GLYCEMIC TARGETS: STANDARDSOF MEDICALCAREINDIABETESD2021
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Diabetes Care Volume 44, Supplement 1, January 2021 S73
6. Glycemic Targets: Standards of American Diabetes Association
Medical Care in Diabetesd2021
Diabetes Care 2021;44(Suppl. 1):S73–S84 | https://doi.org/10.2337/dc21-S006
The American Diabetes Association (ADA) “Standards of Medical Care in Diabetes”
includes the 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
6. GLYCEMIC TARGETS
Committee, a multidisciplinary expert committee (https://doi.org/10.2337/dc21-
SPPC), 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 (https://doi
.org/10.2337/dc21-SINT). 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 control is assessed by the A1C measurement, continuous glucose monitoring
(CGM), and self-monitoring of blood glucose (SMBG). A1C is the metric used to date
in clinical trials demonstrating the benefits of improved glycemic control. Patient
SMBG can be used with self-management and medication adjustment, partic-
ularly in individuals taking insulin. CGM serves an important role in assessing the
effectiveness and safety of treatment in many patients with type 1 diabetes, including
prevention of hypoglycemia, and in selected patients with type 2 diabetes, such as
in those on intensive insulin regimens and in those on regimens associated with
hypoglycemia.
Glycemic Assessment
Recommendations
6.1 Assess glycemic status (A1C or other glycemic measurement) at least two
times a year in patients who are meeting treatment goals (and who have
stable glycemic control). E
6.2 Assess glycemic status at least quarterly, and as needed, in patients whose
therapy has recently changed and/or who are not meeting glycemic goals. E
A1C reflects average glycemia over approximately 3 months. The performance of the
test is generally excellent for National Glycohemoglobin Standardization Program Suggested citation: American Diabetes Associa-
(NGSP)-certified assays (see www.ngsp.org). The test is the primary tool for assessing tion. 6. Glycemic targets: Standards of Medical
glycemic control and has strong predictive value for diabetes complications (1–3). Care in Diabetesd2021. Diabetes Care 2021;
Thus, A1C testing should be performed routinely in all patients with diabetes at initial 44(Suppl. 1):S73–S84
assessment and as part of continuing care. Measurement approximately every © 2020 by the American Diabetes Association.
3 months determines whether patients’ glycemic targets have been reached and Readers may use this article as long as the work is
properly cited, the use is educational and not for
maintained. The frequency of A1C testing should depend on the clinical situation, the profit, and the work is not altered. More infor-
treatment regimen, and the clinician’s judgment. The use of point-of-care A1C testing mation is available at https://www.diabetesjournals
may provide an opportunity for more timely treatment changes during encounters .org/content/license.S74 Glycemic Targets Diabetes Care Volume 44, Supplement 1, January 2021
between patients and providers. Pa- accuracy of the patient’s CGM or meter Table 6.1—Estimated average glucose
tients with type 2 diabetes with stable (or the patient’s reported SMBG re- (eAG)
glycemia well within target may do well sults) and the adequacy of the SMBG A1C (%) mg/dL* mmol/L
with A1C testing or other glucose as- monitoring.
5 97 (76–120) 5.4 (4.2–6.7)
sessment only twice per year. Unstable
Correlation Between SMBG and A1C 6 126 (100–152) 7.0 (5.5–8.5)
or intensively managed patients or
people not at goal with treatment ad- Table 6.1 shows the correlation be- 7 154 (123–185) 8.6 (6.8–10.3)
justments may require testing more tween A1C levels and mean glucose 8 183 (147–217) 10.2 (8.1–12.1)
frequently (every 3 months with interim levels based on the international A1C- 9 212 (170–249) 11.8 (9.4–13.9)
assessments as needed) (4). Derived Average Glucose (ADAG) study, 10 240 (193–282) 13.4 (10.7–15.7)
which assessed the correlation between 11 269 (217–314) 14.9 (12.0–17.5)
A1C Limitations A1C and frequent SMBG and CGM in 12 298 (240–347) 16.5 (13.3–19.3)
The A1C test is an indirect measure of 507 adults (83% non-Hispanic Whites)
average glycemia and, as such, is subject Data in parentheses are 95% CI. A calculator
with type 1, type 2, and no diabetes (6), for converting A1C results into eAG, in
to limitations. As with any laboratory and an empirical study of the average either mg/dL or mmol/L, is available at
test, there is variability in the measure- blood glucose levels at premeal, post- professional.diabetes.org/eAG. *These
ment of A1C. Although A1C variability is estimates are based on ADAG data of ;2,700
meal, and bedtime associated with spec- glucose measurements over 3 months per
lower on an intraindividual basis than ified A1C levels using data from the ADAG A1C measurement in 507 adults with type 1,
that of blood glucose measurements, trial (7). The American Diabetes Associa- type 2, or no diabetes. The correlation
clinicians should exercise judgment when tion (ADA) and the American Association between A1C and average glucose was 0.92
using A1C as the sole basis for assess- (6,7). Adapted from Nathan et al. (6).
for Clinical Chemistry have determined
ing glycemic control, particularly if the that the correlation (r 5 0.92) in the
result is close to the threshold that ADAG trial is strong enough to justify
might prompt a change in medication reporting both the A1C result and the
therapy. For example, conditions that estimated average glucose (eAG) result variants. Other assays have statistically
affect red blood cell turnover (hemolytic when a clinician orders the A1C test. significant interference, but the differ-
and other anemias, glucose-6-phosphate Clinicians should note that the mean ence is not clinically significant. Use of an
dehydrogenase deficiency, recent blood plasma glucose numbers in Table 6.1 assay with such statistically significant
transfusion, use of drugs that stimulate are based on ;2,700 readings per A1C interference may explain a report that
erythropoesis, end-stage kidney disease, in the ADAG trial. In a recent report, for any level of mean glycemia, African
and pregnancy) may result in discrep- mean glucose measured with CGM versus Americans heterozygous for the com-
ancies between the A1C result and the central laboratory–measured A1C in 387 mon hemoglobin variant HbS had lower
patient’s true mean glycemia. Hemoglo- participants in three randomized trials A1C by about 0.3 percentage points
bin variants must be considered, par- demonstrated that A1C may underesti- when compared with those without the
ticularly when the A1C result does not mate or overestimate mean glucose in trait (10,11). Another genetic variant,
correlate with the patient’s CGM or individuals (5). Thus, as suggested, a pa- X-linked glucose-6-phosphate dehydro-
SMBG levels. However, most assays in tient’s SMBG or CGM profile has consid- genase G202A, carried by 11% of African
use in the U.S. are accurate in individuals erable potential for optimizing his or Americans, was associated with a de-
heterozygous for the most common her glycemic management (5). crease in A1C of about 0.8% in hemi-
variants (see www.ngsp.org/interf.asp). zygous men and 0.7% in homozygous
Other measures of average glycemia A1C Differences in Ethnic Populations women compared with those without
such as fructosamine and 1,5-anhydro- and Children the trait (12).
