xxxxx

1.0.0.0 AS­SESS­MENT OF GLYCEMIC CON­TROL

The Amer­i­can Di­a­betes As­so­ci­a­tion (ADA) “Stan­dards of Med­i­cal Care in Di­a­betes” in­cludes ADA’s cur­rent clin­i­cal prac­tice rec­om­men­da­tions and is in­tend­ed to pro­vide the com­po­nents of di­a­betes care, gen­er­al treat­ment goals and guide­lines, and tools to eval­u­ate qual­i­ty of care. Mem­bers of the ADA Pro­fes­sion­al Prac­tice Com­mit­tee, a mul­ti­dis­ci­plinary ex­pert com­mit­tee, are re­spon­si­ble for up­dat­ing the Stan­dards of Care an­nu­al­ly, or more fre­quent­ly as war­rant­ed. For a de­tailed de­scrip­tion of ADA stan­dards, state­ments, and re­ports, as well as the ev­i­dence-‍grad­ing sys­tem for ADA’s clin­i­cal prac­tice rec­om­men­da­tions, please refer to the Stan­dards of Care In­tro­duc­tion. Read­ers who wish to com­ment on the Stan­dards of Care are in­vit­ed to do so at pro­fes­sion­al.di­a­betes.org/‍SOC.

AS­SESS­MENT OF GLYCEMIC CON­TROL

Glycemic man­age­ment is pri­mar­i­ly as­sessed with the A1C test, which was the mea­sure stud­ied in clin­i­cal tri­als demon­strat­ing the benefits of im­proved glycemic con­trol. Pa­tient self-‍mon­i­tor­ing of blood glu­cose (SMBG) may help with self-‍man­age­ment and med­i­ca­tion ad­just­ment, par­tic­u­lar­ly in in­di­vid­u­als tak­ing in­sulin. Con­tin­u­ous glu­cose mon­i­tor­ing (CGM) also has an im­por­tant role in as­sess­ing the ef­fec­tive­ness and safe­ty of treat­ment in many pa­tients with type 1 di­a­betes, and lim­it­ed data sug­gest it may also be help­ful in se­lect­ed pa­tients with type 2 di­a­betes, such as those on in­ten­sive in­sulin reg­i­mens (1).

Sug­gest­ed ci­ta­tion: Amer­i­can Di­a­betes As­so­ci­a­tion. 6. Glycemic tar­gets: Stan­dards of Med­i­cal Care in Di­a­betesd2019. Di­a­betes Care 2019; 42(Suppl. 1):S61–S70 © 2018 by the Amer­i­can Di­a­betes As­so­ci­a­tion. Read­ers may use this ar­ti­cle as long as the work is prop­er­ly cited, the use is ed­u­ca­tion­al and not for prof­it, and the work is not al­tered. More in­for­ma­tion is avail­able at http://www.di­a­betesjournals.org/‍con­tent/‍license.

xxxxx

2.0.0.0 A1C Test­ing

2.1.0.0 Rec­om­men­da­tions

Rec­om­men­da­tions

Per­form the A1C test at least two times a year in pa­tients who are meet­ing treat­ment goals (and who have sta­ble glycemic con­trol). E

Per­form the A1C test quar­ter­ly in pa­tients whose ther­a­py has changed or who are not meet­ing glycemic goals. E

Point-‍of-‍care test­ing for A1C pro­vides the op­por­tu­ni­ty for more time­ly treat­ment changes. E

A1C reflects av­er­age glycemia over ap­prox­i­mate­ly 3 months. The per­for­mance of the test is gen­er­ally ex­cel­lent for NGSP-‍certified as­says (www.ngsp.org). The test is the major tool for as­sess­ing glycemic con­trol and has strong pre­dic­tive value for di­a­betes com­pli­ca­tions (1-3). Thus, A1C test­ing should be per­formed rou­tine­ly in all pa­tients with di­a­betesdat ini­tial as­sess­ment and as part of con­tin­u­ing care. Mea­sure­ment ap­prox­i­mate­ly every 3 months de­ter­mines whether pa­tients’ glycemic tar­gets have been reached and main­tained. The fre­quen­cy of A1C test­ing should de­pend on the clin­i­cal sit­u­a­tion, the treat­ment reg­i­men, and the clin­i­cian’s judg­ment. The use of point-‍of-‍care A1C test­ing may pro­vide an op­por­tu­ni­ty for more time­ly treat­ment changes dur­ing en­coun­ters be­tween pa­tients and pro­viders. Pa­tients with type 2 di­a­betes with sta­ble glycemia well with­in tar­get may do well with A1C test­ing only twice per year. Unsta­ble or in­ten­sively man­aged pa­tients (e.g., preg­nant women with type 1 di­a­betes) may re­quire test­ing more fre­quent­ly than every 3 months (4).

xxxxx

2.2.0.0 A1C Lim­i­ta­tions

The A1C test is an in­di­rect mea­sure of av­er­age glycemia and, as such, is sub­ject to lim­i­ta­tions. As with any lab­o­ra­to­ry test, there is vari­abil­i­ty in the mea­surement of A1C. Al­though such vari­abil­i­ty is less on an in­train­di­vid­u­al basis than that of blood glu­cose mea­surements, clin­i­cians should ex­er­cise judg­ment when using A1C as the sole basis for as­sess­ing glycemic con­trol, par­tic­u­lar­ly if the re­sult is close to the thresh­old that might prompt a change in med­i­ca­tion ther­a­py. Con­di­tions that af­fect red blood cell turnover (hemolyt­ic and other ane­mias, glu­cose-‍6-‍phos­phate de­hy­dro­ge­nase deficien­cy, re­cent blood trans­fu­sion, use of drugs that stim­u­late ery­thro­poe­sis, end-‍stage kid­ney dis­ease, and preg­nan­cy) may re­sult in dis­crep­an­cies be­tween the A1C re­sult and the pa­tient’s true mean glycemia. Hemoglobin vari­ants must be con­sid­ered, par­tic­u­lar­ly when the A1C re­sult does not cor­re­late with the pa­tient’s SMBG lev­els. How­ev­er, most as­says in use in the U.S. are ac­cu­rate in in­di­vid­u­als het­erozy­gous for the most com­mon vari­ants ( www.ngsp.org/‍interf.asp). Other mea­sures of av­er­age glycemia such as fruc­tosamine and 1,5-‍an­hy­droglu­ci­tol are avail­able, but their trans­la­tion into av­er­age glu­cose lev­els and their prog­nos­tic significance are not as clear as for A1C. Though some vari­abil­i­ty in the re­la­tion­ship be­tween av­er­age glu­cose lev­els and A1C ex­ists among dif­fer­ent in­di­vid­u­als, gen­er­ally the as­so­ci­a­tion be­tween mean glu­cose and A1C with­in an in­di­vid­u­al cor­re­lates over time (5).

A1C does not pro­vide a mea­sure of glycemic vari­abil­i­ty or hy­po­glycemia. For pa­tients prone to glycemic vari­abil­i­ty, es­pe­cial­ly pa­tients with type 1 di­a­betes or type 2 di­a­betes with se­vere in­sulin deficien­cy, glycemic con­trol is best eval­u­ated by the com­bi­na­tion of re­sults from SMBG or CGM and A1C. A1C may also in­form the ac­cu­ra­cy of the pa­tient’s meter (or the pa­tient’s re­port­ed SMBG re­sults) and the ad­e­qua­cy of the SMBG test­ing schedule.

xxxxx

2.3.0.0 A1C and Mean Glu­cose

Table 6.1 shows the cor­re­la­tion be­tween A1C lev­els and mean glu­cose lev­els based on two stud­ies: the in­ter­na­tion­al A1C-‍De­rived Av­er­age Glu­cose (ADAG) study, which as­sessed the cor­re­la­tion be­tween A1C and fre­quent SMBG and CGM in 507 adults (83% non-‍His­pan­ic whites) with type 1, type 2, and no di­a­betes (6), and an em­pir­i­cal study of the av­er­age blood glu­cose lev­els at pre­meal, post­meal, and bed­time as­so­ci­at­ed with specified A1C lev­els using data from the ADAG trial (7). The Amer­i­can Di­a­betes As­so­ci­a­tion (ADA) and the Amer­i­can As­so­ci­a­tion for Clin­i­cal Chem­istry have de­ter­mined that the cor­re­la­tion (r = 0.92) in the ADAG trial is strong enough to jus­ti­fy re­port­ing both the A1C re­sult and the es­ti­mat­ed av­er­age glu­cose (eAG) re­sult when a clin­i­cian or­ders the A1C test. Clin­i­cians should note that the mean plas­ma glu­cose num­bers in the table are based on

˜2,700 read­ings per A1C in the ADAG trial. In a re­cent re­port, mean glu­cose mea­sured with CGM ver­sus cen­tral lab­o­ra­to­ry–mea­sured A1C in 387 par­tic­i­pants in three ran­dom­ized tri­als demon­strat­ed that A1C may un­der­es­ti­mate or over­es­ti­mate mean glu­cose (5). Thus, as sug­gested, a pa­tient’s CGM profile has con­sid­er­able po­ten­tial for op­ti­miz­ing his or her glycemic man­age­ment (5).

xxxxx

2.4.0.0 A1C Dif­fer­ences in Eth­nic Pop­u­la­tions and Chil­dren

In the ADAG study, there were no significant dif­fer­ences among racial and eth­nic groups in the re­gres­sion lines be­tween A1C and mean glu­cose, al­though the study was un­der­pow­ered to de­tect a dif­fer­ence and there was a trend to­ward a dif­fer­ence be­tween the African/‍African Amer­i­can and non-‍His­pan­ic white co­horts, with high­er A1C val­ues ob­served in Africans/‍African Amer­i­cans com­pared with non-‍His­pan­ic whites for a given mean glu­cose. Other stud­ies have also demon­strat­ed high­er A1C lev­els in African Amer­i­cans than in whites at a given mean glu­cose con­cen­tra­tion (8,9).

