Current issues of ACP Journal Club are published in Annals of Internal Medicine


Apolipoproteins were predictive of coronary artery disease in men

ACP J Club. 1992 Sept-Oct;117:60. doi:10.7326/ACPJC-1992-117-2-060

Source Citation

Sigurdsson G, Baldursdottir A, Sigvaldason H, et al. Predictive value of apolipoproteins in a prospective survey of coronary artery disease in men. Am J Cardiol. 1992 May 15;69:1251-4.



To determine whether elevated levels of apolipoproteins (apo[a], apo-B, and apo-AI) are independent risk factors for myocardial infarction and coronary death in men aged 45 to 72 years.


Cohort analytic study with a mean follow-up of 8.6 years.


Reykjavik, Iceland.


1332 Icelandic men (65% of a population-based sample who were invited, mean age 58 y) who participated in a health survey from 1979 to 1981.

Assessment of risk factors

Serum samples were frozen for 8 years and subsequently thawed and analyzed for apo(a), apo-B, and apo-AI. Apo(a) and apo-B were determined with a 2-site immunoradiometric assay using 2 different monoclonal antibodies in excess (Pharmacia Diagnostics AB, Uppsala, Sweden). Apo-AI was determined with a competitive radioimmunoassay using specific monoclonal antibodies (Pharmacia Diagnostics AB).

Main outcome measures

Fatal and nonfatal myocardial infarction. A registry that contained all episodes of acute myocardial infarction for all of Iceland from 1980 was used. The diagnostic criteria included symptoms, electrocardiograms, enzymes, and necropsy findings. Causes of death were determined by the review of death certificates and, when available autopsy reports (59% of deaths).

Main results

104 of the 1332 participants (8%) had a fatal or nonfatal myocardial infarction. Proportional hazards analysis, considering apo(a), apo-B, apo-AI, total cholesterol, triglycerides, systolic blood pressure, and age as independent variables, showed that apo(a) was an independent risk factor for coronary artery disease (odds ratio [OR] 1.22 per 1 standard deviation above or below the mean value, P = 0.014) and that apo-AI was a strong negative risk factor (OR 0.70/1 SD, P < 0.001). A strong correlation was found between apo-B and total serum cholesterol which resulted in apo-B being an independent risk factor for coronary artery disease only when substituted for total cholesterol (OR 1.32/1 SD, P < 0.001).


Apolipoproteins, apo-AI and apo(a), were independent risk factors for myocardial infarction in men aged 45 to 72 years. Apolipoprotein-B was highly correlated with total serum cholesterol and did not provide any additional predictive value.

Source of funding: Not stated.

Address for article reprint: Dr. G. Sigurdsson, Department of Medicine, Reykjavik City Hospital, 108 Reykjavik, Iceland.


The study by Sigurdsson and colleagues is one of the first large cohort studies to evaluate the role of lipoprotein(a) (Lp[a]), measured as apo(a), as an independent risk factor for myocardial infarction. The study also looked at apo-AI, which accounts for 75% of the protein component of high-density-lipoprotein (HDL) cholesterol, and at apo-B, the main apolipoprotein associated with non-HDL cholesterol, as risk factors for myocardial infarction. Some concern about the stability of apo(a) frozen at -20°C for 8 years is warranted in considering the validity of the study.

The study shows the importance of Lp(a) as an independent risk factor for coronary heart disease. At present no established therapy exists to lower Lp(a) levels. Gavish and colleagues (1) and Carlson and colleagues (2), however, showed that N-acetylcysteine and niacin, respectively, lowered Lp(a) in a small number of patients.

Because apo-AI was highly correlated with HDL (r = 0.7) and apo-B with total cholesterol (r = 0.7) in the study, measurement of apolipoprotein levels provided no clear benefit over standard lipid profiles for predicting the risk of myocardial infarction in these middle-aged men. These findings may not be true for women because Kweiterovich and colleagues (3) showed plasma apo-B levels to be much better independent predictors of coronary heart disease in women than in men.

Measurement of total cholesterol, triglycerides, and HDL cholesterol provides adequate screening for men at high risk for coronary heart disease, but for women the best predictors are less clear. Until an adequate treatment for elevated lipoprotein(a) levels is available, measurement of apo(a) or Lp(a) should remain a test for specialized laboratories and research protocols.

Charles B. Eaton, MD, MS
Brown University School of MedicinePawtucket, Rhode Island, USA

Charles B. Eaton, MD, MS
Brown University School of Medicine
Pawtucket, Rhode Island, USA


1. Gavish D, Breslow JL. Lancet. 1991;337:203-4.

2. Carlson LA, Hamsten A, Asplund A. J Intern Med. 1989;226:271-6.

3. Kwiterovich PO, Coresh J, Smith HH, et al. Am J Cardiol. 1992;69:1015-21