| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
Departments of1
Laboratory Medicine and2
Medicine University of California San Francisco San Francisco, CA 94110
3 Clinical Laboratory University of California San Francisco Medical Center San Francisco, CA 94110
aAuthor for correspondence.
To the Editor:
We report here the failure of the Synchron LX (Beckman) and Vitros 950 (Johnson & Johnson) assays to detect HDL-cholesterol (HDL-C) in the presence of a paraprotein.
HDL-C was repeatedly undetectable (<50 mg/L) in a male who historically had HDL-C values in the 600–800 mg/L range, as determined by manual dextran sulfate precipitation followed by measurement of HDL-C on a BMC (Roche) Hitachi 747, during a time period that coincided with a change in laboratory HDL-C test methodology to the Synchron LX HDL-C (Beckman). The specimens with undetectable HDL-C values in the Synchron LX assay were not hemolyzed or icteric, two conditions known to decrease HDL-C values in the Synchron LX assay. A review of laboratory test results for this patient at the most recent visit for which his HDL-C was undetectable revealed a newly identified paraprotein (IgG
; Table 1
, specimen 1a) and a new diagnosis of multiple myeloma.
|
HDL-C values (heparin–manganese precipitation method) have been noted to generally be lower in the sera of patients with multiple myeloma than in age- and sex-matched controls (1). Paraproteins themselves have been noted to interfere with HDL-C determinations in the Roche HDL-C Plus assay (polyethylene glycol-modified enzymes, sulfated 
-cyclodextrin, and dextran sulfate method), and the interference is not predictable or correlative with paraprotein isotype or quantity (2). We were unable to locate published information regarding paraprotein interference with the Synchron LX HDL-C assay.
The frequency and magnitude of paraprotein interference was determined with a panel of 19 paraprotein-containing specimens tested on the Synchron LX (Beckman; preferential detergent solubilization and release of HDL-associated cholesterol) and Vitros 950 (Johnson & Johnson; magnetic separation of non-HDL lipoproteins precipitated by dextran sulfate and magnesium chloride, followed by measurement of HDL-C in the supernatant) assays (Table 1
, specimens 1a and 2–19). Of note, three specimens differing in paraprotein isotype and quantity (specimens 1a, 11, and 17) yielded undetectable HDL-C values with the Synchron LX method and different values (including an undetectable HDL-C for specimen 1a) with the Vitros 950 method. Differences in HDL-C values between the two assays were observed for the remaining 16 specimens, were not correlated with paraprotein isotype or quantity, and sometimes exceeded the 5–9% interlaboratory interassay CV (3) and 10–20% physiologic variability (4).
The index patient returned for laboratory evaluation after successful treatment for multiple myeloma. His previously undetectable HDL-C was now readily detectable and concordant with his previous values (Table 1
, specimen 1b), and his paraprotein concentration was dramatically reduced. The temporal association and inverse correlation between his paraprotein and HDL-C values support the hypothesis that his paraprotein was directly responsible for the interference in the HDL-C assay.
Paraproteins are a well-recognized cause of lipoprotein abnormalities, including hyper- and hypolipidemia associated with low LDL and HDL. In one study, 60% of patients with monoclonal gammopathies had an extra band on plasma lipoprotein electrophoresis, which was attributed to an immunoglobulin–lipid complex (5). Another group reported that HDL particles from myeloma patients had decreased apolipoprotein AI and AII and contained an additional protein identified as serum amyloid A (6). These observations suggest that some paraproteins may associate with and alter the physicochemical characteristics of HDL particles, affecting their behavior in assays designed to directly assay HDL-C.
Undetectable HDL-C values are an extremely rare occurrence; only 16 (0.06%) of 24711 HDL-C measurements in the University of California San Francisco Medical Center Clinical Laboratory for calendar year 2002 yielded undetectable values. In contrast, we identified paraprotein interference associated with undetectable HDL-C values in 3 (16%) of 19 paraprotein-containing specimens, a much higher (250-fold) prevalence than observed in the general population. Thus, it is very likely that a specimen with a low or undetectable HDL-C may be from a patient with paraproteinemia.
Our investigation prompted us to implement an internal laboratory protocol in which any specimen with an undetectable HDL-C automatically undergoes evaluation for a monoclonal gammopathy. Since implementation of this protocol, we have identified an unsuspected monoclonal gammopathy in a patient who had multiple low HDL-C values (range, 90–160 mg/L) and one undetectable HDL-C value as measured on the Synchron LX, when simultaneous measurements by other HDL-C methods, i.e., heparin–manganese precipitation or proprietary HDL-C measurements on serum fractionated by ultracentrifugation [personal communication from Atherotech (Birmingham, AL) regarding the VAP cholesterol test®], yielded values ranging from 470 to 580 mg/L. This individual had an unrecognized IgM paraprotein at a concentration of 21 g/L, comprising 24% of his total serum protein (86 g/L). A clinical hematologic evaluation, including a bone marrow examination that revealed 7% plasma cells, led to the diagnosis of monoclonal gammopathy of unknown significance.
Our limited study does not allow us to speculate on the frequency at which low, but still detectable, HDL-C concentrations are attributable to interference from unsuspected paraproteins. We are concerned, however, that this interference may occur more frequently than suspected, given that the two assays we identified as susceptible to paraprotein interference are used by 396 (10%) and 213 (5%), respectively, of the 3971 laboratories participating in the College of American Pathologists 2003 C-B HDL-C survey (7).
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. A. Schifferli, R. B. Goldberg, A. J. Mendez, and M. R. Murali Case 40-2006: Anemia and Low HDL Cholesterol N. Engl. J. Med., May 3, 2007; 356(18): 1893 - 1895. [Full Text] [PDF]
|
|
|
|
|
|
M. R. Murali, A. Kratz, and K. E. Finberg Case 40-2006 -- A 64-Year-Old Man with Anemia and a Low Level of HDL Cholesterol N. Engl. J. Med., December 28, 2006; 355(26): 2772 - 2779. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Garcia-Hejl, P. Vest, C. Renard, A. Merens-Gonthier, A. Boukhira, and H. Thefenne-Astier Falsely Low LDL Cholesterol Results and Cholestasis. Clin. Chem., November 1, 2006; 52(11): 2125 - 2127. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