A1C as­says are avail­able that do not demon­strate a sta­tis­ti­cal­ly significant dif­fer­ence in in­di­vid­u­als with hemoglobin vari­ants. Other as­says have sta­tis­ti­cal­ly significant in­ter­fer­ence, but the dif­fer­ence is not clin­i­cally significant. Use of an assay with such sta­tis­ti­cal­ly significant in­ter­fer­ence may ex­plain a re­port that for any level of mean glycemia, African Amer­i­cans het­erozy­gous for the com­mon hemoglobin vari­ant HbS had lower A1C by about 0.3 per­cent­age points when com­pared with those with­out the trait (10,11). An­oth­er ge­net­ic vari­ant, X-‍linked glu­cose-‍6-‍phos­phate de­hy­dro­ge­nase G202A, car­ried by 11% of African Amer­i­cans, was as­so­ci­at­ed with a de­crease in A1C of about 0.8% in hem­izy­gous men and 0.7% in ho­mozy­gous women com­pared with those with­out the trait (12).

A small study com­par­ing A1C to CGM data in chil­dren with type 1 di­a­betes found a high­ly sta­tis­ti­cal­ly significant cor­re­la­tion be­tween A1C and mean blood glu­cose, al­though the cor­re­la­tion (r = 0.7) was significant­ly lower than in the ADAG trial (13). Whether there are clin­i­cally mean­ing­ful dif­fer­ences in how A1C re­lates to av­er­age glu­cose in chil­dren or in dif­fer­ent eth­nicities is an area for fur­ther study (8,14,15). Until fur­ther ev­i­dence is avail­able, it seems pru­dent to es­tab­lish A1C goals in these pop­u­la­tions with con­sid­er­a­tion of both in­di­vid­u­alized SMBG and A1C re­sults.

xxxxx

3.0.0.0 Glu­cose As­sess­ment

For many peo­ple with di­a­betes, glu­cose mon­i­tor­ing is key for the achieve­ment of glycemic tar­gets. Major clin­i­cal tri­als of in­sulin-treat­ed pa­tients have in­clud­ed SMBG as part of mul­ti­fac­to­ri­al in­ter­ven­tions to demon­strate the benefit of in­ten­sive glycemic con­trol on di­a­betes com­pli­ca­tions (16). SMBG is thus an in­te­gral com­po­nent of ef­fec­tive ther­a­py of pa­tients tak­ing in­sulin. In re­cent years, CGM has emerged as a com­ple­men­tary method for the as­sess­ment of glu­cose lev­els. Glu­cose mon­i­tor­ing al­lows pa­tients to eval­u­ate their in­di­vid­u­al re­sponse to ther­a­py and as­sess whether glycemic tar­gets are being safe­ly achieved. In­te­grat­ing re­sults into di­a­betes man­age­ment can be a use­ful tool for guid­ing med­i­cal nu­tri­tion ther­a­py and phys­i­cal ac­tiv­i­ty, pre­vent­ing hy­po­glycemia, and ad­just­ing med­i­ca­tions (par­tic­u­lar­ly pran­di­al in­sulin doses). The pa­tient’s specific needs and goals should dic­tate SMBG fre­quen­cy and tim­ing or the con­sid­er­a­tion of CGM use. Please refer to Sec­tion 7 “Di­a­betes Tech­nol­o­gy” for a fuller dis­cus­sion of the use of SMBG and CGM.

xxxxx

4.0.0.0 A1C GOALS

4.1.0.0 Overview

For glycemic goals in older adults, please refer to Sec­tion 12 “Older Adults.” For glycemic goals in chil­dren, please refer to Sec­tion 13 “Chil­dren and Ado­les­cents.” For glycemic goals in preg­nant women, please refer to Sec­tion 14 “Man­age­ment of Di­a­betes in Preg­nan­cy.”

Rec­om­men­da­tions

A rea­son­able A1C goal for many nonpreg­nant adults is <7% (53 mmol/‍mol). A

Providers might rea­son­ably sug­gest more strin­gent A1C goals (such as <6.5% [48 mmol/‍mol]) for se­lect­ed in­di­vid­u­al pa­tients if this can be achieved with­out significant hy­po­glycemia or other ad­verse ef­fects of treat­ment (i.e., polyphar­ma­cy). Ap­pro­pri­ate pa­tients might in­clude those with short du­ra­tion of di­a­betes, type 2 di­a­betes treat­ed with lifestyle or met­formin only, long life ex­pectan­cy, or no significant car­dio­vas­cu­lar dis­ease. C

Less strin­gent A1C goals (such as <8% [64 mmol/‍mol]) may be ap­pro­pri­ate for pa­tients with a his­to­ry of se­vere hy­po­glycemia, lim­it­ed life ex­pectan­cy, ad­vanced mi­crovas­cu­lar or macrovas­cu­lar com­pli­ca­tions, ex­ten­sive co­mor­bid con­di­tions, or long-‍stand­ing di­a­betes in whom the goal is difficult to achieve de­spite di­a­betes self-‍man­age­ment ed­u­ca­tion, ap­pro­pri­ate glu­cose mon­i­tor­ing, and ef­fec­tive doses of mul­ti­ple glu­cose-‍low­er­ing agents in­clud­ing in­sulin. B

Reas­sess glycemic tar­gets over time based on the cri­te­ria in Fig. 6.1 or, in older adults,Table 12.1. E

Table 6.1—Mean glu­cose lev­els for specified A1C lev­els (6,7)

xxxxx

4.2.0.0 A1C and Mi­crovas­cu­lar Com­pli­ca­tions

Hy­per­glycemia defines di­a­betes, and glycemic con­trol is fun­da­men­tal to di­a­betes man­age­ment. The Di­a­betes Con­trol and Com­pli­ca­tions Trial (DCCT) (16), a prospec­tive ran­dom­ized con­trolled trial of in­ten­sive (mean A1C about 7% [53 mmol/‍mol]) ver­sus stan­dard (mean A1C about 9% [75 mmol/‍mol]) glycemic con­trol in pa­tients with type 1 di­a­betes, showed defini­tive­ly that bet­ter glycemic con­trol is as­so­ci­at­ed with 50–76% re­duc­tions in rates of de­vel­op­ment and pro­gres­sion of mi­crovas­cu­lar (retinopa­thy, neu­ropa­thy, and di­a­bet­ic kid­ney dis­ease) com­pli­ca­tions. Fol­low-‍up of the DCCT co­horts in the Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions (EDIC) study (17,18) demon­strat­ed per­sis­tence of these mi­crovas­cu­lar benefits over two decades de­spite the fact that the glycemic sep­a­ra­tion be­tween the treat­ment groups di­min­ished and dis­ap­peared dur­ing fol­low-‍up.

The Ku­mamo­to Study (19) and UK Prospec­tive Di­a­betes Study (UKPDS) (20,21) confirmed that in­ten­sive glycemic con­trol significant­ly de­creased rates of mi­crovas­cu­lar com­pli­ca­tions in pa­tients with short-‍du­ra­tion type 2 di­a­betes. Long-‍term fol­low-‍up of the UKPDS co­horts showed endur­ing ef­fects of early glycemic con­trol on most mi­crovas­cu­lar com­pli­ca­tions (22).

