Lp(a) Lipoprotein, Vascular Disease, and Mortality in the Elderly
http://www.100md.com
《新英格兰医药杂志》
To the Editor: Ariyo et al. (Nov. 27 issue)1 show that Lp(a) lipoprotein is an important risk factor for stroke and death in elderly men. However, Lp(a) lipoprotein did not prove to be predictive of coronary artery disease, which is surprising for an atherosclerotic and prothrombotic risk factor. One explanation may be that patients with established coronary heart disease were excluded from the study.
We have used the opposite approach by including only patients with angiographically proven coronary artery disease. In our series of 451 nondiabetic patients, Lp(a) lipoprotein proved to be independently predictive of overall death as well as of death from cardiac causes. In the subgroup of 201 patients older than 65 years of age, Lp(a) lipoprotein was a significant predictor of death from any cause (odds ratio, 1.86 ; P=0.005) and of death from cardiac causes (odds ratio, 1.98 ; P=0.039) in as short a period of follow-up as 2.3 years (Figure 1). To the important data presented by Ariyo et al. we therefore can add that Lp(a) lipoprotein in elderly patients with coronary disease is an independent predictor of death from cardiac causes and death from any cause.
Figure 1. Standardized Adjusted Odds Ratios (and 95 Percent Confidence Intervals) for Lp(a) Lipoprotein as a Predictor of Death from Any Cause and Death from Cardiac Causes in Nondiabetic Patients 65 Years of Age or Older.
Christoph H. Saely, M.D.
Thomas Marte, M.D.
Heinz Drexel, M.D.
Vorarlberg Institute for Vascular Investigation and Treatment
6800 Feldkirch, Austria
vivit@lkhf.at
References
Ariyo AA, Thach C, Tracy R. Lp(a) lipoprotein, vascular disease, and mortality in the elderly. N Engl J Med 2003;349:2108-2115.
To the Editor: In the study by Ariyo et al., the serum level of Lp(a) lipoprotein predicted death from any cause, death from cardiovascular disease, and stroke in elderly men but not women. A possible explanation for this discrepancy is that the serum level of Lp(a) lipoprotein alone may not be sufficient to characterize fully the cardiovascular risk associated with Lp(a) lipoprotein. In a study of patients undergoing dialysis, a low-molecular-weight apolipoprotein (a) phenotype, but not the level of Lp(a) lipoprotein, was associated with the risk of death.1 Small apolipoprotein (a) isoforms were also a predictor of advanced atherosclerotic lesions in a population-based study.2 Small apolipoprotein (a) isoforms were shown to bind more strongly to fibrin than larger isoforms3; hence, the size of the apolipoprotein (a) isoform, in addition to the level of Lp(a) lipoprotein, may play a role in cardiovascular disease. When exploring outcomes related to Lp(a) lipoprotein, apolipoprotein (a) isoforms should also be measured, since different apolipoprotein (a) phenotypes could result in heterogeneous effects, even in the presence of similar serum Lp(a) lipoprotein levels.
Csaba P. Kovesdy, M.D.
Salem Veterans Affairs Medical Center
Salem, VA 24153
csaba.kovesdy@med.va.gov
References
Longenecker JC, Klag MJ, Marcovina SM, et al. Small apolipoprotein(a) size predicts mortality in end-stage renal disease: the CHOICE study. Circulation 2002;106:2812-2818.
Kronenberg F, Kronenberg MF, Kiechl S, et al. Role of lipoprotein(a) and apolipoprotein(a) phenotype in atherogenesis: prospective results from the Bruneck study. Circulation 1999;100:1154-1160.
Hervio L, Girard-Globa A, Durlach V, Angle-Cano E. The antifibrinolytic effect of lipoprotein(a) in heterozygous subjects is modulated by the relative concentration of each of the apolipoprotein(a) isoforms and their affinity for fibrin. Eur J Clin Invest 1996;26:411-417.
The authors reply: In results in line with those we report, Dr. Saely and colleagues found that Lp(a) lipoprotein among the elderly constitutes an independent risk factor for vascular diseases and death from any cause. We excluded participants with established vascular diseases (including coronary disease) from the analyses in order to explore the predictive ability of Lp(a) lipoprotein in healthy people. We have not performed an analysis of recurrent events or events in people with established cardiovascular disease.
Because of the genetics of Lp(a) lipoprotein, the profile of inflammatory cytokines produced, or both, Lp(a) lipoprotein levels fluctuate in disease states1,2 and may increase after cardiac events.2 Given this complexity, patients with angiographically confirmed coronary disease may have elevated Lp(a) lipoprotein levels as a result of their extensive coronary disease or a recent cardiac event. On the basis of the data presented by Dr. Saely and colleagues, it is difficult to assess the temporal relation between the Lp(a) lipoprotein level and the development of coronary disease. Prospective studies, such as ours, are needed to assess the temporal relation between exposure and the outcome after exposure. Although their study and ours differ in terms of study design and population, it is reassuring that among elderly persons with established coronary disease, they found that the Lp(a) lipoprotein level was consistently and positively associated with death from cardiac causes and death from all causes.
