Monoclonal and Polyclonal Immunoglobulin Interference in a

Monoclonal and Polyclonal Immunoglobulin Interference in a Conjugated Bilirubin Assay Lu Song, PhD; Kathleen A. Kelly, PhD; Anthony W. Butch, PhD

Context.—Monoclonal proteins can interfere with the conjugated bilirubin assay on the Beckman Coulter AU5400 and AU2700 instruments and produce spurious results. Protein depletion eliminates the interference, suggesting that monoclonal proteins cause the problem. Objectives.—To determine the interference rate with the Beckman Coulter AU5400 and AU2700 conjugated bilirubin assay and to identify the interfering substance. Design.—Beckman Coulter bilirubin results from 33 720 samples analyzed during 6 months were evaluated for interference. On a subset of samples, protein G columns were used to specifically remove immunoglobulin G (IgG) to determine the cause of the interference. Another 117 samples containing a (known) monoclonal protein were selected to determine the interference rate. Results.—From 26 different patients, 103 samples had conjugated bilirubin results greater than the total bilirubin for a false-positive rate of 0.3%. Removal of IgG from a subset of those samples with IgG monoclonal protein and

increased polyclonal IgG eliminated the conjugated bilirubin interference. In separate, selected samples containing monoclonal proteins, the false-positive interference rate was 7.7% (9 of 117) and was greatest when the monoclonal protein concentration was greater than 4 g/dL. Another 11 of the 117 samples (9.4%) with monoclonal proteins exhibited assay interference by producing false-negative results. The overall interference rate, therefore, was 17.1%. Conclusion.—The false-positive interference rate for the Beckman Coulter conjugated bilirubin assay was 0.3% for routinely analyzed serum/plasma samples. In addition to monoclonal proteins, we found that polyclonal immunoglobulins can also interfere with the conjugated bilirubin assay. The overall interference rate was 17.1% for samples with a monoclonal protein. (Arch Pathol Lab Med. 2014;138:950–954; doi: 10.5858/ arpa.2013-0042-OA)

S

polyclonal immunoglobulin concentrations have not been reported to exhibit interference.4 In 2005, Nauti et al2 reported that serum from 3 patients with a monoclonal protein interfered with the Beckman Coulter AU2700 (Beckman Coulter, Brea, California; formerly Olympus Diagnostics, Melville, New York) conjugated bilirubin assay and produced falsely increased values, exceeding the total bilirubin concentration. Although they did not identify the interfering substance, removal of proteins by ultrafiltration (molecular mass, .30 kDa) eliminated the interference, suggesting that monoclonal proteins were the most likely cause of the interference. Another group1 confirmed that monoclonal proteins can interfere with the Beckman Coulter AU2700 conjugated bilirubin assay, which occurred frequently, at a rate of 44%. In this study, we investigated the conjugated bilirubin interference rate on the Beckman Coulter AU5400/2700 chemistry analyzers at the Ronald Reagan Medical Center at UCLA during a 6 month period. In addition to monoclonal proteins, we discovered that samples with elevated polyclonal immunoglobulins could also interfere with the conjugated bilirubin assay on Beckman Coulter AU chemistry analyzers. Selective removal of immunoglobulin (Ig) G from samples exhibiting the interference demonstrated that monoclonal proteins and polyclonal immunoglobulins caused interference with the Beckman Coulter conjugated bilirubin assay.

erum and plasma samples containing monoclonal proteins (also called paraproteins and monoclonal immunoglobulins) have been shown to interfere with several chemistry assays on a variety of automated chemistry analyzers and can produce spurious test results, which can be either falsely increased or decreased. Chemistry tests prone to spurious results from samples containing a monoclonal protein include conjugated bilirubin,1,2 total bilirubin,1,3–5 creatinine,6,7 phosphate,8,9 calcium,10 glucose,11,12 urea nitrogen,13 uric acid,14 c-glutamyltransferase,12 iron,15,16 low-density lipoprotein cholesterol,17 and high-density lipoprotein cholesterol.1,3,17,18 When samples with known monoclonal proteins were evaluated, interference rates have been reported as high as 46%, depending on the concentration and isotype of the monoclonal protein, and the specific test and assay methodology being evaluated.1,2,4,5,8 Samples with increased Accepted for publication August 8, 2013. From the Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Los Angeles, California. The authors have no relevant financial interest in the products or companies described in this article. Reprints: Lu Song, PhD, Department of Pathology and Laboratory Medicine, Ronald Reagan Medical Center, University of California, Los Angeles, Box 957418, Room B403L, Los Angeles, CA 900957418 (e-mail: [email protected]). 950 Arch Pathol Lab Med—Vol 138, July 2014

