new diabetes guidelines from ADA

the Am Diabetes Assn just released their 2016 update on standards of medical care (see dm diabetes care summary 2016 in dropbox), which, of perhaps some interest in terms of their exposure, is now in the annals of internal medicine instead of the journal diabetes care.  and, if you want the full 119 page document, see http://care.diabetesjournals.org/content/suppl/2015/12/21/39.Supplement_1.DC2/2016-Standards-of-Care.pdf . the main points in the new guidelines are (i am limiting to type 2 diabetes  -- DM2, and not including the brief section on in-hospital management):

--diagnosis of diabetes: not much change: A1c >=6.5, fasting >=126 mg/dL, oral GTT >200 mg/dL or 11.1 mmol/L (should be confirmed by repeat testing), random >200 mg/dL. (the only minor difference is that the fasting >=126 used to require confirmation unless unequivocal symptoms of hyperglycemia, and same with the oral GTT >200 mg/dL . the random >200 only diagnoses DM2 if there are classic symptoms of hyperglycemia [i must admit: this always struck me as a bit weird. an oral GTT of 201 mg/dL, now needing a repeat perhaps showing 202 mg/dL  qualifies for DM2, but an asymptomatic patient with blood sugar of 450 mg/dL does not....].  women with gestational diabetes should get screened for diabetes 6-12 weeks after delivery and at least every 3 years subsequently. not much change in the definitions of prediabetes, with A1c 5.7-6.4, fasting 100-125, oral GTT of 140-199.

--glycemic targets: 

    ​--self-monitoring of blood glucose (SMBG, or glucometer) is important especially in patients on intensive insulin regimens. insufficient evidence on others. and using glucometers is not associated with improved glucose unless this is part of and integrated clinical and self-management plan. [i think this last point is important. my sense, as in a review in BMJ a few years ago, is that clinicians often do not really integrate glucometer readings into active DM2 management, leading the BMJ to conclude glucometers are not very useful]

    --A1C. testing recs not change significantly. do a minimum of 2x/year and every 3 months if patient not meeting goal or changing regimens. best to use it in conjunction with SMBG (need SMBG to assess variability of control). also A1C levels are tied to red cell life: those with hemolysis have shortened RBC life, less time for the RBCs to be glycated nonenzymatically, and lower A1C levels for the same blood sugars; those with a skew to older cells (eg, they have iron-deficiency and fewer new cells created) would have higher A1Cs since their cells are hanging out longer with the glucose in the blood [my addition, but see JAMA. 2016;315(6):605 for more of a discussion].

    --A1C goals (nonpregnant adults): for most, goal of 7%. consider 6.5% in those with short duration of DM2, healthy people with long life expectancy. and goal of <8% in those with history of severe hypoglycemia, limited life expectancy [though, i would add, that i have several patients where the best we can do safely is a tad higher than 8%...]

--hypoglycemia. patients at high risk should get glucagon and a close contact should learn how to give it.

    --hypoglycemia unawareness happens sometimes in patients with tight control, where they lose their counterregulatory hormone release and have decreased autonomic responses. for the first time i've seen, they recommend these patients "be advised to increase their glycemic targets for at least several weeks to partially reverse hypoglycemia unawareness and reduce the risk for future episodes".  [seems pretty reasonable to me, given that the first symptom of hypoglycemia in these patients can be coma, or death]

--medical management of diabetes

    --need to address behavioral, dietary, lifestyle and drugs [i am surprised how often we as providers, perhaps a bit inured/numbed by the frequency we see diabetic patients, do not step back and talk to the patient about the really profound effects that diabetes has on their whole lives, both psychosocially (effects on their functioning individually, psychologically and physically, and in the setting of their families/communities) and medically (with the profound comorbidities of diabetes). i certainly support the lifestyle recommendations regarding exercise (>150 min/week), and nutrition (which they prefer being done by nutritionist, though i'm not sure that is always best), and self-management programs.  the psychosocial piece is not strongly enough represented here...]

    ​--initial therapy -- start with lifestyle changes, including losing 5% of body weight if overweight/obese. metformin is first drug choice, and they appropriately expanded their renal cutoff down to 30-45 mL/min with decreased doses of metformin (see http://gmodestmedblogs.blogspot.com/2015/01/metformin-in-renal-failure.html  for further discussion). a second agent can be added as needed, and can be any of six: sulfonylurea, thiazolidinediones (TZDs), dipeptidyl peptidase-4 (DPP-4) inhibitors, sodium-glucose cotransporter 2  (SGLT-2) inhibitors, glucagon-like peptide-1 (GLP-1) agonists, or basal insulin. they recommend starting 2 agents if A1C>9% [i personally almost never start 2 agents at once because the risk of medication adverse effects increases, and because in those with DM2, they have very slowly developed high blood sugars and i see no reason physiologically to decrease the blood sugar very rapidly, with concern about potentially adverse effects of more rapid intracellular/osmotic changes -- ie, potential risk and i'm not sure there is any benefit]. they do recommend more prominently combo therapy of long-acting insulin with GLP-1 agonists (which i have used quite successfully,a s well as the combo of GLP-1 agonist plus metformin)

