USPSTF recommendations: statins in primary prevention

 the USPSTF just published their revised recommendations on statin use in the primary prevention of cardiovascular disease (see statin primary prev uspstf recs jama2022 in dropbox, or doi:10.1001/jama.2022.13044)

Recommendations:

-- Adults aged 40-75 who have 1 or more cardiovascular risk factors (dyslipidemia, diabetes, hypertension, or smoking) and an estimated 10-year cardiovascular disease (CVD) risk of 10% or greater: clinicians should prescribe a statin for primary prevention of CVD. Grade B recommendation

    -- Grade B recommendation: The USPSTF recommends the service. There is high certainty that the net benefit is moderate or there is moderate certainty that net benefit is moderate to substantial.

-- Adults aged 40-75 who have 1 or more cardiovascular risk factor and an estimated 10-year risk of 7.5% to 10% : clinicians should "selectively offer a statin for the primary preventions of CVD". Grade C recommendation

    -- Grade C recommendation: The USPSTF recommends selectively offering or providing this service to individual patients based on professional judgment and patient preferences. There is at least moderate certainty that the net benefit is small.

--Adults >75yo: current evidence is insufficient to assess the balance of benefits and harms or initiating a statin for the primary prevention of CVD events and mortality". Grade I (insufficient data) recommendation

 

-- overall results of 22 primary prevention studies comparing statins to placebo:

    -- all-cause mortality (18 trials, n=85,816 people): decreased 8%, RR 0.92 [0.87 to 0.98]; absolute risk difference −0.35%

    -- fatal or nonfatal stroke (15 trials; n = 76,610): decreased 22%; RR 0.78 [0.68 to 0.90]; ARD −0.39%

    -- fatal or nonfatal myocardial infarction (12 trials; n = 76,498): decreased 33%, RR 0.67 [0.60 to 0.75]; ARD −0.89%

    -- composite cardiovascular outcomes (n = 74,390): decreased 28%,  RR 0.72 [0.64 to 0.81]; ARD −1.28% 

 

Their comments:

-- in 2018 in the US, adults >65yo had heart disease and cardiovascular disease reported as the first and fourth leading causes of death.  they comment that data were not reported in those >75yo, but the CDC report of 2019 found that in those >85yo, heart disease was the major cause of death (28.7%), followed by "other" at 27.4% (and, pretty high likelihood that a fair number of these "others" were in fact cardiovascular...). of note, the relative incidence of reported "heart disease" deaths increased over time: 20.9% in those aged 45-64, 25.1% in those >65, and 28.7% in those >85yo. see https://www.cdc.gov/nchs/data/nvsr/nvsr70/nvsr70-09-508.pdf 

-- they note that the risk prediction equations "generally show higher risk for Black persons than White persons" but that "race is a social construct and an imperfect proxy for social determinants of health and the effects of structural racism". And that  "limited evidence also suggests underprediction in disadvantaged communities." Good to acknowledge that, but this all suggests that the valued risk prediction equations are not so great

    -- specifically, they comment that inequalities based on demographic and socioeconomic factors (race, ethnicity, sex, area poverty level, income level) are important: Black adults are less likely to get statins, as well as those without health insurance, or those with multiple vulnerabilities (over 65yo, being Black, poverty level of 10% or greater, no health insurance)

-- they also note that calculations of the absolute risk reduction from statin use "is imprecise based on the currently available risk estimation tools". see below for more on this. [but, again, these tools are the basis of their recommendations]

-- they comment that there are limited data on the intensity of statin treatment prescribed, stating that moderate intensity statin therapy was reasonable for primary prevention

-- statins are well-tolerated, as they note: "statins are not associated with an increased risk of myalgia, elevated alanine aminotransferase level, or cognitive harms as compared to placebo" and the increased incidence of diabetes found in the JUPITER trial was limited to those with metabolic syndrome. a few cohort studies did find an increased risk of diabetes (though even if true mathematical analyses suggested continued benefit of statins, given their significant lowering of cardiovascular outcomes). but there are limited data in those >75yo.

    -- by coincidence the Lancet just published an evaluation of the Cholesterol Treatment Trialists' Collaboration, a huge meta-analysis of randomized trials on statins, finding that there was a very small excess of mostly mild muscle pain with statins (>90% of muscle symptoms in those on statins were NOT due to the statins (see statin myopathy mild and unusual lancet2022 in dropbox, or doi.org/10.1016/ S0140-6736(22)01643-9)

-- there were several research gaps that should be addressed: improving risk prediction of the current instruments; benefit/harms in primary prevention in those >75yo; longer term efficacy of statins in those <40yo; reducing disparities in statin use; relative roles of prescribing statins based on statin intensity vs treating LDL to target; more definitive studies on risk of diabetes with statins

 

My concerns:

1. atherosclerotic cardiovascular disease (ASCVD) starts in the young in our society (and, in general, in industrialized societies)

