Statins for primary CAD prevention in the elderly, and review of NICE guidelines
a recent
review assessed the importance of primary prevention of cardiovascular disease
in the elderly, but noted significant
discrepancies among the major international recommendations (see Mortensen MB. JACC 2018; 71910; 85-94)
Details:
--there are pretty different recommendations from 5 international societies, though all based on the same primary studies, as follows:
--the 2013 Am Coll Cardiol/Am Heart Assn (ACC/AHA) guidelines
--the 2014 UK National Institute for Health and Care Excellence (NICE) guidelines
--the 2016 Canadian Cardiovasc Society (CCS guidelines)
--the 2016 USPSTF guidelines
--the 2016 European Society of Cardiol/European Atherosclerosis Society (ESC/EAS) guidelines
--some of the differences (in the order of the above list):
--age ranges: 40-75; 30-84; 30-75; 40-75; 40-65
--risk thresholds (they all used different risk models, other than ACC/AHA and USPSTF, both using the pooled-cohort equation): >7.5%; >10%; 10-19% for intermediate and >20% high risk; >10%; and 5-10% for high risk and >10% very high risk
--risk factor requirements: no; no; yes if 10-20% risk but no if >20%; >1; no
--LDLs before treatment: 70-189; no; >135 if 10-20% risk and not if >20%; <190; >155 if high risk and >100 if >10% risk
--LDL treatment targets: no; high intensity >40% reduction for non-HDL; <77mg per dl or >50% decrease; no; <100mg per dl or >50% decrease in high risk and <70mg per dl or >50% decrease if very high risk
--high risk conditions:
--familial hypercholesterolemia/high cholesterol: LDL>190 and >21 yo; no; LDL>190; no; total chol>310
--diabetes: 40-75 yo and LDL>70; no; >40 yo; no; >40 yo
--CKD (eGFR): no; <60; <60; no; 30-59 high risk and <30 very high risk
--breaking down the recommendations by age groups:
--40-65 yo
--all guidelines have class 1, strong recommendations
--66-75 yo
--4 of the 5 guidelines recommend (other than ESC/EAS, which uses risk model only til age 65). ESC/EAS is also a bit unclear about recommendations for the elderly, though they think it should be considered particularly in those with risk factors. It should be noted that there are good studies showing benefit in primary prevention in this age group, finding similar benefit levels to those found in younger patients (eg 40% risk reduction for MI and 24% for stroke).
-->75 yo
--only the NICE guidelines includes this age group. and by their QRISK2 risk model, all over 75yo have >10% risk, so there is a universal recommendation for statins, specifically atorvastatin 20mg. But there are really no studies looking at primary prevention in those >75. Efficacy of statins is pretty clear in secondary prevention studies, though they should be used "cautiously, considering comorbidity, polypharmacy, potential side effects, and limited life expectancy" [I should comment here: NICE guidelines, as summarized below, actually state that all over 85yo would qualify for statins by their risk calculator, though people over 75 “should be considered at increased risk of CVD, particularly people who smoke or have raised blood pressure”]
--a Danish study of 1,399 apparently healthy elderly patients who had a first MI from 2010-12, 777 of whom were >65yo, found that the NICE algorithm would have led to 57% of those >75yo to be on a statin, vs essentially none of them by the 4 other guidelines
Commentary:
--the burden of atherosclerotic disease is mostly borne by the elderly, and
--the proportion and number of elderly is increasing worldwide
--at age 65, life expectancy in high-income countries is >20 years for women and >17 yrs for men
--the anticipated increase in coronary artery disease in the US is 43% (5 million more people) by 2030, simply from the demographic shift to elderly
--the Copenhagen General Population Study of 48,814 people >40yo found that in apparently healthy people, the event rate for hard ASCVD endpoints (fatal CHD and stroke, plus nonfatal MI and stroke) increased dramatically with age: 6.5/1000 person-yrs at age 56-60, and increasing parabolically to 17.8/1000 person-yrs age 71-75, then to 29.7/1000 person-yrs if >75.
--and, as an important perspective, statin therapy confers a 30-40% relative risk reduction for all: those at lower as well as those at higher risk. which means that the absolute benefit for treating older persons (given their much higher risk) should translate into a greater benefit for more people than for treating low-risk younger persons
--the direct medical costs will increase by 198% ($70 billion)
--there are some concerns about increased statin adverse effects with age: especially,
--myalgias (limited data exist here, though some suggest that this is age-independent. and i would add, there is a reasonable suggestion that vitamin D supplementation, which may be useful anyway, can dramatically improve statin-induced myalgias: see http://gmodestmedblogs.blogspot.com/2017/05/statin-myopathy-and-vitamin-d-deficiency.html ). simvastatin, esp in the 80mg dosage, may be a worse-offender, but we do not use that dose in the US
--diabetes (which is mostly found in those already predisposed to diabetes, eg have metabolic syndrome) may be an issue for elderly, given that it might necessitate more polypharmacy. though, likely as with younger people, the benefits of statins probably outweigh the diabetes risk
--one interesting comment made by the authors is that in apparently lower risk elderly, it might make sense (needs to be tested) that we consider "de-risking" them: eg, using something like coronary artery calcium scoring to identify truly lower risk elderly and not using statins uniformly on them.
