Am Coll of Cardiol diabetes rx recommendations

The American College of Cardiology just released their expert consensus on novel therapies for cardiovascular risk reduction in patients with type II diabetes and clinical atherosclerotic disease (see dm cardiovasc reduction meds ACC consensus2018 in dropbox, or doi.org/10.1016/j.jacc.2018.09.020. The following is a remarkably short (for me) distillation of this 24 page document, largely because i have sent out so many blogs on these agents in the past.

Major points:
-- they acknowledged the very important point (I think) that we should be looking beyond the hemoglobin A1c as the target of diabetes management, but instead look at medications that decrease cardiovascular risk, since the vast majority of diabetics die from cardiovascular causes, and the macrovascular benefits of just lowering A1c are less clear than microvascular ones
-- the 2 medications they focus on are the SGLT-2 inhibitors and GLP-1 receptor agonists, both of which have demonstrated cardiovascular benefit. It is important to remember that all of the studies on these meds were done with metformin as a background medication
-- SGLT-2 inhibitors:
    -- empagliflozin is their preferred agent. It has overall better cardiovascular outcomes than the other drugs (though there are no direct comparisons), as well as apparently fewer of the more severe adverse effects. In general these medications show about a 14% reduction in their primary cardiovascular outcomes

--GLP-1 agonists:
    -- their preferred one is liraglutide, which also has impressive cardiovascular benefit and has been around the longest. It is a daily injection as opposed to the other 2 GLP-1's studied for cardiovascular benefits, semaglutide and exenatide, which are weekly injections. Though these studies found different levels of cardiovascular benefit, their primary cardiovascular outcomes decreased 26%
-- overall there is reasonable equipoise in the consensus document about which of the agents should be used, noting that the SGLT-2s seem to prevent heart failure hospitalizations and perhaps should be preferred in patients with heart failure. also they are oral as opposed to injection drugs. The GLP-1s have far fewer significant adverse effects.

Commentary:
--they do emphasize the importance of screening for type II diabetes in all patients with ASCVD, noting that a 2010 study found only 13% of ASCVD outpatients seen by cardiologists had A1c levels checked
--see my recent blog http://gmodestmedblogs.blogspot.com/2018/11/sglt-2-inhibitors-twice-risk-of.html which details a large database study on the comparative risks of SGLT-2 inhibitors (much higher) vs GLP-1 agonists, including twice the risk of amputation; it also critiques the recent study on cardiovascular benefits of the SGLT-2 inhibitor dapagliflozin
--one concern about these types of guidelines is that they tend to reduce the studies' conclusions to just some of their major statistics (eg: % of benefit) and do not comment on the potential flaws in the study design which could challenge the validity of the stated benefts


    -- for example, as noted in my blog critiquing the SGLT-2 empagliflozin study (see http://gmodestmedblogs.blogspot.com/2015/12/empagliflozin-good-and-bad.html ), there were several major concerns which could significantly distort the reported benefit:
        -- this was a placebo controlled study. in order to maintain some parity in terms of glucose control in the placebo group, those on placebo got medications which can negatively impact cardiovascular outcomes: TZDs (they did not mention which ones used, either in the article or supplement. Could have been the now-known cardiotoxic rosiglitazone), as well as increases in sulfonylureas and insulin, both of which have been associated with increased cardiovascular events.  so, was it benefit of empagliflozin or increased harm in the placebo group??
        -- And, with these other drugs in the placebo group, there still remained a significantly empagliflozin benefit in weight (86 vs 84 kg), systolic blood pressure (135 vs 132 mmHg), as well as a pretty large 0.5% difference in A1c levels
        -- Also, the very significant difference in cardiovascular deaths was largely in the “other cardiovascular death” category, which, hidden in the supplement, “includes fatal cases that were not assessable due to lack of information and were presumed to be cardiovascular deaths as per conventional definition”
--one interesting benefit of SGLT-2s is decreased hospitalizations for heart failure, presumably because they have natriuretic effects when administered with loop diuretics. However, I wonder that if these patients were aggressively monitored and treated for heart failure symptoms and BNP levels with adjustment of their loop diuretics, if the SGLT-2s would add anything?? ie, is the decrease in heart failure hospitalizations simply because patients were not treated well enough for their heart failure?? (after all, this was a trial just of the SGLT-2's, had a limited focus, and they were not optimizing overall treatment for the patients. which brings up another limitation of these types of studies: they do not really reflect what we do in clinical practice in taking care of the whole patient)