There­fore, achiev­ing A1C tar­gets of <7% (53 mmol/‍mol) has been shown to re­duce mi­crovas­cu­lar com­pli­ca­tions of type 1 and type 2 di­a­betes when in­sti­tut­ed early in the course of dis­ease. Epi­demi­o­log­ic anal­y­ses of the DCCT (16) and UKPDS (23) demon­strate a curvi­lin­ear re­la­tion­ship be­tween A1C and mi­crovas­cu­lar com­pli­ca­tions. Such anal­y­ses sug­gest that, on a pop­u­la­tion level, the great­est num­ber of com­pli­ca­tions will be avert­ed by tak­ing pa­tients from very poor con­trol to fair/‍good con­trol. These anal­y­ses also sug­gest that fur­ther low­er­ing of A1C from 7% to 6% [53 mmol/‍mol to 42 mmol/‍mol] is as­so­ci­at­ed with fur­ther re­duc­tion in the risk of mi­crovas­cu­lar com­pli­ca­tions, al­though the ab­so­lute risk re­duc­tions be­come much small­er. Given the sub­stan­tial­ly in­creased risk of hy­po­glycemia in type 1 di­a­betes tri­als and with polyphar­ma­cy in type 2 di­a­betes, the risks of lower glycemic tar­gets may out­weigh the po­ten­tial benefits on mi­crovas­cu­lar com­pli­ca­tions.

Three land­mark tri­als (Ac­tion to Con­trol Car­dio­vas­cu­lar Risk in Di­a­betes [AC­CORD], Ac­tion in Di­a­betes and Vas­cu­lar Dis­ease: Preter­ax and Di­ami­cron MR Con­trolled Eval­u­a­tion [AD­VANCE], and Vet­er­ans Af­fairs Di­a­betes Trial [VADT]) were con­duct­ed to test the ef­fects of near nor­mal­iza­tion of blood glu­cose on car­dio­vas­cu­lar out­comes in in­di­vid­u­als with long-‍stand­ing type 2 di­a­betes and ei­ther known car­dio­vas­cu­lar dis­ease (CVD) or high car­dio­vas­cu­lar risk. These tri­als showed that lower A1C lev­els were as­so­ci­at­ed with re­duced onset or pro­gres­sion of some mi­crovas­cu­lar com­pli­ca­tions (24-26).

The con­cern­ing mor­tal­i­ty find­ings in the AC­CORD trial (27), dis­cussed below, and the rel­a­tive­ly in­tense ef­forts re­quired to achieve near eu­g­lycemia should also be con­sid­ered when set­ting glycemic tar­gets for in­di­vid­u­als with long-‍stand­ing di­a­betes such as those stud­ied in AC­CORD, AD­VANCE, and VADT.

Find­ings from these stud­ies sug­gest cau­tion is need­ed in treat­ing di­a­betes ag­gres­sive­ly to near-‍nor­mal A1C goals in peo­ple with long-‍stand­ing type 2 di­a­betes with or at significant risk of CVD. How­ev­er, on the basis of physi­cian judg­ment and pa­tient pref­er­ences, se­lect pa­tients, es­pe­cial­ly those with lit­tle co­mor­bidity and long life ex­pectan­cy, may benefit from adopt­ing more in­ten­sive glycemic tar­gets (e.g., A1C tar­get <6.5% [48 mmol/‍mol]) if they can achieve it safe­ly with­out hy­po­glycemia or significant ther­a­peu­tic bur­den.

Fig­ure 6.1—De­pict­ed are pa­tient and dis­ease fac­tors used to de­ter­mine op­ti­mal A1C tar­gets. Char­ac­ter­is­tics and predica­ments to­ward the left jus­ti­fy more strin­gent ef­forts to lower A1C; those to­ward the right sug­gest less strin­gent ef­forts. A1C 7% = 53 mmol/‍mol. Adapt­ed with per­mis­sion from In­zuc­chi et al. (40).

xxxxx

4.3.0.0 A1C and Car­dio­vas­cu­lar Dis­ease Out­comes

4.3.1.0 Car­dio­vas­cu­lar Dis­ease and Type 1 Di­a­betes

CVD is a more com­mon cause of death than mi­crovas­cu­lar com­pli­ca­tions in pop­u­la­tions with di­a­betes. There is ev­i­dence for a car­dio­vas­cu­lar benefit of in­ten­sive glycemic con­trol after long-‍term fol­low-‍up of co­horts treat­ed early in the course of type 1 di­a­betes. In the DCCT, there was a trend to­ward lower risk of CVD events with in­ten­sive con­trol. In the 9-year post-‍DCCT fol­low-‍up of the EDIC co­hort, par­tic­i­pants pre­vi­ous­ly ran­dom­ized to the in­ten­sive arm had a significant 57% re­duc­tion in the risk of non­fa­tal my­ocar­dial in­farc­tion (MI), stroke, or car­dio­vas­cu­lar death com­pared with those pre­vi­ous­ly ran­dom­ized to the stan­dard arm (28). The benefit of in­ten­sive glycemic con­trol in this co­hort with type 1 di­a­betes has been shown to per­sist for sev­er­al decades (29) and to be as­so­ci­at­ed with a mod­est re­duc­tion in all-‍cause mor­tal­i­ty (30).

xxxxx

4.3.2.0 Car­dio­vas­cu­lar Dis­ease and Type 2 Di­a­betes

In type 2 di­a­betes, there is ev­i­dence that more in­ten­sive treat­ment of glycemia in newly di­ag­nosed pa­tients may re­duce long-‍term CVD rates. Dur­ing the UKPDS, there was a 16% re­duc­tion in CVD events (com­bined fatal or non­fa­tal MI and sud­den death) in the in­ten­sive glycemic con­trol arm that did not reach sta­tis­ti­cal significance (P = 0.052), and there was no sug­gestion of benefit on other CVD out­comes (e.g., stroke). How­ev­er, after 10 years of ob­ser­va­tion­al fol­low-‍up, those orig­i­nal­ly ran­dom­ized to in­ten­sive glycemic con­trol had significant long-‍term re­duc­tions in MI (15% with sul­fony­lurea or in­sulin as ini­tial pharmacother­a­py, 33% with met­formin as ini­tial pharmacother­a­py) and in all-‍cause mor­tal­i­ty (13% and 27%, re­spec­tive­ly) (22). AC­CORD, AD­VANCE, and VADT sug­gested no significant re­duc­tion in CVD out­comes with in­ten­sive glycemic con­trol in par­tic­i­pants fol­lowed for short­er du­ra­tions (3.5–5.6 years) and who had more ad­vanced type 2 di­a­betes than UKPDS par­tic­i­pants. All three tri­als were con­duct­ed in rel­a­tive­ly older par­tic­i­pants with longer known du­ra­tion of di­a­betes (mean du­ra­tion 8–11 years) and ei­ther CVD or mul­ti­ple car­dio­vas­cu­lar risk fac­tors. The tar­get A1C among in­ten­sive-‍con­trol sub­jects was <6% (42 mmol/‍mol) in AC­CORD, <6.5% (48 mmol/‍mol) in AD­VANCE, and a 1.5% re­duc­tion in A1C com­pared with con­trol sub­jects in VADT, with achieved A1C of 6.4% vs. 7.5% (46 mmol/‍mol vs. 58 mmol/‍mol) in AC­CORD, 6.5% vs. 7.3% (48 mmol/‍mol vs. 56 mmol/‍mol) in AD­VANCE, and 6.9% vs. 8.4% (52 mmol/‍mol vs. 68 mmol/‍mol) in VADT. De­tails of these stud­ies are re­viewed ex­ten­sively in “In­ten­sive Glycemic Con­trol and the Pre­ven­tion of Car­dio­vas­cu­lar Events: Im­pli­ca­tions of the AC­CORD, AD­VANCE, and VA Di­a­betes Tri­als” (31).

The glycemic con­trol com­par­i­son in AC­CORD was halt­ed early due to an in­creased mor­tal­i­ty rate in the in­ten­sive com­pared with the stan­dard treat­ment arm (1.41% vs. 1.14% per year; haz­ard ratio 1.22 [95% CI 1.01–1.46]), with a sim­i­lar in­crease in car­dio­vas­cu­lar deaths. Anal­y­sis of the AC­CORD data did not iden­ti­fy a clear ex­pla­na­tion for the ex­cess mor­tal­i­ty in the in­ten­sive treat­ment arm (27).

Longer-‍term fol­low-‍up has shown no ev­i­dence of car­dio­vas­cu­lar benefit or harm in the AD­VANCE trial (32). The end-‍stage renal dis­ease rate was lower in the in­ten­sive treat­ment group over fol­low-‍up. How­ev­er, 10-year fol­low-‍up of the VADT co­hort (33) showed a re­duc­tion in the risk of car­dio­vas­cu­lar events (52.7 [con­trol group] vs. 44.1 [in­ter­ven­tion group] events per 1,000 per­son-‍years) with no benefit in car­dio­vas­cu­lar or over­all mor­tal­i­ty. Het­ero­gene­ity of mor­tal­i­ty ef­fects across stud­ies was noted, which may reflect dif­fer­ences in glycemic tar­gets, ther­a­peu­tic ap­proach­es, and pop­u­la­tion char­ac­ter­is­tics (34).