glucitol are available, but their translation In the ADAG study, there were no sig- A small study comparing A1C to CGM
into average glucose levels and their nificant differences among racial and ethnic data in children with type 1 diabetes
prognostic significance are not as clear groups in the regression lines between found a highly statistically significant
as for A1C and CGM. Though some A1C and mean glucose, although the study correlation between A1C and mean
variability in the relationship between was underpowered to detect a difference blood glucose, although the correlation
average glucose levels and A1C exists and there was a trend toward a difference (r 5 0.7) was significantly lower than in
among different individuals, generally between the African and African American the ADAG trial (13). Whether there are
the association between mean glucose and the non-Hispanic White cohorts, with clinically meaningful differences in how
and A1C within an individual correlates higher A1C values observed in Africans A1C relates to average glucose in children
over time (5). and African Americans compared with non- or in different ethnicities is an area for
A1C does not provide a measure of Hispanic Whites for a given mean glucose. further study (8,14,15). Until further
glycemic variability or hypoglycemia. For Other studies have also demonstrated evidence is available, it seems prudent
patients prone to glycemic variability, higher A1C levels in African Americans to establish A1C goals in these popula-
especially patients with type 1 diabetes or than in Whites at a given mean glucose tions with consideration of individual-
type 2 diabetes with severe insulin de- concentration (8,9). ized CGM, SMBG, and A1C results. This
ficiency, glycemic control is best evaluated A1C assays are available that do not limitation does not interfere with the
by the combination of results from SMBG demonstrate a statistically significant usefulness of CGM for insulin dose
or CGM and A1C. A1C may also inform the difference in individuals with hemoglobin adjustments.care.diabetesjournals.org Glycemic Targets S75
Glucose Assessment by Continuous to therapy and assess whether glycemic goals in pregnant women, please refer
Glucose Monitoring targets are being safely achieved. The to Section 14 “Management of Diabetes
international consensus on TIR provides in Pregnancy” (https://doi.org/10.2337/
Recommendations
guidance on standardized CGM metrics dc21-S014). Overall, regardless of the
6.3 Standardized, single-page glucose
(see Table 6.2) and considerations for population being served, it is critical for
reports from continuous glucose
clinical interpretation and care (26). To the glycemic targets to be woven into the
monitoring (CGM) devices with
make these metrics more actionable, overall patient-centered strategy. For ex-
visual cues, such as the ambula-
standardized reports with visual cues, ample, in a very young child safety and
tory glucose profile (AGP), should
such as the ambulatory glucose profile simplicity may outweigh the need for
be considered as a standard print-
(see Fig. 6.1), are recommended (26) and perfect control in the short run. Simpli-
out for all CGM devices. E
may help the patient and the provider fication may decrease parental anxiety
6.4 Time in range (TIR) is associated
better interpret the data to guide treat- and build trust and confidence, which
with the risk of microvascular
ment decisions (16,19). SMBG and CGM could support further strengthening of
complications, should be an ac-
can be useful to guide medical nutrition glycemic targets and self-efficacy. Simi-
ceptable end point for clinical
therapy and physical activity, prevent larly, in healthy older adults, there is no
trials moving forward, and can be
hypoglycemia, and aid medication man- empiric need to loosen control. However,
used for assessment of glycemic
agement. While A1C is currently the the provider needs to work with an
control. Additionally, time below
primary measure to guide glucose man- individual and should consider adjusting
target (,70 and ,54 mg/dL [3.9
agement and a valuable risk marker for targets or simplifying the regimen if this
and 3.0 mmol/L]) and time above
developing diabetes complications, the change is needed to improve safety and
target (.180 mg/dL [10.0 mmol/L])
glucose management indicator (GMI) adherence.
are useful parameters for reeval-
along with the other CGM metrics pro-
uation of the treatment regimen. C Recommendations
vide for a more personalized diabetes
management plan. The incorporation of 6.5a An A1C goal for many nonpreg-
CGM is rapidly improving diabetes man- these metrics into clinical practice is in nant adults of ,7% (53 mmol/mol)
agement. As stated in the recommen- evolution, and optimization and harmo- without significant hypoglycemia
dations, time in range (TIR) is a useful nization of CGM terminology will evolve is appropriate. A
metric of glycemic control and glucose to suit patient and provider needs. The 6.5b If using ambulatory glucose pro-
patterns and it correlates well with A1C patient’s specific needs and goals should file/glucose management indica-
in most studies (16–21). New data sup- dictate SMBG frequency and timing and tor to assess glycemia, a parallel
port that increased TIR correlates with consideration of CGM use. Please refer to goal is a time in range of .70%
the risk of complications. The studies Section 7 “Diabetes Technology” (https:// with time below range ,4%
supporting this assertion are reviewed doi.org/10.2337/dc21-S007) for a ful- (Fig. 6.1). B
in more detail in Section 7 “Diabetes ler discussion of the use of SMBG and 6.6 On the basis of provider judg-
Technology” (https://doi.org/10.2337/ CGM. ment and patient preference,
dc21-S007); they include cross-sectional With the advent of new technology, achievement of lower A1C lev-
data and cohort studies (22–24) dem- CGM has evolved rapidly in both accuracy els than the goal of 7% may be
onstrating TIR as an acceptable end point and affordability. As such, many patients acceptable, and even beneficial,
for clinical trials moving forward and have these data available to assist with if it can be achieved safely with-
that it can be used for assessment of both self-management and assessment out significant hypoglycemia or
glycemic control. Additionally, time be- by providers. Reports can be generated other adverse effects of treat-
low target (,70 and ,54 mg/dL [3.9 and from CGM that will allow the provider to ment. C
3.0 mmol/L]) and time above target determine TIR and to assess hypoglyce- 6.7 Less stringent A1C goals (such
(.180 mg/dL [10.0 mmol/L]) are useful mia, hyperglycemia, and glycemic vari- as ,8% [64 mmol/mol]) may be
parameters for reevaluation of the treat- ability. As discussed in a recent consensus appropriate for patients with
ment regimen. document, a report formatted as shown in limited life expectancy, or where
For many people with diabetes, glu- Fig. 6.1 can be generated (26). Published the harms of treatment are greater
cose monitoring is key for achieving data suggest a strong correlation between than the benefits. B
glycemic targets. Major clinical trials of TIR and A1C, with a goal of 70% TIR 6.8 Reassess glycemic targets over
insulin-treated patients have included aligning with an A1C of ;7% in two time based on the criteria in Fig.
