Very low LDL levels: benefit without harm

​A recent meta-analysis of lipid-lowering shows that remarkably low levels of LDL cholesterol provide clinical cardiovascular benefit with no clear increased harm (see lipids very low LDL benefit JAMAcardiol2018 in dropbox, or doi:10.1001/jamacardio.2018.2258).

Details:

-- 4 trials were looked at:

    -- the Cholesterol Treatment Trialists Collaboration (CTTC), a meta-analysis of several RCTs that compared different LDLs achieved through statins, median follow-up of 4.9 years, 1922 vascular events

    -- 3 RCTs evaluated non-statin LDL lowering therapies:

        -- IMPROVE-IT study with ezetimibe (n= 18,144), median follow-up 6.0 years, 5104 vascular events

        -- FOURIER study with the PCSK9 inhibitor evolocumab (n=2,034), median follow-up 2.1 years, 184 vascular events

        -- REVEAL study with the cholesteryl ester transfer protein (CETP) inhibitor anacetrapib (n=30,449), median follow-up 4.1 years, 4282 vascular events

Results:

-- statin therapy:

    -- CTTC: in the subset of patients with a mean baseline LDL of 66 mg/dL, the relative risk for major vascular events per 1-mmol/L (38.7 mg/dL) further reduction in LDL by more intensive therapy was 22%, RR 0.78 (0.65-0.94) (see lipid CTTC low LDL lancet2010 in dropbox, or Lancet. 2010; 376(9753):1670.)

-- non-statin therapy:

    -- IMPROVE-IT: baseline LDL of 70 mg/dL, reduced to 57 mg/dL with ezetimibe, relative risk reduced by 7.5%

    -- FOURIER: baseline LDL of 66 mg/dL, reduced to 42 mg/dL with evolocumab, relative risk of major vascular events decreased 22%

    -- REVEAL: baseline LDL of 63 mg/dL, reduced to 52 mg/dL with anacetrapib, a decline in relative risk of major vascular events by 7%

-- all studies revealed a linear 5-year decrease in estimated major vascular event rates by lower LDLs, all the way down to an LDL of 21 mg/dL

-- the decrease in each of the 4 studies, as percent of events per year, varied from 20 to 23%, with an overall relative risk of 0.79 (0.71-0.87)

-- LDL lowering was not associated with an increased risk of any serious adverse events, myalgias/myositis, elevated aminotransferases, new onset diabetes, hemorrhagic stroke, or cancer both in the individual studies as well as in the meta-analysis

Commentary:

-- though there are many studies done on cholesterol-lowering, these were the only studies that had sufficient granular information on the specific LDL values (e.g., some studies grouped all the patients with an LDL <70 mg/dL, instead of enumerating exact LDL levels achieved)

-- it is certainly reassuring that such low levels of LDL were achieved with no significant adverse events. But, it is important also to realize that these were short-term studies, and several adverse events in the past (such as diabetes) were not discovered until much later (ie, there may be some upcoming surprises...). And, we do have cholesterol for a reason (eg, cell membranes, including neurons…). will there be longer-term cognitive decline with more intensive LDL lowering???

-- it is clear from several studies that the percent lowering of LDL confers protection. For example, the full ​CTTC study found that patients with a median LDL of 132 mg/dL had a similar 22% reduction in major vascular events per 1-mmol/L (38.7 mg/dL) reduction in LDL, the same % reduction they found with the subgroup starting with an LDL of 66 mg/dL

-- but there is a significant literature suggesting that treating LDL to target also makes sense:

    --the Treating to New Targets (TNT) trial comparing atorvastatin 10 mg vs 80 mg  (see N Engl J Med 2005;352:1425-35), which on secondary analysis found that the achieved LDL predicted cardiac events; it did not matter if that target were achieved with the higher or lower atorvastatin dose. [as a perhaps relevant aside, the study found that there was an interplay between the achieved LDL and HDL, with those patients achieving an LDL >100 mg/dL but also having a higher HDL >55 mg/dL had a similar cardiovasc event rate as those with an LDL <70 mg/dL but a lower HDL <38 mg/dL]

    -- observational data show that there is a progressive decrease in coronary atherosclerotic plaque size as well as major vascular events with decreases of LDL down to 7 mg/dL

    -- and there is a general linear relationship found in many studies between the achieved LDL and cardiovasc events

    -- Also, it is important to realize that the absolute cardiovasc risk reduction does depend on the baseline LDL. The example provided in the paper was: initial LDL of 100 mg/dL, achieving a 60% decrease in LDL (as typically done by PCSK9 inhibitors) and decreasing the LDL by 60 mg/dL reduces the relative risk by 31%. Assuming a 5 year major vascular event rate of 25% in secondary prevention, this would yield an absolute risk reduction of 7.8%. However if the initial LDL were 70 mg/dL, decreasing the LDL 60% (by 42 mg/dL) would lead to a significantly lower absolute risk reduction of 5.8%

-- one interesting finding in the above study was that the 21% relative risk reduction was independent of the type of medication (statin, or the 3 non-statins), though the actual percent reduction in clinical events did vary significantly by which agent was used. So, it really seems to be that the achieved LDL is important, since all of these drugs lower the LDL by different mechanisms, though with different intensities)

So,

--the big question remains: how much LDL lowering is appropriate (what is the target???), given the dramatic increase in cost and medicalization as more drugs keep being added (especially the PCSK9 inhibitors). Certainly, the absolute benefit varies considerably by pretest probability of a future cardiovascular event. Those at very high risk would be candidates for lower LDL levels than those at the lower risk.

--One reasonable interpretation based on this study is that the lower the LDL the better, especially since there does not seem to be much increase in adverse events. Many clinicians have been hesitant to lower the LDL below 40 or 50 mg/dL (myself included), given lack of document benefit and fear of increased adverse events. So, one change i will make based on this study is to be more aggressive -- i had avoided the “high-intensity” statin approach to everyone at higher risk, since some patients had dramatic responses to lower doses: i had several patients achieving an LDL in the 40-50 mg/dL range on a medium dose statin, so i often decreased the statin dose to get in the 60-70 range. now i will put high risk patients more routinely on higher doses of statins.  But, based on this study, should we routinely be using only rosuvastatin 40mg, since that is the most potent statin???

--The exact risk-benefit analysis of when to start a 2^nd agent, or which one to use, is not clear. Ezetimibe, for example, only achieves a small improvement over a statin (and the IMPROVE-IT trial did not even optimize the statin, using only simvastatin 40mg). is that enough of a reduction in LDL? And, moving to the big gun (ie PCSK9 inhibitors) is associated with significant increase in invasiveness (injections, with some injection site reactions), and major cost (especially if these are much more widely prescribed)

--we really do need well-defined RCTs with different LDL goals to assess the incremental clinical benefit for each LDL target, to determine the appropriate one. And this needs to be done for both those at very high cardiovascular risk (secondary prevention, or close to it) and those at more moderate risk (primary prevention)

--but this study overall, i think, really does support more aggressive lipid-lowering

​ 

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 blogs since 8/15/17:

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 type in a name in the search box and get all the blogs with that name in them

to access older blogs from the BMJ website, from October 2013 until 8/15/17: go to http://blogs.bmj.com/bmjebmspotlight/category/archive/ 

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