new AHA guidelines for risk assessment/statins

as has been hitting the papers, new guidelines for cholesterol management by am heart assn.  Sorry for this really, really, really long e-mail, but it is a lot shorter than the papers.... and, raises lots of questions

1. one guideline is on the assessment of cardiovascular risk (see chol guideline cardiovasc risk assess circ 2013 in dropbox, or DOI: 10.1161/01.cir.0000437741.48606.98). this brief 50 page paper reviews the evidence for developing a new comprehensive 10 year risk assessment for ASCVD in non-Hispanic African American and non-Hispanic white men and women aged 40-79.  There is a link to a website to provide this calculator (see http://my.americanheart.org/professional/StatementsGuidelines/PreventionGuidelines/Prevention-Guidelines_UCM_457698_SubHomePage.jsp) This guideline also looked at:

         -- the role of various other markers including hs-CRP, apoB, GFR, microalbumin, family history, cardiorespiratory fitness, ABI, carotid intima–media thickness, and coronary artery calcium (CAC) scores as additional factors for risk assessment

         -- whether the models that exist can assess the long term (greater than 15 year) risk of a first cardiovascular event). 

Their conclusions:

    --Use the race and sex specific ASCVD risk assessment tool for non-Hispanic African American, and non-Hispanic whites aged 40-79 years of age (level B evidence).  use of this tool in other groups is reasonable by expert opinion (level C evidence)

    --if after quantitative risk assessment a treatment decision is uncertain, assessment of one of the more of the following may inform the decision to treat: Family history, hs-CRP, CAC score, or ABI (level B evidence).  They do note that measuring CAC is most likely to be useful.

    --No recommendation for using apoB, chronic kidney disease, albuminuria, or cardiorespiratory fitness (not enough data).  Recommend against using carotid intima-media thickness.

    --it is reasonable to assess traditional atherosclerotic risk factors and the 10 year ASCVD risk analysis every 4-6 years in adults aged 20-79 who were initially free from ASCVD (level B evidence)

    --in terms of measuring lifetime or long-term risk, they do note that developing CAD is a long slow process, and that looking at the 10 year risk for a relatively healthy 40-year-old male not be very useful clinically for that person, since they may have a low risk in the short-term but high risk long-term.  Therefore with a level C recommendation they do recommend assessing the 30 year or lifetime risk based on traditional risk factors (age, sex, total and HDL cholesterol, systolic blood pressure, use of antihypertensive therapy, diabetes, and current smoking). this is included in the calculator at the website above.

    --They recommend repeating the risk assessment after 4-6 years in patients at low 10 year risk (less than 7.5%), which is a bit at odds with the recommendations below, where people with a 5-7.5% should be offered statin therapy)

2. the major guideline is on statin use in treating high risk patients (see chol guideline AHA circ 2013 in dropbox, or DOI: 10.1161/01.cir.0000437738.63853.7a for the 84 page document).

the big issue here is that the guideline writers felt that they could only make strong recommendations based on RCTs, so their recommendations are defined by the limitations of these studies.  for example, the preponderance of evidence from RCTs comes from fixed dose statin therapy, so hard to make recommendations based on specific targeted LDL levels (ie, majority of studies are on giving high risk patients a fixed dose of a statin and not titrating the statin to a targeted LDL or non-HDL goal, so hard to recommend one specific target over another.  we do get insight into this by looking at achieved LDL levels for different cohorts, but this is a secondary analysis and not as statistically rigorous. see below for some of this data). so, based on the inherent limitations of the medical literature, they identified that statins might benefit four groups for ASCVD reduction in primary and secondary prevention, and had 2 basic targets of statin intensity.

statin intensity regimens:

High intensity regimens (anticipated LDL reduction greater than 50%) include atorvastatin 40-80 mg, rosuvastatin 20-40 mg.

Moderate intensity (anticipated LDL lowering of 30-50%): Atorvastatin 10-20 mg, rosuvastatin5/2/10, simvastatin 20-40, pravastatin 40-80, lovastatin 40.

Note: they do not differentiate within this groups – ie, generic atorvastatin is as good as brand-name crestor

Four groups to consider are:

    --those with documented ASCVD, where clinical ASCVD is defined by acute coronary syndrome, history of MI, angina, stroke, TIA, PAD (Note: unlike prior recommendations, they add stroke and TIA, which seems quite reasonable to me) (grade A recommendation)

    --Adults greater than 21 with LDL greater than 190 should be treated with high intensity statin (grade B recommendation)  -- see my note at the end on HDL, which challenges this recommendation.