We thank Dr. Kovesdy for his comments and agree that research on Lp(a) lipoprotein isoforms may prove informative. Unfortunately, when we performed our assays, isoform analysis was not readily available. The pathogenic effect of Lp(a) lipoprotein is complex and multifactorial. For example, we have learned that it may be misleading to attribute its atherogenicity to fibrin binding alone, since enzymatically cleaved fragments of Lp(a) lipoprotein are now known to bind to matrix proteins. Another component of Lp(a) lipoprotein — the small, dense, low-density lipoprotein — has also been shown to be an independent risk factor for coronary disease.3
Furthermore, Dr. Kovesdy cites a study4 that found that small apolipoprotein (a) isoforms, but not the Lp(a) lipoprotein level, predict death. The study was conducted among patients undergoing dialysis, and the relation of this group to otherwise healthy older adults is uncertain. Moreover, the same authors have also reported data from the same cohort that point to an association between the Lp(a) lipoprotein level and coronary artery disease.5
Obviously, the pathobiology of Lp(a) lipoprotein is complicated. In his accompanying Perspective article, Dr. Scanu6 describes in detail the elaborate heterogeneity of Lp(a) lipoprotein as found in plasma and tissues. Further research that focuses on Lp(a) lipoprotein and all its components, not just apolipoprotein (a), will help improve our understanding of this complex particle.
Abraham A. Ariyo, M.D., M.P.H.
HeartMasters
Dallas, TX 75219
Russell P. Tracy, Ph.D.
University of Vermont
Cochester, VT 05446
Chau Thach, Ph.D.
Merck
Rahway, NJ 07065
References
Andreassen AK, Berg K, Torsvik H. Changes in Lp(a) lipoprotein and other plasma proteins during acute myocardial infarction. Clin Genet 1994;46:410-416.
Min WK, Lee JO, Huh JW. Relation between lipoprotein(a) concentrations in patients with acute-phase response and risk analysis for coronary heart disease. Clin Chem 1997;43:1891-1895.
Austin MA, Breslow JL, Hennekens CH, Buring JE, Willett WC, Krauss RM. Low-density lipoprotein subclass patterns and risk of myocardial infarction. JAMA 1988;260:1917-1921.
Longenecker JC, Klag MJ, Marcovina SM, et al. Small apolipoprotein(a) size predicts mortality in end-stage renal disease: the CHOICE study. Circulation 2002;106:2812-2818.
Longenecker JC, Coresh J, Marcovina SM, et al. Lipoprotein(a) and prevalent cardiovascular disease in a dialysis population: the Choices for Healthy Outcomes in Caring for ESRD (CHOICE) study. Am J Kidney Dis 2003;42:108-116.
Scanu AM. Lp(a) lipoprotein -- coping with heterogeneity. N Engl J Med 2003;349:2089-2090.
We have used the opposite approach by including only patients with angiographically proven coronary artery disease. In our series of 451 nondiabetic patients, Lp(a) lipoprotein proved to be independently predictive of overall death as well as of death from cardiac causes. In the subgroup of 201 patients older than 65 years of age, Lp(a) lipoprotein was a significant predictor of death from any cause (odds ratio, 1.86 ; P=0.005) and of death from cardiac causes (odds ratio, 1.98 ; P=0.039) in as short a period of follow-up as 2.3 years (Figure 1). To the important data presented by Ariyo et al. we therefore can add that Lp(a) lipoprotein in elderly patients with coronary disease is an independent predictor of death from cardiac causes and death from any cause.
Figure 1. Standardized Adjusted Odds Ratios (and 95 Percent Confidence Intervals) for Lp(a) Lipoprotein as a Predictor of Death from Any Cause and Death from Cardiac Causes in Nondiabetic Patients 65 Years of Age or Older.
Christoph H. Saely, M.D.
Thomas Marte, M.D.
Heinz Drexel, M.D.
Vorarlberg Institute for Vascular Investigation and Treatment
6800 Feldkirch, Austria
vivit@lkhf.at
References
Ariyo AA, Thach C, Tracy R. Lp(a) lipoprotein, vascular disease, and mortality in the elderly. N Engl J Med 2003;349:2108-2115.
To the Editor: In the study by Ariyo et al., the serum level of Lp(a) lipoprotein predicted death from any cause, death from cardiovascular disease, and stroke in elderly men but not women. A possible explanation for this discrepancy is that the serum level of Lp(a) lipoprotein alone may not be sufficient to characterize fully the cardiovascular risk associated with Lp(a) lipoprotein. In a study of patients undergoing dialysis, a low-molecular-weight apolipoprotein (a) phenotype, but not the level of Lp(a) lipoprotein, was associated with the risk of death.1 Small apolipoprotein (a) isoforms were also a predictor of advanced atherosclerotic lesions in a population-based study.2 Small apolipoprotein (a) isoforms were shown to bind more strongly to fibrin than larger isoforms3; hence, the size of the apolipoprotein (a) isoform, in addition to the level of Lp(a) lipoprotein, may play a role in cardiovascular disease. When exploring outcomes related to Lp(a) lipoprotein, apolipoprotein (a) isoforms should also be measured, since different apolipoprotein (a) phenotypes could result in heterogeneous effects, even in the presence of similar serum Lp(a) lipoprotein levels.
