2019 diabetes guidelines

The Am Diabetic Assn just released its Diabetes Care 2019 guidelines.  Will summarize and briefly comment on the relevant primary care sections (and mostly focus on type II diabetes), using bold type for the newer recommendations (I realize this is a long blog, but it is lots shorter than the paper….). these guidelines are more in line with the established guidelines from the European Association for the Study of Diabetes (EASD) and NICE (National Institute for Health and Care Excellence in the UK). my commentary is mostly embedded in the details.

for access to the reecommendations, see:

--dm diabetes care summary 2019 in dropbox, or http://care.diabetesjournals.org/content/42/Supplement_1/S4.full-text.pdf

--dm diabetes care meds2019 in dropbox, or http://care.diabetesjournals.org/content/42/Supplement_1/S90.full-text.pdf

or go to http://care.diabetesjournals.org/content/42/Supplement_1 to access all of the sections of the new guidelines in detail

Diagnostic tests

--testing should be done for anyone with BMI >25 (>23 in Asian Americans) and at least one risk factor (includes HDL<35, BP>140/90, physical inactivity, African-Americans/Latino/Native American/Asian American/Pacific Islander)

--Everyone >45 yo should be tested

--If normal, repeat in 3 years

--no change from past ranges (eg: diabetes is A1c>6.5%, FPG>126 mg/dL, 2-h PG with 75-g OGTT >200 mg/dL)

--A1c is most convenient, less day-to-day variation, but also less sensitive (A1c >6.5% diagnoses only 30% of diabetes cases, as compared to collectively testing A1c, FPG or 2-h PG)

    --African-Americans heterozygous for sickle cell trait: at any level of glycemia may have A1c 0.3% lower than those without HbS

    --those with x-linked G6PD G202A variant (11% of African-Americans) have an A1C decreased by 0.8% in homozygous men and 0.7% in homozygous women

    --African-Americans without these tend to have a higher A1C level with similar FPG and 2-hr PG levels than non-Hispanic whites (this may be true also for fructosamine or glycated albumin assessments, but conflicting data)

    --A1C is also less reliable in those with increased RBC turnover (eg sickle cell), pregnancy, post-partum, hymodialysis: only FPG should be used

--diagnosis should be confirmed by 2 tests (can be 2 on the same sample)

Prevention or Delay of diabetes

--in those with pre-diabetes, at least annual monitoring for diabetes

--lifestyle interventions help: overweight should lose/maintain 7% decreased weight, 150 min/wk moderate-intensity exercise, diet (Mediterranean, and low-calorie low-fat eating plan). [see my comments below on the low carb diets]

--consider metformin in those with prediabetes, esp if BMI>35, <60 yo, and women with prior gestational diabetes

  

Comprehensive Medical Evaluation

--patient-centered and team-based care

--develop care management plan

--use 10-yr ASCVD risk calculator (see below, on cardiovascular disease)

--usual vaccinations (routine, plus PPSV23). They also mention routine hep B vaccine til age 59, and consider in those older (?higher risk of transmission)

--fracture risk is higher in diabetics, independent of BMD. Make sure vitamin D is replete. Regularly assess fracture history and risk factors and recommend BMD as appropriate for age and sex. Avoid TZDs and SGLT-2 inhibitors

--annual depression screen

Lifestyle Management

--focus on patient-centered self-management education and support 

--nutrition: achieve and maintain body weight goals

    --any of several diets are fine, depends on what works best for the patient (DASH diet, plant-based diet, Mediterranean diet, low carbohydrate diet). They do comment on one study using the SGLT-2 inhibitor luseogliflozin, that there may be more risk for DKA in those on low carbohydrate diets. They do also acknowledge that the main dietary issue is diet maintenance, hard with any diet, though some longer studies do show more A1c and lipid improvements with low carbohydrate, low glycemic index diets

    -- sodium: restrict to <2,300 mg/d

    -- protein: no evidence of need to adjust intake, unless kidney disease (see below)

    -- fats:  controversial. Mediterranean diet is rich in poly- and mono-unsaturated fats and is healthy. But dietary supplement with n-3 fatty acids does not seem to help

    -- micronutrients: no clear benefit from herbal or nonherbal (vitamin or mineral) supplements

    -- alcohol: moderate alcohol intake does not affect glucose control much

    -- non-nutritive sweeteners: inconsistent results, with some studies showing weight loss and some weight gain.  Emphasis should be to stop non-nuitritive sweeteners, as well as sugar-sweetened beverages, and increase water intake

--exercise: 150 min of moderate-to-vigorous exercise per week, with no more than 2 consecutive days without activity

--prolonged sitting should be interrupted every 30 minutes for improved glucose control

--screen for depression, distress from diabetes, disordered eating, cognitive capabilities, with patient-appropriate validated tools.