SMBG as part of multifactorial inter- prospective studies (18,27). 6.2 and in older adults (Table
ventions to demonstrate the benefit of 12.1). E
intensive glycemic control on diabetes
complications (25). SMBG is thus an in- GLYCEMIC GOALS A1C and Microvascular Complications
tegral component of effective therapy of For glycemic goals in older adults, please Hyperglycemia defines diabetes, and gly-
patients taking insulin. In recent years, refer to Section 12, “Older Adults” (https:// cemic control is fundamental to diabetes
CGM has emerged as a complementary doi.org/10.2337/dc21-S012). For glycemic management. The Diabetes Control and
method for assessing glucose levels. Both goals in children, please refer to Section 13 Complications Trial (DCCT) (25), a pro-
approaches to glucose monitoring allow “Children and Adolescents” (https://doi spective randomized controlled trial of in-
patients to evaluate individual response .org/10.2337/dc21-S013). For glycemic tensive(meanA1Cabout7%[53mmol/mol])S76 Glycemic Targets Diabetes Care Volume 44, Supplement 1, January 2021
Table 6.2—Standardized CGM metrics for clinical care enduring effects of early glycemic con-
trol on most microvascular complications
1. Number of days CGM device is worn (recommend 14 days) (33).
2. Percentage of time CGM device is active Therefore, achieving A1C targets of
(recommend 70% of data from 14 days) ,7% (53 mmol/mol) has been shown
3. Mean glucose to reduce microvascular complications
4. Glucose management indicator of type 1 and type 2 diabetes when
5. Glycemic variability (%CV) target #36%* instituted early in the course of disease
6. TAR: % of readings and time .250 mg/dL (1,34). Epidemiologic analyses of the
(.13.9 mmol/L) Level 2 hyperglycemia DCCT (25) and UKPDS (35) demonstrate a
7. TAR: % of readings and time 181–250 mg/dL curvilinear relationship between A1C and
(10.1–13.9 mmol/L) Level 1 hyperglycemia
microvascular complications. Such anal-
8. TIR: % of readings and time 70–180 mg/dL (3.9–10.0 mmol/L) In range yses suggest that, on a population level,
9. TBR: % of readings and time 54–69 mg/dL (3.0–3.8 mmol/L) Level 1 hypoglycemia the greatest number of complications
10. TBR: % of readings and time ,54 mg/dL (,3.0 mmol/L) Level 2 hypoglycemia will be averted by taking patients from
CGM, continuous glucose monitoring; CV, coefficient of variation; TAR, time above range; TBR, very poor control to fair/good control.
time below range; TIR, time in range. *Some studies suggest that lower %CV targets (,33%) These analyses also suggest that further
provide additional protection against hypoglycemia for those receiving insulin or sulfonylureas.
Adapted from Battelino et al. (26).
lowering of A1C from 7% to 6% [53 mmol/
mol to 42 mmol/mol] is associated with
further reduction in the risk of microvas-
cular complications, although the abso-
versus standard (mean A1C about 9% these microvascular benefits over two lute risk reductions become much smaller.
[75 mmol/mol]) glycemic control in pa- decades despite the fact that the glycemic The implication of these findings is that
tients with type 1 diabetes, showed de- separation between the treatment groups there is no need to deintensify therapy for
finitively that better glycemic control is diminished and disappeared during an individual with an A1C between 6% and
associated with 50–76% reductions in follow-up. 7% and low hypoglycemia risk with a long
rates of development and progression of The Kumamoto Study (30) and UK life expectancy. There are now newer
microvascular (retinopathy, neuropathy, Prospective Diabetes Study (UKPDS) (31,32) agents that do not cause hypoglycemia,
and diabetic kidney disease) complica- confirmed that intensive glycemic con- making it possible to maintain glucose
tions. Follow-up of the DCCT cohorts in trol significantly decreased rates of mi- control without risk of hypoglycemia (see
the Epidemiology of Diabetes Interven- crovascular complications in patients with Section 9 “Pharmacologic Approaches to
tions and Complications (EDIC) study short-duration type 2 diabetes. Long-term Glycemic Treatment,” https://doi.org/10
(28,29) demonstrated persistence of follow-up of the UKPDS cohorts showed .2337/dc21-S009).
Figure 6.1—Key points included in standard ambulatory glucose profile (AGP) report. Adapted from Battelino et al. (26).care.diabetesjournals.org Glycemic Targets S77
populations with diabetes. There is ev-
idence 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 random-
ized 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 stan-
dard arm (41). The benefit of intensive
glycemic control in this cohort with type 1
diabetes has been shown to persist for
several decades (42) and to be associated
with a modest reduction in all-cause
mortality (43).
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
Figure 6.2—Patient and disease factors used to determine optimal glycemic targets. Character-
long-term CVD rates. In addition, data
istics and predicaments toward the left justify more stringent efforts to lower A1C; those toward
the right suggest less stringent efforts. A1C 7% 5 53 mmol/mol. Adapted with permission from from the Swedish National Diabetes Reg-
Inzucchi et al. (59). istry (44) and the Joint Asia Diabetes
Evaluation (JADE) demonstrate greater
Given the substantially increased risk A1C goals in people with long-standing proportions of people with diabetes be-
of hypoglycemia in type 1 diabetes and type 2 diabetes with or at significant risk of ing diagnosed at ,40 years of age and a
with polypharmacy in type 2 diabetes, CVD. These landmark studies need to be demonstrably increased burden of heart
the risks of lower glycemic targets may considered with an important caveat; glu- disease and years of life lost in people
outweigh the potential benefits on micro- cagon like peptide 1 (GLP-1) receptor ago- diagnosed at a younger age (45–48).
vascular complications. Three landmark nists and sodium–glucose cotransporter Thus, for prevention of both microvas-
trials (Action to Control Cardiovascular 2 (SGLT2) inhibitors were not approved at cular and macrovascular complications of
Risk in Diabetes [ACCORD], Action in the time of these trials. As such, these diabetes, there is a major call to over-
Diabetes and Vascular Disease: Preterax agents with established cardiovascular and come therapeutic inertia and treat to
and Diamicron MR Controlled Evaluation renal benefit appear to be safe in this group target for an individual patient (47,49).
[ADVANCE], and Veterans Affairs Diabe- of high-risk patients. Clinical trials examining During the UKPDS, there was a 16%
tes Trial [VADT]) were conducted to test these agents for cardiovascular safety were reduction in CVD events (combined fatal
the effects of near normalization of blood notdesignedtotesthigherversuslowerA1C; or nonfatal MI and sudden death) in the
glucose on cardiovascular outcomes in therefore, beyond post hoc analysis of these intensive glycemic control arm that did
individuals with long-standing type 2 di- trials, we do not have evidence that it is the not reach statistical significance (P 5
abetes and either known cardiovascular glucose lowering by these agents that con- 0.052), and there was no suggestion of
disease (CVD) or high cardiovascular risk. fers the CVD and renal benefit (40). As such, benefit on other CVD outcomes (e.g.,
These trials showed that lower A1C levels on the basis of physician judgment and stroke). Similar to the DCCT/EDIC, after
were associated with reduced onset or patient preferences, select patients, es- 10 years of observational follow-up,
progression of some microvascular com- pecially those with little comorbidity and those originally randomized to intensive
plications (36–38). long life expectancy, may benefit from glycemic control had significant long-term
The concerning mortality findings in adopting more intensive glycemic targets reductions in MI (15% with sulfonylurea
the ACCORD trial (39), discussed below, if they can achieve them safely without or insulin as initial pharmacotherapy,
and the relatively intense efforts required hypoglycemia or significant therapeutic 33% with metformin as initial pharma-
to achieve near euglycemia should also be burden. cotherapy) and in all-cause mortality (13%
considered when setting glycemic targets and 27%, respectively) (33).