Csaba P. Kovesdy, M.D.
Salem Veterans Affairs Medical Center
Salem, VA 24153
csaba.kovesdy@med.va.gov
References
Longenecker JC, Klag MJ, Marcovina SM, et al. Small apolipoprotein(a) size predicts mortality in end-stage renal disease: the CHOICE study. Circulation 2002;106:2812-2818.
Kronenberg F, Kronenberg MF, Kiechl S, et al. Role of lipoprotein(a) and apolipoprotein(a) phenotype in atherogenesis: prospective results from the Bruneck study. Circulation 1999;100:1154-1160.
Hervio L, Girard-Globa A, Durlach V, Angle-Cano E. The antifibrinolytic effect of lipoprotein(a) in heterozygous subjects is modulated by the relative concentration of each of the apolipoprotein(a) isoforms and their affinity for fibrin. Eur J Clin Invest 1996;26:411-417.
The authors reply: In results in line with those we report, Dr. Saely and colleagues found that Lp(a) lipoprotein among the elderly constitutes an independent risk factor for vascular diseases and death from any cause. We excluded participants with established vascular diseases (including coronary disease) from the analyses in order to explore the predictive ability of Lp(a) lipoprotein in healthy people. We have not performed an analysis of recurrent events or events in people with established cardiovascular disease.
Because of the genetics of Lp(a) lipoprotein, the profile of inflammatory cytokines produced, or both, Lp(a) lipoprotein levels fluctuate in disease states1,2 and may increase after cardiac events.2 Given this complexity, patients with angiographically confirmed coronary disease may have elevated Lp(a) lipoprotein levels as a result of their extensive coronary disease or a recent cardiac event. On the basis of the data presented by Dr. Saely and colleagues, it is difficult to assess the temporal relation between the Lp(a) lipoprotein level and the development of coronary disease. Prospective studies, such as ours, are needed to assess the temporal relation between exposure and the outcome after exposure. Although their study and ours differ in terms of study design and population, it is reassuring that among elderly persons with established coronary disease, they found that the Lp(a) lipoprotein level was consistently and positively associated with death from cardiac causes and death from all causes.
We thank Dr. Kovesdy for his comments and agree that research on Lp(a) lipoprotein isoforms may prove informative. Unfortunately, when we performed our assays, isoform analysis was not readily available. The pathogenic effect of Lp(a) lipoprotein is complex and multifactorial. For example, we have learned that it may be misleading to attribute its atherogenicity to fibrin binding alone, since enzymatically cleaved fragments of Lp(a) lipoprotein are now known to bind to matrix proteins. Another component of Lp(a) lipoprotein — the small, dense, low-density lipoprotein — has also been shown to be an independent risk factor for coronary disease.3
Furthermore, Dr. Kovesdy cites a study4 that found that small apolipoprotein (a) isoforms, but not the Lp(a) lipoprotein level, predict death. The study was conducted among patients undergoing dialysis, and the relation of this group to otherwise healthy older adults is uncertain. Moreover, the same authors have also reported data from the same cohort that point to an association between the Lp(a) lipoprotein level and coronary artery disease.5
Obviously, the pathobiology of Lp(a) lipoprotein is complicated. In his accompanying Perspective article, Dr. Scanu6 describes in detail the elaborate heterogeneity of Lp(a) lipoprotein as found in plasma and tissues. Further research that focuses on Lp(a) lipoprotein and all its components, not just apolipoprotein (a), will help improve our understanding of this complex particle.
Abraham A. Ariyo, M.D., M.P.H.
HeartMasters
Dallas, TX 75219
Russell P. Tracy, Ph.D.
University of Vermont
Cochester, VT 05446
Chau Thach, Ph.D.
Merck
Rahway, NJ 07065
References
Andreassen AK, Berg K, Torsvik H. Changes in Lp(a) lipoprotein and other plasma proteins during acute myocardial infarction. Clin Genet 1994;46:410-416.
Min WK, Lee JO, Huh JW. Relation between lipoprotein(a) concentrations in patients with acute-phase response and risk analysis for coronary heart disease. Clin Chem 1997;43:1891-1895.
Austin MA, Breslow JL, Hennekens CH, Buring JE, Willett WC, Krauss RM. Low-density lipoprotein subclass patterns and risk of myocardial infarction. JAMA 1988;260:1917-1921.
Longenecker JC, Klag MJ, Marcovina SM, et al. Small apolipoprotein(a) size predicts mortality in end-stage renal disease: the CHOICE study. Circulation 2002;106:2812-2818.
Longenecker JC, Coresh J, Marcovina SM, et al. Lipoprotein(a) and prevalent cardiovascular disease in a dialysis population: the Choices for Healthy Outcomes in Caring for ESRD (CHOICE) study. Am J Kidney Dis 2003;42:108-116.
Scanu AM. Lp(a) lipoprotein -- coping with heterogeneity. N Engl J Med 2003;349:2089-2090.