Interference in a Conjugated Bilirubin Assay—Song et al

MATERIALS AND METHODS Patient Data and Specimens Conjugated and total bilirubin results for 33 720 patient serum/ plasma samples from the Clinical Laboratory of the Ronald Reagan Medical Center at UCLA during a 6-month period, were reviewed retrospectively for falsely increased conjugated bilirubin concentrations by programming the laboratory information management system to identify falsely increased conjugated bilirubin results. A conjugated bilirubin concentration greater than the total bilirubin was considered to be falsely increased because of interference. In addition, plasma samples from 2 patients identified to have falsely increased conjugated bilirubin concentrations were further investigated to determine the cause of the interference. Patient 1 was a 67-year-old woman who had previously received a liver transplant for primary biliary cirrhosis. She had impaired renal function and normal liver function. Patient 2 was a 62-year-old woman with anemia, hypertension, diabetes, hepatitis C, and end-stage renal disease. Liver function test results were within reference intervals, except for a serum IgG concentration of 3500 mg/dL (reference interval, 690–1660 mg/dL; to convert to grams per liter, multiply by 0.01) in the absence of a monoclonal protein. An additional 117 serum samples were selected with a monoclonal protein, based on protein electrophoresis, to determine the interference rate in the conjugated bilirubin assay at our institution. The 117 samples had monoclonal proteins ranging from 0.1 g/dL to 6.0 g/dL (to convert to grams per liter, multiply by 10), based on protein electrophoresis. Institutional review was not required for this study because the study was performed using deidentified discarded blood samples that were obtained for patient care and not for research purposes.

Test Methods and Instrumentation Conjugated bilirubin and total bilirubin concentrations were routinely measured on 2 automated Beckman Coulter AU5400 chemistry analyzers and one AU2700 chemistry analyzer. The conjugated bilirubin assay is an endpoint assay that couples conjugated bilirubin (direct bilirubin) with the diazonium salt of 3,5-dichloroaniline at low pH to form azobilirubin. Conjugated bilirubin in the serum or plasma sample is directly proportional to the formation of azobilirubin, which is measured bichromatically at 540/600 nm. The total bilirubin assay uses caffeine and surfactant to stabilize unconjugated bilirubin and to accelerate the reaction. Conjugated and unconjugated bilirubin react with 3,5-dichlorophenyldiazonium tetrafluoroborate to form azobilirubin, which is measured bichromatically at 570/660 nm. Both bilirubin assays use a second cuvette as a sample blank and subtract the absorbance from the absorbance of the reaction cuvette. Total protein and creatinine were measured on Beckman Coulter AU5400/2700 analyzers by the biuret and Jaffe methods, respectively. IgC, IgA, and IgM concentrations were determined by immunonephelometry on the Dimension Vista 1500 (Siemens Healthcare Diagnostics, Tarrytown, New York). Serum protein electrophoresis and immunofixation electrophoresis were performed using the Sebia HYDRASIS system (Sebia, Inc, Norcross, Georgia). The monoclonal protein concentration was determined by multiplying the relative density of the monoclonal protein band in the electrophoresis gel by the total serum protein concentration.

IgG Depletion Protein G was used to deplete IgG because it binds only to IgG and not IgM, IgA, or IgD.19 A column containing approximately 1 mL of protein G–agarose slurry (product No. 20397, Pierce Biotechnology, Rockford, Illinois) was prepared, and 1 mL of plasma was allowed to flow through the column 3 times. Plasma trapped in the column was recovered by gently overlaying the column with phosphate-buffered saline and collecting the eluate until a total volume of 1 mL was obtained. This process was repeated on a second protein G column. Conjugated and total bilirubin, IgG, IgA, IgM, and creatinine were measured before and Arch Pathol Lab Med—Vol 138, July 2014

after protein G treatment. The ratio of posttreatment to pretreatment creatinine concentrations was used to correct for dilution effects during the column chromatography procedure.