Mor­tal­i­ty find­ings in AC­CORD (27) and sub­group anal­y­ses of VADT (35) sug­gest that the po­ten­tial risks of in­ten­sive glycemic con­trol may out­weigh its benefits in high­er-‍risk pa­tients. In all three tri­als, se­vere hy­po­glycemia was significant­ly more like­ly in par­tic­i­pants who were ran­dom­ly as­signed to the in­ten­sive glycemic con­trol arm. Those pa­tients with long du­ra­tion of di­a­betes, a known his­to­ry of hy­po­glycemia, ad­vanced atheroscle­ro­sis, or ad­vanced age/‍frailty may benefit from less ag­gres­sive tar­gets (36,37).

As dis­cussed fur­ther below, se­vere hy­po­glycemia is a po­tent mark­er of high ab­so­lute risk of car­dio­vas­cu­lar events and mor­tal­i­ty (38). Providers should be vig­i­lant in pre­vent­ing hy­po­glycemia and should not ag­gres­sive­ly at­tempt to achieve near-‍nor­mal A1C lev­els in pa­tients in whom such tar­gets can­not be safe­ly and rea­son­ably achieved. As dis­cussed in Sec­tion 9 “Phar­ma­co-‍logic Ap­proach­es to Glycemic Treat­ment,” ad­di­tion of specific sodi­um–glu­cose co­trans­porter 2 in­hibitors (SGLT2i) or glucagon-‍like pep­tide 1 re­cep­tor ag­o­nists (GLP-1 RA) to im­prove car­dio­vas­cu­lar out­comes in pa­tients with es­tab­lished CVD is in­di­cat­ed with con­sid­er­a­tion of glycemic goals. If the pa­tient is not at A1C tar­get, con­tin­ue met­formin un­less contrain­di­cat­ed and add SGLT2i or GLP-1 RA with proven car­dio­vas­cu­lar benefit. If the pa­tient is meet­ing A1C tar­get, con­siderone of three strate­gies (39):

    If al­ready on dual ther­a­py or mul­ti­ple glu­cose-‍low­er­ing ther­a­pies and not on an SGLT2i or GLP-1 RA, con­sid­er switch­ing to one of these agents with proven car­dio­vas­cu­lar benefit.

    Recon­sid­er/lower in­di­vid­u­alized A1C tar­get and in­tro­duce SGLT2i or GLP-1 RA.

    Reas­sess A1C at 3-‍month in­ter­vals and add SGLT2i or GLP-1 RA if A1C goes above tar­get.

xxxxx

4.4.0.0 Set­ting and Mod­i­fy­ing A1C Goals

Nu­mer­ous fac­tors must be con­sid­ered when set­ting glycemic tar­gets. The ADA pro­pos­es gen­er­al tar­gets ap­pro­pri­ate for many pa­tients but em­pha­sizes the im­por­tance of in­di­vid­u­alization based on key pa­tient char­ac­ter­is­tics. Glycemic tar­gets must be in­di­vid­u­alized in the con­text of shared de­ci­sion mak­ing to ad­dress the needs and pref­er­ences of each pa­tient and the in­di­vid­u­al char­ac­ter­is­tics that influence risks and benefits of ther­a­py for each pa­tient.

The fac­tors to con­sid­er in in­di­vid­u­alizing goals are de­pict­ed in Fig. 6.1. Fig­ure 6.1 is not de­signed to be ap­plied rigid­ly but to be used as a broad con­struct to guide clin­i­cal de­ci­sion mak­ing (40) in both type 1 and type 2 di­a­betes. More strin­gent con­trol (such as an A1C of 6.5% [48 mmol/‍mol] or <7% [53 mmol/‍mol]) may be rec­om­mend­ed if it can be achieved safe­ly and with ac­cept­able bur­den of ther­a­py and if life ex­pectan­cy is sufficient to reap benefits of tight con­trol. Less strin­gent con­trol (A1C up to 8% [64 mmol/‍mol]) may be rec­om­mend­ed if the life ex­pectan­cy of the pa­tient is such that the benefits of an in­ten­sive goal may not be re­al­ized, or if the risks and bur­dens out­weigh the po­ten­tial benefits. Se­vere or fre­quent hy­po­glycemia is an ab­so­lute in­di­ca­tion for the modification of treat­ment reg­i­mens, in­clud­ing set­ting high­er glycemic goals.

Di­a­betes is a chron­ic dis­ease that pro­gress­es over decades. Thus, a goal that might be ap­pro­pri­ate for an in­di­vid­u­al early in the course of the dis­ease may change over time. Newly di­ag­nosed pa­tients and/‍or those with­out co­mor­bidities that limit life ex­pectan­cy may benefit from in­ten­sive con­trol proven to pre­vent mi­crovas­cu­lar com­pli­ca­tions. Both DCCT/EDIC and UKPDS demon­strat­ed metabol­ic mem­o­ry, or a lega­cy ef­fect, in which a finite pe­ri­od of in­ten­sive con­trol yield­ed benefits that ex­tend­ed for decades after that con­trol ended. Thus, a finite pe­ri­od of in­ten­sive con­trol to near-‍nor­mal A1C may yield endur­ing benefits even if con­trol is sub­se­quent­ly dein­ten­sified as pa­tient char­ac­ter­is­tics change. Over time, co­mor­bidities may emerge, de­creas­ing life ex­pectan­cy and the po­ten­tial to reap benefits from in­ten­sive con­trol. Also, with longer du­ra­tion of dis­ease, di­a­betes may be­come more difficult to con­trol, with in­creas­ing risks and bur­dens of ther­a­py. Thus, A1C tar­gets should be reeval­u­ated over time to bal­ance the risks and benefits as pa­tient fac­tors change.

Rec­om­mend­ed glycemic tar­gets for many nonpreg­nant adults are shown in Table 6.2. The rec­om­men­da­tions in­clude blood glu­cose lev­els that ap­pear to cor­re­late with achieve­ment of an A1C of <7% (53 mmol/‍mol). The issue of prepran­di­al ver­sus postpran­di­al SMBG tar­gets is com­plex (41). El­e­vat­ed post-‍chal­lenge (2-h oral glu­cose tol­er­ance test) glu­cose val­ues have been as­so­ci­at­ed with in­creased car­dio­vas­cu­lar risk inde­pendent of fast­ing plas­ma glu­cose in some epi­demi­o­log­ic stud­ies, but in­ter­ven­tion tri­als have not shown postpran-‍dial glu­cose to be a car­dio­vas­cu­lar risk fac­tor inde­pendent of A1C. In sub­jects with di­a­betes, sur­ro­gate mea­sures of vas­cu­lar pathol­o­gy, such as en­dothe­lial dys­func­tion, are neg­a­tive­ly af­fected by postpran­di­al hy­per­glycemia. It is clear that postpran­di­al hy­per­glycemia, like prepran­di­al hy­per­glycemia, con­tributes to el­e­vat­ed A1C lev­els, with its rel­a­tive con­tri­bu­tion being greater at A1C lev­els that are clos­er to 7% (53 mmol/‍mol). How­ev­er, out­come stud­ies have clear­ly shown A1C to be the pri­ma­ry pre­dic­tor of com­pli­ca­tions, and land­mark tri­als of glycemic con­trol such as the DCCT and UKPDS re­lied over­whelm­ing­ly on pre-‍pran­di­al SMBG. Ad­di­tion­al­ly, a ran­dom­ized con­trolled trial in pa­tients with known CVD found no CVD benefit of in­sulin reg­i­mens tar­geting postpran­di­al glu­cose com­pared with those tar­geting prepran­di­al glu­cose (42). There­fore, it is rea­son­able for postpran­di­al test­ing to be rec­om­mend­ed for in­di­vid­u­als who have pre­meal glu­cose val­ues with­in tar­get but have A1C val­ues above tar­get. Mea­sur­ing postpran­di­al plas­ma glu­cose 1–2 h after the start of a meal and using treat­ments aimed at re­duc­ing postpran­di­al plas­ma glu­cose val­ues to <180 mg/dL (10.0 mmol/‍L) may help to lower A1C.

An anal­y­sis of data from 470 par­tic­i­pants in the ADAG study (237 with type 1 di­a­betes and 147 with type 2 di­a­betes) found that ac­tu­al av­er­age glu­cose lev­els as­so­ci­at­ed with con­ven­tion­al A1C tar­gets were high­er than older DCCT and ADA tar­gets (Table 6.1) (7,43). These find­ings sup­port that pre­meal glu­cose tar­gets may be re­laxed with­out un­der­min­ing over­all glycemic con­trol as mea­sured by A1C. These data prompt­ed the re­vi­sion in the ADA-‍rec­om­mend­ed pre­meal glu­cose tar­get to 80–130 mg/dL (4.4–7.2 mmol/‍L) but did not af­fect the def­i­ni­tion of hy­po­glycemia.