-----------------------------------------------------------------------------------------------------------------------------
A few comments on the 2014 NICE guidelines on cardiovascular disease, which includes lipid modification (see: https://www.nice.org.uk/guidance/cg181/resources/cardiovascular-disease-risk-assessment-and-reduction-including-lipid-mod
ification-pdf-35109807660997 ):
--they recommend more aggressive cutpoint for treatment, using their QRISK2 risk assessment tool, now using the cutpoint of 10% risk in 10 years vs prior 2008 guideline of 20%. see blog http://gmodestmedblogs.blogspot.com/2017/08/a-new-atherosclerosis-risk-calculator_2.html which reviews/critiques the QRISK-3 calculator, the upgraded tool)
--they do recommend treating to target, using a 40% decrease in non-HDL cholesterol (there are several studies showing that non-HDL cholesterol is a better marker of clinical cardiovascular risk than LDL, especially in those on statins). for a blog arguing that treating-to-target makes sense, see
--overall, for those in the primary prevention group (including all people >85yo), use atorvastatin 20mg; for secondary prevention, use atorvastatin 80mg. they really like atorvastatin because it is cheap, especially well-tolerated, and strong (though does have significant drug-drug interactions through the CYP3A4 system: see http://gmodestmedblogs.blogspot.com/2016/11/cardiac-drug-interactions-with-statins.html for a review of the drugs and their drug-drug interactions0.
--they do rate statins by intensity, though they consider high intensity statins, those likely to decrease the non-HDL cholesterol by at least 40%, to include: simvastatin 80mg (which we do not use in the US any more), atorvastatin 20-80mg (ie, they include 20mg, noting a 43% decrease, though the ACC/AHA does not consider this dose to be high intensity), and rosuvastatin 10-40mg. they encourage atorvastatin 80 mg in those with known CVD, and in their analysis, this decreases the non-HDL the most at 55% (though rosuvastatin 40mg is 53%; looking just at LDL, i have certainly seen more impressive reductions with rosuvastatin 40 vs atorvastatin 80)
--for labs: they suggest measuring total cholesterol and HDL, to calculate the non-HDL level. this does not require a fasting sample (see blog referenced below about the overall clinical value of non-fasting lipid evaluation). repeat the lipids at 3 months and titrate statin as needed to get the >40% reduction of non-HDL level. also measure baseline LFTs, repeat at 3 months then at 12 months. do nothing unless they are above 3x upper limit of normal. do not measure CK levels at baseline or in asymptomatic patients on statins (unless they have baseline generalized unexplained muscle pains)
--they state "consider people over 85 or older to be at increased risk of CVD because of age alone, particularly people who smoke or have raised blood pressure"
--and they even allow for the "gestalt" that i use in my original critique of the 2013 AHA guidelines (see http://gmodestmedblogs.blogspot.com/2013/11/new-aha-guidelines-for-risk.html for my initial critique, and the above blog on the QRISK3 calculator): the point being that these CVD risk calculators do not include many items which might increase an individual's risk of cardiovascular events (eg, peripheral vascular disease, CKD, albuminuria, migraine....), though these risk factors might not show up on population studies because their prevalence is too low, but in the individual's cardiovascular risk assessment may well put the person above the threshold to treat
--and, as per usual in NICE guidelines, they do repeatedly emphasize the importance of lifestyle modification and patient empowerment
so, as i have stated in prior statin blogs, i have had many elderly in their late 80s to late 90s on statins, even higher intensity ones, mostly for secondary prevention and tolerating them well. At least from studies in younger people, the benefit of statins typically becomes manifest within 6-12 months (secondary prevention studies), so statins may be beneficial in elderly people even if the person has only a 1-2 year life expectancy. Fertile ground for a discussion with the patient and/or family (shared decision-making) about the appropriateness of adding the drug. and, as a perspective in this discussion, there are studies suggesting that elderly persons are more concerned about morbidity issues (stroke, heart attack), whereas younger people more with mortality (ie, the fear of a disabling heart attack or stroke, more likely to be disabling in the elderly, may tilt them to more aggressive treatment)
see blog http://gmodestmedblogs.blogspot.com/2016/05/fasting-lipids-not-so-fast.html which argues for using non-fasting lipids
see http://gmodestmedblogs.blogspot.com/2017/07/a-post-hoc-secondary-analysis-of-data.html for prior blog on recommending using statins in the elderly (though mean age in this study was 73)
Geoffrey Modest, MD
Details:
--there are pretty different recommendations from 5 international societies, though all based on the same primary studies, as follows:
--the 2013 Am Coll Cardiol/Am Heart Assn (ACC/AHA) guidelines
