high chol in young people: high cardiovasc risk

A recent large international study found that non-HDL cholesterol was an excellent predictor of cardiac events in the distant future in patients initially free from cardiovascular disease (see lipid nonHDL lancet2019 in dropbox, or /doi.org/10.1016/ S0140-6736(19)32519-X)

Details:
-- 398,846 individuals without baseline cardiovascular disease, from 38 cohorts in the Multinational Cardiovascular Risk Consortium, which includes data from 19 countries across Europe, Australia, and North America
-- 49% women, median age 51
-- there were both  derivation (199,415 individuals) and validation (199,431 individuals) cohorts within this group, the first to determine the relative risks and the second to predict them.  
    --they created a tool to estimate the probabilities of a cardiovascular disease event by the age of 75, dependent on age, sex, and cardiovascular risk factors, as well as modeling the effect of a 50% reduction in non-HDL cholesterol on predicted events
-- median follow-up was 13.5 years, with maximum follow-up of 43.6 years
-- primary composite endpoint: an array of major atherosclerotic cardiovascular events, both those associated with coronary heart disease as well as with ischemic stroke

Results:
-- 54,542 had a cardiovascular endpoint
-- there was a progressively higher 30-year cardiovascular disease event rate with increasing non-HDL cholesterol categories (in women: from 7.7% for non-HDL <100 mg/dL, to 33.7% for non-HDL > 220 mg/dL; in men: from 12.8% to 43.6%)
-- lifetime sex-specific hazard ratios for nonfatal and fatal cardiovascular disease, by multivariable adjusted models, with non-HDL cholesterol < 100 mg/dL as the reference:
    -- women < 45 years old:
        -- non-HDL cholesterol 100-145 mg/dL: HR 1.2 (0.9-1.6)
        -- non-HDL cholesterol 145-185 mg/dL: HR 1.8 (1.3-2.4)
        -- non-HDL cholesterol 185-220 mg/dL: HR 2.6 (1.8-3.6)
        -- non-HDL cholesterol > 220 mg/dL: HR 4.3 (3.0-6.1)
    -- women 45-59 years old:
        -- non-HDL cholesterol 100-145 mg/dL: HR 1.0 (0.8-1.3))
        -- non-HDL cholesterol 145-185 mg/dL: HR 1.4 (1.1 -1.7)
        -- non-HDL cholesterol 185-220 mg/dL: HR 1.7 (1.4-2.2)
        -- non-HDL cholesterol > 220 mg/dL: HR 2.3 (1.8-2.9)
    -- women >60 years old:
        -- non-HDL cholesterol 100-145 mg/dL: HR 1.1 (0.9-1.4)
        -- non-HDL cholesterol 145-185 mg/dL: HR 1.2 (0.9-1.4)
        -- non-HDL cholesterol 185-220 mg/dL: HR 1.3 (1.0-1.6)
        -- non-HDL cholesterol > 220 mg/dL: HR 1.4 (1.1-1.7)

    -- men < 45 years old:
        -- non-HDL cholesterol 100-145 mg/dL: HR 1.4 (1.0-2.0)
        -- non-HDL cholesterol 145-185 mg/dL: HR 2.0 (1.4 - 2.8)
        -- non-HDL cholesterol 185-220 mg/dL: HR 3.2 (2.3-4.5)
        -- non-HDL cholesterol > 220 mg/dL: HR 4.6 (3.3-6.5)
    -- men 45-59 years old:
        -- non-HDL cholesterol 100-145 mg/dL: HR 1.0 (0.8-1.2)
        -- non-HDL cholesterol 145-185 mg/dL: HR 1.3 (1.1-1.5)
        -- non-HDL cholesterol 185-220 mg/dL: HR 1.6 (1.3-1.9)
        -- non-HDL cholesterol > 220 mg/dL: HR 2.1 (1.8-2.6)
    -- men <>60 years old:
        -- non-HDL cholesterol 100-145 mg/dL: HR 1.1 (1.0-1.4)
        -- non-HDL cholesterol 145-185 mg/dL: HR 1.3 (1.1-1.6)
        -- non-HDL cholesterol 185-220 mg/dL: HR 1.5 (1.3-1.8)
        -- non-HDL cholesterol > 220 mg/dL: HR 1.8 (1.5-2.2)

-- for both men and women, the 30-year cardiovascular disease rates were approximately 3 to 4 times higher in those in the highest non-HDL category
-- the steepest increase in the relative hazard associated with non-HDL was in individuals <45 yo
-- there was high comparability between the derivation and validation cohorts (ie, the latter did well in predicting long-term risk)

-- in terms of actual numbers-needed-to treat (NNT) to prevent one cardiovascular event at age 75, (see their table 3):
    -- for women:
        -- <45 years old with 0-1 cardiac risk factors (blood pressure, diabetes, obesity, and smoking) and non-HDL <100, NNT=32 to prevent one cardiovascular event; for those with >1 risk factor and non-HDL >220, the NNT=4.6
        -- 45-59 years old with non-HDL <100 and 0-1 risk factors, NNT = 38.8; for those with >1 risk factor and non-HDL >220, the NNT=5.6
        -- > 60 years old with non-HDL <100 and 0-1 risk factors, NNT = 63.4; for those with >1 risk factor and non-HDL >220, the NNT=8.3
   --  for men:
        --; for those with >1 risk factor and non-HDL >220, the NNT=2.6
        -- 45-59 years old with non-HD<45 years old with non-HDL <100 and 0-1 risk factors, NNT=15.5L <100 and 0-1 risk factors, NNT=1.4; for those with >1 risk factor and non-HDL >220, the NNT=4.3
        -- > 60 years old with non-HDL <100 and 0-1 risk factors, NNT=39.8; for those with >1 risk factor and non-HDL >220, the NNT=6.9
-- for a 30% reduction the NNT for non-HDL is about 1.5 times that as compared to the 50% reduction (see their table S10 for all of the calculations)
-- their figure 4 is a visual including the age, sex, and number of cardiac risk factors, showing the probability of a cardiac event at age 75, and the huge decreases that would happen with a 50% reduction in non-HDL