for individuals with long-standing diabe- A1C and Cardiovascular Disease ACCORD, ADVANCE, and VADT sug-
tes, such as those studied in ACCORD, Outcomes gested no significant reduction in CVD
ADVANCE, and VADT. Findings from these Cardiovascular Disease and Type 1 Diabetes outcomes with intensive glycemic con-
studies suggest caution is needed in treat- CVD is a more common cause of death trol in participants followed for shorter
ing diabetes aggressively to near-normal than microvascular complications in durations (3.5–5.6 years) and who hadS78 Glycemic Targets Diabetes Care Volume 44, Supplement 1, January 2021
more advanced type 2 diabetes than age/frailty may benefit from less aggres- The factors to consider in individual-
UKPDS participants. All three trials were sive targets (55,56). izing goals are depicted in Fig. 6.2. This
conducted in relatively older participants As discussed further below, severe figure is not designed to be applied rigidly
with longer known duration of diabetes hypoglycemia is a potent marker of high but to be used as a broad construct to
(mean duration 8–11 years) and either absolute risk of cardiovascular events and guide clinical decision-making (59) and
CVD or multiple cardiovascular risk fac- mortality (57). Providers should be vigilant in engage in shared decision-making in
tors. The target A1C among intensive- preventing hypoglycemia and should not people with type 1 and type 2 diabetes.
control subjects was ,6% (42 mmol/mol) aggressively attempt to achieve near-normal More stringent targets may be recom-
in ACCORD, ,6.5% (48 mmol/mol) in A1C levels in patients in whom such mended if they can be achieved safely
ADVANCE, and a 1.5% reduction in A1C targets cannot be safely and reasonably and with acceptable burden of therapy
compared with control subjects in VADT, achieved. As discussed in Section 9 “Phar- and if life expectancy is sufficient to reap
with achieved A1C of 6.4% vs. 7.5% macologic Approaches to Glycemic Treat- benefits of stringent targets. Less strin-
(46 mmol/mol vs. 58 mmol/mol) in AC- ment” (https://doi.org/10.2337/dc21-S009), gent targets (A1C up to 8% [64 mmol/
CORD, 6.5% vs. 7.3% (48 mmol/mol vs. addition of specific (SGLT2) inhibitors or mol]) may be recommended if the life
56 mmol/mol) in ADVANCE, and 6.9% vs. GLP-1 receptor agonists that have dem- expectancy of the patient is such that the
8.4% (52 mmol/mol vs. 68 mmol/mol) in onstrated CVD benefit is recommended benefits of an intensive goal may not be
VADT. Details of these studies are re- for use in patients with established CVD, realized, or if the risks and burdens
viewed extensively in the joint ADA po- chronic kidney disease, and heart failure. outweigh the potential benefits. Severe
sition statement, “Intensive Glycemic As outlined in more detail in Section 9 or frequent hypoglycemia is an absolute
Control and the Prevention of Cardio- “Pharmacologic Approaches to Glycemic indication for the modification of treat-
vascular Events: Implications of the Treatment” (https://doi.org/10.2337/dc21- ment regimens, including setting higher
ACCORD, ADVANCE, and VA Diabetes S009) and Section 10 “Cardiovascular Dis- glycemic goals.
Trials” (50). ease and Risk Management” (https://doi Diabetes is a chronic disease that pro-
The glycemic control comparison in .org/10.2337/dc21-S010), the cardiovas- gresses over decades. Thus, a goal that
ACCORD was halted early due to an in- cular benefits of SGLT2 inhibitors or GLP-1 might be appropriate for an individual
creased mortality rate in the intensive receptor agonists are not dependent upon early in the course of their diabetes may
compared with the standard treatment A1C lowering; therefore, initiation can be change over time. Newly diagnosed pa-
arm (1.41% vs. 1.14% per year; hazard considered in people with type 2 diabetes tients and/or those without comorbidities
ratio 1.22 [95% CI 1.01–1.46]), with a and CVD independent of the current A1C that limit life expectancy may benefit from
similar increase in cardiovascular deaths. or A1C goal or metformin therapy. Based intensive control proven to prevent mi-
Analysis of the ACCORD data did not on these considerations, the following crovascular complications. Both DCCT/
identify a clear explanation for the two strategies are offered (58): EDIC and UKPDS demonstrated metabolic
excess mortality in the intensive treat- memory, or a legacy effect, in which a
ment arm (39). 1. If already on dual therapy or multiple finite period of intensive control yielded
Longer-term follow-up has shown no glucose-lowering therapies and not on benefits that extended for decades after
evidence of cardiovascular benefit or harm an SGLT2 inhibitor or GLP-1 receptor that control ended. Thus, a finite period
in the ADVANCE trial (51). The end-stage agonist, consider switching to one of of intensive control to near-normal A1C
renal disease rate was lower in the intensive these agents with proven cardiovas- may yield enduring benefits even if con-
treatment group over follow-up. However, cular benefit. trol is subsequently deintensified as pa-
10-year follow-up of the VADT cohort (52) 2. Introduce SGLT2 inhibitors or GLP-1 tient characteristics change. Over time,
showed a reduction in the risk of cardio- receptor agonists in patients with CVD comorbidities may emerge, decreasing
vascularevents(52.7[controlgroup]vs.44.1 at A1C goal (independent of metformin) life expectancy and thereby decreasing
[intervention group] events per 1,000 person- for cardiovascular benefit, independent the potential to reap benefits from in-
years) with no benefit in cardiovascular or of baseline A1C or individualized A1C tensive control. Also, with longer dura-
overall mortality. Heterogeneity of mortality target. tion of disease, diabetes may become
effects across studies was noted, which more difficult to control, with increasing
may reflect differences in glycemic targets, Setting and Modifying A1C Goals risks and burdens of therapy. Thus, A1C
therapeutic approaches, and, importantly, Numerous factors must be considered targets should be reevaluated over time
population characteristics (53). when setting glycemic targets. The ADA to balance the risks and benefits as patient
Mortality findings in ACCORD (39) and proposes general targets appropriate for factors change.
subgroup analyses of VADT (54) suggest many patients but emphasizes the im- Recommended glycemic targets for
that the potential risks of intensive glyce- portance of individualization based on many nonpregnant adults are shown in
mic control may outweigh its benefits key patient characteristics. Glycemic tar- Table 6.3. The recommendations include
in higher-risk patients. In all three trials, gets must be individualized in the context blood glucose levels that appear to cor-
severe hypoglycemia was significantly more of shared decision-making to address the relate with achievement of an A1C
likely in participants who were randomly needs and preferences of each patient of ,7% (53 mmol/mol). Pregnancy rec-
assigned to the intensive glycemic control and the individual characteristics that ommendations are discussed in more
arm. Those patients with long duration of influence risks and benefits of therapy for detail in Section 14 “Management
diabetes, a known history of hypoglycemia, each patient in order to optimize patient of Diabetes in Pregnancy” (https://doi
advanced atherosclerosis, or advanced engagement and self-efficacy. .org/10.2337/dc21-S014).care.diabetesjournals.org Glycemic Targets S79
The issue of preprandial versus post- findings support that premeal glucose
and reevaluation of the treat-
prandial SMBG targets is complex (60). targets may be relaxed without under-
ment regimen. E
Elevated postchallenge (2-h oral glucose mining overall glycemic control as mea-
6.13 Insulin-treated patients with hy-
tolerance test) glucose values have been sured by A1C. These data prompted the
poglycemia unawareness, one
associated with increased cardiovascular revision in the ADA-recommended pre-
level 3 hypoglycemic event, or
risk independent of fasting plasma glu- meal glucose target to 80–130 mg/dL
a pattern of unexplained level 2
cose in some epidemiologic studies, (4.4–7.2 mmol/L) but did not affect the
hypoglycemia should be advised
whereas intervention trials have not shown definition of hypoglycemia.
