suboptimal response to statins and increased CVD

A recent prospective cohort study in the UK found the more than 50% of patients on statins have less than the recommended 40% decrease in LDL levels, associated with more cardiac events  (see statin suboptimal response heart2019 in dropbox, or doi:10.1136/heartjnl-2018-314253).

Details:

-- 165,411 primary-care patients from the UK Clinical Practice Research Datalink who were initially free of clinical CAD and had at least one pre-statin and one post-statin treatment LDL levels, from 1990-2016

-- mean age 62 years, 49% women, baseline LDL 159 mg/dL, posttreatment LDL 101, BMI 29, blood pressure 143/83, alcohol “misuse” 0.7%, unknown smoking status 96%, social economic status well-distributed in their 5 strata, diabetes 83%, atrial fibrillation 3%, chronic kidney disease 3%, family history of CAD 11%, treated hypertension 26%, low potency statins 24%/medium 81%/high 5%

-- they assessed the LDL reduction within 24 months of statin therapy, with the goal of <40% reduction

-- a larger percentage of patients with suboptimal response were on lower potency statins

-- they then assessed CVD events in this population, with median follow-up 6.2 years

Results:

-- 84,609 (51.2%) had a suboptimal LDL response to statin therapy within 24 months

-- 22,798 total CVD events (13.8%), during 1,077,299 person-years of follow-up:

    -- 12,142 in suboptimal responders

    -- 10,656 in optimal responders

-- of these CVD events, coronary artery disease was in 13,184 (8%), stroke/TIA 4865 (3%), PVD 3097 (2%) CVD-related death 1694 (1%)

-- rate of CVD: 22.6/1000 person-yrs in the suboptimal group, and 19.7/1000 person-yrs in the optimal LDL responder group

-- cumulative incidence of CVD, for suboptimal responder vs optimal responders:

    -- 13% increased risk (unadjusted), HR 1.13 (1.10-1.16), p<0.001

    -- 19% increased risk (adjusted for age, baseline untreated LDL level, and presumably the other demographic/clinical variables), HR 1.19 (1.16-1.23), p<0.001

-- these differences in suboptimal vs optimal responders were consistently found in each of the individual CVD subgroups

-- a 38.67 mg/dL (1 mmol/L) reduction in LDL was found to be associated with a significant 6% decrease in any CVD event, OR 0.94 (0.91-0.98) in patients with suboptimal decreases in LDL

-- though, in those with optimal decreases in LDL, there was an even a higher protective effect of the 38.67 mg/dL reduction: a 13% decrease in future CVD events per 37 mg/dL reduction in LDL

Commentary:

-- the UK National Institute for Health and Care Excellence (NICE) guidelines aim for a >40% reduction in non-HDL cholesterol, whereas the 2013 ACC/AHA guidelines suggest a fixed intensity of statin for each risk category, with an intended LDL reduction of 30-49% and > 50% for moderate and high intensity statins, respectively.

-- There is well-documented genetic variability in individual LDL response to statin therapy. And, of course, variability associated with medication nonadherence. Studies have suggested that LDL reductions after administration of statins can range from 5-70%

-- statin adherence is clearly a major factor; for example a New Jersey Medicaid study found that 79% of patients were taking their statins in the 1^st 3 months, falling to 56% in the 2^nd 3 months, 50% after 12 months, and further decreases over years. Adherence was lower in those with no known heart disease, but was still significant in the group with evident CAD (e.g. CABG or PTCA) as well as those having had an acute MI. A more recent study found similar numbers: persistence of therapy was around 64% in those without CHD or diabetes, 67% in those with diabetes but without CHD, and 78% in those following an MI (see https://www.ahajournals.org/doi/full/10.1161/JAHA.118.010376). These numbers are similar to adherence rates found in patients on antihypertensives

-- part of the reason I bring up the study is that I think it undercuts a few of the recommendations in the 2003 ACC/AHA cholesterol guidelines (see chol guideline AHA circ 2013 in dropbox, or go to http://circ.ahajournals.org/content/early/2013/11/11/01.cir.0000437738.63853.7a.citation ):

    -- these guidelines suggest a specific statin intensity for patients at different CAD risks. The above UK study and others have suggested that there is quite a bit of individual variability in response to statins (as I suspect we all see in our clinical practice, where patients with low to moderate dose statins may well have profound lipid-lowering effects and vice versa)

    -- in terms of checking lipid levels: they suggest checking a lipid panel 1 to 3 months after starting a statin to assess adherence (again, even adherent patients may have quite variable responses), with further lipid assessments “as clinically indicated”. Studies pretty consistently suggest that there is waning adherence in general, supporting doing regular checks on lipid levels (after all, statin therapy is one of the most effective therapies we have for preventing the most common causes of mortality in the US and much of the Western world, and increasingly in other areas)

    -- this study, as well as others, do suggest that an absolute decrease in LDL, and not just the percent change, is associated with differences in CAD events. For a more detailed argument on this point, see http://gmodestmedblogs.blogspot.com/2016/11/ldl-more-data-supporting-treat-to-target.html

-- Clearly, this large data-mining study has huge benefits in terms of its size, its broad representation of the UK population, and the quality of the databases. However, much information is missing, including data on patient nonadherence to therapy. Also they did not have data for smoking. They did not explicitly say that they adjusted for all of the underlying risk factors, but that was implied.

so, this study, and others mentioned above, reinforces a few important clinical points:

-- the clinical effectiveness of statins does reflect their effects on LDL levels (the achieved LDL level), with those having suboptimal responses having more cardiovascular events

-- it probably does make sense to check lipid levels on patients regularly (? annually or so) to assess adherence with this important therapy, given the evidence of decreasing medication adherence over time

-- it also likely does make sense to have specific LDL targets varying by the anticipated risk level for a cardiovasc event, especially in light of studies suggesting that even very low LDL levels provide clinical benefit without much harm, especially in those at high risk (see http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html)

geoff

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

For access to the dropbox, go to link: https://www.dropbox.com/sh/0bmvtita8mzms11/XDTwHySFFg

Then go to "clinic", then to either "clinical stuff" for articles, or "powerpt presentations" for the powerpoint presentations

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 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