-- sensitivity analyses: similar results were found in diabetics and nondiabetics, those receiving lipid-lowering therapy at baseline, and with further adjustment HDL levels (the group where HDL played the biggest part was in men < 45yo)
-- there was also a linear association with LDL levels on a log-hazard scale, very similar to that found with non-HDL levels

Commentary:
-- this study used non-HDL cholesterol as their basis, since that would combine all of the pro-atherogenic lipoproteins containing apolipoprotein B [i.e., VLDL and their metabolic remnants, intermediate density lipoproteins, lipoprotein (a), and LDL]. The non-HDL is roughly the LDL+30 mg/dL. Current guidelines in both the US and Europe do suggest this as a better estimate for cardiovascular risk factor assessment
-- this study adds lots of new information:
    -- most of the data on lipid-lowering therapy is in high-risk individuals or for secondary prevention, with unequivocally positive findings. The data are less clear for primary prevention, especially in younger people
    -- the conventional guidelines assessing cardiovasc risk use the 10-year risk calculators. These are not terribly useful for patients who have longer life expectancies because:
        -- for younger patients, their 10-year life expectancy in the aggregate is going to be quite high, even if they have terrible non-HDL (or LDL) levels
        -- however, as noted in this study, the non-HDL cholesterol levels in those <45yo has the highest relative risk of cardiovascular disease 30 years later. This relative risk does decrease as patients age, though it is important to remember that the absolute risk does increase with age (older people actually do have more cardiovasc events attributable to the non-HDL, though the relative risk is lower)
        -- so, this study adds tremendously to the concept that we should be testing people and treating them at an early age. The 2013 AHA/ACC cardiovascular disease guidelines does bring up a 30 year cardiovascular event horizon, which I think was one of the most important advances of that guideline, but this current study extends this through having a much larger database
    -- this study confirms the importance of addressing lipids in young people, especially since there are such limited data on primary prevention in this group, but:
        -- it is clear that cardiovascular disease starts early: fatty streaks are present in essentially everyone aged 15-34
        -- advanced atherosclerotic lesions are found in 2% of men and 0% in women aged 15-19; though by age 30-34, they are found in 20% of men and 8% of women
        -- and, we know that statins in particular do stabilize atherosclerotic plaques, though with essentially no regression of those plaques

        -- of note, the American Academy of Pediatrics recommends that all children should have cholesterol screening once between the ages of 9 and 11 and once between 17 to 21 (see https://www.aappublications.org/content/33/2/1.3 ); and, even though some adult guidelines do suggest checking lipids in pretty much everyone, the USPSTF suggests screening men at age 35 and women at age 45 (though starting at age 20 if they are at increased risk)

        -- so, this study does reinforce the concept of testing and treating early, especially since epidemiologic studies suggest that non-HDL cholesterol levels do not change markedly over time (e.g. the Framingham Offspring Study)

-- what are the implications of aggressively treating those <45yo for high non-HDL levels?
    -- It is pretty clear from this data that these people are the most likely to benefit
    -- it is also clear that men have close to twice the risk of women, especially in the younger cohort
    -- we also know that statins or other cholesterol-lowering medication should not be given to women who might become pregnant when taking them; also, menstruating women in general have a very low incidence of cardiovascular disease, which unfortunately increases dramatically with menopause
    -- and, as per many prior blogs, the real focus should be on lifestyle changes which promote decreases in non-HDL (e.g. diet, exercise, achieving a normal body weight) but also seems to prevent an array of diseases including diabetes, cancer, Alzheimer’s, and premature death

-- limitations of the study include: this is a compilation of medical databases with differing methodologies and endpoints from many different countries; there were only data on baseline lipid levels and not any dynamic changes (though, most people’s non-HDL levels do track consistently, unless there are significant lifestyle changes); this involved people of European ancestry from high income countries, the results may not be generalizable to others; the models for the 30% or 50% reductions in non-HDL are mathematical ones and not real world interventions; these models assume that those on lipid-lowering agents take them regularly and long-term (a real issue for many asymptomatic people); and there are no granular data about the specific other cardiovascular risk factors: they make untrue assumptions that their risk is all the same (which also brings up the question of the interplay of multiple risk factors, which may be more than additive in some cases)

for some other recent relevant blogs:

-- http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html , finding benefit to lowering LDL as low as 21mg/dL

-- http://gmodestmedblogs.blogspot.com/2018/11/new-cholesterol-guidelines-2018.html reviews the 2018 cholesterol guidelines

-- http://gmodestmedblogs.blogspot.com/2019/11/stroke-treating-ldl-to-target.html argues that we should be treating atherosclerotic disease with a target LDL, and references to many other relevant blogs

so, seems like an important article, which highlights the very high cardiovasc risk of young people with high to very high non-HDL levels. By using a long-term (not 10-year) CVD risk assessment, this study helps counter several analyses questioning the role of treating lipids as primary prevention in those without known cardiovascular disease. Of course, this does not mean that everyone should be on a statin, but it does mean that we clinicians and our patients should understand that a high cholesterol is really really bad in younger people. and that we treat this all very seriously, engaging the patient in trying to modify lifestyle issues, with or without the use of meds.

geoff​

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list