Glycemic Targets 

--A1c should be checked at lease 2x/year in those with stable glycemic control, 4x/year otherwise

--“reasonable goal”: <7%

-- “might reasonably suggest”: <6.5% if can be achieved without significant hypoglycemia or other adverse effects. Esp in those with short duration diabetes, easily treated with minimal meds, long life expectancy, no significant cardiovascular disease [I would note: I do have several older patients with long-standing diabetes, heart disease, occasional hypoglycemia, etc who had A1c in the 10+% range, were put on a GLP-1 agonist, peeled back on their sulfonylurea and cut their insulin, and have A1c in the 6.5-7% range without hypoglycemia or other adverse effects. and they and i are all happy....]

-- “less stringent” goal:  <8%, if severe hypoglycemia, limited life expectancy, advanced micro or macrovasc complications, and difficult to get lower goal

Diabetes technology

--patients on intensive insulin treatment should monitor blood glucose prior to meals and snacks, at bedtime, and occasionally postprandially

--for those on less intensive treatments (eg basal insulin with or without oral agents), evidence less clear but probably should assess fasting glucose to help determine dose adjustments

--for those not on insulin, unclear that routine monitoring helps, other than if patient feels hypoglycemic [I do find that patients can use the glucose monitor after eating new foods/changes in eating patterns to get direct feedback about how those foods or changes are helping or hurting their glucose control, and I do think that this can be useful information to them, and to me]

Obesity management

--for obese individuals, diets should be individualized, and any that provide the same caloric restriction are equally effective [again, I would argue that the low glucose-index diet has some advantages, including that fact that it does not create a trigger of undernutrition to the body as happens with a high glucose-index diet, where the body lowers the basal rate to conserve energy, seehttp://gmodestmedblogs.blogspot.com/2018/09/high-glycemic-index-diet-causes-obesity.html ]

--consider diabetes meds that help lose weight: eg metformin, a-glucosidase inhibitors, SGLT-2 inhibitors, GLP-1 agonists, and amylin mimetics. Try to avoid insulin secretagogues (eg sulfonylureas), insulin, TZDs

--review all meds and try to decrease those associated with weight gain (eg antipsychotics, some antidepressants, gabapentin, etc)

--can consider weight-loss meds, but only short-term and stop them if not working: eg phentermine or other FDA-approved meds (orlistat, lorcaserin, topiramate,  liraglutide, naltrexone/buproprion) [my experience here is really limited. Too many potential adverse effects for me to prescribe them. And benefits do not seem to be long-standing]

--surgery: “should be recommended as an option” if BMI >40 (>37.5 in Asian Americans), and in those who do not achieve durable weight loss/improvement in comorbidities with BMI 35-40 (32.5-37.5 in Asian Americans)

    --may be an option in that group with BMI 30-35 (27.5-32.5 in Asian Americans)

Drug Therapy 

-- type I diabetes: no significant change. Insulin is mainstay. No consensus on choosing which form of insulin, though the analogues may be associated with less hypoglycemia, less weight gain, and lower A1c. Adding oral drugs does not add much. Metformin helps a bit with body weight and lipid levels but does not change A1c.

-- Type II diabetes:

    -- Metformin: still the preferred initial drug. Renal dosing, as per blog noted below. Vitamin B12 deficiency associated with long-term metformin use, should be checked periodically. [older studies have shown that metformin 1000mg/d is about 85% as good as 2000 mg/d.  my pretty extensive experience is that people get very dramatic benefit with fewer adverse effects with just 500mg once a day. and occasionally with 250mg once a day if intolerable GI distress. and there is less GI distress if take with meals}

    -- adding second agent: each lowers A1c approximately 0.7-1.0%. Any of the six treatment options added to metformin are reasonable: sulfonylureas, thiazolidinedione, DPP-4, SGLT2, GLP-1, or basal insulin

        -- no empiric evidence for which agent to choose if no CAD or CKD.