to raise their glycemic targets to
postprandial glucose to be a cardiovas-
strictly avoid hypoglycemia for
cular risk factor independent of A1C. In HYPOGLYCEMIA
at least several weeks in order
people with diabetes, surrogate measures
Recommendations to partially reverse hypoglyce-
of vascular pathology, such as endothelial
6.9 Occurrence and risk for hypo- mia unawareness and reduce
dysfunction, are negatively affected by
glycemia should be reviewed at risk of future episodes. A
postprandial hyperglycemia. It is clear
every encounter and investigated 6.14 Ongoing assessment of cogni-
that postprandial hyperglycemia, like pre-
as indicated. C tive function is suggested with
prandial hyperglycemia, contributes to
6.10 Glucose (approximately 15–20 g) increased vigilance for hypogly-
elevated A1C levels, with its relative
is the preferred treatment for cemia by the clinician, patient,
contribution being greater at A1C levels
the conscious individual with blood and caregivers if low cognition
that are closer to 7% (53 mmol/mol).
glucose ,70 mg/dL (3.9 mmol/L], or declining cognition is found. B
However, outcome studies have clearly
although any form of carbohy-
shown A1C to be the primary predictor of
drate that contains glucose may
complications, and landmark trials of Hypoglycemia is the major limiting fac-
be used. Fifteen minutes after
glycemic control such as the DCCT and tor in the glycemic management of type 1
treatment, if self-monitoring of
UKPDS relied overwhelmingly on pre- and type 2 diabetes. Recommendations
blood glucose (SMBG) shows
prandial SMBG. Additionally, a random- regarding the classification of hypogly-
continued hypoglycemia, the
ized controlled trial in patients with cemia are outlined in Table 6.4 (63–68).
treatment should be repeated.
known CVD found no CVD benefit of Level 1 hypoglycemia is defined as a
Once the SMBG or glucose pattern
insulin regimens targeting postprandial measurable glucose concentration ,70
is trending up, the individual
glucose compared with those targeting mg/dL (3.9 mmol/L) but $54 mg/dL (3.0
should consume a meal or snack
preprandial glucose (61). Therefore, it is mmol/L). A blood glucose concentra-
to prevent recurrence of hypogly-
reasonable for postprandial testing to be tion of 70 mg/dL (3.9 mmol/L) has been
cemia. B
recommended for individuals who have recognized as a threshold for neuroendo-
6.11 Glucagon should be prescribed
premeal glucose values within target but crine responses to falling glucose in peo-
for all individuals at increased
A1C values above target. In addition, ple without diabetes. Because many
risk of level 2 or 3 hypoglycemia
when intensifying insulin therapy, mea- people with diabetes demonstrate im-
so that it is available should it be
suring postprandial plasma glucose 1– paired counterregulatory responses to
needed. Caregivers, school per-
2 h after the start of a meal and using hypoglycemia and/or experience hypo-
sonnel, or family members of
treatments aimed at reducing post- glycemia unawareness, a measured glu-
these individuals should know
prandial plasma glucose values to cose level ,70 mg/dL (3.9 mmol/L) is
where it is and when and how to
,180 mg/dL (10.0 mmol/L) may help considered clinically important (indepen-
administer it. Glucagon admin-
to lower A1C. dent of the severity of acute hypoglycemic
istration is not limited to health
An analysis of data from 470 partici- symptoms). Level 2 hypoglycemia (de-
care professionals. E
pants in the ADAG study (237 with type 1 fined as a blood glucose concentration
6.12 Hypoglycemia unawareness or
diabetes and 147 with type 2 diabetes) ,54 mg/dL [3.0 mmol/L]) is the thresh-
one or more episodes of level 3
found that the glucose ranges highlighted old at which neuroglycopenic symptoms
hypoglycemia should trigger hy-
in Table 6.1 are adequate to meet targets begin to occur and requires immediate
poglycemia avoidance education
and decrease hypoglycemia (7,62). These action to resolve the hypoglycemic event.
If a patient has level 2 hypoglycemia
without adrenergic or neuroglycopenic
Table 6.3—Summary of glycemic recommendations for many nonpregnant symptoms, they likely have hypoglyce-
adults with diabetes
mia unawareness (discussed further
A1C ,7.0% (53 mmol/mol)*#
below). This clinical scenario warrants
Preprandial capillary plasma glucose 80–130 mg/dL* (4.4–7.2 mmol/L)
investigation and review of the medical
Peak postprandial capillary plasma glucose† ,180 mg/dL* (10.0 mmol/L)
regimen. Lastly, level 3 hypoglycemia is
*More or less stringent glycemic goals may be appropriate for individual patients. #CGM may be defined as a severe event characterized
used to assess glycemic target as noted in Recommendation 6.5b and Fig. 6.1. Goals should be
individualized based on duration of diabetes, age/life expectancy, comorbid conditions, known
by altered mental and/or physical func-
CVD or advanced microvascular complications, hypoglycemia unawareness, and individual patient tioning that requires assistance from
considerations (as per Fig. 6.2). †Postprandial glucose may be targeted if A1C goals are not met another person for recovery.
despite reaching preprandial glucose goals. Postprandial glucose measurements should be made Symptoms of hypoglycemia include,
1–2 h after the beginning of the meal, generally peak levels in patients with diabetes.
but are not limited to, shakiness,S80 Glycemic Targets Diabetes Care Volume 44, Supplement 1, January 2021
Table 6.4—Classification of hypoglycemia overtreatment and provide a safety mar-
Glycemic criteria/description
gin in patients titrating glucose-lowering
drugs such as insulin to glycemic targets.
Level 1 Glucose ,70 mg/dL (3.9 mmol/L) and $54 mg/dL
(3.0 mmol/L)
Hypoglycemia Treatment
Level 2 Glucose ,54 mg/dL (3.0 mmol/L)
Providers should continue to counsel pa-
Level 3 A severe event characterized by altered mental and/or
tients to treat hypoglycemia with fast-
physical status requiring assistance for treatment of
hypoglycemia acting carbohydrates at the hypoglycemia
alert value of 70 mg/dL (3.9 mmol/L) or
Reprinted from Agiostratidou et al. (63).
less. This should be reviewed at each
patient visit. Hypoglycemia treatment re-
quires ingestion of glucose- or carbohydrate-
irritability, confusion, tachycardia, and insulin use, poor or moderate versus containing foods (86–88). The acute
hunger. Hypoglycemia may be inconve- good glycemic control, albuminuria, and glycemic response correlates better
nient or frightening to patients with di- poor cognitive function (77). Level 3 hy- with the glucose content of food than
abetes. Level 3 hypoglycemia may be poglycemia was associated with mortality with the carbohydrate content of food.
recognized or unrecognized and can in participants in both the standard and Pure glucose is the preferred treatment,
progress to loss of consciousness, sei- the intensive glycemia arms of the AC- but any form of carbohydrate that con-
zure, coma, or death. Hypoglycemia is CORD trial, but the relationships between tains glucose will raise blood glucose.