--cardiovascular risk factor management:

    ​--blood pressure goal:140/90 mmHg. be careful in older adults if BP< 130/70. try to use either ACE-I or ARB (but not both), if needed beyond lifestyle therapy

    --lipids: (some new recommendations): lifestyle interventions, plus

        --<40yo: moderate or high intensity statin if ASCVD risk factors (LDL >100 mg/dl, hypertension, smoking, overweight/obese, family history); high intensity statin if ASCVD

        --40-75yo: no risk factors -- moderate intensity statin; ASCVD risk factors or ASCVD -- high intensity statin; acute coronary syndrome, LDL>50, or inability to take high intensity statin -- moderate plus ezetimibe;

        ​-->75yo: same as 40-75 yo, but moderate or high intensity ​statin if only has ASCVD risk factors. 

        --high-intensity statin = atorvastatin 40-80; rosuvastatin 20-40; most of rest=moderate-intensity [note: if you are inclined to target LDL levels, as i am, there really is a pretty big difference between atorvastatin 40 and rosuvastatin 40]

        --they do not recommend combining a statin with a fibrate, but should be considered in men if triglycerides >204 mg/dL and HDL <35 mg/dL. [given lack of data of cardiovascular benefit of the combination, i tend to combine the 2 only if the triglycerides are even higher, like >500 mg/dL, and only use fenofibrate given fewer adverse events than the combo with gemfibrozil, though the latter is probably clinically more effective. but based on the TNT trial (N Engl J Med 2007;357:1301​), i do target a lower LDL in patients who have lower HDL levels]

    --antiplatelet agents:

        ​--aspirin 75-162 mg "recommended" (note: this is a somewhat higher endorsement of aspirin than in the past few years of these guidelines) if 10-year cardiovasc risk >10%, and not if <5%. clopidogrel if aspirin-allergic

--management of microvascular disease:

        --kidney: obtain best A1c you can, annual screening with microalbumin on spot urine and eGFR; and if microalbumin is high, repeat at least once to confirm [i would add that microalbumin can be high in setting of hyperglycemia. so check it when the patient is at baseline blood sugars]. use ACE-I, ARB if eGFR<60 and urine alb/creat ratio >300 [for some reason, they did not include the prior recommendation to consider these meds if ratio >30 mg/g, which i still would endorse and actually is suggested in the full text article noted above]

        --eyes: obtain best A1c, blood pressure, and lipid control. annual eye exam, and NOT substitute retinal photographs [retinal photographs are trashed in the summary, though in the full text article, they do think it has value though is not a substitute for a full eye exam when that is available. also there are some data suggesting that meds, esp ACE-I/ARBs help here, not sure why they are not mentioned]

        --nerves: obtain best A1c, preferred drugs are pregabalin, duloxetine, tapentadol. but others used include tricyclics, gabapentin, venlafaxine, carbamazepine, topical capsaicin, tramadol. [i have had remarkable success with low-dose desipramine or nortriptylline over the decades, and these are usually the first drugs i try]

        ​--feet: annual 10-g monofilament test plus pinprick, vibration, or ankle reflex check annually. also general foot exam to check skin, deformities, pulses [in my experience, i am occasionally surprised to find major ulcers/other problems without the patient knowing it. so i do check feet pretty regularly]. higher risk of problems if PAD, peripheral neuropathy, smoking. consider screening for PAD at age 50 or if >10 years of DM 

--in terms of conflicts of interest, 2 of the 4 writers get $$ from industry, one MD and one PA do not. the chair of the committee did not, and 21/26 other committee members did not

so, a few additional points:

--as mentioned in many prior blogs on diabetes (see http://gmodestmedblogs.blogspot.com/search/label/diabetes ), we are treating patients not lab values. these recommendations are a bit better than previous ones, since they still focus on A1C levels but incorporate SMBG determinations to assess blood sugar variability. so, one of my patients with an A1c of 12 on a good day, but has SBGM levels ranging from 40-400, should not get a cranking up of her diabetes regimen (or the 40 will become an unconscious 20). the trick is to find out why there is such variability (usually either dietary or medication nonadherence) and try to work on that. but sometimes one can't get the A1c down much more....

--the above link to diabetes blogs includes several concerns with some of the newer agents (eg severe DKA with  SGLT-2 inhibitors; joint pains with DPP-4 inhibitors, etc). the only ones i use clinically are the older ones (pioglitazone, which has some data suggesting cardioprotection; and the GLP-1 agonists, which are more targeted than the DPP-4 inhibitors, are more effective, have been around for awhile, are associated with weight loss and seem to me to be pretty physiologic).

--i remain somewhat underwhelmed by the utility of ezetimibe (see blog: http://gmodestmedblogs.blogspot.com/2015/06/improve-it-trial-ezetimibe.html ) and have not used it to date

--these guidelines are a bit different from the recent ones by the Am Assn of Clinical Endocrinologits and Am College of Endocrinology (see http://gmodestmedblogs.blogspot.com/2016/01/new-dm-management-algorithm.html  , with my more extensive comments