--the 2014 UK National Institute for Health and Care Excellence (NICE) guidelines
--the 2016 Canadian Cardiovasc Society (CCS guidelines)
--the 2016 USPSTF guidelines
--the 2016 European Society of Cardiol/European Atherosclerosis Society (ESC/EAS) guidelines
--some of the differences (in the order of the above list):
--age ranges: 40-75; 30-84; 30-75; 40-75; 40-65
--risk thresholds (they all used different risk models, other than ACC/AHA and USPSTF, both using the pooled-cohort equation): >7.5%; >10%; 10-19% for intermediate and >20% high risk; >10%; and 5-10% for high risk and >10% very high risk
--risk factor requirements: no; no; yes if 10-20% risk but no if >20%; >1; no
--LDLs before treatment: 70-189; no; >135 if 10-20% risk and not if >20%; <190; >155 if high risk and >100 if >10% risk
--LDL treatment targets: no; high intensity >40% reduction for non-HDL; <77mg per dl or >50% decrease; no; <100mg per dl or >50% decrease in high risk and <70mg per dl or >50% decrease if very high risk
--high risk conditions:
--familial hypercholesterolemia/high cholesterol: LDL>190 and >21 yo; no; LDL>190; no; total chol>310
--diabetes: 40-75 yo and LDL>70; no; >40 yo; no; >40 yo
--CKD (eGFR): no; <60; <60; no; 30-59 high risk and <30 very high risk
--breaking down the recommendations by age groups:
--40-65 yo
--all guidelines have class 1, strong recommendations
--66-75 yo
--4 of the 5 guidelines recommend (other than ESC/EAS, which uses risk model only til age 65). ESC/EAS is also a bit unclear about recommendations for the elderly, though they think it should be considered particularly in those with risk factors. It should be noted that there are good studies showing benefit in primary prevention in this age group, finding similar benefit levels to those found in younger patients (eg 40% risk reduction for MI and 24% for stroke).
-->75 yo
--only the NICE guidelines includes this age group. and by their QRISK2 risk model, all over 75yo have >10% risk, so there is a universal recommendation for statins, specifically atorvastatin 20mg. But there are really no studies looking at primary prevention in those >75. Efficacy of statins is pretty clear in secondary prevention studies, though they should be used "cautiously, considering comorbidity, polypharmacy, potential side effects, and limited life expectancy" [I should comment here: NICE guidelines, as summarized below, actually state that all over 85yo would qualify for statins by their risk calculator, though people over 75 “should be considered at increased risk of CVD, particularly people who smoke or have raised blood pressure”]
--a Danish study of 1,399 apparently healthy elderly patients who had a first MI from 2010-12, 777 of whom were >65yo, found that the NICE algorithm would have led to 57% of those >75yo to be on a statin, vs essentially none of them by the 4 other guidelines
Commentary:
--the burden of atherosclerotic disease is mostly borne by the elderly, and
--the proportion and number of elderly is increasing worldwide
--at age 65, life expectancy in high-income countries is >20 years for women and >17 yrs for men
--the anticipated increase in coronary artery disease in the US is 43% (5 million more people) by 2030, simply from the demographic shift to elderly
--the Copenhagen General Population Study of 48,814 people >40yo found that in apparently healthy people, the event rate for hard ASCVD endpoints (fatal CHD and stroke, plus nonfatal MI and stroke) increased dramatically with age: 6.5/1000 person-yrs at age 56-60, and increasing parabolically to 17.8/1000 person-yrs age 71-75, then to 29.7/1000 person-yrs if >75.
--and, as an important perspective, statin therapy confers a 30-40% relative risk reduction for all: those at lower as well as those at higher risk. which means that the absolute benefit for treating older persons (given their much higher risk) should translate into a greater benefit for more people than for treating low-risk younger persons
--the direct medical costs will increase by 198% ($70 billion)
--there are some concerns about increased statin adverse effects with age: especially,
--myalgias (limited data exist here, though some suggest that this is age-independent. and i would add, there is a reasonable suggestion that vitamin D supplementation, which may be useful anyway, can dramatically improve statin-induced myalgias: see http://gmodestmedblogs.blogspot.com/2017/05/statin-myopathy-and-vitamin-d-deficiency.html ). simvastatin, esp in the 80mg dosage, may be a worse-offender, but we do not use that dose in the US
--diabetes (which is mostly found in those already predisposed to diabetes, eg have metabolic syndrome) may be an issue for elderly, given that it might necessitate more polypharmacy. though, likely as with younger people, the benefits of statins probably outweigh the diabetes risk
--one interesting comment made by the authors is that in apparently lower risk elderly, it might make sense (needs to be tested) that we consider "de-risking" them: eg, using something like coronary artery calcium scoring to identify truly lower risk elderly and not using statins uniformly on them.