RESULTS Incidence of Conjugated Bilirubin Assay Interference on the Beckman Coulter AU5400 and AU2700 Chemistry Analyzers Of the 33 720 conjugated bilirubin tests routinely performed on the Beckman Coulter AU5400 and AU2700 chemistry analyzers during a 6-month period, there were 103 samples with falsely elevated conjugated bilirubin results (conjugated bilirubin . total bilirubin), producing a false positive rate of 0.3% for these samples. The interference was observed in 26 different patients with total bilirubin concentrations ranging from 0.1 to 6.9 mg/dL (to convert to micromoles per liter, multiply by 17.1). When repeat specimens from the same patient (collected at different times) were excluded, the false-positive rate was less than 0.1%. In all cases of interference, the conjugated bilirubin result was not reported because the laboratory information system was programmed to alert the operator when a serum or plasma sample has a conjugated bilirubin that exceeded the total bilirubin concentration. Because samples with a monoclonal protein can interfere with the conjugated bilirubin assay on the Beckman Coulter AU2700,1,2 we reviewed additional laboratory data to determine whether any of the samples from the 26 patients with the interference had a monoclonal protein. Six of the 26 patients (23%) had been evaluated by protein electrophoresis, and 4 of the 6 (67%) had monoclonal proteins of 0.5, 2.5, 4.5, and 6.5 g/dL. The remaining 2 patients (33%) had negative results for a monoclonal protein but had elevated IgG concentrations of 2900 and 3500 mg/dL (reference interval, 690–1660 mg/dL). Characterization of the Interference Of the 26 patients with samples exhibiting interference, plasma samples from 2 patients (8%) exhibiting the conjugated bilirubin interference were further examined to determine whether immunoglobulins were the cause of the interference. Patient 1 had an IgG k-monoclonal protein of 2500 mg/dL, and patient 2 had an elevated IgG concentration of 3500 mg/dL, which was not monoclonal (Table). Both patients had total protein concentrations within reference range. When the samples were tested on 3 different Beckman Coulter instruments (2 AU5400s and one AU2700), the interference was consistently present, but the conjugated bilirubin results were not reproducible, with some results being preceded by a minus sign. Conjugated bilirubin concentrations ranged from 0.6 to 1.1 mg/dL and 0.2 to 1.0 mg/dL for the same plasma sample from patients 1 and 2, respectively (Table). Similar widely varying conjugated bilirubin results were observed when the samples were measured in triplicate on the same instrument (data not shown). The 2 samples were not icteric, and total bilirubin concentrations were reproducible when measured on different instruments. To remove IgG from the samples, a protein-G affinity column was used because that material binds only to human IgG and to none of the other immunoglobulin classes.19 Each plasma sample was applied to separate protein G columns, and the eluates from the columns were collected and tested. As shown in the Table, specific depletion of IgG Interference in a Conjugated Bilirubin Assay—Song et al 951

Removal of Immunoglobulin (Ig) G Eliminates Conjugated Bilirubin Interference Patient 1: IgG-k Monoclonal Protein Testa Conjugated bilirubin, mg/dL Total bilirubin, mg/dL IgG, mg/dL IgA, mg/dL IgM, mg/dL Total protein, g/dL

Untreatedb 1.1, 0.1, 0.3 3070 327 213 7.9

0.6

Patient 2: No Monoclonal Protein

Treated With Protein Gc

Untreatedb

Treated With Protein Gc

,0.1, ,0.1, ,0.1 0.3 ,35 521 278

1.0, 0.2, 0.7 0.6 3500 679 232 8.1

0.2, 0.2, 0.2 0.5 ,35 789 224

a

Reference intervals: conjugated bilirubin, 0.0-0.2 mg/dL (to convert to micromoles per liter, multiply by 17.1); total bilirubin, 0.2–1.1 mg/dL (to convert to micromoles per liter, multiply by 17.1); IgG, 690–1660 mg/dL; IgA, 80–400 mg/dL (to convert to milligrams per liter, multiply by 10); IgM, 37–318 mg/dL (to convert to milligrams per liter, multiply by 10); and total protein, 6.2–8.6 g/dL (to convert to grams per liter, multiply by 10). b Results are presented from all 3 chemistry instruments. c After protein G treatment test results were normalized to the pretreatment creatinine concentration to correct for dilutional effects.

