Gordon A. Francis MD, FRCPC, FAHA (biography and disclosures)
Disclosures: Dr. Francis is an advisory board and speaker’s board member for Alexion, Amgen, and Sanofi. Mitigating potential bias: only published trial data is presented, recommendations are consistent with published European and Canadian Lipid Guidelines, recommendations are consistent with current practice patterns.
What I did before
Young onset or “premature” ischemic cardiovascular disease (CVD) (i.e., angina, documented coronary disease, myocardial infarction, stroke or TIA occurring in men before age 55 or women before age 65) in patients or in their families is always striking, and requires a careful review of risk factors and a search for additional reasons for the early disease where they are not obvious. The standard approach to cardiovascular risk assessment includes review of the known major risk factors including hyperlipidemia, in particular elevated LDL-cholesterol level, smoking, hypertension, diabetes, family history of early cardiovascular disease, level of life stress, diet and exercise routines. In most cases, the causes of premature or even later onset CVD are not mysterious or hard to identify; however, in many cases an additional inherited risk factor, elevated lipoprotein (a) [aka “Lp little a” or Lp(a)] is present and can explain the premature heart disease or stroke 1. In some cases high Lp(a) is the only obvious risk factor that can be identified, or it helps to explain early onset CVD in patients or their families with only a single other major risk factor or in whom risk factors are present but are mild. In addition, high Lp(a) may explain recurrent CVD events in secondary prevention patients.
What changed my practice
Lp(a) is a lipoprotein particle not reported in the standard lipid profile, but which can be ordered by all physicians in BC with no charge to the patient if a diagnosis is provided [write “Lp(a)” under “Other Tests”, and indicate a diagnosis]. It is composed of an LDL particle with an additional protein, apolipoprotein (a), attached to apolipoprotein B100 of LDL. The increased cardiovascular risk associated with high Lp(a) is clear 2, but not fully understood. Lp(a) is thought to be atherogenic like LDL by inducing plaque formation, and may also increase clotting tendency. In British Columbia the normal range of Lp(a) is < 300 mg/L. Lp(a) levels were found to be above 500 mg/L in 20% of subjects in the Copenhagen General Population Study, a mainly Caucasian population 3. While a prevalence study has not been done in Canada or BC, these data and other population studies suggest that high Lp(a) is likely present in at least 20% of the BC population, i.e., it is a very common CVD risk factor. Odds ratios for the major risk factors as a cause of first myocardial infarction in the INTERHEART Study 4 were: highest to lowest apoB/apoA1 ratio, roughly LDL/HDL (3.25), current smoking versus never smoking (2.87), psychosocial stress (2.67), diabetes mellitus (2.37), and hypertension (1.91); in comparison the hazard ratios of elevated Lp(a) in the Copenhagen City Heart Study were 1.6 for Lp(a) 300-759 mg/L, 1.9 for Lp(a) 760-1169 mg/L, and 2.6 for Lp(a) >1170 mg/L, after adjustment for LDL and other major risk factors 3. Mendelian randomization and genome-wide association studies have confirmed Lp(a) is an independent heritable risk factor for premature cardiovascular disease 5. While niacin in doses of 1000-2000 mg/d lowers Lp(a) on average 30% (INSERT NEW REFERENCE: Banach M. Lipoprotein (a) – We know so much yet still have much to learn. Journal of the American Heart Association 2016;5:e003597, DOI 10.1161/JAHA.116.003597) lowering Lp(a) with niacin has not yet been proven to reduce risk of cardiovascular events 6. However, most of the risk of high Lp(a) appears to be removed by lowering LDL-C with statins, even though statins do not lower Lp(a) 7.
