Diabetes: less of a cardiovascular risk factor than before

  

A new data-mining study found that though diabetes continues to be a risk factor for cardiovascular disease, it is not an equivalent risk factor to having had prior cardiovascular disease as had been found in older studies (see dm cardiovasc risk dec jama2022 in dropbox, or doi:10.1001/jama.2022.14914 

 

Details: 

-- a retrospective population-based Canadian study, relying on administrative healthcare data, including roughly 2 million people aged 20 to 84, compared cardiovascular mortality in 1994, 1999, 2004, 2009, and 2014 

-- 25% of each cohort were followed for up to five years 

-- these cohorts were divided into those with diabetes, those with prior cardiovascular disease (CVD), or both; the reference group was those with neither diabetes nor CVD

-- there was a general shift to a somewhat older populations over these time periods, reflecting the aging population. but overall about 20% were age 20 to 29, 20% age 30 to 39, 20% 40 to 49, 17% 50 to 59, 12% 60 to 69, and 10% >70 years old 

-- 51% female, socioeconomic status: 20% in each of the five quintiles of income 

-- comorbidities: diabetes 6%, cardiovascular disease 4%, hypertension 18%, heart failure 2%, chronic kidney disease 1% 

-- primary outcome: cardiovascular events, defined as a composite of hospitalization for acute MI or stroke, or all-cause mortality 

    -- outcome adjusted for prespecified covariates of age, sex, social economic status 

 

Results: 

-- comparing to those with no diabetes or cardiovascular disease at baseline: 

    --  1994 cohort:  

        -- diabetes: twice the risk, 28.4 versus 12.7 cardiovascular events per 1000 person-years 

            -- relative risk: 2.06 (2.02-2.10) 

            -- absolute risk increase: 4.4% (4.2%-4.5%) 

        -- prior history of CVD: 36.1 events per 1000 person-years 
            -- relative risk: 2.16 (2.12-2.21) 

            -- absolute risk increase: 5.1% (4.9%-5.2%) 

        -- both diabetes and CVD: 74.0 events per 1000 person-years

            -- relative risk: 3.81 (3.69-3.93)  

            -- absolute risk increase: 12.0% (11.5% -12.5%)  

 

    --  2014 cohort:  

        -- diabetes: twice the risk, 14.0 versus 8.0 cardiovascular events per 1000 person-years 

            -- relative risk: 1.58 (1.56-1.61)

            -- absolute risk increase: 2.0% (1.9%-2.0%) 

        -- prior history of CVD: 23.9 events per 1000 person-years 

            -- relative risk: 2.06 (2.02 -2.10)

            -- absolute risk increase: 3.7% (3.6% -3.9%)  

        -- both diabetes and CVD: 51.3 events per 1000 person-years 

            -- relative risk: 3.10 (3.04-3.17) 

            -- absolute risk increase: 7.6% (7.3%-7.8%) 

-- Overall:

 

Commentary: 

-- this data shows that overall there was improvement in cardiovascular risk in patients with diabetes, prior cardiovascular disease, or both over time, though the cardiovascular risk from diabetes decreased more so than that from prior cardiovascular disease

    -- in this study, the cardiovascular event rate decreased from 1994 cohort to the 2014 one, with rate ratios of 0.49 for diabetes but much less at 0.66 for prior CVD

    -- the data suggest that in both the 1994 and 2014 cohorts, prior history of cardiovascular disease in fact conferred a somewhat higher risk for subsequent events than diabetes (non-overlapping confidence intervals), which had been found in other earlier studies as well, though they were closer to the same in the older studies

  

 -- the evolution found in this study that diabetes is a less profound risk factor than a prior cardiovascular event, suggests a few things: 

    -- we are likely doing a better job of improving diabetes control then we used to, based on several of the newer medications 

        -- my own experience, and those of many others with whom I work, is that prescribing GLP-1 agonists in particular has resulted in remarkable changes in diabetic control, with several people with A1c levels in the 13 to 15% range now having them in the 6 to 7% range. And, improved diabetes control itself is associated with decreased cardiovascular events, even noted before these new medications have been available (see dm improved control dec CVD HeartinDiab2013 in dropbox, or https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3920786/#:~:text=Intensified%20treatment%20for%20type%202,%3C%200.001)%20(12). 

        -- and, the older agents (especially insulin and sulfonylureas) may themselves be associated with increased cardiovascular mortality, likely related to the adverse effects of high circulating insulin levels (hyperinsulinemia, associated with insulin resistance), which are associated with pro-thrombotic effects (increases in plasminogen activator inhibitor-1), increased inflammation (see https://pubmed.ncbi.nlm.nih.gov/33775061/ ), endothelial dysfunction (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3594115/#:~:text=Hyperinsulinemia%20stimulates%20ET%2D1%20secretion,%2C%20hypertension%2C%20and%20diabetes%20mellitus. ), and platelet activation (see https://diabetesjournals.org/diabetes/article/37/6/780/7247/Insulin-Directly-Reduces-Platelet-Sensitivity-to ) 

        -- and, the newer agents (specifically GLP-1 agonists and SGLT-2 inhibitors), and likely pioglitazone, are cardioprotective (which reinforces that the quality improvement guidelines for diabetes should really NOT target A1c values, but a combination of the A1c and taking cardioprotective meds, since >2/3 of diabetics die/have morbidity from cardiovascular causes)

        -- in addition, of course, there is much more aggressive use of statins overall now than previously, which undoubtedly decrease cardiovascular risk on the order of 30% (though the decrease in CVD events in this study was more profoundly found in diabetics than in those with prior CVD. ??why. because diabetics may be treated with statins more aggressively? or are followed more closely leading to better medication adherence? or perhaps diabetics may do more lifestyle changes than those with CVD?