-----------------------------------------------------------------------------------------------------------------------------
A few comments on the 2014 NICE guidelines on cardiovascular disease, which includes lipid modification (see: https://www.nice.org.uk/guidance/cg181/resources/cardiovascular-disease-risk-assessment-and-reduction-including-lipid-mod
ification-pdf-35109807660997 ):
--they recommend more aggressive cutpoint for treatment, using their QRISK2 risk assessment tool, now using the cutpoint of 10% risk in 10 years vs prior 2008 guideline of 20%. see blog http://gmodestmedblogs.blogspot.com/2017/08/a-new-atherosclerosis-risk-calculator_2.html which reviews/critiques the QRISK-3 calculator, the upgraded tool)
--they do recommend treating to target, using a 40% decrease in non-HDL cholesterol (there are several studies showing that non-HDL cholesterol is a better marker of clinical cardiovascular risk than LDL, especially in those on statins). for a blog arguing that treating-to-target makes sense, see
--overall, for those in the primary prevention group (including all people >85yo), use atorvastatin 20mg; for secondary prevention, use atorvastatin 80mg. they really like atorvastatin because it is cheap, especially well-tolerated, and strong (though does have significant drug-drug interactions through the CYP3A4 system: see http://gmodestmedblogs.blogspot.com/2016/11/cardiac-drug-interactions-with-statins.html for a review of the drugs and their drug-drug interactions0.
--they do rate statins by intensity, though they consider high intensity statins, those likely to decrease the non-HDL cholesterol by at least 40%, to include: simvastatin 80mg (which we do not use in the US any more), atorvastatin 20-80mg (ie, they include 20mg, noting a 43% decrease, though the ACC/AHA does not consider this dose to be high intensity), and rosuvastatin 10-40mg. they encourage atorvastatin 80 mg in those with known CVD, and in their analysis, this decreases the non-HDL the most at 55% (though rosuvastatin 40mg is 53%; looking just at LDL, i have certainly seen more impressive reductions with rosuvastatin 40 vs atorvastatin 80)
--for labs: they suggest measuring total cholesterol and HDL, to calculate the non-HDL level. this does not require a fasting sample (see blog referenced below about the overall clinical value of non-fasting lipid evaluation). repeat the lipids at 3 months and titrate statin as needed to get the >40% reduction of non-HDL level. also measure baseline LFTs, repeat at 3 months then at 12 months. do nothing unless they are above 3x upper limit of normal. do not measure CK levels at baseline or in asymptomatic patients on statins (unless they have baseline generalized unexplained muscle pains)
--they state "consider people over 85 or older to be at increased risk of CVD because of age alone, particularly people who smoke or have raised blood pressure"
--and they even allow for the "gestalt" that i use in my original critique of the 2013 AHA guidelines (see http://gmodestmedblogs.blogspot.com/2013/11/new-aha-guidelines-for-risk.html for my initial critique, and the above blog on the QRISK3 calculator): the point being that these CVD risk calculators do not include many items which might increase an individual's risk of cardiovascular events (eg, peripheral vascular disease, CKD, albuminuria, migraine....), though these risk factors might not show up on population studies because their prevalence is too low, but in the individual's cardiovascular risk assessment may well put the person above the threshold to treat
--and, as per usual in NICE guidelines, they do repeatedly emphasize the importance of lifestyle modification and patient empowerment
so, as i have stated in prior statin blogs, i have had many elderly in their late 80s to late 90s on statins, even higher intensity ones, mostly for secondary prevention and tolerating them well. At least from studies in younger people, the benefit of statins typically becomes manifest within 6-12 months (secondary prevention studies), so statins may be beneficial in elderly people even if the person has only a 1-2 year life expectancy. Fertile ground for a discussion with the patient and/or family (shared decision-making) about the appropriateness of adding the drug. and, as a perspective in this discussion, there are studies suggesting that elderly persons are more concerned about morbidity issues (stroke, heart attack), whereas younger people more with mortality (ie, the fear of a disabling heart attack or stroke, more likely to be disabling in the elderly, may tilt them to more aggressive treatment)
see blog http://gmodestmedblogs.blogspot.com/2016/05/fasting-lipids-not-so-fast.html which argues for using non-fasting lipids
see http://gmodestmedblogs.blogspot.com/2017/07/a-post-hoc-secondary-analysis-of-data.html for prior blog on recommending using statins in the elderly (though mean age in this study was 73)
Geoffrey Modest, MD
to be added to the email list: contact me at gmodest@uphams.org
Comments
Post a Comment
if you would like to receive the near-daily emails regularly, please email me at gmodest@uphams.org