from the 2 plasma samples eliminated the conjugated bilirubin assay interference. After treatment, conjugated bilirubin concentrations were reproducible across instruments; furthermore, the concentrations became significantly less than the total bilirubin concentration. Immunoglobulin G concentrations were undetectable after protein G treatment, indicating that the procedure had eliminated all of the IgG from the samples. In contrast, IgA and IgM concentrations were not lowered by protein G treatment (Table). The untreated plasma sample from patient 2 was also analyzed by the MicroSlide method on the Ortho Vitros 350 (Ortho Clinical Diagnostics, Raritan, New Jersey), and the conjugated bilirubin concentration was 0.2 mg/dL, which was the same as the concentration after protein G treatment. Because the MicroSlide method is unaffected by elevated immunoglobulin concentrations,20 those data demonstrate that the conjugated bilirubin interference was eliminated by removal of IgG whether the immunoglobulins were monoclonal or polyclonal. Incidence of Monoclonal Immunoglobulin Interference Serum samples with known monoclonal proteins ranging from 0.1 to 6.0 g/dL were selected and tested for interference in the Beckman Coulter conjugated bilirubin assay. As shown in the Figure, 9 out of 117 samples (7.7%) exhibited the interference when measured in singleton test, based on a conjugated bilirubin greater than the total bilirubin concentration. As the monoclonal protein concen-

tration increased so did the false-positive interference rate. For instance, an interference rate of 75% (6 of 8) was found when the monoclonal protein concentration was more than 4.0 g/dL. An additional 11 of the 117 samples (9.4%) with monoclonal proteins ranging from 0.8 to 6.0 g/dL had falsenegative conjugated bilirubin concentrations secondary to interference. These falsely negative samples increased the overall interference rate to 17.1% (20 of 117). Of the 20 serum samples with the interference, the isotype of the monoclonal protein was IgG, IgA, and IgM in 16 (80%), 1 (5%), and 1 (5%) of the samples, respectively. The isotype of the monoclonal protein in the other 2 samples (10%) was unknown. COMMENT Blood samples containing a monoclonal protein can produce erroneous test results for many common laboratory tests, including conjugated bilirubin.1,2,21 We found that the conjugated bilirubin false-positive interference rate on the Beckman Coulter AU5400 and AU2700 automated chemistry analyzers was 0.3% during a 6-month period based on the analysis of 33 720 routinely tested serum and plasma samples. The Beckman Coulter conjugated bilirubin interference rate will most likely vary widely among laboratories depending on the prevalence of monoclonal gammopathies and abnormal immunoglobulin synthesis in the patient samples being studied. When samples with a monoclonal protein were previously studied, the interference rate with

The incidence of monoclonal protein interference in the conjugated bilirubin assay. The 117 samples were tested and grouped by monoclonal protein concentration. The monoclonal protein was known in 53 (45%) of the samples: 35 of the 53 (66%) were immunoglobulin (Ig) G, 8 (15%) were IgA, and 10 (19%) were IgM. The solid and shaded bars represent the total number of samples and the number of samples in each group with the interference, respectively.

