Projects in Knowledge How to Create a Powerpoint Webcast

Report
Meeting Goals of Therapy in Type 2
Diabetes Mellitus
Zachary Bloomgarden, MD
Clinical Professor
Mount Sinai School of Medicine
New York, New York
Questions for Discussion

Is there evidence of adverse effects of
diabetes associated with glycemia?

How durable is glycemic control?

What is the relationship of HbA1c to
glycemia?

What are the benefits and risks of
treatment?

What are the benefits and risks of intensive
glycemic control?

Should treatment goals be standardized or
individualized?
Is there evidence of adverse effects
of diabetes associated with
glycemia?
Age-Adjusted Mortality Rate
per 10,000 Person/Years
Mortality in Multiple Risk Factor
Intervention Trial (MRFIT)
RR = 2.5
RR = 3.0
RR = 3.2
RR = 2.8
RR = 2.3
Relative risk adjusted for age, race, income, systolic BP, and smoking.
Abbreviations: CHD, coronary heart disease; CVD, cardiovascular disease; RR, relative risk.
Stamler J, et al. Diabetes Care. 1993;16:434-444.
Legacy Effect of Earlier Glucose Control
After median 8.5
years posttrial
follow-up
Aggregate Endpoint
19971
20072
Any diabetes-related endpoint
RRR:
P:
12%
.029
9%
.040
Microvascular disease
RRR:
P:
25%
.0099
24%
.001
Myocardial infarction
RRR:
P:
16%
.052
15%
.01
All-cause mortality
RRR:
P:
6%
.44
13%
.007
Abbreviation: RRR, relative risk reduction.
1. Turner R, et al, for the UKPDS Group. Lancet. 1998;352:837-853. 2. Holman RR, et al. N Engl J Med.
2008;359:1577-1589.
How durable is glycemic control?
Loss of Glycemic Control
in the UKPDS

Overweight cohort from UKPDS
– Newly diagnosed T2DM treated with conventional
dietary modifications, a sulfonylurea, insulin, or
metformin

Treatment produced initial decreases in HbA1c, and
over 10 years of follow-up, HbA1c remained lower with
treatment vs conventional therapy, but…
– HbA1c rose steadily over time in all arms
– After about 4–6 years, median HbA1c had
exceeded 7.0% in all treatment arms
Turner RC, et al, for UKPDS Study Group. Lancet. 1998;352:854-865.
ADOPT
Durability of Glycemia
Therapy
Annual 95% Confidence
Increase
Interval
in HbA1c
Years to Reach
Baseline Level
Glyburide
0.24%
0.23–0.26
~3.5
Metformin
0.14%
0.13–0.16
~5
Rosiglitazone
0.07%
0.06–0.09
>5
Kahn SE, et al. N Engl J Med. 2006;355:2427-2443.
RECORD Trial
Mean HbA1c (%) Over Time
Background Metformin
Background Sulfonylurea
HbA1c (%)
HbA1c (%)
8.0
Sulfonylurea (n = 1084)
P <.0001
7.6
7.2
Rosiglitazone (n = 1106)
6.8
8.0
Metformin (n = 1096)
P <.0001
7.6
Rosiglitazone (n = 1083)
7.2
6.8
0
1
2
3
Time (years)
4
5
0
1
2
3
4
Time (years)
Model-Adjusted Mean (%, SE).
With permission from Beck-Nielsen H. Presented at: 69th ADA; June 5-9, 2009.
5
UKPDS Posttrial Changes in HbA1c
Glycated Hemoglobin (%)
10 P = .008 P = .14 P = .82 P = .84 P = .99 P = .71
9
Conventional therapy
8
Sulfonylurea-insulin
7
0
1997
1998
1999
2000
2001
2002
Holman RR, et al. N Engl J Med. 2008;359:1577-1589. Copyright © 2008 Massachusetts Medical Society.
All rights reserved.
What is the relationship of
HbA1c to glycemia?
Factors Influencing HbA1c
GLYCATION
3
4
2
HEMOGLOBIN
1
ERYTHROCYTE
DESTRUCTION
ERYTHROPOIESIS
5
ASSAY
With permission from Gallagher EJ, et al. J Diabetes. 2009;1:9-17.
Factors Influencing HbA1c
1.
Erythropoiesis
–
–
2.
Altered hemoglobin
–
3.
Reductions in iron, vitamin B12, and erythropoiesis
increase HbA1c
Use of erythropoietin, iron, B12; reticulocytosis; and
chronic liver disease decrease HbA1C
Hemoglobinopathy, fetal hemoglobin, and
methemoglobin have variable effects on HbA1c
Glycation
–
–
–
Alcoholism, chronic renal failure, decreased intraerythrocyte pH increase HbA1c
Use of aspirin, vitamin C and E; certain
hemoglobinopathies, and increased intra-erythrocyte pH
decrease HbA1c
Genetic determinants have variable effects on HbA1c
Gallagher EJ, et al. J Diabetes. 2009;1:9-17.
Factors Influencing HbA1c
4.
Erythrocyte destruction
–
–
5.
Longer erythrocyte life span (splenectomy) increases
HbA1c
Shorter erythrocyte life span (hemoglobinopathies;
splenomegaly; rheumatoid arthritis; use of
antiretrovirals, ribavirin, dapsone) decreases HbA1c
Assays
–
–
–
Hyperbilirubinemia, carbamylated hemoglobin,
alcoholism, large doses of aspirin, chronic opiate use
increase HbA1c
Hypertriglyceridemia decreases HbA1c
Hemoglobinopathies have variable effects on HbA1c
Gallagher EJ, et al. J Diabetes. 2009;1:9-17.
Intrapatient Variability of HbA1c