    -- autopsy studies have found early lesions in young people;  many studies done on young men who died in war have found fatty streaks (precursor of more advanced lesions). For example, in US soldiers killed in Korea, 3/4 had fatty streaks at average age of 22. A more recent autopsy study in US medical centers found that intimal lesions appeared in all of the aortas and more than half of the right coronary arteries in men and women aged 15-19yo, with clinically significant raised lesions increasing rapidly in prevalence and extent through age 30-34yo (see https://pubmed.ncbi.nlm.nih.gov/10052443/ )

    -- based in large part on these studies and the fact that 1 in 500 are heterozygous for familial hypercholesterolemia, the American Association of Pediatrics recommends routine lipid screening for all average risk pediatric patients at age 9-11 and again age 17-21 (see https://journals.lww.com/jaapa/Fulltext/2015/03000/Updated_guidelines_for_lipid_screening_in_children.4.aspx  and  https://publications.aap.org/pediatrics/article-abstract/130/2/353/81649/Guidelines-for-Lipid-Screening-in-Children-and?redirectedFrom=fulltext ). They dismissed targeted screening of kids who have family history of heart disease because several studies have found that only about 1/2 of kids with significant hyperlipidemia are picked up with this targeted screening. see https://publications.aap.org/pediatrics/article-abstract/126/2/260/68541/Universal-Versus-Targeted-Blood-Cholesterol?redirectedFrom=fulltext . (ie, all those at average risk should be screened independent of risk factors. And screened earlier if higher risk). and, by the way, the 20-year outcome of statins in kids with familial hypercholesterolemia is that they decrease the risk of cardiovascular disease... (see https://pubmed.ncbi.nlm.nih.gov/31618540/ )

        -- needless to say, USPSTF does not endorse lipid screening in kids. for a critique of that, see http://gmodestmedblogs.blogspot.com/2016/08/a-response-to-uspstf-does-not-back.html

2. ASCVD mortality continues to increase in those >75yo

-- in terms of those >75yo, for whom the USPSTF makes no recommendation, deaths attributed to heart disease continue to increase in this population (statistics above), the benefits of statins in many studies (though mostly secondary prevention and in younger people) seem to occur within 6 months to 1-2 years, the adverse effects of statins in older people is not so high, so it might be very reasonable to still consider statins in this group. And, anecdotally, my experience is that even higher potency statins are really well-tolerated in my many patients in their 90s.... For a large retrospective review of the benefits of statins in the elderly, see http://gmodestmedblogs.blogspot.com/2020/07/elderly-statins-help.html

3. once clinical ASCVD is established (secondary prevention), statins are less effective in preventing further atherosclerotic disease

-- though the relative risk of cardiovascular disease prevention is pretty similar for primary and secondary prevention; the absolute risks are very different: those with established heart disease have a much higher risk of another cardiovascular event even if taking statins (ie, seems better to treat earlier when we can prevent the advanced lesions)

    --and, the risk reduction by statins may not be linear over time: it may well be that the absolute benefit for primary prevention will increase significantly over time (ie, starting a statin in a 40yo may lead to a huge absolute risk reduction by age 75-80). we need much longer studies to know if this is true

-- it also seems to me that there is a fine line between those at very high risk (eg smokers with hypertension, diabetes, hyperlipidemia,...) who have a remarkably high likelihood of ASCVD, and those who had an actual clinical ASCVD event, and that aggressive treatment of these really high-risk people seems warranted. Especially when considering the minimal long-term risk of statins

4. the cardiovascular risk calculators used by USPSTF as the basis of their recommendations, and more generally in clinical practice, are not so great

    -- the USPSTF guidelines only comment on the 4 typical risk factors (diabetes, smoking, hypertension, lipids) and they (and the risk calculators) do not include other important risk factors, such as chronic inflammation (which occurs in some chronic infections including well-treated HIV, other chronic inflammatory conditions such as RA and SLE, but also with the remarkably common issues of metabolic syndrome, central/visceral obesity, stress)

        -- the 2013 Am Heart /Assn guidelines do raise this issue and therefore has a 30-year risk calculator, still not great for a 30-40 year old, though the accuracy of this prediction is unclear given the lack of really long-term, high-quality observational studies, as they acknowledge (see http://gmodestmedblogs.blogspot.com/2013/11/new-aha-guidelines-for-risk.html, or http://circ.ahajournals.org/content/early/2013/11/11/01.cir.0000437738.63853.7a.citation .

        -- and, for the USPSTF review above, there were 22 trials that assessed the benefit of statins for primary prevention: mean duration of follow-up was 3.3 years!!!  hardly a lifetime risk for most of us....