Table 6.2—Sum­ma­ry of glycemic rec­om­men­da­tions for many nonpreg­nant adults with di­a­betes

xxxxx

5.0.0.0 HY­PO­GLYCEMIA

5.1.0.0 Rec­om­men­da­tions

Rec­om­men­da­tions

In­di­vid­u­als at risk for hy­po­glycemia should be asked about symp­tomat­ic and asymp­tomat­ic hy­po­glycemia at each en­counter. C

Glu­cose(15–20 g) is thep­re­ferred treat­ment for the con­scious in­di­vid­u­al with blood glu­cose

<70 mg/dL [3.9 mmol/‍L]), al­though any form of car­bo­hy­drate that con­tains glu­cose may be used. Fif­teen min­utes after treat­ment, if SMBG shows con­tin­ued hy­po­glycemia, the treat­ment should be re­peat­ed. Once SMBG re­turns to nor­mal, the in­di­vid­u­al should con­sume a meal or snack to pre­vent re­cur­rence of hy­po­glycemia. E

Glucagon should be pre­scribed for all in­di­vid­u­als at in­creased risk of level 2 hy­po­glycemia, defined as blood glu­cose <54 mg/dL (3.0 mmol/‍L), so it is avail­able should it be need­ed. Care­givers, school per­son­nel, or fam­i­ly mem­bers of these in­di­vid­u­als should know where it is and when and how to ad­min­is­ter it. Glucagon ad­min­is­tra­tion is not lim­it­ed to health care pro­fes­sion­als. E

Hy­po­glycemia un­aware­ness or one or more episodes of level 3 hy­po­glycemia should trig­ger ree­va­lua­ti­on of the treat­ment reg­i­men. E

Insulin-treat­ed pa­tients with hy­po­glycemia un­aware­ness or an episode of level 2 hy­po­glycemia should be ad­vised to raise their glycemic tar­gets to strict­ly avoid hy­po­glycemia for at least sev­er­al weeks in order to par­tial­ly re­verse hy­po­glycemia un­aware­ness and re­duce risk of fu­ture episodes. A

On­go­ing as­sess­ment of cog­ni­tive func­tion is sug­gested with

in­creased vig­i­lance for hy­po­glycemia by the clin­i­cian, pa­tient, and care­givers if low cog­ni­tion or de­clin­ing cog­ni­tion is found. B

Table 6.3—Clas­sification of hy­po­glycemia (44)

Hy­po­glycemia is the major lim­it­ing fac­tor in the glycemic man­age­ment of type 1 and type 2 di­a­betes. Rec­om­men­da­tions re­gard­ing the clas­sification of hy­po­glycemia are out­lined in Table 6.3 (44). Level 1 hy­po­glycemia is defined as a mea­sur­able glu­cose con­cen­tra­tion <70 mg/dL (3.9 mmol/‍L) but ≥54 mg/dL (3.0 mmol/‍L). A blood glu­cose con­cen­tra­tion of 70 mg/dL (3.9 mmol/‍L) has been rec­og­nized as a thresh­old for neu­roen­docrine re­sponses to falling glu­cose in peo­ple with­out di­a­betes. Be­cause many peo­ple with di­a­betes demon­strate im­paired coun­ter­reg­u­la­to­ry re­sponses to hy­po­glycemia and/‍or ex­pe­ri­ence hy­po­glycemia un­aware­ness, a mea­sured glu­cose level <70 mg/dL (3.9 mmol/‍L) is con­sid­ered clin­i­cally im­por­tant, inde­pendent of the sever­i­ty of acute hy­po­glyce­mic symp­toms. Level 2 hy­po­glycemia (defined as a blood glu­cose con­cen­tra­tion <54 mg/dL [3.0 mmol/‍L]) is the thresh­old at which neu­ro­gly­copenic symp­toms begin to occur and re­quires im­me­di­ate ac­tion to re­solve the hy­po­glyce­mic event. Last­ly, level 3 hy­po­glycemia is defined as a se­vere event char­ac­ter­ized by al­tered men­tal and/‍or phys­i­cal func­tioning that re­quires as­sis­tance from an­oth­er per­son for re­cov­ery. Stud­ies of rates of level 3 hy­po­glycemia that rely on claims data for hos­pi­tal­iza­tion, emer­gen­cy de­part­ment vis­its, and am­bu­lance use sub­stan­tial­ly un­der­es­ti­mate rates of level 3 hy­po­glycemia (45), yet find high bur­den of hy­po­glycemia in adults over 60 years of age in the com­mu­ni­ty (46). African Amer­i­cans are at sub­stan­tial­ly in­creased risk of level 3 hy­po­glycemia (46,47). In ad­di­tion to age and race, other im­por­tant risk fac­tors found in a com­mu­ni­ty-‍based epi­demi­o­log­ic co­hort of older black and white adults with type 2 di­a­betes in­clude in­sulin use, poor or mod­er­ate ver­sus good glycemic con­trol, al­bu­min­uria, and poor cog­ni­tive func­tion (46).

Symp­toms of hy­po­glycemia in­clude, but are not lim­it­ed to, shak­i­ness, ir­ri­tabil­i­ty, con­fu­sion, tachy­car­dia, and hunger. Hy­po­glycemia may be in­con­ve­nient or fright­en­ing to pa­tients with di­a­betes. Level 3 hy­po­glycemia may be rec­og­nized or unrec­og­nized and can progress to loss of con­sciousness, seizure, coma, or death. It is re­versed by ad­min­is­tra­tion of rapid-‍act­ing glu­cose or glucagon. Hy­po­glycemia can cause acute harm to the per­son with di­a­betes or oth­ers, es­pe­cial­ly if it caus­es falls, motor ve­hi­cle ac­ci­dents, or other in­jury. A large co­hort study sug­gested that among older adults with type 2 di­a­betes, a his­to­ry of level 3 hy­po­glycemia was as­so­ci­at­ed with greater risk of de­men­tia (48). Con­verse­ly, in a sub­study of the AC­CORD trial, cog­ni­tive im­pair­ment at base­line or de­cline in cog­ni­tive func­tion dur­ing the trial was significant­ly as­so­ci­at­ed with sub­se­quent episodes of level 3 hy­po­glycemia (49). Ev­i­dence from DCCT/EDIC, which in­volved ado­les­cents and younger adults with type 1 di­a­betes, found no as­so­ci­a­tion be­tween fre­quen­cy of level 3 hy­po­glycemia and cog­ni­tive de­cline (50), as dis­cussed in Sec­tion 13 “Chil­dren and Ado­les­cents.”

Level 3 hy­po­glycemia was as­so­ci­at­ed with mor­tal­i­ty in par­tic­i­pants in both the stan­dard and the in­ten­sive glycemia arms of the AC­CORD trial, but the re­la­tion­ships be­tween hy­po­glycemia, achieved A1C, and treat­ment in­ten­si­ty were not straight-‍for­ward. An as­so­ci­a­tion of level 3 hy­po­glycemia with mor­tal­i­ty was also found in the AD­VANCE trial (51). An as­so­ci­a­tion be­tween self-‍re­port­ed level 3 hy­po­glycemia and 5-year mor­tal­i­ty has also been re­port­ed in clin­i­cal prac­tice (52).

Young chil­dren with type 1 di­a­betes and the el­der­ly, in­clud­ing those with type 1 and type 2 di­a­betes (48,53), are noted as par­tic­u­lar­ly vul­ner­a­ble to hy­po­glycemia be­cause of their re­duced abil­i­ty to rec­og­nize hy­po­glyce­mic symp­toms and ef­fec­tively com­mu­ni­cate their needs. In­di­vid­u­al­ized glu­cose tar­gets, pa­tient ed­u­ca­tion, di­etary in­ter­ven­tion (e.g., bed­time snack to pre­vent overnight hy­po­glycemia when specifically need­ed to treat low blood glu­cose), ex­er­cise man­age­ment, med­i­ca­tion ad­just­ment, glu­cose mon­i­tor­ing, and rou­tine clin­i­cal surveil­lance may im­prove pa­tient out­comes (54). CGM with au­to­mat­ed low glu­cose sus­pend has been shown to be ef­fec­tive in re­duc­ing hy­po­glycemia in type 1 di­a­betes (55). For pa­tients with type 1 di­a­betes with level 3 hy­po­glycemia and hy­po­glycemia un­aware­ness that per­sists de­spite med­i­cal treat­ment, human islet trans­plan­ta­tion may be an op­tion, but the ap­proach re­mains experimen­tal (56,57).

