Correspondence

Case 40-2006: Anemia and Low HDL Cholesterol

N Engl J Med 2007; 356:1893-1895May 3, 2007

Article

To the Editor:

In the Case Record of a patient with anemia and a low level of high-density lipoprotein (HDL) cholesterol, presented by Murali and colleagues (Dec. 28 issue),1 the discussant states that on rare occasions the presence of a paraprotein interferes with the measurement of HDL cholesterol, wrongly suggesting a low level. The discussant also states that the low levels of HDL cholesterol were due to genetic factors or were secondary to specific hormones, cigarette smoking, obesity, a low-fat diet, and drugs. Although all these secondary factors are found frequently, the most frequent cause of a low level of HDL cholesterol in hospitalized patients is simply an acute or chronic severe inflammatory disease with a marked acute-phase reaction. Like albumin and transferrin, HDL cholesterol drops and returns to normal within days after the end of the acute-phase reaction. Negative tests for acute-phase proteins are often overlooked as excellent markers of ongoing inflammation, and too often, a low albumin level is immediately regarded as evidence of malnutrition and a low HDL cholesterol level as evidence of a cardiovascular risk factor necessitating intervention. One should remember the complex changes that occur in lipoproteins during the acute-phase reaction in a patient with inflammation or infection.2,3

Jürg A. Schifferli, M.D.
University Hospital Basel, 4031 Basel, Switzerland
j.schifferli@unibas.ch

3 References

1. 1

   Case Records of the Massachusetts General Hospital (Case 40-2006). N Engl J Med 2006;355:2772-2779
   Full Text | Web of Science | Medline

2. 2

   Van Lenten BJ, Reddy ST, Navab M, Fogelman AM. Understanding changes in high density lipoproteins during the acute phase response. Arterioscler Thromb Vasc Biol 2006;26:1687-1688
   CrossRef | Web of Science | Medline

3. 3

   Khovidhunkit W, Memon RA, Feingold KR, Grunfeld C. Infection and inflammation-induced proatherogenic changes of lipoproteins. J Infect Dis 2000;181:Suppl 3:S462-S472
   CrossRef | Web of Science | Medline

To the Editor:

Murali and colleagues report on a patient with a paraproteinemia who was found to have an artifactually unmeasurable level of HDL cholesterol on a liquid-based HDL-cholesterol assay. Although the results were normal with the use of a non–liquid-based assay — presumably, involving the precipitation technique — they refer to a case report suggesting that both kinds of assays may produce artifacts, since readings of very low HDL cholesterol levels were reported with the use of both methods in patients with paraproteinemia.1 Since an independent method such as ultracentrifugation was not used, it is possible that the case report they cite and others1,2 represent true HDL cholesterol deficiency associated with paraproteinemia, a phenomenon that Murali et al. do not consider in their discussion. In 2002, we reported on two patients with paraproteinemia and very low HDL cholesterol levels that were confirmed on the basis of ultracentrifugation, electrophoresis, and apolipoprotein A-I levels.3 We showed that immunoglobulin was bound to apolipoprotein A-I and that the immunoglobulin–HDL complexes were rapidly degraded by cultured macrophages (Figure 1Figure 1Binding of HDL to IgG in a Patient with Paraproteinemia but Not in a Control Subject.). Thus, the finding of a very low HDL cholesterol level in a patient with paraproteinemia may in some cases not be artifactual.

Ronald B. Goldberg, M.D.
Armando J. Mendez, Ph.D.
University of Miami Miller School of Medicine, Miami, FL 33136
rgoldber@med.miami.edu

3 References

1. 1

   Baca A, Haber RJ, 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;50:255-256
   CrossRef | Web of Science | Medline

2. 2

   Hachem H, Favre G, Ghalim N, Puchois P, Fruchart JC, Soula G. Quantitative abnormalities of lipoprotein particles in multiple myeloma. J Clin Chem Clin Biochem 1987;25:675-679
   Medline

3. 3

   Mendez AJ, Goldberg RB, Arnold PI, Schultz DR. Acquired HDL deficiency associated with apolipoprotein A-I reactive monoclonal immunoglobulins. Arterioscler Thromb Vasc Biol 2002;22:1740-1741
   CrossRef | Web of Science

Author/Editor Response

Schifferli appropriately comments that acute or chronic inflammatory diseases, with a marked acute-phase response, can lower the level of HDL cholesterol. Our patient had no clinical or laboratory features of acute or chronic inflammation. However, Schifferli's comment does reinforce the increasing awareness of lipid abnormalities in inflammatory processes.1

Figure 2 in the Supplementary Appendix of this Case Record highlights the mechanism of falsely low levels of HDL cholesterol in liquid-based assays but not in solid-phase assays. Since the discrepancy was resolved in this Case Record, the need for ultracentrifugal studies, as suggested by Goldberg and Mendez, did not arise.

In contrast to the patients described by Mendez and colleagues,2 who had IgG monoclonal gammopathy and low levels of HDL cholesterol, our patient had IgM monoclonal gammopathy. Since macrophages have receptors for the Fc portion of IgG but not for IgM, enhanced catabolism of apolipoprotein-A-I–IgG complexes by macrophages does not explain the low HDL cholesterol levels in our patient. In fact, Goldberg reinforces our observation that when confronted with unexplained low levels of HDL cholesterol, clinicians and clinical pathologists need to rule out a paraproteinemia,2 using appropriate laboratory methods to ensure that the observed low HDL cholesterol levels are not due to paraprotein-associated interactions or interference in assay systems.

Mandakolathur R. Murali, M.D.
Massachusetts General Hospital, Boston, MA 02114

2 References

1. 1

   Munford RS. Statins and the acute-phase response. N Engl J Med 2001;344:2016-2018
   Full Text | Web of Science | Medline

2. 2

   Mendez AJ, Goldberg RB, Arnold PI, Schultz DR. Acquired HDL deficiency associated with apolipoprotein A-I reactive monoclonal immunoglobulins. Arterioscler Thromb Vasc Biol 2002;22:1740-1741
   CrossRef | Web of Science

Citing Articles (1)

Citing Articles

1. 1

   Edwin O. F. Gorselen, Theo Diekman, Jan Hessels, Patrick M. J. Verhorst, Clemens Birgelen. (2010) Artifactual measurement of low serum HDL-cholesterol due to paraproteinemia. Clinical Research in Cardiology 99:9, 599-602
   CrossRef