reversed byadministration of rapid-acting hypoglycemia, achieved A1C, and treat- Added fat may retard and then prolong
glucose or glucagon. Hypoglycemia can ment intensity were not straightforward. the acute glycemic response. In type 2
cause acute harm to the person with An association of level 3 hypoglycemia diabetes, ingested protein may increase
diabetes or others, especially if it causes with mortality was also found in the insulin response without increasing
falls, motor vehicle accidents, or other ADVANCE trial (79). An association be- plasma glucose concentrations (89). There-
injury. Recurrent level 2 hypoglycemia tween self-reported level 3 hypoglycemia fore, carbohydrate sources high in protein
and/or level 3 hypoglycemia is an urgent and 5-year mortality has also been re- should not be used to treat or prevent
medical issue and requires interven- ported in clinical practice (80). hypoglycemia. Ongoing insulin activity or
tion with medical regimen adjustment, Young children with type 1 diabetes insulin secretagogues may lead to recur-
behavioral intervention, and, in some and the elderly, including those with rent hypoglycemia unless more food is
cases, use of technology to assist with type 1 and type 2 diabetes (73,81), are ingested after recovery. Once the glucose
hypoglycemia prevention and identifica- noted as particularly vulnerable to hypo- returns to normal, the individual should
tion (64,69–72). A large cohort study glycemia because of their reduced ability be counseled to eat a meal or snack to
suggested that among older adults with to recognize hypoglycemic symptoms and prevent recurrent hypoglycemia.
type 2 diabetes, a history of level 3 hy- effectively communicate their needs. In- Glucagon
poglycemia was associated with greater dividualized glucose targets, patient ed- The use of glucagon is indicated for the
risk of dementia (73). Conversely, in a ucation, dietary intervention (e.g., bedtime treatment of hypoglycemia in people
substudy of the ACCORD trial, cognitive snack to prevent overnight hypoglycemia unable or unwilling to consume carbo-
impairment at baseline or decline in when specifically needed to treat low hydrates by mouth. Those in close con-
cognitive function during the trial was blood glucose), exercise management, tact with, or having custodial care of,
significantly associated with subsequent medication adjustment, glucose moni- people with hypoglycemia-prone diabetes
episodes of level 3 hypoglycemia (74). toring, and routine clinical surveillance (family members, roommates, school
Evidence from DCCT/EDIC, which in- may improve patient outcomes (82). personnel, childcare providers, correc-
volved adolescents and younger adults CGM with automated low glucose sus- tional institution staff, or coworkers)
with type 1 diabetes, found no associa- pend has been shown to be effective in should be instructed on the use of glu-
tion between frequency of level 3 hypo- reducing hypoglycemia in type 1 diabetes cagon, including where the glucagon
glycemia and cognitive decline (75). (83). For patients with type 1 diabetes product is kept and when and how to
Studies of rates of level 3 hypoglycemia with level 3 hypoglycemia and hypogly- administer it. An individual does not need
that rely on claims data for hospitaliza- cemia unawareness that persists despite to be a health care professional to safely
tion, emergency department visits, and medical treatment, human islet trans- administer glucagon. In addition to
ambulance use substantially underesti- plantation may be an option, but the traditional glucagon injection powder
mate rates of level 3 hypoglycemia (76) approach remains experimental (84,85). that requires reconstitution prior to
yet reveal a high burden of hypoglycemia In 2015, the ADA changed its prepran- injection, intranasal glucagon and glu-
in adults over 60 years of age in the dial glycemic target from 70–130 mg/dL cagon solution for subcutaneous injec-
community (77). African Americans are (3.9–7.2 mmol/L) to 80–130 mg/dL (4.4– tion are available. Care should be taken
at substantially increased risk of level 7.2 mmol/L). This change reflects the to ensure that glucagon products are
3 hypoglycemia (77,78). In addition to results of the ADAG study, which dem- not expired.
age and race, other important risk factors onstrated that higher glycemic targets
found in a community-based epidemio- corresponded to A1C goals (7). An ad- Hypoglycemia Prevention
logic cohort of older Black and White ditional goal of raising the lower range Hypoglycemia prevention is a critical
adults with type 2 diabetes include of the glycemic target was to limit component of diabetes management.care.diabetesjournals.org Glycemic Targets S81
SMBG and, for some patients, CGM are targetinghypoglycemia,moststudiesdem- A physician with expertise in diabetes
essential tools to assess therapy and onstrated a significant reduction in time management should treat the hospital-
detect incipient hypoglycemia. Patients spent between 54 and 70 mg/dL. A recent ized patient. For further information on
should understand situations that in- report in people with type 1 diabetes the management of diabetic ketoacidosis
crease their risk of hypoglycemia, such as over the age of 60 years revealed a small and the nonketotic hyperglycemic hyper-
when fasting for tests or procedures, when but statistically significant decrease in hy- osmolar state, please refer to the ADA
meals are delayed, during and after the poglycemia (110). No study to date has consensus report “Hyperglycemic Crises
consumption of alcohol, during and after reported a decrease in level 3 hypogly- in Adult Patients With Diabetes” (119).
intense exercise, and during sleep. Hypo- cemia. In a single study using intermit-
glycemia may increase the risk of harm to tently scanned CGM, adults with type 1 References
self or others, such as with driving. Teach- diabetes with A1C near goal and impaired 1. Laiteerapong N, Ham SA, Gao Y, et al. The
ing people with diabetes to balance insulin awareness of hypoglycemia demonstrated legacy effect in type 2 diabetes: impact of early
use and carbohydrate intake and exercise no change in A1C and decreased level glycemic control on future complications (The
are necessary, but these strategies are not 2 hypoglycemia (100). For people with Diabetes & Aging Study). Diabetes Care 2019;42:
416–426
always sufficient for prevention. type 2 diabetes, studies examining the 2. Stratton IM, Adler AI, Neil HAW, et al. Asso-
In type 1 diabetes and severely insulin- impact of CGM on hypoglycemic events ciation of glycaemia with macrovascular and
deficient type 2 diabetes, hypoglycemia are limited; a recent meta-analysis does microvascular complications of type 2 diabetes
unawareness (or hypoglycemia-associ- not reflect a significant impact on hypo- (UKPDS 35): prospective observational study.
ated autonomic failure) can severely glycemic events in type 2 diabetes (111), BMJ 2000;321:405–412
3. Little RR, Rohlfing CL, Sacks DB; National
compromise stringent diabetes control whereas improvements in A1C were Glycohemoglobin Standardization Program (NGSP)
and quality of life. This syndrome is observed in most studies (111–117). Over- Steering Committee. Status of hemoglobin A1c
characterized by deficient counterregu- all, real-time CGM appears to be a useful measurement and goals for improvement: from
latory hormone release, especially in tool for decreasing time spent in a hypo- chaos to order for improving diabetes care. Clin
Chem 2011;57:205–214
older adults, and a diminished autonomic glycemic range in people with impaired
4. Jovanovič L, Savas H, Mehta M, Trujillo A,
response, which are both risk factors for, awareness. For type 2 diabetes, other Pettitt DJ. Frequent monitoring of A1C during
and caused by, hypoglycemia. A corollary strategies to assist patients with insulin pregnancy as a treatment tool to guide therapy.
