglobin was increased by 4.1%, and methemoglobin was increased by 2.3%. At about the time of this report, we observed that an occasional heparinized blood sample analyzed by the AVL Omni would give a valid total hemoglobin result but report “interference” for the oxy-, carboxy-, and methemoglobin measurements. We associated these results to patients who were receiving Propofol, a sedative used in intensive care settings and administered as a turbid emulsion. In discussions with AVL Scientific Corporation, they believed the interference was attributable to turbidity and had developed an algorithm to correct for turbidity. Because perflubron emulsion is also turbid, we investigated whether the algorithm for the AVL Omni would also correct the interferences caused by perflubron. We obtained perflubron-based emulsion (600 g/L; product no. AF0144) from Alliance Pharmaceutical Corporation, San Diego, CA. The blood gas and CO-oximeter analyzers were an AVL Omni 9 blood gas analyzer and a CO-oximeter from AVL Scientific, and a Radiometer OSM3 hemoximeter attached to an ABL 505 blood gas analyzer from Radiometer America.

137

Clinical Chemistry 46, No. 1, 2000

Table 1. Effect of perflubron emulsion on blood gas and CO-oximeter measurements by the AVL Omni. Perflubron concentration, g/La Test

10

30

40

60

pH PCO2, kPab PO2, kPa Hematocrit, % Total Hb,c g/L Oxy-Hb, %d Carboxy-Hb, %d Met-Hb, %d

⫺0.003 ⫾ 0.006 0.0 ⫾ 0.26 0.1 ⫾ 0.6 ⫺0.23 ⫾ 1.7 ⫺1 ⫾ 6 ⫺1.1 ⫾ 2.2 0.2 ⫾ 0.19 0.1 ⫾ 0.25

⫺0.018 ⫾ 0.007 0.12 ⫾ 0.28 ⫺0.1 ⫾ 0.8 ⫺0.7 ⫾ 2.0 ⫺3 ⫾ 7 ⫺0.5 ⫾ 1.2 0.6 ⫾ 0.35 0.1 ⫾ 0.27

⫺0.018 ⫾ 0.007 0.25 ⫾ 0.32 ⫺0.1 ⫾ 0.9 ⫺0.6 ⫾ 1.9 ⫺2 ⫾ 6 ⫺1.1 ⫾ 1.0 0.6 ⫾ 0.4 0.3 ⫾ 0.3

⫺0.027 ⫾ 0.020 0.35 ⫾ 0.19 0.6 ⫾ 0.8 ⫺1.7 ⫾ 1.6 ⫺6 ⫾ 5 ⫺1.0 ⫾ 0.7 0.8 ⫾ 0.5 0.3 ⫾ 0.2

a Data shown are the mean ⫾ SD of the differences in measurements between blood with saline and blood with the indicated concentration of perflubron. Each data pair represents results for 15 samples. b kPa ⫻ 7.5 ⫽ mmHg. c Hb, hemoglobin. d Percentage of total hemoglobin fraction.

To heparinized whole blood samples from patients in operating rooms or intensive care, we added perflubron within 5 min after the sample was analyzed for blood gases. We anaerobically transferred 0.2 mL of either saline or perflubron mixed with saline to 0.8 mL of each sample of whole blood to give final perflubron concentrations of 10, 30, 40, and 60 g/L. Each specimen was kept thoroughly mixed. We made no other adjustments to the samples, which had Po2 values ranging from 5.1 to 30.7 kPa (38 to 231 mmHg) and oxyhemoglobin values of 87–98%. The difference in the CO-oximeter or blood gas results between the sample mixed with saline and the same sample mixed with perflubron was used to determine the analytical effects of perflubron. To minimize changes in the sample during the analyses, we analyzed individual blood samples to which we added saline and perflubron rather than a series of samples prepared from a larger pool of blood. With only ⬃2 mL of blood available in each leftover sample, we used 60 different samples for the study: 15 different blood samples for each of the four concentrations of perflubron studied. We used the paired t-test to determine the significance of differences in means between results for samples with saline and those with perflubron (2 ). The use of perflubron has the therapeutic effect of both replacing blood

volume and supplementing the oxygen and carbon dioxide transport capabilities of blood. The concentrations of perflubron we studied covered the range expected in clinical practice. Based on an expected dosage of 0.9 –2.7 g of perflubron per kilogram of body weight, the concentration of perflubron in blood is estimated to be 10 – 40 g/L (1 ). Perflubron up to 40 g/L produced no significant changes in the blood gas or CO-oximeter measurements on the AVL (Table 1). At 60 g/L, all changes were statistically significant (all P values ⬍0.005), and changes for pH, Pco2, Po2, hematocrit, and total hemoglobin were greater than the SDs of the respective methods. These differences at 60 g/L represent changes of marginal clinical significance. Our results on the Radiometer OSM3 (not shown) confirmed the reported effects (1 ) on oxyhemoglobin, carboxyhemoglobin, and methemoglobin. The algorithm for turbidity correction has been incorporated in our analyzers for nearly 1 year. With this algorithm in place, we have had only two or three samples that have given error messages. Therefore, we conclude that this algorithm has significantly improved the ability of the AVL Omni to report valid COoximeter data in turbid samples and has virtually eliminated the effect of perflubron emulsion in blood at concentrations up to 40 g/L.

