Prediabetes: increases risk heart failure if AFib

 A Hong Kong study found that patients with atrial fibrillation (AF) who had prediabetes had a higher incidence of developing heart failure (HF): see dm prediabetes inc risk heart failure DiabCare2022 in dropbox or​ doi.org/10.2337/dc22-1188 

 

Details: 

-- 70,943 patients with newly diagnosed atrial fibrillation from 2015 to 2018 were divided into three groups: normoglycemia, prediabetes, and type II diabetes 

-- exclusions included those with valvular heart disease, rheumatic heart disease, previous valve surgery, as well as those with hyperthyroidism 

-- data was from the CDARS system (Clinical Data Analysis Reporting System) in Hong Kong, where >90% of the local population is under the care of the public system and has relevant data in this database 

-- mean age 75, 47% female, smoking 8%, CHA₂DS₂-VASc score 4.0 (with a score of at least 2 in 79%), fasting blood glucose 115, A1c 6.9 (5.4% in those with normoglycemia, 6.0% in those with prediabetes, and 7.3% in those with diabetes), eGFR 65 

-- hypertension in 31%, ischemic stroke 12%, CAD 13%, autoimmune disease 15%, anemia 10%, cancer 16%, dyslipidemia 10% 

-- meds: NOAC 52%, warfarin 17%, statins 40%, aspirin 40%, beta blockers 32%, ACE/ARB 48% 

-- 3711 (21%) had prediabetes, 10,127 (56%) had diabetes at baseline 

-- primary outcome: incident heart failure 

-- median follow-up 4.7 years 

 

Results: 

-- development of heart failure:  

    -- normoglycemia:  518 patients (14%), incidence rate/100-person years: 3.02 

    -- prediabetes: 646 patients (15.7%), incidence rate/100-person years: 3.14 

    -- diabetes: 1795 patients (17.7%), incidence rate/100-person years: 3.38 

 

-- Adjusting for confounding factors, risk of HF, as compared to those normoglycemic: 

    -- prediabetes: 12% higher, HR 1.12 (1.03-1.22), p =0.02 

    -- diabetes: 30% higher, HR 1.30 (1.19-1.39), p <0.01 

 

-- with A1c as a continuous variable: each 1% increase in A1c was associated with an 8% increased risk of HF, HR 1.08 (1.02-1.14), p< 0.01 

 

-- patients with prediabetes at baseline, after two years: 

    -- progression to diabetes: 403 patients (11.1%) 

    -- reversal to normoglycemia: 311 patients (8.6%) 

-- risk of heart failure in these subgroups, as compared to those remaining prediabetic: 

    -- progressed to diabetes: 50% increased risk of HF, HR 1.50 (1.13-1.97) 

    -- reverted to normoglycemia: 39% decreased risk: HR 0.61 (0.42-0.94) 

-- no significant difference was found in the risk of all outcomes of atherosclerotic cardiovascular disease by the different definitions of prediabetes (see below) 

 

  graph A: cumulative incidence of HF among the 3 groups at baseline; graph B cumulative risk of HF for prediabetics if they progress to diabetes or revert to normoglycemia

Commentary: 

-- per the American Diabetes Association: prediabetes is defined as impaired fasting glucose (IFG), between 100-125 mg/dL; impaired glucose tolerance (IGT) with glucose 140-199 after 75g oral glucose load; or A1c between 5.7-6.5%. Studies have suggested that A1c is the best single marker 

-- prediabetes is increasing globally, with scary numbers: by 2030 it is estimated to occur in >470 million people, and >70% of prediabetics develop diabetes 

-- In the US, the prevalence of prediabetes is 34.4%!!!!. in China it is 15.5% (though in Hong Kong, where this study was done, the rate may be different from the overall Chinese average) 

-- atrial fibrillation is increasing as well, since many of the risk factors are increasing (aging population, hypertension, diabetes, CAD, obesity) 

-- heart failure is the most frequent nonfatal cardiovascular event in those with AF, as well as the cause of death (probably because so many patients at high risk are on anticoagulation, decreasing systemic emboli leading to bad outcomes) 

-- other studies have found that those with prediabetes progress to overt diabetes about 5-10% annually (much higher in those with A1c closer to 6.5%), and the same general proportion revert to normoglycemia 

 

-- in this context of the study, I would like to highlight a 2020 updated meta-analysis that assessed the association between prediabetes and risk of all-cause mortality and cardiovascular disease (see dm prediabetes inc risk CVD mortality BMJ2020 in dropbox, or doi.org/10.1136/bmj.m2297): 

    -- 129 studies identified, all having adjusted for other risk factors in these numbers: more than 10 million individuals followed a median of 9.8 years; as compared to the general population, prediabetes versus normoglycemia was associated with (the relative risks were calculated over a follow-up of 9.8 years; absolute risk follow-up of 3.2 years): 

        -- all-cause mortality: 13% increased risk, RR 1.13 (1.10-1.17) 

            -- absolute risk: 7.36 per 10,000 person-years 

        -- composite cardiovascular disease: 15% increased risk, RR 1.15 (1.11-1.18) [composite cardiovascular disease was not defined and ??may have varied by the different individual studies ]

            -- absolute risk: 8.75 per 10,000 person-years 

        -- coronary heart disease: 16% increased risk, RR 1.16 (1.11-1.21) 