to this “vicious cycle” is that several dosing can improve A1C with minimal Diabetes Care 2011;34:53–54
weeks of avoidance of hypoglycemia has hypoglycemia (118). 5. Beck RW, Connor CG, Mullen DM, Wesley
DM, Bergenstal RM. The fallacy of average: how
been demonstrated to improve counter-
using HbA1c alone to assess glycemic control can
regulation and hypoglycemia awareness be misleading. Diabetes Care 2017;40:994–
in many patients (90). Hence, patients INTERCURRENT ILLNESS 999
with one or more episodes of clinically For further information on management 6. Nathan DM, Kuenen J, Borg R, Zheng H,
significant hypoglycemia may benefit of patients with hyperglycemia in the Schoenfeld D, Heine RJ; A1c-Derived Average
Glucose Study Group. Translating the A1C assay
from at least short-term relaxation of hospital, see Section 15 “Diabetes Care into estimated average glucose values. Diabe-
glycemic targets and availability of glu- in the Hospital” (https://doi.org/10.2337/ tes Care 2008;31:1473–1478
cagon (91). dc21-S015). 7. Wei N, Zheng H, Nathan DM. Empirically
Stressful events (e.g., illness, trauma, establishing blood glucose targets to achieve
Use of CGM Technology in Hypoglycemia surgery, etc.) may worsen glycemic con- HbA1c goals. Diabetes Care 2014;37:1048–1051
8. Selvin E. Are there clinical implications of
Prevention trol and precipitate diabetic ketoacidosis racial differences in HbA1c? A difference, to
With the advent of CGM and CGM- or nonketotic hyperglycemic hyperosmolar be a difference, must make a difference. Di-
assisted pump therapy, there has been state, life-threatening conditions that re- abetes Care 2016;39:1462–1467
a promise of alarm-based prevention of quire immediate medical care to prevent 9. Bergenstal RM, Gal RL, Connor CG, et al.; T1D
hypoglycemia (92,93). To date, there Exchange Racial Differences Study Group. Racial
complications and death. Any condition
differences in the relationship of glucose con-
have been a number of randomized leading to deterioration in glycemic control centrations and hemoglobin A1C levels. Ann
controlled trials in adults with type 1 necessitates more frequent monitoring Intern Med 2017;167:95–102
diabetes and studies in adults and chil- of blood glucose; ketosis-prone patients 10. Lacy ME, Wellenius GA, Sumner AE, et al.
dren with type 1 diabetes using real-time also require urine or blood ketone moni- Association of sickle cell trait with hemoglobin
A1C in African Americans. JAMA 2017;317:
CGM (see Section 7 “Diabetes Technol- toring. If accompanied by ketosis, vomiting,
507–515
ogy,” https://doi.org/10.2337/dc21-S007). or alteration in the level of consciousness, 11. Rohlfing C, Hanson S, Little RR. Measure-
These studies had differing A1C at entry marked hyperglycemia requires tempo- ment of hemoglobin A1c in patients with sickle
and differing primary end points and thus rary adjustment of the treatment regi- cell trait. JAMA 2017;317:2237–2238
must be interpreted carefully. Real-time men and immediate interaction with the 12. Wheeler E, Leong A, Liu C-T, et al.; EPIC-CVD
Consortium; EPIC-InterAct Consortium; Lifelines
CGM studies can be divided into studies diabetes care team. The patient treated Cohort Study. Impact of common genetic deter-
with elevated A1C with the primary end with noninsulin therapies or medical nu- minants of hemoglobin A1c on type 2 diabetes
point of A1C reduction and studies with trition therapy alone may require insulin. risk and diagnosis in ancestrally diverse popu-
A1C near target with the primary end point Adequate fluid and caloric intake must be lations: a transethnic genome-wide meta-analysis.
of reduction in hypoglycemia (93–109). In ensured. Infection or dehydration is more PLoS Med 2017;14:e1002383
13. Wilson DM, Kollman; Diabetes Research in
people with type 1 and type 2 diabetes with likely to necessitate hospitalization of Children Network (DirecNet) Study Group. Re-
A1C above target, CGM improved A1C individuals with diabetes versus those lationship of A1C to glucose concentrations in
between 0.3% and 0.6%. For studies without diabetes. children with type 1 diabetes: assessments byS82 Glycemic Targets Diabetes Care Volume 44, Supplement 1, January 2021
high-frequency glucose determinations by sen- PA, et al. Effect of intensive diabetes therapy on disease in patients with type 1 diabetes. N Engl J
sors. Diabetes Care 2008;31:381–385 the progression of diabetic retinopathy in pa- Med 2005;353:2643–2653
14. Buse JB, Kaufman FR, Linder B, Hirst K, El tients with type 1 diabetes: 18 years of follow-up 42. Nathan DM, Zinman B, Cleary PA, et al.;
Ghormli L, Willi S; HEALTHY Study Group. in the DCCT/EDIC. Diabetes 2015;64:631–642 Diabetes Control and Complications Trial/
Diabetes screening with hemoglobin A1c ver- 29. Lachin JM, Genuth S, Cleary P, Davis MD, Epidemiology of Diabetes Interventions and
sus fasting plasma glucose in a multiethnic Nathan DM; Diabetes Control and Complications Complications (DCCT/EDIC) Research Group. Modern-
middle-school cohort. Diabetes Care 2013;36: Trial/Epidemiology of Diabetes Interventions day clinical course of type 1 diabetes mellitus
429–435 and Complications Research Group. Retinopathy after 30 years’ duration: the Diabetes Control and
15. Kamps JL, Hempe JM, Chalew SA. Racial and nephropathy in patients with type 1 diabetes Complications Trial/Epidemiology of Diabetes In-
disparity in A1C independent of mean blood four years after a trial of intensive therapy. N Engl terventions and Complications and Pittsburgh Ep-
glucose in children with type 1 diabetes. Diabetes J Med 2000;342:381–389 idemiology of Diabetes Complications experience
Care 2010;33:1025–1027 30. Ohkubo Y, Kishikawa H, Araki E, et al. In- (1983-2005). Arch Intern Med 2009;169:1307–
16. Advani A. Positioning time in range in diabetes tensive insulin therapy prevents the progression 1316
management. Diabetologia 2020;63:242–252 of diabetic microvascular complications in Jap- 43. Orchard TJ, Nathan DM, Zinman B, et al.;
17. Avari P, Uduku C, George D, Herrero P, Reddy anese patients with non-insulin-dependent di- Writing Group for the DCCT/EDIC Research
M, Oliver N. Differences for percentage times in abetes mellitus: a randomized prospective 6-year Group. Association between 7 years of intensive
glycemic range between continuous glucose study. Diabetes Res Clin Pract 1995;28:103–117 treatment of type 1 diabetes and long-term
monitoring and capillary blood glucose moni- 31. UK Prospective Diabetes Study (UKPDS) mortality. JAMA 2015;313:45–53
toring in adults with type 1 diabetes: analysis of Group. Effect of intensive blood-glucose control 44. Di Angelantonio E, Kaptoge S, Wormser D,
the REPLACE-BG dataset. Diabetes Technol with metformin on complications in overweight et al.; Emerging Risk Factors Collaboration. As-
Ther 2020;22:222–227 patients with type 2 diabetes (UKPDS 34). Lancet sociation of cardiometabolic multimorbidity with
18. Vigersky RA, McMahon C. The relationship of 1998;352:854–865 mortality. JAMA 2015;314:52–60
hemoglobin A1C to time-in-range in patients 32. UK Prospective Diabetes Study (UKPDS) 45. Yeung RO, Zhang Y, Luk A, et al. Metabolic
with diabetes. Diabetes Technol Ther 2019;21: Group. Intensive blood-glucose control with profiles and treatment gaps in young-onset
81–85 sulphonylureas or insulin compared with con- type 2 diabetes in Asia (the JADE programme):
19. Kröger J, Reichel A, Siegmund T, Ziegler R. ventional treatment and risk of complications a cross-sectional study of a prospective cohort.