    --diabetics age 40-75 with LDL in the 70-189 range and without clinical ASCVD should be treated with moderate intensity therapy (grade B). high intensity if also has 10-year risk of >7.5%. For diabetics younger than 40 or older than 75 it is reasonable to evaluate the potential benefits, paying attention to adverse effects, drug drug interactions.  (Grade C recommendation) -- see my note below regarding am diab assn recommendations

    --People without clinical ASCVD or diabetes with LDL 70-189 but an estimated 10 year ASCVD risk of greater or equal to 7.5% --  to treat with moderate to high intensity statins (grade A). if in the 5-7.5% range, offer moderate intensity therapy, (grade B recommendation).  this is really the questionable one. i think it is great that they use a global risk assessment instead of just the lipids (this has been pretty clear for decades from the framingham data that multiple risk factors interact with each other in more than an additive manner in predicting CAD risk). turns out, however, by their calculator that a 70 year old white male with pristine lipids (total cholesterol of 130 and HDL of 70), normal systolic bp (130) not on meds, no diabetes, but smokes 1-2 cigarettes/day has a 10-year risk of 15.7% (sounds like they should be in hospice, not just on high intensity statins......, and might make sense to focus on the smoking rather than the lipids????).  where are the RCTs for this one??? even a 60 year old white male nonsmoker with systolic blood pressure of 130 on meds and the same remarkable lipids (total cholesterol of 130 and HDL of 70) has a 5.5% risk, qualifying for moderate intensity statin????? and if he happens to be an African American with the same situation, the risk is 11% and qualifies for high intensity statins.  in fact, a 45 yo Af-Am male nonsmoker with controlled bp (SBP of 130 on meds) and the above pristine lipids has a risk of 5.6%, qualifying for moderate intensity statins!!!

one group probably less aggressively treated by this guideline is white women: the previous guidelines which focused on LDL tended to overtreat white women. however African American women and men tended to be undertreated because they have a higher global ASCVD risk for the same LDL level.

For patients greater than 75 years old they recommend moderate intensity statin therapy, noting that there were very few of these individuals in RCTs of high versus moderate intensity statins.  However they do suggest that statin therapy should be individualized for people over 75 with clinical ASCVD, including the possible use of high intensity statins. my experience is that i have lots of older people with diabetes or ASCVD who are on high intensity statins, monitored a bit more closely, but without any problems

for heart failure and hemodialysis, no recommendations are made.  The issue with heart failure seems to be that the relatively few trials done did not discriminate those with atherosclerotic disease and those without.  The data on end-stage renal disease is also unclear, and a bit perplexing to me because these patients are extraordinarily high risk of atherosclerotic disease deaths, yet statins do seem to be less protectiv

Non-statin therapies: No data to support using the routine use of nonstatin drugs with statin therapy to reduce ASCVD events.  In addition they state that there were no RCTs which assessed ASCVD outcomes in statin-intolerant patients on nonstatin lipid lowering agents (although the old Coronary Drug Project did find that niacin did decrease events after many years). However, at a later point in the recommendations, they do give a level C recommendation to adding a non-statin cholesterol-lowering drug but that preference should be given to drugs known to reduce ASCVD events.

(i personally feel that there is absolutely no place for therapies which do not have clear clinical benefit.  the most egregious of them, i think, is ezetimibe, with several studies suggesting that although LDL is lowered a lot, clinical endpoints may even be worse!!!, yet ezetimibe is a multi-billion dollar moneymaker for the drug company)

Evaluation of patients for statins:

Initial evaluation of the patients not on statin therapy should include a fasting lipid panel, ALT, CK if indicated, and considering secondary causes.  Secondary causes include: diet (saturated/trans fats), drugs (diuretics), nephrotic syndrome such or biliary obstruction, hypothyroidism, obesity

 Monitoring statin therapy:

    -- they suggest using fasting lipid panel for the initial lipid assessment, followed by a second lipid panel 4-12 weeks after initiation of therapy.  Thereafter every 3-12 months as clinically indicated. (note: i have sent around prior articles arguing that non-fasting lipids are basically as predictive as fasting and much easier for patients. however, i suspect they are pushing for fasting lipids because the preponderance of RCTs based their recommendations on fasting values).

    --CK should not be routinely measured in individuals receiving statin therapy (level A recommendation) but could be measured as a baseline in individuals believed to be at increased risk for adverse muscle events based on personal or family history of statin intolerance or muscle disease, or concomitant drug therapy that might increase the risk of myopathy (level C evidence)

    -- measure ALT before starting therapy and afterwards if symptoms suggest hepatotoxicity (level C recommendation),  did not recommend monitoring ALT while on statin therapy because there was no significant difference in ALT elevations between the statin and placebo treatment groups.

    --monitor for new onset diabetes according to the diabetes screening guidelines. note that for those who developed diabetes on statins, should continue the statins but engage in activities to decrease the glucose intolerance through nonpharmacologic means.  The rate of excess diabetes by statin therapy in those on moderate intensity statins is approximately 0.1 case per 100 statin treatment and individuals.  This increases to 0.3 per 100 in those on high intensity statins

     --in terms of safety of statins, there are certain patient characteristics that are important including previous serious comorbidities such as impaired renal or hepatic function, concomitant use of drugs which affect statin metabolism, history of hemorrhagic stroke, and greater than 75 years of age. These patients should be monitored more closely and may need reduced dose of statin. Also commented Asian ancestry might influence initial choice of statin intensity.  