What I do now
The presence of high Lp(a) should be suspected in all cases of premature coronary heart disease or stroke, particularly where 2 or more other obvious coronary risk factors like genetically high LDL or smoking are not present. I now measure Lp(a) in all cases of premature CVD where obvious risk factors are not present, or where there is a strong family history of premature CAD or stroke in multiple individuals. In addition, high Lp(a) should be suspected in individuals with recurrent CVD events, especially if they already have effective lowering of LDL-C with statins +/- ezetimibe. Lp(a) is frequently high in people with familial hypercholesterolemia (FH), and can further increase their CVD risk 8; Lp(a) should therefore be measured in individuals with likely FH, i.e., LDL-C higher than 4.5-5 mmol/L despite careful lifestyle habits. The Lp(a) test using the recommended non-isoform-specific assay currently costs around $26 in BC, and only needs to be measured once, as the level is not affected by changes in lifestyle or treatment with statins. Since high Lp(a) is >90% heritable, family members of affected individuals should also be tested. In addition to reducing other CVD risk factors, the approach to reducing risk of high Lp(a) is to lower LDL-C using statins 5. High Lp(a) may suggest a need for earlier initiation of statin therapy, especially in cases with strong family history of premature CVD, and a lower LDL-C target for a given level of CVD risk, particularly in secondary prevention patients. In statin-intolerant patients, other approaches include using a low dose of a statin on alternate days and addition of ezetimibe and/or a bile acid-binding resin (e.g., colestipol or colesevelam).
Conclusion: High Lp(a) is a major CVD risk factor that should be measured and acted upon in patients and families where there is history of premature CV events but lack of clear risk factors, and in patients with known CVD and recurrent events despite treatment to LDL-C target. For primary prevention patients with high Lp(a) and family history of premature CVD, discussion and mutual decision making is recommended regarding earlier use of statin therapy to reduce the risk of the high Lp(a) 2. For primary prevention in patients with familial hypercholesterolemia, the presence of high Lp(a) heightens the need to lower LDL-C and consider achieving high risk targets (LDL-C < 2 mmol/L) in addition to >50% reduction of LDL-C. For secondary prevention patients with high Lp(a) and recurrent events despite LDL-C being treated to target with statins/ezetimibe, it remains to be seen whether addition of a PCSK9 inhibitor will reduce CVD events related to high Lp(a) in addition to lowering LDL-C. For these patients a discussion of evolving treatment recommendations or referral to a prevention/lipid clinic is warranted.
References
- Anderson T, Gregoire J, Pearson G, et al. 2016 Canadian Cardiovascular Society guidelines for the management of dyslipidemia for the prevention of cardiovascular disease in the adult. Can. J. Cardiol. 2016;32:1263-1282. (View with CPSBC or UBC) DOI: 10.1016/j.cjca.2016.07.510
- Boffa MB, Koschinsky ML. Update on Lipoprotein(a) as a Cardiovascular Risk Factor and Mediator. Curr. Atheroscler. Rep. 2013;15:1-10. (Request with CPSBC or view UBC) DOI: 10.1007/s11883-013-0360-6
- Kamstrup PR, Tybjaerg-Hansen A, Steffensen R, Nordestgaard BG. Genetically elevated lipoprotein(a) and increased risk of myocardial infarction. Jama. 2009;301:2331-2339 (Request with CPSBC or view UBC) DOI: 10.1001/jama.2009.801.
- Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, McQueen M, Budaj A, Pais P, Varigos J, Lisheng L. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet. 2004;364:937-952. (View with CPSBC or UBC) DOI: 1016/S0140-6736(04)17018-9
- Nordestgaard BG, Chapman MJ, Ray K, Boren J, Andreotti F, Watts GF, Ginsberg H, Amarenco P, Catapano A, Descamps OS, Fisher E, Kovanen PT, Kuivenhoven JA, Lesnik P, Masana L, Reiner Z, Taskinen M, Tokgozoglu L, Tybjaerg-Hansen A. European Atherosclerosis Society Consensus P. Lipoprotein(a) as a cardiovascular risk factor: Current status. European heart journal. 2010;31:2844-2853 View DOI: 10.1093/eurheartj/ehq386
- Albers JJ, Slee A, O’Brien KD, Robinson JG, Kashyap ML, Kwiterovich PO, Jr., Xu P, Marcovina SM. Relationship of apolipoproteins a-1 and b, and lipoprotein(a) to cardiovascular outcomes: The aim-high trial (atherothrombosis intervention in metabolic syndrome with low hdl/high triglyceride and impact on global health outcomes). Journal of the American College of Cardiology. 2013;62:1575-1579 (View with CPSBC or UBC) DOI: 10.1016/j.jacc.2013.06.051
- Maher VM, Brown BG, Marcovina SM, Hillger LA, Zhao XQ, Albers JJ. Effects of lowering elevated ldl cholesterol on the cardiovascular risk of lipoprotein(a). Jama. 1995;274:1771-1774.(Request with CPSBC view UBC) DOI: 10.1001/jama.1995.03530220037029
- Alonso R, Andres E, Mata N, Fuentes-Jimenez F, Badimon L, Lopez-Miranda J, Padro T, Muniz O, Diaz-Diaz JL, Mauri M, Ordovas JM, Mata P. Lipoprotein(a) levels in familial hypercholesterolemia: An important predictor of cardiovascular disease independent of the type of ldl receptor mutation. Journal of the American College of Cardiology. 2014;63:1982-1989. (View with CPSBC or UBC) DOI:10.1016/j.jacc.2014.01.063
Loading ...