In 2015, the ADA changed its prepran­di­al glycemic tar­get from 70–130 mg/dL (3.9–7.2 mmol/‍L) to 80–130 mg/dL (4.4–7.2 mmol/‍L). This change reflects the re­sults of the ADAG study, which demon­strat­ed that high­er glycemic tar­gets cor­re­spond­ed to A1C goals (7). An ad­di­tional goal of rais­ing the lower range of the glycemic tar­get was to limit overtreat­ment and pro­vide a safe­ty mar­gin in pa­tients titrat­ing glu­cose-‍low­er­ing drugs such as in­sulin to glycemic tar­gets.

xxxxx

5.2.0.0 Hy­po­glycemia Treat­ment

xxxxx

5.2.1.0 Overview

Providers should con­tin­ue to coun­sel pa­tients to treat hy­po­glycemia with fast-‍act­ing car­bo­hy­drates at the hy­po­glycemia alert value of 70 mg/dL (3.9 mmol/‍L) or less. Hy­po­glycemia treat­ment re­quires in­ges­tion of glu­cose-‍ or car­bo­hy­drate-containing foods. The acute glycemic re­sponse cor­re­lates bet­ter with the glu­cose con­tent of food than with the car­bo­hy­drate con­tent of food. Pure glu­cose is the pre­ferred treat­ment, but any form of car­bo­hy­drate that con­tains glu­cose will raise blood glu­cose. Added fat may re­tard and then pro­long the acute glycemic re­sponse. In type 2 di­a­betes, in­gest­ed pro­tein may in­crease in­sulin re­sponse with­out in­creas­ing plas­ma glu­cose con­cen­tra­tions (58). There­fore, car­bo­hy­drate sources high in pro­tein should not be used to treat or pre­vent hy­po­glycemia. On­go­ing in­sulin ac­tiv­i­ty or in­sulin sec­re­t­a­gogues may lead to recur­rent hy­po­glycemia un­less more food is in­gest­ed after re­cov­ery. Once the glu­cose re­turns to nor­mal, the in­di­vid­u­al should be coun­seled to eat a meal or snack to pre­vent recur­rent hy­po­glycemia.

5.2.2.0 Glucagon

The use of glucagon is in­di­cat­ed for the treat­ment of hy­po­glycemia in peo­ple un­able or un­will­ing to con­sume car­bo­hy­drates by mouth. Those in close con­tact with, or hav­ing cus­to­di­al care of, peo­ple with hy­po­glycemia-‍prone di­a­betes (fam­i­ly mem­bers, room­mates, school per­son­nel, child care pro­viders, cor­rec­tion­al in­sti­tu­tion staff, or cowork­ers) should be in­struct­ed on the use of glucagon kits, in­clud­ing where the kit is and when and how to ad­min­is­ter glucagon. An in­di­vid­u­al does not need to be a health care pro­fes­sion­al to safe­ly ad­min­is­ter glucagon. Care should be taken to en­sure that glucagon kits are not ex­pired.

xxxxx

5.3.0.0 Hy­po­glycemia Pre­ven­tion

Hy­po­glycemia pre­vention is a crit­i­cal com­po­nent of di­a­betes man­age­ment. SMBG and, for some pa­tients, CGM are es­sen­tial tools to as­sess ther­a­py and de­tect in­cip­i­ent hy­po­glycemia. Pa­tients should un­der­stand sit­u­a­tions that in­crease their risk of hy­po­glycemia, such as when fast­ing for tests or pro­ce­dures, when meals are de­layed, dur­ing and after the con­sump­tion of al­co­hol, dur­ing and after in­tense ex­er­cise, and dur­ing sleep. Hy­po­glycemia may in­crease the risk of harm to self or oth­ers, such as with driv­ing. Teach­ing peo­ple with di­a­betes to bal­ance in­sulin use and car­bo­hy­drate in­take and ex­er­cise are nec­es­sary, but these strate­gies are not al­ways sufficient for pre­vention.

In type 1 di­a­betes and se­verely in­sulindeficient type 2 di­a­betes, hy­po­glycemia un­aware­ness (or hy­po­glycemia-‍as­so­ci­at­ed au­to­nom­ic fail­ure) can se­verely com­pro­mise strin­gent di­a­betes con­trol and qual­i­ty of life. This syn­drome is char­ac­ter­ized by deficient coun­ter­reg­u­la­to­ry hor­mone re­lease, es­pe­cial­ly in older adults, and a di­min­ished au­to­nom­ic re­sponse, which both are risk fac­tors for, and caused by, hy­po­glycemia. A corol­lary to this “vi­cious cycle” is that sev­er­al weeks of avoid­ance of hy­po­glycemia has been demon­strat­ed to im­prove coun­ter­reg­u­la­tion and hy­po­glycemia aware­ness in many pa­tients (59). Hence, pa­tients with one or more episodes of clin­i­cally significant hy­po­glycemia may benefit from at least short-‍term re­lax­ation of glycemic tar­gets.

xxxxx

6.0.0.0 IN­TER­CUR­RENT ILL­NESS

For fur­ther in­for­ma­tion on man­age­ment of pa­tients with hy­per­glycemia in the hos­pi­tal, please refer to Sec­tion 15 “Di­a­betes Care in the Hos­pi­tal.

Stress­ful events (e.g., ill­ness, trau­ma, surgery, etc.) may wors­en glycemic con­trol and pre­cip­i­tate di­a­bet­ic ke­toaci­do­sis or non­ke­tot­ic hy­per­glycemic hy­per­os­mo­lar state, life-‍threat­en­ing con­di­tions that re­quire im­me­di­ate med­i­cal care to pre­vent com­pli­ca­tions and death. Any con­di­tion lead­ing to de­te­ri­o­ra­tion in glycemic con­trol ne­ces­si­tates more fre­quent mon­i­tor­ing of blood glu­cose; ke­to­sis-‍prone pa­tients also re­quire urine or blood ke­tone mon­i­tor­ing. If ac­com­pa­nied by ke­to­sis, vom­it­ing, or al­ter­ation in the level of con­sciousness, marked hy­per­glycemia re­quires tem­po­rary ad­just­ment of the treat­ment reg­i­men and im­me­di­ate interac­tion with the di­a­betes care team. The pa­tient treat­ed with nonin­sulin ther­a­pies or med­i­cal nu­tri­tion ther­a­py alone may re­quire in­sulin. Ad­e­quate fluid and caloric in­take must be en­sured. In­fec­tion or de­hy­dra­tion is more like­ly to ne­ces­si­tate hos­pi­tal­iza­tion of the per­son with di­a­betes than the per­son with­out di­a­betes.

A physi­cian with ex­pertise in di­a­betes man­age­ment should treat the hos­pi­talized pa­tient. For fur­ther in­for­ma­tion on the man­age­ment of di­a­bet­ic ke­toaci­do­sis and the non­ke­tot­ic hy­per­glycemic hy­per­os­mo­lar state, please refer to the ADA con­sen­sus re­port “Hy­per­glycemic Crises in Adult Pa­tients With Di­a­betes” (60).

xxxxx

7.0.0.0 Ref­er­ences

  1. Laiteer­apong N, Ham SA, Gao Y, et al. The lega­cy ef­fect in type 2 di­a­betes: im­pact of early glycemic con­trol on fu­ture com­pli­ca­tions (the Di­a­betes & Aging Study). Di­a­betes Care. In press

  2. Strat­ton IM, Adler AI, Neil HAW, et al. As­so­ci­a­tion of gly­caemia with macrovas­cu­lar and mi­crovas­cu­lar com­pli­ca­tions of type 2 di­a­betes (UKPDS 35): prospec­tive ob­ser­va­tion­al study. BMJ 2000;321:405–412

  3. Lit­tle RR, Rohlfing CL, Sacks DB; Na­tion­al Gly­co­he­moglobin Stan­dard­iza­tion Pro­gram (NGSP) Steer­ing Com­mit­tee. Sta­tus of hemoglobin A1c mea­surement and goals for im­provement: from chaos to order for im­prov­ing di­a­betes care. Clin Chem 2011;57:205–214

  4. Jo­vanovicˇ L, Savas H, Mehta M, Tru­jil­lo A,

    Pet­titt DJ. Fre­quent mon­i­tor­ing of A1C dur­ing preg­nan­cy as a treat­ment tool to guide ther­a­py. Di­a­betes Care 2011;34:53–54

  5. Beck RW, Con­nor CG, Mullen DM, Wes­ley DM, Bergen­stal RM. The fal­la­cy of av­er­age: how using HbA1c alone to as­sess glycemic con­trol can be mislead­ing. Di­a­betes Care 2017;40:994–999

  6. Nathan DM, Kue­nen J, Borg R, Zheng H, Schoen­feld D, Heine RJ; A1c-‍De­rived Av­er­age Glu­cose Study Group. Trans­lat­ing the A1C assay into es­ti­mat­ed av­er­age glu­cose val­ues. Di­a­betes Care 2008;31:1473–1478

  7. Wei N, Zheng H, Nathan DM. Em­pir­i­cal­ly es­tab­lishing blood glu­cose tar­gets to achieve HbA1c goals. Di­a­betes Care 2014;37:1048–1051

  8. Selvin E. Are there clin­i­cal im­pli­ca­tions of racial dif­fer­ences in HbA1c? A dif­fer­ence, to be a dif­fer­ence, must make a dif­fer­ence. Di­a­betes Care 2016;39:1462–1467

  9. Bergen­stal RM, Gal RL, Con­nor CG, et al.; T1D Ex­change Racial Dif­fer­ences Study Group. Racial dif­fer­ences in the re­la­tion­ship of glu­cose con­cen­tra­tions and hemoglobin A1c lev­els. Ann In­tern Med 2017;167:95–102