            -- absolute risk: 6.59 per 10,000 person-years 

        -- stroke: 14% increased risk, RR 1.14 (1.08-1.20) 

            -- absolute risk: 3.68 per 10,000 person-years 

    -- in patients with pre-existing atherosclerotic cardiovascular disease, prediabetes was associated with: 

        -- all-cause mortality: 36% increased risk, RR 1.36 (1.21-1.54) 

           -- absolute risk: 66.19 per 10,000 person-years 

        -- composite cardiovascular disease: 37% increased risk, RR 1.37 (1.23-1.53) 

            -- absolute risk: 189.77 per 10,000 person-years 

        -- coronary heart disease: 15% increased risk, RR 1.15 (1.02-1.29) 

            -- absolute risk: 8.54 years 

        -- stroke: 3.68: no significant difference 

-- there have been a few studies finding significant increases in cardiovascular disease in those with even lower A1c's: eg a European study found that in men, compared to an A1c <5%, those with A1c of 5-5.4% had a 56% increase in coronary heart disease, those 5.5-5.9% had relative risk RR 2.00, 6-6.4% had RR 2.13, 6.5-6.9% RR 3.44, >7% RR 7.07 (see dm A1C and cardiovasc dz annals 2004 in dropbox, or Khaw K, AnnInternMed 2004; 141: 413-420

 

Limitations: 

-- little information about the specifics: what was the actual blood pressure? What were the lipid values? 

-- this was a pretty healthy population with relatively few comorbidities. Would these results apply to other countries with more cardiovascular risk than in Hong Kong? 

    -- also, this group of diabetics had pretty well-controlled diabetes (A1c=7.3%). would these results apply to those with less well-controlled diabetes?

-- what were the factors that led to the changes from prediabetes to either diabetes or normoglycemia? Lifestyle changes (weight, diet, exercise…), medications, more general family/community support? 

    -- it would be great to have info on the weight/diet/exercise/smoking over time, not just at baseline 

-- no information on the types of AF (permanent, paroxysmal??), or the treatments done (medical/interventional)?, or what antiarrhythmics were used (some are associated with adverse cardiac events)? or if the treatments led to rate control or rhythm control 

-- no echocardiographic info: was there valvular disease in some? What was their LV ejection fraction and left atrial size in the subgroups?  

-- it seems very odd that only 40% were on statins given the age (75), pretty frequent comorbidities, and presence of diabetes in 40% and prediabetes in 16%; the number should probably be pretty close to 100%. This could limit the interpretation of their results

-- heart failure can be precipitated by tachycardia, and tachycardia-related heart failure is pretty commonly associated with patients who have rapid ventricular responses to the AF (i have seen several cases in my own panel of patients….). And perhaps there was cardiac autonomic neuropathy leading to more tachycardia/heart failure in those with diabetes and possibly even with prediabetes. No data provided on this 

 

So, the point here is really that prediabetes is a significant risk factor for bad cardiovascular outcomes: the development of heart failure in the first study, and the development of many atherosclerotic complications in the second. And this all suggests a few conclusions: 

-- we clinicians should be testing A1c levels regularly in patients, especially if there is any change in their weight, exercise, or if prior tests were near the prediabetes cutoff of 6.5% or increasingly getting nearer to that

-- prediabetes should be a strong signal for us and patients taking the situation very seriously: 

    -- it should be emphasized by us and internalized by patients that this “pre” form of diabetes is really serious, though the “pre” makes it sound like it is just a minor predilection 

    -- we should encourage patients to engage in necessary life style changes, especially weight loss, decreased carbohydrates in diet, regular exercise, stopping smoking. The Diabetes Prevention Program in the US and similar ones in Finland and China found that these types of lifestyle interventions were the best (even better than meds): see http://gmodestmedblogs.blogspot.com/2017/11/diabetes-lifestyle-changes-longer.html 

    -- we should consider meds for prediabetes if necessary: metformin has the “strongest evidence base” per the American Diabetes Association (see https://diabetesjournals.org/clinical/article/40/1/10/139035/Standards-of-Medical-Care-in-Diabetes-2022 ) 

    -- we should add prediabetes into our calculation of overall CVD risk, since it seems to add to the risk. 

        -- I should note that adding prediabetes to the standard 10-year risk calculators has not been found to change their accuracy much, but it is clear that the risk of bad CVD outcomes tracks with levels of A1c, and there may be a time lag:  those with prediabetes may take longer to manifest their CVD than those with higher A1c who have frank diabetes, so the 10-year risk calculator is not necessarily a good predictive instrument in most prediabetics with life expectancies beyond 10 years (see http://gmodestmedblogs.blogspot.com/2022/09/uspstf-recommendations-statins-in.html for a critique of the 10-year calculators)

        -- and, there are no long-term studies suggesting that aggressive treatment of prediabetes averts cardiovascular outcomes (though, seems pretty likely….., and the data in this study did find that reversion to normoglycemia did decrease the risk). And nonpharmacologic interventions do have a slew of other benefits for people anyway. but would be great to have long-term high quality studies 

        -- I personally do add prediabetes into my mental calculation of adding statin therapy and shooting for a lower LDL target, especially if the patient has other risk factors (smoking, hypertension, obesity, etc)  

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