        -- analyses of the 10-year ASCVD risk predicted in the 2013 ACC guidelines risk predictor, seems to be dramatically off: one study found a 75-150% deviation of predicted risk over the actual risk in 3 cohorts: see http://gmodestmedblogs.blogspot.com/2018/06/acc-guidelines-overestimate-cad-risk.html 

            -- one issue here may be that the small dense LDL particles are 3x as atherogenic as the big fluffy ones (both are categorized as LDL by electrophoresis), and some conditions (eg diabetes, metabolic syndrome, and others) are more associated with the small LDLs. but the risk calculators do not discriminate this well (eg see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6341753/ )

5. Their assertion that "moderate intensity" statins are reasonable for primary prevention is not-at-all evidence based

    -- i think there is a pretty strong argument that we should be treating to LDL target, see http://gmodestmedblogs.blogspot.com/2021/12/ldl-less-than-40-in-high-risk-patients.html . And, my practice confirms this regularly and repeatedly: there is very wide divergence in achieved LDL in different patients on the same statin and same dose. And this treat-to-target LDL is strongly suggested in observational studies: http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html 

--so, this would all argue for screening early (teens or before) and addressing hyperlipidemia aggressively (initially by lifestyle changes, then meds along with the lifestyle issues as needed). And, also for continued screening and lipid management in those >75

Why do guidelines have such disparate recommendations??

-- different guidelines have different metrics on how they evaluate the medical literature and what factors go into their decisions (is it just the results of RCTs? or the near- and long-term costs of implementing the recommendations? or the increased medicalization of patients? or psychological adverse effects of finding an abnormality, or some combination of factors??)

-- the USPSTF tends to require more specific and higher quality studies before they come to their recommendations, typically large and well-done RCTs, but this has its downsides:

    -- there are often no high-quality studies done (which is a problem in much of clinical practice)

    -- studies are often funded by drug companies, and they typically design studies to maximize the likelihood that they will benefit the drug companies. these studies may raise more questions than they answer.

    -- studies may use surrogate markers or risk calculators that may not be so useful. for example, using the 10-year risk for ASCVD as above in a 40 year old may not be very useful if a disease begins early in life (as does ASCVD), is progressive (as is ASCVD), once established has much higher mortality risk even with treatment (as found with ASCVD), those who have a cardiovascular event may live the rest of their lives with disability though mortality is what was measured (disabilities such as heart failure, stroke sequelae, etc) assuming they survive the initial event, and a 40-year-old person might otherwise live another 50+ years. the 10-years risk may well be very low even if the person is perhaps destined to die from ASCVD at the young age of 50-60.... And, as noted above, these 10-year risk predictors may not be so accurate.

    -- and, even the best RCTs may be irrelevant to practice: many of the studies done exclude patients with CKD, or other serious comorbidities.  or exclude those >80 years old. or those who are frail. and few have the breadth of cultural diversity (and attendant differences in diet/exercise/social supports/etc/etc) to be relevant. Does a new study apply to my 85 yo pretty healthy Capeverdean patient who has some interstitial lung disease and stage 3a CKD??? Maybe. But she is nothing like the patients studied....

    -- i should also mention that there are some significant inconsistencies in the actual USPSTF approach:

        -- eg, low-dose CT scanning in smokers: they based their original recommendations on a single study with 3-years of screening in those 55-74yo, but then recommended prolonged screening, up to 25 years, and also increasing the age range to 55-80: see http://gmodestmedblogs.blogspot.com/2014/10/uspstf-lung-cancer-screening-revisited.html , as well as http://gmodestmedblogs.blogspot.com/2016/07/lung-cancer-screening-for-smokers.html.  this raises the evident contradiction that they are hesitant to recommend statins for primary prevention in the elderly (though massive data on statin use in the elderly confirm few risks of therapy, and there is very likely benefit based on some studies, as noted above), yet these LDCT recommendations for smokers really exceed the boundaries of the single study used in their recommendations

-- and, by the way, there does not seem to be a deluge of high quality information: data from the American and European Heart Associations (some of the largest and best funded medical specialty societies in the world) does not seem to be tracking towards higher quality studies on which to base their recommendations: there was no significant increase in evidence-based cardiology guidelines from 2008 to 2018 from both the US and Europe: see http://gmodestmedblogs.blogspot.com/2019/04/guidelines-lacking-evidence-based.html. still about 1/2 of recommendations by the pre-eminent US and European cardiology societies are based on expert opinion. And, often these "experts" are the most admired, well-funded researchers in their fields and not so much the regular old clinicians like us who are actually treating these patients......

-- so, guidelines are guidelines. they all have their biases and may or may not be particularly accurate or appropriate for the individual patient in front of us. this all raises the issue that it is important to understand the biases of the guideline makers and the quality of the information on which they base guidelines  (eg, information on the effects of statins in primary prevention which has an average follow-up of 3.3 years may not be a good basis for broad long-term recommendations...)

-- so, my bottom line on primary prevention of ASCVD: screen everyone for lipids (and nonfasting lipids is just fine), beginning as teenagers. If their lipids are high, try to engage them in lifestyle changes to decrease their cardiovascular risk (diet, exercise, not smoking, decreasing overweight, stress reduction). if that does not work (the patient has tried unsuccessfully for months, with our support, or the patient is not in a position to develop healthier lifestyles), i do have a low threshold to start statins, at least in males. premenopausal women are at lower risk overall, unless they have diabetes or severe hyperlipidemia, and there are potentially adverse fetal effects of statins if they become pregnant.

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 (2 options):

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

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

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