  10. Lacy ME, Welle­nius GA, Sum­n­er AE, et al. As­so­ci­a­tion of sick­le cell trait with hemoglobin A1c in African Amer­i­cans. JAMA 2017;317:507– 515

  11. Rohlfing C, Han­son S, Lit­tle RR. Mea­sure­ment of hemoglobin A1c in pa­tients with sick­le cell trait. JAMA 2017;317:2237

  12. Wheel­er E, Leong A, Liu C-T, et al.; EPIC-‍CVD Con­sor­tium; EPIC-‍InterAct Con­sor­tium; Life­lines Co­hort Study. Im­pact of com­mon ge­net­ic de­ter­mi­nants of hemoglobin A1c on type 2 di­a­betes risk and di­ag­no­sis in an­ces­tral­ly di­verse pop­u­la­tions: a transeth­nic genome-‍wide metaanal­y­sis. PLoS Med 2017;14:e1002383

  13. Wil­son DM, Koll­man; Di­a­betes Re­search in Chil­dren Net­work (Di­rec­Net) Study Group. Re­la­tion­ship of A1C to glu­cose con­cen­tra­tions in chil­dren with type 1 di­a­betes: as­sess­ments by high-‍fre­quen­cy glu­cose de­ter­mi­na­tions by sen­sors. Di­a­betes Care 2008;31:381–385

  14. Buse JB, Kauf­man FR, Lin­der B, Hirst K, El Ghorm­li L, Willi S; HEALTHY Study Group. Di­a­betes screen­ing with hemoglobin A1c ver­sus fast­ing plas­ma glu­cose in a multieth­nic mid­dle-‍school co­hort. Di­a­betes Care 2013;36:429–435

  15. Kamps JL, Hempe JM, Chalew SA. Racial dis­par­i­ty in A1C inde­pendent of mean blood glu­cose in chil­dren with type 1 di­a­betes. Di­a­betes Care 2010;33:1025–1027

  16. Di­a­betes Con­trol and Com­pli­ca­tions Trial Re­search Group. The ef­fect of in­ten­sive treat­ment of di­a­betes on the de­vel­op­ment and pro­gres­sion of long-‍term com­pli­ca­tions in in­sulin-de­pendent di­a­betes mel­li­tus. N Engl J Med 1993;329:977–986

  17. Di­a­betes Con­trol and Com­pli­ca­tions Trial (DCCT)/Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions (EDIC) Re­search Group. Ef­fect of in­ten­sive di­a­betes ther­a­py on the pro­gres­sion of di­a­bet­ic retinopa­thy in pa­tients with type 1 di­a­betes: 18 years of fol­low-‍up in the DCCT/EDIC. Di­a­betes 2015;64:631–642

  18. Lachin JM, Genuth S, Cleary P, Davis MD, Nathan DM; Di­a­betes Con­trol and Com­pli­ca­tions Trial/‍Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions Re­search Group. Retinopa­thy and nephropa­thy in pa­tients with type 1 di­a­betes four years after a trial of in­ten­sive ther­a­py [pub­lished cor­rec­tion ap­pears in N Engl J Med 2000;342:1376]. N Engl J Med 2000;342:381–389

  19. Ohkubo Y, Kishikawa H, Araki E, et al. In­ten­sive in­sulin ther­a­py pre­vents the pro­gres­sion of di­a­bet­ic mi­crovas­cu­lar com­pli­ca­tions in Japanese pa­tients with non-in­sulin-de­pendent di­a­betes mel­li­tus: a ran­dom­ized prospec­tive 6-year study. Di­a­betes Res Clin Pract 1995;28: 103–117

  20. UK Prospec­tive Di­a­betes Study (UKPDS) Group. Ef­fect of in­ten­sive blood-‍glu­cose con­trol with met­formin on com­pli­ca­tions in over­weight pa­tients with type 2 di­a­betes (UKPDS 34). Lancet 1998;352:854–865

  21. UK Prospec­tive Di­a­betes Study (UKPDS) Group. In­ten­sive blood-‍glu­cose con­trol with sulpho­ny­lureas or in­sulin com­pared with con­ven­tion­al treat­ment and risk of com­pli­ca­tions in pa­tients with type 2 di­a­betes (UKPDS 33). Lancet 1998;352:837–853

  22. Hol­man RR, Paul SK, Bethel MA, Matthews DR, Neil HAW. 10-year fol­low-‍up of in­ten­sive glu­cose con­trol in type 2 di­a­betes. N Engl J Med 2008;359:1577–1589

  23. Adler AI, Strat­ton IM, Neil HAW, et al. As­so­ci­a­tion of sys­tolic blood pres­sure with macrovas­cu­lar and mi­crovas­cu­lar com­pli­ca­tions of type 2 di­a­betes (UKPDS 36): prospec­tive ob­ser­va­tion­al study. BMJ 2000;321:412–419

  24. Duck­worth W, Abraira C, Moritz T, et al.; VADT In­ves­ti­ga­tors. Glu­cose con­trol and vas­cu­lar com­pli­ca­tions in vet­er­ans with type 2 di­a­betes. N Engl J Med 2009;360:129–139

  25. Patel A, MacMa­hon S, Chalmers J, et al.; AD­VANCE Col­lab­o­ra­tive Group. In­ten­sive blood glu­cose con­trol and vas­cu­lar out­comes in pa­tients with type 2 di­a­betes. N Engl J Med 2008; 358:2560–2572

  26. Is­mail-‍Beigi F, Craven T, Baner­ji MA, et al.; AC­CORD trial group. Ef­fect of in­ten­sive treat­ment of hypergly­caemia on mi­crovas­cu­lar out­comes in type 2 di­a­betes: an anal­y­sis of the AC­CORD ran­domised trial. Lancet 2010; 376:419–430

  27. Ger­stein HC, Miller ME, By­ing­ton RP, et al.; Ac­tion to Con­trol Car­dio­vas­cu­lar Risk in Di­a­betes Study Group. Ef­fects of in­ten­sive glu­cose low­er­ing in type 2 di­a­betes. N Engl J Med 2008; 358:2545–2559

  28. Nathan DM, Cleary PA, Back­lund J-YC, et al.; Di­a­betes Con­trol and Com­pli­ca­tions Trial/‍ Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions (DCCT/‍EDIC) Study Re­search Group. In­ten­sive di­a­betes treat­ment and car­dio­vas­cu­lar dis­ease in pa­tients with type 1 di­a­betes. N Engl J Med 2005;353:2643–2653

  29. Nathan DM, Zin­man B, Cleary PA, et al.; Di­a­betes Con­trol and Com­pli­ca­tions Trial/‍ Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions (DCCT/‍EDIC) Re­search Group. Mod­ern-‍day clin­i­cal course of type 1 di­a­betes mel­li­tus after 30 years’ du­ra­tion: the Di­a­betes Con­trol and Com­pli­ca­tions Trial/‍Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions and Pitts­burgh Epi­demi­ol­o­gy of Di­a­betes Com­pli­ca­tions ex­pe­ri­ence (1983-2005). Arch In­tern Med 2009;169:1307–1316

  30. Or­chard TJ, Nathan DM, Zin­man B, et al.; Writ­ing Group for the DCCT/EDIC Re­search Group. As­so­ci­a­tion be­tween 7 years of in­ten­sive treat­ment of type 1 di­a­betes and long-‍term mor­tal­i­ty. JAMA 2015;313:45–53

  31. Skyler JS, Bergen­stal R, Bonow RO, et al.; Amer­i­can Di­a­betes As­so­ci­a­tion; Amer­i­can

    Col­lege of Car­di­ol­o­gy Foun­da­tion; Amer­i­can Heart As­so­ci­a­tion. In­ten­sive glycemic con­trol and the pre­vention of car­dio­vas­cu­lar events: im­pli­ca­tions of the AC­CORD, AD­VANCE, and VA di­a­betes tri­als: a po­si­tion state­ment of the Amer­i­can Di­a­betes As­so­ci­a­tion and a sci­en­tific state­ment of the Amer­i­can Col­lege of Car­di­ol­o­gy Foun­da­tion and the Amer­i­can Heart As­so­ci­a­tion. Di­a­betes Care 2009;32:187–192

  32. Zoun­gas S, Chalmers J, Neal B, et al.; AD­VANCE-‍ON Col­lab­o­ra­tive Group. Fol­low-‍up of blood-‍pres­sure low­er­ing and glu­cose con­trol in type 2 di­a­betes. N Engl J Med 2014;371:1392– 1406

  33. Hay­ward RA, Reav­en PD, Wi­ita­la WL, et al.; VADT In­ves­ti­ga­tors. Fol­low-‍up of glycemic con­trol and car­dio­vas­cu­lar out­comes in type 2 di­a­betes. N Engl J Med 2015;372:2197–2206