    -- when A1c is greater than 1.5 percent above glycemic target, many patients will require dual therapy [i also question this: many patients with very high blood sugars do initially need metformin with insulin to get the blood sugar down, given that the high blood sugars themselves cause "glucotoxicity" where the body's insulin is less effective in promoting glucose uptake into muscle. but with decreasing blood sugars and enhanced insulin sensitivity, some do well just with metformin, and i have had several patients who are ultimately extremely well-controlled just with diet/exercise]

    -- consider insulin if catabolic features are present (weight loss, hypertriglyceridemia, ketosis), though can often simplify to oral agents after glucotoxicity resolves. Though insulin analogs may reduce symptomatic and nocturnal hypoglycemia compared to human insolence,” in practice they made not affect the development of hypoglycemia compared with NPH insulin”. [And much lower cost, more on this tomorrow]

--“ in most patients who need the greater glucose lowering effect of an injectable medication, GLP-1 receptor agonists are preferred to insulin” [i completely agree, though this does contradict their bullet above suggesting that one can choose any of the 6 treatment options without further guidance]

--“ among patients with type II diabetes who have established atherosclerotic cardiovascular disease, SGLT-2 inhibitors or GLP-1 receptor agonists have demonstrated cardiovascular disease benefit and are recommended as part of the anti-hyperglycemic regimen”

--“ among patients with atherosclerotic cardiovascular disease at high risk of heart failure or in whom heart failure coexists, SGLT-2 inhibitors are preferred". [see my comments below]

Cardiovascular Disease

-- they recommend using the ACC/AHA ASCVD risk calculator (Risk Estimator Plus) for estimating the 10 year risk

-- prediabetes is a cardiac risk factor; so screen for and treat other modifiable risk factors

-- hypertension:

    -- hypertension is defined as greater than 140/90 mmHg, though they do comment that a target of less than 130/80 mmHg may be reasonable in diabetics with hypertension and at higher cardiovascular risk, either existing ASCVD or 10-year risk >15% by calculator [see new hypertension recommendations that do reinforce the lower goal: http://gmodestmedblogs.blogspot.com/2017/11/new-aha-hypertension-guidelines.html ]

    -- they also do recommend both home blood pressure self-monitoring and 24-hour ambulatory blood pressure monitoring (see above htn guidelines, as well as http://gmodestmedblogs.blogspot.com/2015/01/uspstf-recs-on-ambulatory-blood.html )

    -- lifestyle management is an important component of treatment, including reducing excess body weight, restricting sodium intake (less than 2300 mg a day) increasing consumption of fruits and vegetables and low-fat dairy products, avoiding excess alcohol, and increasing activity levels (see above section for more details)

    -- pharmacologic interventions

        -- treat if office-based blood pressure when confirmed is > 140/90

        -- use 2 drug therapy initially if >160/100 [i think this is overstated, both because there is no real harm in lowering the blood pressure more slowly since it has very likely been quite high for awhile and there may be compensatory vascular remolding which might lower the effective pressure more to end-organs, and because there really is a lot of variability in patient response to even low doses of BP meds. the harm of overtreatment to me outweighs the unlikely benefit, assuming that the patient is followed closely]

       -- use ACE inhibitors or ARBs if microalbuminuria present (though no benefit of these drugs as compared to thiazide-like diuretics or dihydropyridine calcium channel blockers in the absence of albuminuria)

       -- consider bedtime dosing (given the prognostic importance of nocturnal blood pressure and the incidence of ASCVD) [an article a few years ago found that evening dosing of BP meds might lower diabetes incidence: see http://gmodestmedblogs.blogspot.com/2015/09/take-blood-pressure-meds-at-night-to.html  ]

-- lipid management

    -- lifestyle: promoting Mediterranean diet or Dietary Approaches to Stop Hypertension (DASH diet), reduction of saturated and trans fats, increased dietary and n-3 fatty acids, viscous fiber, and plant sterols/sterols, increased physical activity, and optimizing glycemic control for those with high triglycerides and low HDL levels

    -- drugs:     

        -- high intensity statin for all patients of all ages with diabetes and ASCVD disease or calculated risk >20% [as i have commented on previously, i find this to be too cookie-cutter: i have had patients with very dramatic LDL decreases on moderate-intensity statins (one recent patient on atorvastatin 20mg went from LDL of 183 to 90), and their binary classification of statin potency does not make sense (the "high-intensity" rosuvastatin 40mg can be much more effective than the "high-intensity" atorvastatin 40mg). so, again as with hypertension, i often start in the moderate range, follow patients closely, then augment the statin til the goal LDL is achieved, occasionally adding other lipid meds]