952 Arch Pathol Lab Med—Vol 138, July 2014


the Beckman Coulter AU2700 conjugated bilirubin assay was reported to be 1.5% and 44% in separate studies.1,2 Protein depletion (molecular mass, .30 kDa) eliminated the conjugated bilirubin interference, thereby providing indirect evidence that monoclonal proteins were the interfering substance.2 In patients with a known monoclonal protein, we found an interference rate of 17.1%. The reason for the vastly different interference rate among studies is most likely due to differing concentrations of monoclonal protein and the criteria used to identify the conjugated bilirubin interference. As pointed out in this study and by Yang et al,1 samples with higher monoclonal protein concentrations are more likely to have erroneous bilirubin results and produce higher interference rates. Interestingly, the mean monoclonal protein concentration in the study by Yang et al,1 which reported a conjugated bilirubin interference rate of 44%, was 3.7 g/dL, whereas in our study, fewer than 10% of the samples (,10%) had a monoclonal protein greater than 3.0 g/dL. This difference in monoclonal protein concentrations most likely contributed to the lower conjugated bilirubin interference rate of 17.1% we identified. In the study by Nauti et al,2 monoclonal protein concentrations were not provided, and they identified the interference based solely on a conjugated bilirubin greater than the total bilirubin. That approach would not identify many spurious results and would produce a falsely low interference rate because the interference can produce negative conjugated bilirubin values in more than half the cases.1 We have demonstrated that the monoclonal protein was the source of the interference by selective removal of the IgG monoclonal protein from 2 different patient samples. Besides IgG, other immunoglobulin isotypes can produce erroneous bilirubin results. For instance, we found that serum samples containing an IgA or IgM monoclonal protein could also interfere with the Beckman Coulter conjugated bilirubin assay. An IgM monoclonal protein has also been reported to interfere with the total bilirubin assay on the Roche Hitachi 917 analyzer (Roche Diagnostics, Indianapolis, Indiana).4 In addition, we found that elevated IgG concentrations, in the absence of a monoclonal protein, could interfere with the conjugated bilirubin assay; to our knowledge this is the first report of polyclonal immunoglobulin interference with a conjugated bilirubin assay. The mechanism for the conjugated bilirubin interference appears to be the precipitation of immunoglobulins after reagent addition, resulting in a white, flocculent turbidity that alters absorbance readings.1,2,4,5,22,23 We manually performed the conjugated bilirubin assay and confirmed the presence of a fine, white precipitate, using a sample containing a monoclonal protein, whereas no precipitate was observed for a sample from a control patient (data not shown). Similar results were also observed for a sample with elevated polyclonal immunoglobulins. Because the absorbance of the sample blank cuvette is subtracted from the absorbance of the reaction cuvette, differences in turbidity between the cuvettes can make the absorbance (reaction) appear either larger or smaller. For example, increased absorbance in the reaction cuvette because of turbidity could result in falsely increased conjugated bilirubin results, whereas increased absorbance in the sample blank cuvette because of turbidity could cause either falsely low or negative conjugated bilirubin results. Because the precipitation of a monoclonal protein and formation of turbidity are not reproducible and vary dramatically between tests, a wide Arch Pathol Lab Med—Vol 138, July 2014

range of errant conjugated bilirubin values can be obtained following repeat analysis. A spurious laboratory result can be confusing or misleading. It may trigger additional laboratory testing, other diagnostic tests, or even a wrong treatment plan. Thus, laboratories must rapidly identify errant test results and ensure clinicians are aware of the potential for incorrect laboratory test results because of interfering substances, such as monoclonal and polyclonal immunoglobulins. One approach to prospectively identify errant conjugated bilirubin results is to program the assay to flag results when an abnormal absorbance at a measurement point in the reaction curve is detected.3 That approach may be feasible for some instruments, such as Roche Modular instrumentation, which allows the introduction of flags in the reaction monitoring function (instruments with open channels) but is not applicable to Beckman Coulter AU5400 and AU2700 chemistry analyzers because the reaction curves are not reviewable by the operator. Because of our previous clinical experience identifying a sample with a conjugated bilirubin that was higher than the total bilirubin, we programmed the laboratory information management system to flag the operator and not to report conjugated bilirubin results for samples with the false-positive interference. Another flag was later introduced to identify falsely decreased conjugated bilirubin results that were negative. After introduction of the second flag, the interference rate for conjugated bilirubin during a 3-month period (16 310 bilirubin tests) increased from 0.3% to 1.1% (179 of 16 310). However, a limitation to this flagging system is that it will not detect the interference if a falsely decreased value is not negative or if a falsely increased value is less than the total bilirubin. To identify that type of interference, additional criteria will need to be established and/or a delta check system will need to be implemented. In conclusion, in addition to monoclonal proteins, polyclonal IgG can interfere with the Beckman Coulter AU5400 and AU2700 conjugated bilirubin assay. To identify errant conjugated bilirubin results, we recommend flagging negative conjugated bilirubin results and conjugated bilirubin results that exceed total bilirubin concentrations. Although this will identify most samples that interfere with the Beckman Coulter conjugated bilirubin assay, a few errant results may not be detected. For cases when the conjugated bilirubin is not flagged and the result is suspicious, repeat testing should be performed.1 In cases of a known interference, a conjugated bilirubin test should be performed by a different method that is known not to be affected by immunoglobulin interference. References 1. Yang Y, Howanitz PJ, Howanitz JH, Gorfajn H, Wong K. Paraproteins are a common cause of interferences with automated chemistry methods. Arch Pathol Lab Med. 2008;132(2):217–223. 2. Nauti A, Barassi A, Merlini G, Melzi d’Eril GV. Paraprotein interference in an assay of conjugated bilirubin. Clin Chem. 2005;51(6):1076–1077. 3. Smogorzewska A, Flood JG, Long WA, Dighe AS. Paraprotein interference in automated chemistry analyzers. Clin Chem. 2004;50(9):1691–1693. 4. Pantanowitz L, Horowitz GL, Upalakalin JN, Beckwith BA. Artifactual hyperbilirubinemia due to paraprotein interference. Arch Pathol Lab Med. 2003; 127(1):55–59. 5. Sheppard CA, Allen RC, Austin GE, Young AN, Ribeiro MA, Fantz CR. Paraprotein interference in automated chemistry analyzers. Clin Chem. 2005; 51(6):1077–1078. 6. Kwok JS-S, Chow KM, Lit LC-W, Chan MH-M. Paraproteinemia-associated pseudohypercreatininemia across different analytical methodologies. Kidney Int. 2010;78(6):621–622. 7. Lankireddy S, Ghandour F. Interference by IgG monoclonal protein in the enzymatic method for creatinine determination. Gundersen Lutheran Med J. 2007;4(2):76–78.