  34. Turn­bull FM, Abraira C, An­der­son RJ, et al.; Con­trol Group. In­ten­sive glu­cose con­trol and macrovas­cu­lar out­comes in type 2 di­a­betes [pub­lished cor­rec­tion ap­pears in Di­a­betolo­gia 2009;52:2470]. Di­a­betolo­gia 2009;52:2288–2298

  35. Duck­worth WC, Abraira C, Moritz TE, et al.; In­ves­ti­ga­tors of the VADT. The du­ra­tion of di­a­betes af­fects the re­sponse to in­ten­sive glu­cose con­trol in type 2 sub­jects: the VA Di­a­betes Trial. J Di­a­betes Com­pli­ca­tions 2011;25:355–361

  36. Lip­s­ka KJ, Ross JS, Miao Y, Shah ND, Lee SJ, Stein­man MA. Po­ten­tial overtreat­ment of di­a­betes mel­li­tus in older adults with tight glycemic con­trol. JAMA In­tern Med 2015;175: 356–362

  37. Vijan S, Suss­man JB, Yud­kin JS, Hay­ward RA. Ef­fect of pa­tients’ risks and pref­er­ences on health gains with plas­ma glu­cose level low­er­ing in type 2 di­a­betes mel­li­tus. JAMA In­tern Med 2014;174:1227–1234

  38. Lee AK, War­ren B, Lee CJ, et al. The as­so­ci­a­tion of se­vere hy­po­glycemia with in­ci­dent car­dio­vas­cu­lar events and mor­tal­i­ty in adults with type 2 di­a­betes. Di­a­betes Care 2018;41: 104–111

  39. Davies MJ, D’Alessio DA, Frad­kin J, et al. Man­age­ment of hy­per­glycemia in type 2 di­a­betes, 2018. A con­sen­sus re­port by the Amer­i­can Di­a­betes As­so­ci­a­tion (ADA) and the Eu­ro­pean As­so­ci­a­tion for the Study of Di­a­betes (EASD). Di­a­betes Care 2018;41:2669–2701

  40. In­zuc­chi SE, Bergen­stal RM, Buse JB, et al. Man­age­ment of hy­per­glycemia in type 2 di­a­betes, 2015: a pa­tient-centered ap­proach: up­date to a po­si­tion state­ment of the Amer­i­can Di­a­betes As­so­ci­a­tion and the Eu­ro­pean As­so­ci­a­tion for the Study of Di­a­betes. Di­a­betes Care 2015;38:140– 149

  41. Amer­i­can Di­a­betes As­so­ci­a­tion. Postpran­di­al blood glu­cose. Di­a­betes Care 2001;24:775– 778

  42. Raz I, Wil­son PWF, Stro­jek K, et al. Ef­fects of pran­di­al ver­sus fast­ing glycemia on car­dio­vas­cu­lar out­comes in type 2 di­a­betes: the HEART2D trial. Di­a­betes Care 2009;32:381– 386

  43. Al­bers JW, Her­man WH, Pop-‍Busui R, et al.; Di­a­betes Con­trol and Com­pli­ca­tions Trial/‍Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions Re­search Group. Ef­fect of prior in­ten­sive in­sulin treat­ment dur­ing the Di­a­betes Con­trol and Com­pli­ca­tions Trial (DCCT) on pe­riph­er­al neu­ropa­thy in type 1 di­a­betes dur­ing the Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and

    Com­pli­ca­tions (EDIC) study. Di­a­betes Care 2010;33:1090–1096

  44. Agios­trati­dou G, An­halt H, Ball D, et al. Stan­dard­iz­ing clin­i­cally mean­ing­ful out­come mea­sures be­yond HbA1c for type 1 di­a­betes: a con­sen­sus re­port of the Amer­i­can As­so­ci­a­tion of Clin­i­cal En­docri­nol­o­gists, the Amer­i­can As­so­ci­a­tion of Di­a­betes Ed­u­ca­tors, the Amer­i­can Di­a­betes As­so­ci­a­tion, the En­docrine So­ci­ety, JDRF In­ternational, The Leona M. and Harry B. Helm­s­ley Char­i­ta­ble Trust, the Pe­di­atric En­docrine So­ci­ety, and the T1D Ex­change. Di­a­betes Care 2017;40:1622–1630

  45. Karter AJ, Mof­fet HH, Liu JY, Lip­s­ka KJ. Surveil­lance of hy­po­glycemiadlim­i­ta­tions of emer­gen­cy de­part­ment and hos­pi­tal uti­liza­tion data. JAMA In­tern Med 2018;178:987–988

  46. Lee AK, Lee CJ, Huang ES, Shar­rett AR, Core­sh J, Selvin E. Risk fac­tors for se­vere hy­po­glycemia in black and white adults with di­a­betes: the Atheroscle­ro­sis Risk in Com­mu­ni­ties (ARIC) study. Di­a­betes Care 2017;40:1661–1667

  47. Karter AJ, Lip­s­ka KJ, O’Con­nor PJ, et al.; SUPREME-‍DM Study Group. High rates of se­vere hy­po­glycemia among African Amer­i­can pa­tients with di­a­betes: the SUrveil­lance, PRE­ven­tion, and Man­agE­ment of Di­a­betes Mel­li­tus (SUPREME-‍DM) net­work. J Di­a­betes Com­pli­ca­tions 2017;31: 869–873

  48. Whit­mer RA, Karter AJ, Yaffe K, Que­sen­ber­ry CP Jr, Selby JV. Hy­po­glycemic episodes and risk of de­men­tia in older pa­tients with type 2 di­a­betes mel­li­tus. JAMA 2009;301:1565–1572

  49. Pun­tha­kee Z, Miller ME, Launer LJ, et al.; AC­CORD Group of In­ves­ti­ga­tors; AC­CORD-‍MIND In­ves­ti­ga­tors. Poor cog­ni­tive func­tion and risk of se­vere hy­po­glycemia in type 2 di­a­betes: post hoc epi­demi­o­log­ic anal­y­sis of the AC­CORD trial. Di­a­betes Care 2012;35: 787–793

  50. Ja­cob­son AM, Musen G, Ryan CM, et al.; Di­a­betes Con­trol and Com­pli­ca­tions Trial/‍ Epi­demi­ol­o­gy of Di­a­betes In­ter­ven­tions and Com­pli­ca­tions Study Re­search Group. Longterm ef­fect of di­a­betes and its treat­ment on cog­ni­tive func­tion. N Engl J Med 2007;356: 1842–1852

  51. Zoun­gas S, Patel A, Chalmers J, et al.; AD­VANCE Col­lab­o­ra­tive Group. Se­vere hy­po­glycemia and risks of vas­cu­lar events and death. N Engl J Med 2010;363:1410–1418

  52. McCoy RG, Van Houten HK, Ziegen­fuss JY, Shah ND, Wer­m­ers RA, Smith SA. In­creased mor­tal­i­ty of pa­tients with di­a­betes re­port­ing se­vere hy­po­glycemia. Di­a­betes Care 2012;35: 1897–1901

  53. Du­Bose SN, We­in­stock RS, Beck RW, et al. Hy­po­glycemia in older adults with type 1 di­a­betes. Di­a­betes Tech­nol Ther 2016;18:765– 771

  54. Seaquist ER, An­der­son J, Childs B, et al. Hy­po­glycemia and di­a­betes: a re­port of a work­group of the Amer­i­can Di­a­betes As­so­ci­a­tion and the En­docrine So­ci­ety. Di­a­betes Care 2013;36: 1384–1395

  55. Bergen­stal RM, Klonoff DC, Garg SK, et al.; AS­PIRE In-‍Home Study Group. Thresh­old-‍based in­sulin-‍pump in­ter­rup­tion for re­duc­tion of hy­po­glycemia. N Engl J Med 2013;369:224– 232

  56. Her­ing BJ, Clarke WR, Bridges ND, et al.; Clin­i­cal Islet Trans­plan­ta­tion Con­sor­tium.

    Phase 3 trial of trans­plan­ta­tion of human islets in type 1 di­a­betes com­pli­cat­ed by se­vere hy­po­glycemia. Di­a­betes Care 2016;39:1230– 1240

  57. Har­lan DM. Islet trans­plan­ta­tion for hy­po­glycemia un­aware­ness/se­vere hy­po­glycemia:

    caveat emp­tor. Di­a­betes Care 2016;39:1072– 1074

  58. Lay­man DK, Clifton P, Gan­non MC, Krauss RM, Nut­tall FQ. Pro­tein in op­ti­mal health: heart dis­ease and type 2 di­a­betes. Am J Clin Nutr 2008; 87:1571S–1575S

  59. Cryer PE. Di­verse caus­es of hy­po­glycemia-‍as­so­ci­at­ed au­to­nom­ic fail­ure in di­a­betes. N Engl J Med 2004;350:2272–2279

  60. Kitabchi AE, Umpier­rez GE, Miles JM, Fish­er JN. Hy­per­glycemic crises in adult pa­tients with di­a­betes. Di­a­betes Care 2009;32:1335–1343