        -- for patients with diabetes <40 years old additional atherosclerotic risk factors, consider moderate-intensity statin

        -- diabetics 40 to 75 years old, as well as greater than 75 without atherosclerotic disease, use moderate-intensity statins

        -- for those with diabetes and atherosclerotic disease, if after LDL >70 on maximally tolerated statins, considering adding ezetimibe or PCSK9-inhibitor

        -- for those with fasting triglycerides >500  mg/dL, look for secondary causes and consider medical therapy to reduce the risk of pancreatitis. For those with lower levels of triglycerides, treat the lifestyle factors (obesity/metabolic syndrome), secondary factors (chronic liver or kidney disease, hypothyroidism), and try to decrease medications that raise triglycerides

    -- antiplatelet drugs: use low-dose aspirin for secondary prevention. Clopidogrel if aspirin allergy

    -- low-dose aspirin “may be considered as a primary prevention strategy in those with diabetes who are increased cardiovascular risk, after discussion with the patient on the benefits versus increased risk of bleeding” [the studies on aspirin in diabetics are inconsistent, hence the "may be considered"]

-- cardiovascular disease:

    -- routine screening is not recommended

    -- if known ASCVD, consider ACE inhibitor or ARB to reduce risk of cardiovascular events

    -- beta blockers should be used for at least two years after an MI

    -- Metformin is okay with stable heart failure but should be avoided in unstable hospitalized patients with heart failure

    -- heart failure (a new section in these guidelines): 50% of patients with type II diabetes may develop heart failure. Some drugs in those with heart failure should be avoided (TZDs), some have mixed results (DPP-4 inhibitors, some with no evidence of benefit (GLP-1 agonists). 

     -- in those who have established ASCVD, SGLT-2 inhibitors or GLP-1 agonists have demonstrated cardiovascular benefits. SGLT-2 inhibitors are preferred in those at high risk of heart failure or already have heart failure (studies have shown decreased heart failure in those with and without a prior history of heart failure). [though one might argue that the SGLT-2 inhibitors act as loop diuretics. And this may not be the best heart failure therapy: perhaps patients would be better servied with a real, titratable loop diuretic, along with ancillary meds as indicated by their type of heart failure???}

Microvascular complications

Kidney:

--the usual: check urinary microalbuminuria annually

--optimize glycemic control to reduce risk or delay progression of chronic kidney disease

--SGLT-2 inhibitors and GLP-1 agonists reduce the risk of chronic kidney disease progression, cardiovascular disease, or both

--optimize BP control

--dietary protein intake in those with non-dialysis dependent CKD should have 0.8 g/kg protein

--ACE or ARBs are recommended if urinary albumin/creatinine ratio >30 and/or eGFR <60. But not for primary prevention of renal disease in those who are not hypertensive

Eye:

--optimize glucose control to reduce risk or slow progression; other risk factors for diabetic ophthalmopathy include hypertension, nephropathy, and dyslipidemia

--if dilated/comprehensive eye exam is normal, exams can be repeated every 1-2 years

--retinopathy is not a contraindication to aspirin therapy

Nerves:

--for distal symmetric polyneuropathy, check temperature or pinprick (small-fiber function), and vibration with 128-Hz tuning fork and 10-g monofilament (large-fiber function)

--annual 10-g monofilament testing to check feet for risk of ulceration/amputation

--optimize glucose control to prevent or delay progression of neuropathy

--initial drug treatment recommended: pregabalin, duloxetine, or gabapentin [I still continue to use tricyclics as first-line, with great success. Venlafaxine also works quite well.  And a recent blog found that gabapentinoids (pregababln and gabapentin) were not very effective: http://gmodestmedblogs.blogspot.com/2017/04/diabetic-peripheral-neuropathy-and-more.html  )

--check for autonomic neuropathy: hypoglycemia unawareness, resting tachycardia, orthostatic hypotension, gastroparesis, constipation, diarrhea, fecal incontinence, erectile dysfuntciton, increased or decreased sweating)

Foot care

--annual comprehensive foot exam

--inspect feet every visit, but only in those with evidence of sensory loss or prior ulceration/amputation