in terms of target to treat strategy, widely used in the past 15 years: it is unclear what the target should be, what the additional ASCVD risk reduction is if one chooses one target lower than another, or what to do if unable to achieve that goal with a statin alone (ie, there is remarkably little data on clinical effect of combo therapy).  Although they are against specific targets of therapy, they do feel that it is reasonable to use as indicators of anticipated therapeutic response to statins: For high intensity and LDL reduction of greater than 50% of moderate intensity and LDL reduction of 30-50%  –the LDL reductions are to be used only to assess response to therapy and adherence and not as a performance standard.

hard to make recommendations for lack of clear data on:

    --the value of risk reduction for those with life expectancies much greater than 10 years (see first guideline above). using the risk calculator, a 50 year old white male with pristine lipids (total chol of 130 and HDL of 70) SBP of 160 off meds but smokes 1-2 cigarettes a day has a low 10-yr risk but a 69% lifetime risk (a 20yo with the same other data has a 50% lifetime risk).  should these guys be on a statin?? They do include this in the risk calculator but give no guidance on therapy.

areas for further guidelines include treatment of hyper-triglyceridemia, whether using non-HDL is useful in treatment decision-making, whether using apo B, Lp(a),or LDL particles are useful for guiding treatment, how to incorporate lifetime ASCVD risk and the optimal age for initiating statin therapy, long-term effects of statin associated new onset diabetes

Note a few points:

-- these guidelines are in accord  with the Am Diab Assn guidelines over the past few years suggesting statins in those over 40 and one or more other risk factors, independent of lipid levels.  there have been a few impressive studies: the CARDS study of 3000 diabetic patients aged 40-75 who were felt not to need lipid-lowering treatment by diabetes specialist were randomized to atorvastatin 10mg versus placebo.  Mean LDL 117 decreased to 72.  Study terminated early with 37% decrease in major events.  Subgroup of patients with an initial LDL of less than 100 had a 26% decrease in events.  Benefits were evident within the year. And, the Heart Protection Study of 21,000 high risk patients, randomized to simvastatin 40 versus placebo, found the 25% risk reduction benefit was independent if the initial LDL was <116 vs >135.  In fact lowering the LDL from 97 to 65 had the same efficacy as if the LDL was initially higher.  these studies, however, were secondary analyses. although i am not sure that i should put my diabetic patient with a baseline LDL of 90 and HDL of 75 (yes, there are occasional patients like this) on a statin, i do think that diabetes is such a high predictor of ASCVD that the overall general approach should be aggressive lipid management

--these guidelines focus exclusively on statin therapy. the Lyon Diet Heart Study a couple of decades ago with 605 pts (91% men) with ASCVD put on a strict Mediterranean diet (bread, root/green veges, fish, poultry, fruit, low sat fats but high fish oils, low cholesterol) vs "prudent" (typical Am Heart Assn diet) and followed 4 years found a 72% (!!!) decrease in combo of cardiac death and nonfatal MI, with continued divergence of the curves after 5 years. basically as good as statins. but, given that almost all studies now are funded by drug companies, these diet studies have not been repeated and have faded into the distant past..... Too bad

--HDL is not included (though it is included in both the Framingham and their own study risk stratifiers) but the Treating to New Targets study, comparing atorvastatin 10 mg versus 80 mg, did find that the five-year risk of major cardiovascular events was similar in those within achieved LDL of greater than 100 if their HDL were greater than 55, as in those within achieved LDL of less than 70 but a HDL of less than 38.  This was a secondary analysis of the large study.  There are other data as well suggesting that HDL adds to assessment of the clinical risk, but there are not specific randomized trials targeting HDL changes on which to base evidence based recommendations.  in terms of primary prevention, the Framingham Study found that the risk of CHD in patients with LDL of 220 with HDL of 85 was 1/3 (!!) that of those with LDL of 100 and HDL of 25. so, i do include HDL in my assessment of primary prevention and (though the ground is shakier) in secondary prevention as well.

--There are essentially no data on primary prevention in those 21-39 years old. we know from autopsy studies that 20% of men and 8% of women have advanced atherosclerotic lesions by age 30-34 (not just lipid streaks, which are pretty much in everyone by that age). no RCTs here, but i'm a bit concerned that these guidelines reinforce not taking cardiovasc risk factors as seriously in those under 40 (though i doubt this was their intention)

--I think they use RCTs inconsistently, dismissing the use of targeted LDLs because there is only one RCT, yet making sweeping recommendations, especially with their calculator, based on very little data

--one concern with all the guidelines over the past 20 years is the increasing representation of researchers and clinicians with ties to the pharmaceutical or med equipment industry. this group was better, with only one of 2 co-chairs and 50% of the task force itself supported by pharma (the chair and other co-chair was not). BUT still brings up potentially very significant conflicts of interest..

so...... hard for me to support much of this. similar to the rather excessive (to my mind) recommendations for aggressive screening and treatment of kids, which came out over the past year (prior email), which actually had even more influence of drug and med supply companies than this board.

congrats, you made it to the end