-- one general issue brought up in this study is that I think we clinicians may assume that all drugs of a certain class are similar (eg, all ACE-I’s are the same). It seems pretty clear that each of the SGLT-2 inhibitors in particular has a somewhat different array of adverse effects (they comment in the above recommendations that "these results suggest the potential for clinically relevant heterogeneity within the class"). so, doesn't this suggest that we should be very hesitant to prescribe the newest kid on the block??? perhaps some of the adverse effects of the newest drug in a class will not become apparent until several years later (eg, canagliflozin came out after empagliflozin and was subsequently black-boxed by the FDA because of the increased risk of amputations. are there other serious adverse effects that might become evident 3 years later???). and if we decide to use one of these drugs, should we stick with one of the earlier ones that has at least been around for awhile???? (which supports using empagliflozin over the others, as also suggested in this consensus statement)
-- and, another general issue: many studies use a composite endpoint as their primary endpoint, since it is easier to reach statistical significance in lumping together individual outcomes into a larger group. but, not all end-points have the same value. some cardiology studies, for example, put together the hard (and bad) endpoints of cardiovascular death, disabling stroke,..., along with percutaneous coronary interventions as a primary outcome. but most people, i would assume, would be happier having coronary stenting than a major stroke. a few points:
    --by making this conglomeration as the primary endpoint, it attributes higher statistical import to that outcome than secondary analyses of the individual components
    --this primary endpoint may be used as the comparator of different studies (as in the above document), though they mya include an array of different endpoints
    --that being said, it does seem that several cardiology studies have evolved a more consistent MACE (major adverse cardiovascular events) to include just 3: cardiovascular death, myocardial infarction, or ischemic stroke.  this is much better, though there are pretty big differences between death or a small MI or a large MI leading to severe heart failure or a major disabling stroke or a small stroke with no sequelae......
        --for example, in the GLP-1 liraglutide study, they found a 13% decrease in the primary composite of death from cardiovascular causes, nonfatal MI and nonfatal stroke. but, on looking at the specific outcomes (though prespecified, in this case) found a 22% decrease in cardiovascular deaths (NNT=77 for 3 yrs) but only a 14% decrease in silent MIs. however, this composite decrease of 13% would be pretty clinically different if all of them were just decreases in silent MIs....
    --so, these composite endpoints may not really reflect the grossly different values a patient may attribute to the different specific components of the endpoint. and patients may well have very different values for each of the specific endpoints.
    --which may mean that there need to be longer studies with more patients to really get the endpoints that are granular enough for patients to really make informed decisions


So, I think it is great that these cardiovascular considerations of diabetes meds are being taken into account. The American Diabetes Association, which long-held to the primary agent should be metformin with no specific recommendations about one 2^nd agent vs another, in their 2018 statement has been more circumspect about the benefits of different agents, noting the importance of cardiovascular benefit (though still not really pushing this so strongly as an issue). they do comment now that we should consider an SGLT2 or GLP-1 if known ASCVD, though they also really push insulin in those with high A1c levels

As mentioned in many prior blogs, I do think the GLP-1 agonists are great. They yield impressive results from lowering A1c, sometimes 4-5 percentage points (and oftentimes much more impressive than the results on high-dose insulin), are usually well-tolerated (even the nausea tends to get better over time or with some dose reduction). These remain my preferred add-on medication after metformin, insurance company permitting.  And, though the above recommendations apply to patients with known ASCVD, I think it makes sense to use them even in all diabetic patients, since those without clinical ASCVD mostly still have it and are still most likely to develop that/die from it. 

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