Clinical recommendations for the use of the in patients with type 2 diabetes (UKPDS 33). Lancet Diabetes Endocrinol 2014;2:935–943
ambulatory glucose profile in diabetes care. J Lancet 1998;352:837–853 46. Sattar N, Rawshani A, Franzén S, et al. Age at
Diabetes Sci Technol 2020;14:586–594 33. Holman RR, Paul SK, Bethel MA, Matthews diagnosis of type 2 diabetes mellitus and asso-
20. Livingstone R, Boyle JG, Petrie JR. How tightly DR, Neil HAW. 10-year follow-up of intensive ciations with cardiovascular and mortality risks.
controlled do fluctuations in blood glucose levels glucose control in type 2 diabetes. N Engl J Med Circulation 2019;139:2228–2237
need to be to reduce the risk of developing 2008;359:1577–1589 47. Zabala A, Darsalia V, Holzmann MJ, et al. Risk
complications in people with type 1 diabetes? 34. Lind M, Pivodic A, Svensson A-M, Ólafsdóttir of first stroke in people with type 2 diabetes and
Diabet Med 2020;37:513–521 AF, Wedel H, Ludvigsson J. HbA1c level as a risk its relation to glycaemic control: a nationwide
21. Messer LH, Berget C, Vigers T, et al. Real factor for retinopathy and nephropathy in chil- observational study. Diabetes Obes Metab 2020;
world hybrid closed-loop discontinuation: pre- dren and adults with type 1 diabetes: Swedish 22:182–190
dictors and perceptions of youth discontinuing population based cohort study. BMJ 2019;366: 48. Zoungas S, Woodward M, Li Q, et al.; AD-
the 670G system in the first 6 months. Pediatr l4894 VANCE Collaborative group. Impact of age, age at
Diabetes 2020;21:319–327 35. Adler AI, Stratton IM, Neil HAW, et al. diagnosis and duration of diabetes on the risk of
22. Mayeda L, Katz R, Ahmad I, et al. Glucose Association of systolic blood pressure with mac- macrovascular and microvascular complications
time in range and peripheral neuropathy in type 2 rovascular and microvascular complications of and death in type 2 diabetes. Diabetologia 2014;
diabetes mellitus and chronic kidney disease. type 2 diabetes (UKPDS 36): prospective obser- 57:2465–2474
BMJ Open Diabetes Res Care 2020;8:e000991 vational study. BMJ 2000;321:412–419 49. Gabbay RA, Beebe C, Cuddeback J, et al.
23. Yoo JH, Choi MS, Ahn J, et al. Association 36. Duckworth W, Abraira C, Moritz T, et al.; Addressing therapeutic inertia in 2020 and be-
between continuous glucose monitoring-derived VADT Investigators. Glucose control and vascular yond: a 3-year initiative of the American Diabetes
time in range, other core metrics, and albumin- complications in veterans with type 2 diabetes. N Association. Clinical Diabetes 2020;38:371–381
uria in type 2 diabetes. Diabetes Technol Ther Engl J Med 2009;360:129–139 50. Skyler JS, Bergenstal R, Bonow RO, et al.;
2020;22:768–776 37. Patel A, MacMahon S, Chalmers J, et al.; American Diabetes Association; American Col-
24. Lu J, Ma X, Shen Y, et al. Time in range is ADVANCE Collaborative Group. Intensive blood lege of Cardiology Foundation; American Heart
associated with carotid intima-media thickness glucose control and vascular outcomes in pa- Association. Intensive glycemic control and the
in type 2 diabetes. Diabetes Technol Ther 2020; tients with type 2 diabetes. N Engl J Med 2008; prevention of cardiovascular events: implica-
22:72–78 358:2560–2572 tions of the ACCORD, ADVANCE, and VA diabetes
25. Nathan DM, Genuth S, Lachin J, et al.; Di- 38. Ismail-Beigi F, Craven T, Banerji MA, et al.; trials: a position statement of the American
abetes Control and Complications Trial Research ACCORD trial group. Effect of intensive treatment Diabetes Association and a scientific statement
Group. The effect of intensive treatment of of hyperglycaemia on microvascular outcomes in of the American College of Cardiology Founda-
diabetes on the development and progression type 2 diabetes: an analysis of the ACCORD tion and the American Heart Association. Di-
of long-term complications in insulin-dependent randomised trial. Lancet 2010;376:419–430 abetes Care 2009;32:187–192
diabetes mellitus. N Engl J Med 1993;329:977– 39. Gerstein HC, Miller ME, Byington RP, et al.; 51. Zoungas S, Chalmers J, Neal B, et al.;
986 Action to Control Cardiovascular Risk in Diabetes ADVANCE-ON Collaborative Group. Follow-up of
26. Battelino T, Danne T, Bergenstal RM, et al. Study Group. Effects of intensive glucose low- blood-pressure lowering and glucose control
Clinical targets for continuous glucose monitor- ering in type 2 diabetes. N Engl J Med 2008;358: in type 2 diabetes. N Engl J Med 2014;371:
ing data interpretation: recommendations from 2545–2559 1392–1406
the international consensus on time in range. 40. Buse JB, Bain SC, Mann JFE, et al.; LEADER 52. Hayward RA, Reaven PD, Wiitala WL, et al.;
Diabetes Care 2019;42:1593–1603 Trial Investigators. Cardiovascular risk reduction VADT Investigators. Follow-up of glycemic con-
27. Beck RW, Bergenstal RM, Cheng P, et al. The with liraglutide: an exploratory mediation anal- trol and cardiovascular outcomes in type 2 di-
relationships between time in range, hypergly- ysis of the LEADER trial. Diabetes Care 2020;43: abetes. N Engl J Med 2015;372:2197–2206
cemia metrics, and HbA1c. J Diabetes Sci Technol 1546–1552 53. CONTROL Group; Turnbull FM, Abraira C,
2019;13:614–626 41. Nathan DM, Cleary PA, Backlund J-YC, et al.; Anderson RJ, et al. Intensive glucose control and
28. Diabetes Control and Complications Trial Diabetes Control and Complications Trial/ macrovascular outcomes in type 2 diabetes.
(DCCT)/Epidemiology of Diabetes Interventions Epidemiology of Diabetes Interventions and Diabetologia 2009;52:2288–2298
and Complications (EDIC) Research Group; La- Complications (DCCT/EDIC) Study Research Group. 54. Duckworth WC, Abraira C, Moritz TE, et al.;
chin JM, White NH, Hainsworth DP, Sun W, Cleary Intensive diabetes treatment and cardiovascular Investigators of the VADT. The duration ofYou can also read
