several people have asked what i consider a reasonable approach to lipid management. i have been following the lipid literature at this point for several decades, since i have always felt that heart disease was a manifestation of our industrialized society and abnormal lipids (from the attendant changes in lifestyle) was a principal component of atherosclerotic disease. my approach over the past decade or so involves developing a general sense of atherosclerotic risk as follows:
–the major part of the overall gestalt of a person’s risk is the traditional risk factors: age, smoking, hypertension, and lipids (with smoking being the strongest remedial component — we can’t do a lot about age).
–existence of atherosclerotic disease to me is an indication for very aggressive lipid management, independent of the lipid levels. in this category i have always included strokes (data on positive effect of statins is at least a decade old), as well as PAD or other atherosclerotic vascular disease (eg, renal artery stenosis, carotid stenosis). although the data are quite robust that LDL is bad and HDL protective, there are important gradations here: some LDL is less bad than others (large LDL in the large Quebec study had 1/3 the atherosclerotic risk of small dense LDL) and a minority of people have a pro-inflammatory HDL (see below for an old post, but HDL can be bad). and, the most important issue is not community data finding that LDL is bad etc, but the individual patient, and a person with atherosclerotic disease is telling me that they are prone to more…. (this is like the value of TSH, as a reflection of the individual person’s thyroid status, as opposed to T4 levels, which are a range reflecting the normal distribution for the community).
–diabetes is a reasonable atherosclerotic equivalent. as mentioned in my really long blog on recent AHA guidelines, there are some patients with very high HDLs and very low LDLs. there are no studies in the literature of this group that i am aware of, and i have not been pushing statins on them, but overall i do support the basic Am Diab Assn guideline to give statins to all diabetics over age 40 with 1 additional risk factor (ie, the default should be to give a statin, since there is such a high risk of ASCVD, except in a few cases, as above). interestingly, the AHA guidelines do not include people with glucose intolerance, though they have almost as high a risk as several other risk factors (eg, a prospective european study found for CAD risk: men in 5-5.4 range had 56% increased risk, 5.5-5.9 100% increase, known diabetes 382%; women not significant til 6.6-4 with 129% increase and 5-fold increase if known diabetes. (see doi:10.7326/0003-4819-141-6-200409210-00006). these numbers for men with A1c in the 5.5-5.9 range are not so different from those with PAD, for example. i have been checking A1c’s for years on patients as part of my cardiovasc risk stratifier, and i do incorporate glucose intolerance into my gestalt of who to treat.
–there are other risk factors to consider, including microalbuminuria in nondiabetics and even early renal disease (creat above 1.4). i think the data on coronary artery calcium score is quite strong (and is an individual marker of disease), but i will not do this given the radiation exposure. hs-crp is also reasonable (the JUPITER study was very impressive, though interestingly, the hs-crp added nothing in the subgroup who were stratified by coronary calcium scores. maybe hs-crp is a cheap and nontoxic alternative to coronary artery calcium???). at this point i have not incorporated hs-crp into my ASCVD risk assessment, and i do not check for microalbuminuria regularly, except in people with hypertension or diabetes. even aortic sclerosis in some studies is associated with increased risk.
–HDL is still important, to my view (interesting that AHA does not discuss HDL, though it is still part of its risk calculator!!). for primary prevention, i usually use the framingham cutpoint of total cholesterol/HDL ratio of 4 (there have been lots of studies validating the total chol/HDL ratio as a much better predictor of clinical events than LDL). for patients on statins, the Treating to New Targets study (TNT) found that an achieved LDL >100 with the highest quintile of HDL, >55, had the same rate of clinical events as those with LDL <70 with the lowest quintile of HDL, <38 — see LINK TO SLIDES BELOW SLIDE #4 and comment below –so, i still treat to an LDL target, which is modified by the achieved HDL, though i realize in some studies that the rate of protection by statins was independent of the initial LDL, but analysis of many studies shows cardioprotection is related to the achieved LDL (one can easily argue that these are secondary anaylses and not as statistically rigorous, but that’s all we have… and i do find it ironic that AHA is such a stickler for LDL goal studies (ie, the lack of multiple studies with differing LDL goals), yet applies their calculator on groups with absolutely no data (eg the 45 yo African-American male as mentioned in the longer blog with a total cholesterol of 130 and HDL of 70, and well-controlled BP on meds (SBP=130) qualifies for moderate intensity statin…..)
–so, based on this, who do i treat? first, pretty much everyone gets the diet and exercise discussion, with pretty good data suggesting that these are helpful in preventing heart disease as well as in secondary prevention. second, the framingham data has shown that the risk of heart disease is higher in patients with a little of several risk factors (the relationship is more than just additive) than in those with even a quite high single risk factor. so, someone who smokes and has glucose intolerance will probably get a statin, unless their lipids are remarkably good (again, if they have any evidence of atherosclerotic disease, they get a statin anyway)
i do realize that this “gestalt” is not exactly quantitative, but probably better (to me) to have some integrated approach which assesses different risk factors (which, from studies, do have attributable risks — eg, risk from different A1c levels is as above) than to put together a quantitative risk calculator which does not have sufficient data but spews out a “number” (eg the 45yo african american with controlled hypertension and pristine lipids).
after i was putting the above together, there was an interview in Medscape Cardiology today of 2 of the AHA committee members, to which i feel compelled to respond (sorry, this post is getting longer…). the passages are from the medscape article:
Dr Donald Lloyd-Jones, the cochair of the guidelines on the assessment of cardiovascular risk, said the evidence for treating to target simply isn’t there, but that doesn’t mean repeated measurements of LDL cholesterol won’t be needed.