We appreciate the help of Shari Morgan and Dr. Peter Keipert of Alliance Pharmaceutical Corporation for providing the perflubron emulsion used in this study. We also appreciate the information provided by Dr. A. P. Shepherd regarding his protocol. Dr Toffaletti receives funding for research from AVL Scientific Corporation. References 1. Shepherd AP, Steinke JM. CO-Oximetry interference by perflubron emulsion: comparison of hemolyzing and nonhemolyzing instruments. Clin Chem 1998;44:2183–90. 2. Barnett RN, ed. Clinical laboratory statistics, 2nd ed. Boston: Little, Brown & Company, 1979:21pp.

John G. Toffaletti* Robert F. Wildermann Clinical Laboratories Department of Pathology Duke University Medical Center Durham, NC 27710 *Address correspondence to this author at: P.O. Box 3015, Duke University Medical Center, Durham, NC 27710. Fax 919-6817786; e-mail [email protected].

Abbreviated Direct and Indirect ELISAs: A New and Simple Format

To the Editor: Many laboratories, including our own, perform ELISAs using immobilized antibody or antigen. A common format is to coat the ELISA plate with

138

antibody for direct assays, or antigen for indirect assays. ELISA plates are subsequently “blocked” to prevent any further nonspecific absorption. Direct assays often are performed by incubation of antibody-coated microtiter plate wells with calibrators or patient samples and then washing the plate and detecting bound antigen with a mouse monoclonal antibody, followed by washing and the addition of anti-mouse immunoglobulin-peroxidase, further washing, and the addition of substrate. We compared our direct sandwich ELISA for sex hormone-binding globulin (SHBG), using this format (1 ), to a procedure that differs in that the supernatant contains both monoclonal antibody to SHBG and the anti-mouse immunoglobulin-peroxidase (1:20 and 1:500, respectively; cat. no. NA 931; Amersham). The combined antibodies are then added to washed polyclonal anti-SHBG antibody-coated ELISA plate wells containing bound calibrator or sample for a 30-min incubation before washing and addition of the substrate. This is in contrast to the two separate 30-min steps— one using a 1:20 dilution for the SHBG monoclonal antibody supernatant, and the second using a 1:1000 dilution of anti-mouse immunoglobulin-peroxidase— before the wells are washed and substrate added. The calibration curves were identical, and the correlation between the one- and two-step protocols is shown in Fig. 1A over a wide physiological range. We have extended this concept to our indirect ELISAs for steroid hormones, using immobilized steroid conjugates. Cortisol and dehydroepiandrosterone sulfate (DHEAS) had been assayed previously by adding plasma, diluted plasma samples, or calibrators (50 ␮L) directly to either cortisol-thyroglobulin-coated or DHEASthyroglobulin-coated plates (2, 3 ). This was followed by the addition of 50 ␮L/well of either cortisol monoclonal antibody supernatant diluted 1:50 or DHEA monoclonal antibody supernatant diluted 1:100 and incubation at room temperature, followed by washing and the addition of anti-mouse immunoglobulin-per-

Letters

Fig. 1. Comparison of plasma SHBG (A), cortisol (B), and DHEAS (C) assayed by the one-step and two-step methods. The equations for the lines are: (A), y ⫽ 1.03x ⫹ 1.6, r 2 ⫽ 0.987; (B), y ⫽ 1.07x ⫺ 45, r 2 ⫽ 0.985; (C), y ⫽ 0.97x ⫺ 0.01, r 2 ⫽ 0.983.

oxidase (100 ␮L/well) at a 1:1000 dilution. After an additional incubation, plates were washed and substrate was added. We now combine the monoclonal antibody supernatant, at a dilution similar to that for