Older Adults

--Screen for early  detection of mild cognitive impairment and dementia in those >65yo (both higher incidence of these in diabetics, and their presence may affect their treatment). Seehttp://gmodestmedblogs.blogspot.com/2017/12/metformin-decreases-dementia-risk.html , a retrospective study suggesting that metformin decreased risk of dementia, vs sulfonylureas

--avoidance of hypoglycemia. More common in elderly (more use of insulin since they are more likely to be insulin deficient, as well as increased renal dysfunction with aging). May lead to de-escalation/simplification of therapy

--screening for diabetic complications may be different in elderly and more individualized, with more focus on those complications that might affect function more

--metformin is still the first line agent, though, as with all patients, should be prescribed at a dose per renal function: see http://gmodestmedblogs.blogspot.com/2018/06/metformin-and-lactic-acidosis-again.html

Children and Adolescents

--screen for type II diabetes esp in those who are overweight (BMI>85^th%) or obese (BMI >95^th%), and have at least one risk factor, as per screening recommendations above. Repeat in 3 years if normal

--start meds and lifestyle counseling at diagnosis of type II diabetes (usually with metformin)

--if more meds are needed, add basal insulin (not much data on other meds in kids)

--consider metabolic (gastric) surgery if markedly obese

--more attention to the psychosocial issues, stresses (patient and family) in kids with diabetes (type I or II)

--counseling is often helpful

--see guidelines for specific recommendations for type I diabetes (http://care.diabetesjournals.org/content/42/Supplement_1/S148 )

Some more general comments (specific ones embedded in text above)

--these guidelines both reflect and perpetuate one of the (many) fundamental problems in the US health care system: lack of coherence/consistency.  unlike the UK's NICE guidelines, for example, we have 2 agencies (American Diabetes Assn and the  Am Heart Assn)  putting out different guidelines for goals of hypertension in diabetics.  it really makes sense for the (underfunded) NIH to continue their prior task of being a major convener and synthesizer of clinical guidelines, which are now booted out to the specialty societies that may well have different approaches and biases in developing these guidelines.  the current system is confusing for patients and clinicians, and i think reflects the fragmented approach we have to health care in general.

--for recommendations on the used os SGLT-2 inhibitors or GLP-1 agonists, the guidelines focus on diabetic patients with ASCVD. the reason for this limited focus is that those were the patients the studies included, finding benefit. the issue here is that 80% of diabetics die from ASCVD, and the general pretest assumption is that diabetics have atherosclerotic disease. But the actual studies done were designed by the drug companies, who want the smallest, most advanced group of patient that would prove benefits for their drug (otherwise would need many more patients and following them for longer). so, we clinicians and patients do not necessarrily get the answers to the important clinical questions we have ("should we use these drugs preferentially in diabetics overall, who may not have manifest but do likely have silent ASCVD???"). This is a fundamental issue with "Evidence-Based Medicine": the evidence we have is distorted by the design of the study, and this design is almost always by large corporations (drug companies) who just want to show enough benefit as quickly and cheaply as possible to get by the FDA. But "EBM" mantra suggests scientific objectivity.

    --i should note here that the ADA guidelines above do promote GLP-1 over insulin when choosing an injectable.  this is a bit inconsistent with their flow chart highlighting this drug in those with ASCVD. And my comments (as per many blogs): these are now largely once-a-week injections, are really easy (unlike the initial long-acting exenatide), hurt probably less than fingersticks for blood sugar,  and are NOT insulin (which patients may be afraid of, and many studies do suggest may promote atherosclerotic disease)

--as another recent guideline perspective, see http://gmodestmedblogs.blogspot.com/2018/12/am-coll-of-cardiol-diabetes-rx.html , which details the Am College of Cardiology perpective on the SGLT-2 inhibitors and GLP-1 receptor agonsits, along with my commentary

--but, overall, i think these new guidelines do reflect an increasing shift away from just looking at the surrogate marker of A1c and more to diabetes control in the context of the overwhelmingly important cardiovascular outcomes, highlighting drugs that are cardioprotective (metformin, GLP-1 agonists, and probably SGLT-2 inhibitors). this last blog  (http://gmodestmedblogs.blogspot.com/2018/12/am-coll-of-cardiol-diabetes-rx.html ) does highlight some of my concerns about the SGLT-2's....

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