topeesp.gif (5672 bytes)

[ Scientific Activities - Actividades Científicas ]

How to Manage the New Cardiovascular Risk Factors?: Lipoprotein (a)

José Pablo Werba, MD
Chief of the Center for Dyslipidemias and Early Detection of Atherosclerosis
ICYCC-Favaloro Foundation and Favaloro University
Buenos Aires, Argentina

Is it neccesary or beneficial to reduce Lp(a) levels?
Is it then possible to favourably modify Lp(a) levels?
Then, why measure Lp(a) and what to do with
those patients with high levels today?

Much has been written about Lipoprotein(a) [Lp(a)] in this ending decade. Since commercial kits to measure Lp(a) in serum became widely available (even though not necessarily a well standarized determination), scientists have found it relatively simple to perform any kind of epidemiological and clinical survey on Lp(a). Although these studies have yielded an enormous bulk of relevant information, it is fair to anticipate in this presentation that there is still no conclusive answer to the main question: how to manage Lp(a)?. Having said that, let me quickly comment on some well-stablished issues about Lp(a) features and modifications, intentionally putting aside the many controversies that this puzzling particle has raised.

Lipoprotein (a) [Lp(a)] has been originally identified as a genetic variant of low density lipoproteins (LDL), which contains apolipoprotein B-100, linked by a disulphide bond to apo(a), a protein structurally related to plasminogen.

The interest in Lp(a) (a lipoprotein found only in primates and hedgehogs) has risen considerably since several case-control studies have demonstrated that elevated plasma Lp(a) levels are associated with an increased risk for coronary, cerebral and peripheral atherosclerosis. The mechanism by which Lp(a) would exert a pathogenic effect has been the subject of intensive study and either pro-atherogenic and/or pro-thrombotic mechanisms have been proposed.

In the general population there is a very wide range of plasma Lp(a) levels (from 0 to more than 300 mg/dl) and the distribution is usually highly skewed, with most people in the low range and with little variation within each individual. This distribution is clearly shifted to the right not only in patients with clinical atherosclerosis but also in subjects with subclinical disease. The Lp(a) value for risk has been settled at 30 mg/dl on the basis of most case-control studies. Some years ago, using this threshold, my colegues and I reported that hyperlipidemic subjects with plasma Lp(a) levels > 30 mg/dl attending an Italian lipid clinic showed a significantly higher common carotid intima-medial thickness (ccIMT) than those with Lp(a) < 30 mg/dl, whatever the number of further traditional risk factors they were exposed to. Most important, a continuation of those studies was recently published showing that Lp(a) is positively associated with carotid IMT exclusively in those subjects with very high LDL levels. These findings are in line with earlier reports showing that the association of Lp(a) with angiographically defined coronary heart disease is limited to subjects with high LDL levels. They also closely agree with recent data from our lipid clinic demonstrating an increased prevalence of CHD in hyperlipidemic subjects with high Lp(a) levels with an OR of 2.62 (1.15-5.99; p = 0.02). The fact that Lp(a) pathogenicity is influenced by LDL levels may be important in terms of lipid management, as I’ll comment later on.

Lp(a) levels are also increased in clinical conditions typically and highly associated with an increased risk of atherosclerotic disease such as NIDDM (mainly if complicated with urinary protein loss) and chronic renal failure (CRF; before and in-dialysis). So, although it is well established that Lp(a) levels are strongly determined by genetic factors (the major factor is the genotype of apo (a) and the minor determinants are the genotypes of apo E and the B:E receptor), some acquired pathologies or even environmental (i.e ethanol consumption may lower Lp(a)) conditions may modify Lp(a) levels.

Metabolic studies suggest that Lp(a) levels are predominantly governed by its rate of synthesis and secretion by the liver and that the kidney tissue may play a considerable role in its catabolism. The latter is supported by the clinical observation that subjects with terminal CRF on haemodialysis show a further increase in Lp(a) levels after the surgical removal or residual kidney for refractory hypertension. This information might be useful for developing strategies aimed at favourably modifying plasma Lp(a) levels.

Is it neccesary or beneficial to reduce Lp(a) levels?

Although we don’t really know, we think it is worth reducing Lp(a) levels on the basis of epidemiological and physiopathological studies. In fact, no trials could be performed yet to investigate the effect of Lp(a) lowering on CV events inasmuch as no single intervention was developed to modify plasma Lp(a) significantly and selectively, with the possible exception of "Lp(a) apheresis with Lp(a) binding columns". However, and to the best of my knowledge, this is a procedure which has not yet been explored in terms of clinical outcomes.

Is it then possible to favourably modify Lp(a) levels?