            -- the American Diabetes Association recommends statins for primary cardiovascular disease prevention (ie those without evident atherosclerotic disease) in all patients aged 40-75 in addition to life-style therapy (Grage A recommendation) (https://diabetesjournals.org/care/article/44/Supplement_1/S125/30445/10-Cardiovascular-Disease-and-Risk-Management ). In addition, for those 20-39yo with additional atherosclerotic risk factors “it may be reasonable to initiate statin therapy” (Grade C recommendation) 

        -- several of the other cardiovascular risk factors also have improved over time leading to fewer CVD events, including decreased smoking, better control of hypertension, use of more heart-healthy antihypertensives, dietary changes including a shift away from carbohydrates (which adversely affect lipids, specifically increasing triglycerides and decreasing HDL levels) and maybe more emphasis by some of us clinicians in encouraging patients to do more exercise, perhaps using more successful techniques such as motivational interviewing 

 

-- One question:  I have seen several diabetic patients with LDLs spontaneously in the 70 mg/dL range or even lower. Should they get statins (as well as helping them with appropriate lifestyle changes, of course)?? The short answer, I think, is “yes”, mostly because they are still at high atherosclerotic risk for an array of reasons: 

    -- most diabetics have other risk factors, such as hypertension or obesity, and diabetes is a proinflammatory condition (which is an independent CVD risk factor). And the combination of risk factors increases CVD risk more than additively (ie, the quantitative increase in CVD risk having 2 risk factors, for example, is greater than the sum of the quantitative values for each if they were single risk factors)

    -- the LDL subtypes in diabetes tend to be more atherogenic: the classification of LDL (and HDL) is an arbitrary cutpoint of different LDLs by their density/sioze, and the recorded values represent several different LDL (and HDL) moieties. small, dense LDLs are 3 times as atherogenic as the bigger fluffy ones in some studies, since they seem to oxidize much more and it seems that the oxidized LDL is what strongly binds to tissue macrophages (and also elicit adhesion molecules and attract platelets, part of the atherosclerotic process) and cause the development of foam cells, which mature over time into lipid streaks and atherosclerosis. the clinical outcomes as stratified by LDL size are somewhat mixed in the studies, but apolipoprotein B levels in several studies correlate more strongly with cardiac events than LDL levels, and apoB levels are inversely related to LDL size (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8540246/ )

    -- and statins are very well tolerated, even into old age, and it seems that the lower the LDL the better (see http://gmodestmedblogs.blogspot.com/2018/08/very-low-ldl-levels-benefit-without-harm.html ) 

Limitations:

-- this was a quick and dirty data-mining study, without lots of details: no information on the role of all of the other risk factors (including smoking, lipids), if patients were being treated for these problems and how, what their values were at the time of the study (eg blood pressure, lipid values)

-- as an observational study, one cannot assume causality, just association, since there could well be residual confounding

-- no information on the all-important lifestyle issues: diet, BMI, exercise, stress levels, etc

-- no data on cause-specific mortality

-- no data on morbidity: mortality is really an easily quantifiable binary number, but morbidity is harder to quantify and is really important in terms of quality of life, quantity of life longer-term, functionality, social relations, etc

so, an interesting study which brings up a few issues:

-- there has been a really significant change over time in our treatments. and, for the better. likely from higher intensity of treatments now, and (perhaps most importantly) the development/proliferation of use of cardioprotective hypoglycemic agents and statins

-- this study should not undercut our consideration of diabetes as a potent macrovascular and microvascular risk factor, and should lead to more aggressive treatment of it for lots of reasons, and most importantly cardiovascular disease, since there is still a 2-4 fold increase in CVD deaths in diabetics

-- diabetics frequently have a slew of other risk factors (hypertension, lipids, obesity), and the Framingham and other observational studies have found that combinations of risk factors lead to higher CVD mortality in a more than additive fashion

-- this all reinforces the importance of aggressively implementing our treatment strategies (lifestyle and meds) to achieve optimal results (ie, more than just saying "stop smoking, lose weight, take these meds and see me in 6 months"). we do know that medication adherence is much less than what it could/should be, on the 50% level for diabetic meds and statins (eg see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4966497/ ). and probably even less so for lifestyle changes

-- but it is always welcoming to find out that our interventions actually help people (at least sometimes).....

 

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