the two-step method, with antimouse immunoglobulin-peroxidase (NA 931) at a 1:500 dilution and then add 50 ␮L to each well for a single 1-h incubation at room temperature, followed by washing and the addition of substrate. The calibration curves are similar to the two-step protocol, with excellent correlation between the one- and two-step formats for cortisol and DHEAS (Fig. 1, B and C, respectively) over a wide physiological range. The final dilution of either monoclonal antibody or anti-mouse immunoglobulin-peroxidase is identical for both the one- and two-step protocols for the indirect formats described. The direct format also uses the same dilution of monoclonal antibody with double the concentration of anti-mouse immunoglobulin-peroxidase. Optimization experiments also showed that doubling the concentrations of monoclonal antibody used in the two-step protocol and/or the anti-mouse immunoglobulinperoxidase produced similar results for both the direct and indirect ELISAs. This suggests that these systems are particularly robust and implies that the anti-mouse immunoglobulin-peroxidase used neither binds to the antigen-binding sites of the monoclonal antibodies nor is in close enough proximity to produce steric hindrance. The finding that the dilutions of antibodies for the onestep methods are similar to those for the two-step method suggests that this could be a relatively common finding, and hence, it could act as a guide when optimizing other assays to the one-step protocol. The obvious caveat is avoiding sodium azide in all reagents. There are many advantages of the procedures described. The need for purifying monoclonal antibody from culture supernatants and covalent conjugation to enzyme is abolished together with the tendency for procedural losses and the need for regular syntheses because of the limited shelf-life of enzyme conjugates. The use of a common and relatively inexpensive anti-mouse immunoglobulin-peroxidase combined with monoclonal antibody supernatant is an

139

Clinical Chemistry 46, No. 1, 2000

attractive, simple alternative that overcomes these difficulties and allows a one-step protocol. This concept has direct application to the laboratory as well as to ELISA kit manufacturers, whose dependence on antibody isolation and enzyme conjugation could be minimized. References 1. Lewis JG, Elder PA. Alternative ELISA for sex hormone-binding globulin in plasma [Letter]. Clin Chem 1998;44:1999. 2. Lewis JG, Manley LM, Whitlow JC, Elder PA. Production of a monoclonal antibody to cortisol: application to a direct enzyme-linked immunosorbent assay of plasma. Steroids 1992;57: 82–5. 3. Lewis JG, Bason LM, Elder PA. Production of monoclonal antibodies to dehydroepiandrosterone sulfate: application to direct enzymelinked immunosorbent assays of dehydroepiandrosterone sulfate and androsterone/ epiandrosterone sulfates in plasma. Steroids 1996;61:682–7.

John G. Lewis* Peter A. Elder Steroid and Immunobiochemistry Laboratory Canterbury Health Laboratories Clinical Biochemistry Unit P.O. Box 151 Christchurch, New Zealand *Author for correspondence. Fax 64-33640-889; e-mail [email protected].

Excessive Phlebotomies

To the Editor: I suggest as obligatory reading for any clinical laboratorian who may have missed it Jocelyn Hicks’ Letter to the Editor (1 ) on excessive volumes of blood sampling or a prominent review in the September issue of Clinical Laboratory News (2 ). Hicks presents a horror story from the “Doctor, BE your Patient” standpoint. Her recital of her prolonged hospital stay in which 10 mL of blood was drawn twice daily for 2 weeks for only a complete blood count and white cell differential (defended as the “computer-dictated” volume) shocked me as much as it did her. I can vouch for the commercial availability of 4-mL blood collec-

tion tubes for hematologic measurements as early as the 1960s, and tubes of varying sizes for other specific uses have become available since then. (Has “Hal” taken over the decisions of the clinical laboratory?) Undoubtedly, the great majority of current clinical laboratorians were not trained in the fading era of universal manual laboratory tests, when patient sample collection was performed not by “phlebotomists”, but by trained laboratory staff. The importance of the improvement between 10 mL and the miniscule samples now drawn for red blood cells and white blood cells in dilution pipettes from ear lobe- or finger-sticks is beyond estimation. Capillary sampling for the hematocrit has supplanted the several milliliters required by the old graduated tubes, and the Drummond system, with its guaranteed-bore small capillaries and high gravity centrifugation (as a percentage of an unmeasured negligible sample), and microscopic readers now permit precise measurements of (arterial) red cells and plasma hematocrit by sharp delineation of the white blood cell pack. Of course, finger-stick sampling has the disadvantage of requiring an adequate puncture to avoid milking the site, and the samples represent capillary blood, giving somewhat higher cell/plasma ratios than venous samples. In addition, precise automated sample dilutors have limited drawn sample requirements for multiple assays. Microchemical methods were being developed rapidly, notably by Manny Sanz in Berne, when they were overtaken by automated, and particularly “multiphasic”, testing for ever larger panels of analytes— most of which were ignored by the physicians who needed only a few critical results. These requirements appear to be perpetuated in excessive draws for modern technologies. Not that we can return to the “good old days”, but there are ways of drastically reducing this awful waste. All of this illustrates the fallacy of the assumption that real progress necessarily correlates positively with time over the years.

References 1. Hicks JM. Excessive blood drawing for laboratory tests [Letter]. N Engl J Med 1999;340: 1690. 2. Sainato D. Do hospitals draw too much blood? Clin Lab News 1999;25(9):1–10.