To some extent, yes. The usual "prudent diet" or even more strict lipid lowering regimens do not modify Lp(a) levels. However, in hypertrigliceridemic subjects, some Lp(a) lowering could be achieved by using omega-3 fatty acids, compounds that decrease the synthesis and secretion of triglyceride-rich lipoproteins by the liver. However, myself and others have reported that triglycerides and Lp(a) are inversely correlated and thus, the condition of both high Lp(a) levels and hypertriglyceridemia is highly unusual.

Nicotinic acid (NA), a drug that mainly affects lipoprotein synthesis in the liver, reduces not only triglycerides (TG), total colesterol (TC) and LDL-C but also Lp(a) significantly, mainly at high doses. However, the use of NA is limited in clinical practice due to its frequent side effects. NA derivatives (acipimox, inositol hexanicotinate, and others) are more tolerable but less active in terms of lipid-lowering action. We have ourselves evaluated acipimox in hyperlipidemic subjects and we didn’ t find any clinically significant Lp(a) changes. Early long-release formulations of NA were more tolerable but highly unrecommended on the basis of the risk of severe liver toxicity. More modern and long release formulations with theoretical better tolerability have been recently developed and launched in the U.S. and might be of value in Lp(a) control, but this needs to be proved.

Commonly used hypolipidemic drugs, including cholestyramine, fibrates (except bezafibrate to some extent) and statins do not significantly modify plasma Lp(a) levels. Furthermore, some reports show that some of the latter may slightly increase them. The biological and clinical significance of this unwanted effect is uncertain, as many of these drugs (including those that like Simvastatin may raise Lp(a)) have shown a clear benefit in terms of CHD morbidity and mortality. Secondarily, the lack of an Lp(a) lowering effect of statins gave a clue to understand that LDL (apoB:E) receptors, which are upregulated by statins, are not mainly involved in Lp(a) catabolism or regulation.

It used to be speculated that the transformation of Lp(a)+ into Lp(a)- particles (without apo (a) and so equivalent to LDL) might have helped to remove Lp(a) from the circulation through apo B:E receptors in the liver. Although initial case reports have shown an Lp(a)-lowering effect by the reducing and mucolytic agent N-acetylcystein, we didn’ t find any change in Lp(a) levels in hyperlipidemic subjects with hyperLp(a) (more than 40 mg/dl), using a schedule of graded doses of the drug. Neither did carbocisteine, a related compound, produce any changes in other studies.

Estrogen-progestin formulations, the former either oral or transdermal, may induce moderate plasma Lp(a) reductions. Many important studies on this topic were performed by Maurizio Soma, a prominent and young investigator and a good friend who recently and suddenly died at the age of 39. This mention is just to pay homage to him. In a curious way, anabolic steroids, frequently indicated in CRF and haemodialisis patients, may also reduce Lp(a). These effects may be an additional metabolic advantage of hormone replacement therapy (HRT) in postmenopausal women who need that treatment for other reasons, and an additive favourable effect of anabolics in hemodialisis patients, who indeed have a high CV risk and frequently have high Lp(a) levels, as was mentioned. Moreover, recent studies have demonstrated that plasma Lp(a) levels are prospectively predictive of CV events in the haemodialisis population. However, this by no means implies that HRT or anabolics steroids may be used in clinical practice for the sole aim to lower Lp(a) levels, even in its target patients. Although renal transplant is followed by a fall in plasma Lp(a) levels, renal recipients still have an increased risk of CV disease in association with other metabolic disorders.

Finally, LDL apheresis, a procedure for the physical removal of LDL from blood, also removes Lp(a). However, this effect is transient and the procedure is very expensive, not widely available and invasive.

Then, why measure Lp(a) and what to do with those patients with high levels today?

Lp(a) values serve to better estimate the risk of the individual patient and it is now generally accepted that patients with elevated Lp(a) levels might be selected for particularly intense LDL lowering therapy, possibly with statins or drug combinations. In my personal view and experience, not every patient is intolerant to nicotinic acid, and increasing doses of this old drug may be tried in subjects with high Lp(a) levels, mainly if low HDL-C levels and/or hypertriglyceridemia and/or mild hypercholesterolemia are present.

In any case, there is common consensus (and it is also common sense, isn’t it?) in that management of patients with high Lp(a) concentrations should be aimed at minimizing all other "modifiable" risk factors for atherosclerotic disease.

Thank you very much for your "virtual" attention.




These companies contributed to the Congress:

klchlogo4.gif (6328 bytes)
adalip.gif (17403 bytes)bayer.gif (2605 bytes)lip_ada.gif (19392 bytes)