“There have been no clinical trials in which they’ve taken an approach where they’ve titrated medication dosing to achieve a certain LDL level,” said Lloyd-Jones. “We just haven’t had those trials designed or performed yet. So we just couldn’t endorse that kind of approach. And yet, we’re not abandoning the measurement of LDL cholesterol, because it’s perhaps our best marker of understanding whether patients are going to achieve as much benefit as they can for the dose of statin they can tolerate. For the clinician, it’s also a very important marker of adherence.”
1. unfortunately, it is true there are not lots of data from studies (which, by the way, are done by drug companies, so they choose which study to do — there is not some overarching coherent strategy to doing studies, unfortunately).
2. interpretation of the existing studies have found a consistent relationship between LDL levels achieved and cardiac events (this is, of course, a secondary analysis, though rather consistent). i have attached a powerpoint link(first slide at link below) of one analysis of the results (from the TNT reference in next point).
3. the Treating to New Targets study, as mentioned, did compare atorvastatin 10 to 80mg (see N Engl J Med 2005;352:1425-35). the second slide shows pretty clearly that clinical events were significantly fewer with atrovastatin 80mg vs 10mg. slide 3 from a subsequent TNT article (see N Engl J Med 2007;357:1301-10) shows pretty clearly on secondary analysis that the achieved LDL was associated with cardiac events (the lower, the better), AND it didn’t matter if on atorvastatin 10mg or 80mg. slide 4 shows that the achieved HDL, when combined with the achieved LDL, significantly affects the subsequent cardiac risk (this is the graph of data from point above on HDL).
Dr Neil Stone, the chair of the expert panel who wrote the guidelines, said there were some problems with treating to goal, specifically in patients who were treated close but not exactly to target.
“In secondary prevention, what if your patient is on high-intensity statin therapy and gets an LDL-cholesterol level of 78 [mg/dL] and is adhering to an excellent lifestyle?” said Stone. “From our point of view, there is a large body of evidence that says he’s actually doing as good a job as he can possibly do. If he has to get to an optional goal of under 70 [mg/dL] as some would advocate, it means adding on medicines for which there is not proven benefit.”
In addition, the panel said that the use of LDL-cholesterol targets might result in the overtreatment of patients with nonstatin drugs. These other agents have not been shown to reduce the risk of atherosclerotic cardiovascular disease.
1. i totally agree that there are no good large scale studies showing benefit of adding anything to a statin, and that some meds (eg ezetimibe) may be harmful. but that does not mean that we should not be treating to an LDL target (perhaps modified some by the HDL, as above), just that we may need to titrate the statin. so, the AHA chart (in my prior blog post) which clumps several “high intensity” regimens together (atorvastatin 40-80mg or rosuvastatin 20-40mg) should perhaps revert to the old way: if atorvastatin 40mg does not achieve the target, we should augment the intensity of therapy, moving to atorvastatin 80mg, or up to rosuvastatin 40mg. then call it quits. (of course, continuing to reinforce wt loss, diet, exercise…)
2. there are some pretty interesting drugs probably coming out which MAY have impressive clinical efficacy (eg, the PCSK9 inhibitors), and we may (???soon) have additional potent drugs in the armamentarium.
sorry if you think i am perseverating on this. i do, however, think these guidelines overall do a disservice (though, as i mentioned in the really long prior post, there are some very positive changes as well). let me know your thoughts.
Although the vast majority of epidemiol studies have found HDL to be cardioprotective, there have always been some concerns. HDL is comprised of a diverse group of lipoproteins with significant metabolic heterogeneity. there were a few older studies finding a “pro-inflammatory HDL”, which predisposed people to heart disease (see my lipid powerpoint, slide 106, which cites small studies from 2003 and 2009). the clinical trial of Torcetrapib, a cholest ester transfer protein inhibitor, dramatically increased HDL but was not cardioprotective. the researchers suggested that the HDL was somehow deformed. (this large torcetrapib trial overwhelmed a meta-analysis last year in BMJ, suggesting no benefit to raising HDL). in any event, there is a likely very illuminating article from harv school pub health. they had found before that there was occasionally a small apolipoprotein (apo C-III) on some lipoproteins causing a pro-inflammatory and atherogenic response. on LDL particles, this apo C-III caused increased coronary athero independent of the LDL itself. they looked at the data from 2 large epidem studies — Nurses health study (NHS, 121K female nurses) and the health professionals followup study (HPFS, 52K males), looked at stored serum and assessed the HDL C-III relation to cardiac events. results:
. 14% of of women in NHS had HDL with apo C-III; 11% of men in HPFS had apo C-III
. Overall, each standard deviation increase in HDL was assoc with a 21% dec in cardiac events; but for patients without apo C-III, there was a 34% decrease in events and for those with apo C-III there was a statistically significant 18% increase.
. looking at the effect of apo C-III in multivariate analysis of other risk factors (all of below statistic signif):
compared to pts with normal wt, overwt and obesity were assoc with 7% and 12% lower levels of HDL without apo C-III — ie, overwt/obesity with lower of the good HDL.
alcohol assoc with 3% higher levels of both HDL types
smokers had 1% higher levels of HDL with apo C-III (the bad one)
premenopausal women had 9% higher levels of HDL without apo C-III, as did postmenop women on estrogen replacement therapy, vs other postmenop women.
per SD increase in triglycerides, 8% lower HDL without apo C-III and 15% increase in HDL with apo C-III
per SD increase in A1C, 4% increase in HDL with apo C-III
(ie, several of these risk factors which depress HDL also lead to more HDL with apo C-III)