Thank you for this interesting summary of evidence which is mostly new to me. This is an exciting idea but I’m not completely certain whether measurement of Lp(a) is “ready for prime time” based on the evidence presented.
In 2015 very sensible lipid management guidelines were published in the Canadian Family Physician Journal (http://www.cfp.ca/content/61/10/857). It was based on good evidence that:
1) Lipid lowering therapy should be based on overall cardiovascular risk, not lipid levels.
2) Treating to target is not evidence based, as statin therapy of a given intensity (i.e. low- medium- or high-dose) confers the same relative risk reduction regardless of initial lipid levels.
With regard to the above article, if we bear in mind the very sensibile and evidence-based principles described in that guideline it seems to me that the following conclusions can be drawn:
– Since secondary prevention patients should always be on the highest intensity statin therapy they can tolerate, there is no need to check the Lp(a) levels in those patients regardless of their age at first MI.
– Non-statin lipid lowering therapies (e.g. Niacin, as mentioned in the article) have never been shown to reduce CVD, another important reminder of the fallacy of the treat-to-target approach.
– the evidence for primary prevention of CVD with statins is very weak in all but the highest risk groups, even though presumably 20% of the patients in those primary prevention trials would have had elevated Lp(a) based on the epidemiological evidence presented above.
Thus, we conclude that 1) patients at low baseline risk will likely not benefit or benefit minimally from statin therapy regardless of Lp(a) level; 2) patients at high risk should already be on maximal statin therapy regardless of Lp(a) level; and 3) the only question is whether measuring Lp(a) in patients at moderate risk who are not currently on a statin might nudge them into the higher risk category and justify starting statin therapy. It seems to me that this question is interesting but far from answered.
Thank you for your feedback. The Canadian Family Physician Lipid Guidelines published in 2015 are aimed at primary prevention patients, and do not address patients with inherited hyperlipidemia of which high Lp(a) is one type. Regardless of the lipid guidelines you choose to follow, the point of the article is that if you encounter premature coronary heart disease or stroke in a patient in whom obvious other reasons are not present, you should consider measuring Lp(a) to help understand the etiology of the early CVD. If you have a patient in whom premature CVD is present in their first degree relatives, and the patient is found to have high Lp(a), consideration should be given to starting the patient on a statin sooner to reduce the risk associated with the high Lp(a). The article does not propose measuring Lp(a) routinely in primary prevention patients, only those with such a history. The article also did not propose measuring Lp(a) routinely in secondary prevention patients, only in those with recurrent events despite adequate treatment with statins +/- other lipid lowering agents such as ezetimibe. In those cases is it optional to consider use of an agent like niacin to reduce the Lp(a), even though a good trial of this approach has not been done. It remains as part of the “art” of medicine and works very well in individual patients. Regarding your other points, while percent reduction of vascular events in patients with a given dose of a statin is similar in patients with varying levels of starting LDL-C, the actual event numbers remain much higher in those with higher starting LDL-C (as evidenced by the Heart Protection Study), suggesting lowering LDL-C further is beneficial. The non-statin therapy ezetimibe has been shown to have clear benefit in reducing CVD events when added to a statin (the IMPROVE-IT Trial, New Engl J Med 2015;372:2387-97). The Cochrane Collaboration concluded in their 2013 review that statins are indeed protective in primary prevention patients, including women (http://www.cochrane.org/CD004816/VASC_statins-primary-prevention-cardiovascular-disease).