Interference in a Conjugated Bilirubin Assay—Song et al 953

8. Sinclair D, Smith H, Woodhead P. Spurious hyperphosphatemia caused by an IgA paraprotein: a topic revisited. Ann Clin Biochem. 2004;41(2):119–124. 9. Cohen AM, Magazanik A, van-der Lijn E, Shaked P, Levinsky H. Pseudohyperphosphataemia incidence in an automatic analyzer. Eur J Clin Chem Biochem. 1994;32(7):559–561. 10. John R, Oleesky D, Issa B, et al. Pseudohypercalcaemia in two patients with IgM paraproteinaemia. Ann Clin Biochem. 1997;34(6):694–696. 11. Tokmakjian S, Moses G, Haines M. Excessive sample blankings in two analyzers generate reports of apparent hypoglycemia and hypophosphatemia in patients with macroglobulinemia. Clin Chem. 1990;36(6):1261–1262. 12. Dimeski G, Carter A. Rare IgM interference with Roche/Hitachi modular glucose and -glutamyltransferase methods in heparin samples. Clin Chem. 2005; 51(11):2202–2204. 13. Smith JD, Nobiletti J, Freed M, Malkus M, Donabedian R. Interference with the Astra 8 and Synchron CX3 assays of urea nitrogen in serum by a high-M(r) inhibitor in a patient with multiple myeloma. Clin Chem. 1992;38(4):598–599. 14. Langman LJ, Allen LC, Romaschin AD. Interference of IgM paraproteins in the Olympus AU800 uric acid assay. Clin Biochem. 1998;31(7):517–521. 15. Bakker AJ. Influence of monoclonal immunoglobulins in direct determinations of iron in serum. Clin Chem. 1991; 37(5):690–695.

954 Arch Pathol Lab Med—Vol 138, July 2014

16. Wu LC, Chuang SS, Lin CN, Su CC, Lin MP. Multiple myeloma uncovered by investigating a negative serum iron level. J Clin Pathol. 2007;60(1):110. 17. Tsai LY, Tsai SM, Lee SC, Liu SF. Falsely low LDL-cholesterol concentrations and artifactual undetectable HDL-cholesterol measured by direct methods in a patient with monoclonal paraprotein. Clin Chim Acta. 2005;358(1–2):192– 195. 18. Baca A, Haber R, Sujishi K, Frost PH, Ng VL. Artifactual undetectable HDL-cholesterol with the Beckman Synchron LX and Vitros 950 assays temporally associated with a paraprotein. Clin Chem. 2004;50(1):255–256. 19. Bjorck L, Kronvall G. Purification and some properties of streptococcal protein G, a novel IgG-binding reagent. J. Immunol. 1984;133(2):969–974. 20. Zaman Z, Sneyers L, Van Orshoven A, Blanckaert N, Mari¨en G. Elimination of paraprotein interference in determination of plasma inorganic phosphate by ammonium molybdate method. Clin Chem. 1995;41(4):609–614. 21. Roy V. Artifactual laboratory abnormalities in patients with paraproteinemia. South Med J. 2009;102(2):167–170. 22. King RI, Florkowski CM. How paraproteins can affect laboratory assay: spurious results and biological effects. Pathology. 2010;42(5):397–401. 23. Bakker AJ, Mtcke M. Gammopathy interference in clinical chemistry ¨ assays: mechanisms, detection and prevention. Clin Chem Lab Med. 2007;45(9): 1240–1243.