Alan Mather 3 Woodvine Lane Lake Wylie, SC 19710

Elution of Hemoglobin ␣Montgomery2␤S2 Hybrid Tetramers by the Variant Apparatus

To the Editor: Hemoglobin (Hb) Montgomery (␣48 Leu3 Arg) is an uncommon variant first reported in 1975 (1 ). We report here a case of the association of homozygous HbS (␤6 Glu3 Val) with this variant identified by the late Dr. T.H.J. Huisman in an 8-year-old black girl with a life-long sickling disorder. We quantified Hb fractions by cation-exchange HPLC with the VariantTM Hemoglobin Testing System Beta-Thalassemia Short Program (Bio-Rad Diagnostics) (2, 3 ). The complete blood count was determined by a Beckman-Coulter STKSTM. Alkaline Hb electrophoresis was performed on cellulose acetate (Helena Laboratories). Solubility testing was done with phosphate-buffered saline using SickleScreenTM (Pacific Hemostasis). The Hb was 88 g/L [reference interval (RI), 110 –160 g/L], mean cell volume was 83.1 fL (RI, 80 –98 fL), and the red cell distribution width was 19.5% (RI, 11.5– 18.0%). HbF, HbA2, HbS, and “HbC” were 8.9%, 3.8%, 67.7%, and 15.8%, respectively, by HPLC (Fig. 1). The blood sample was solubility positive. Alkaline Hb electrophoresis had an “HbSCF” pattern with an abnormal HbA2 band. Although Hb Montgomery is not rare in African Americans (4 ), it is less common than HbG-Philadelphia (␣68 Asn3 Lys); hence, the combination of homozygous HbS and Hb Montgomery in this patient (A. Kutlar, personal communication) is very

140

Letters

which also migrate to the HbC position (6 ).

References

Fig. 1. HPLC elution pattern of a whole-blood hemolysate. F, HbF elution peak; A2, HbA2; S, HbS; M-S, Hb ␣Montgomery2␤S2 hybrid tetramers. A0 may reflect glycated HbS components that elute with HbA0.

uncommon, although an accurate prevalence estimate of this condition is unknown to us. The lack of HbA by both HPLC and alkaline electrophoresis confirms homozygosity for HbS; therefore, the HbC window by HPLC and the band on cellulose acetate are ␣Montgomery2␤S2 hybrid tetramers. The elution of these tetramers in the HbC window on the Variant has not been reported. This window cannot contain Hb Montgomery (␣Mont-

gomery2␤2) because only ␤S chains were synthesized. Hb Montgomery is a stable variant of the ␣2-globin gene (␣x␣/␣␣) (5 ); the relatively high concentration of hybrid tetramers (15.8%; ␣Montgomery2␤S2) here reflects the high degree of ␣-chain variant stability and its affinity for the ␤S chains (5, 6 ). In alkaline Hb electrophoresis, the ␣Montgomery2␤S2 hybrid tetramers behave similarly to ␣GPhiladelphia2␤S2 hybrid tetramers,

1. Brimhall B, Jones RT, Schneider RG, Hosty TS, Tomlin G, Atkins R. Hemoglobin Alabama [␤39(C5)Gln3 Lys] and hemoglobin Montgomery [␣48(CD6)Leu3 Arg]. Biochim Biophys Acta 1975;379:28 –32. 2. Suh DD, Krauss JS, Bures K. Influence of hemoglobin S adducts on hemoglobin A2 quantification by HPLC. Clin Chem 1996;42: 1113– 4. 3. Thomas S. Sample dilution to resolve mistaken identification of haemoglobin D as haemoglobin E using the VariantTM automated system. J Clin Pathol 1998;51:253– 4. 4. Zeng Y-T, Huang S-Z, Dong Z-Y, Lam H, Wilson JB, Huisman THJ. Hemoglobin Montgomery (␣248Leu3 Arg ␤2) in a Chinese family. Hemoglobin 1982;6:213–5. 5. Molchanova TP, Pobedimskaya DD, Huisman THJ. The differences in quantities of ␣2- and ␣1-globin gene variants in heterozygotes. Br J Haematol 1994;88:300 – 6. 6. Felice A, Abraham EC, Miller A, Cope N, Gravely M, Huisman THJ. Post-translational control of human hemoglobin synthesis; the number of ␣ chain genes and the synthesis of HbS. In: Brewer GJ, ed. The red cell. New York: AR Liss, 1978:131–53.

Jonathan S. Krauss* Keith Bures Elizabeth Kenimer Department of Pathology Section of Clinical Pathology Medical College of Georgia Augusta, GA 30912 *Address correspondence to this author at: BIH-222B, MCG, Pathology, 1120 15th St., Augusta GA 30912-3620. Fax 706721-7837; e-mail [email protected].