Comparison of HbA1c and mean blood glucose
(MBG) in 128 children with type 1 diabetes

Dilemma: For any given MBG, HbA1c varied widely
– 95% CI 7%–15% at MBG 200 mg/dL

Linear correlation found

In any given individual, the linear correlation is
much stronger than in the group as a whole

Concept: There may be high and low Hb glycation
phenotypes
Hempe JM, et al. J Diabetes Complications. 2002;16:313-320.
Effect of Iron Deficiency Anemia on
HbA1c in Nondiabetics
Abbreviation: IDA, iron deficiency anemia.
Coban E, et al. Acta Haematol. 2004;112:126–128.
Ferritin (ng/dL)
HbA1c (%)
HbA1c (%)
Hemoglobin (g/dL)
P <.001
No significant differences in
fasting or postprandial glucose
between groups.
Effects of Age on HbA1c
Framingham Offspring Study and NHANES
HbA1c (%)
“…the uniform
results between
FOS and
NHANES
establish clearly
that A1C
increases with age
even after
multivariate
adjustments for
sex, fasting, and
2-h postload
glucose.”
Age (years)
Pani LN, et al. Diabetes Care. 2008;31:1991-1996.
Effects of Ethnicity on HbA1c
Diabetes Prevention Program
Adjusted Mean HbA1c (%)
All groups P <.0001 vs whites
Data adjusted for: age, gender, systolic blood pressure, diastolic blood pressure, BMI, fasting glucose,
glucose AUC, corrected insulin response, and insulin resistance
Herman WH, et al. Diabetes Care. 2007;30:2453-2457.
What are the benefits and risks
of treatment?
Effect of Lifestyle Interventions on
20-Year Mortality in IGT

577 Chinese adults with impaired glucose tolerance randomized
in 1986 to 6 years of lifestyle interventions or control

20-year mortality rate
– Intervention: 102/438 (23%)
– Control: 40/138 (29%)

Cumulative cardiovascular disease (CVD) mortality and
all-cause mortality were lowered by lifestyle changes, but
not significantly so
– CVD death: 11.6% vs 17.4% (HR 0.73; CI 0.42–1.26)
– All-cause: 21.7% vs 26.6% (HR 0.90; CI 0.59–1.37)

Conclusion: Lifestyle changes may help, but we have insufficient
evidence to be certain
Li G, et al. Lancet. 2008;371:1783-1789.
Benefits of T2DM Medications
AACE/ACE Algorithm
PPG
lowering
+
++
+++
++
++
+
+
++
+++
+++
FPG
lowering
++
+
+
++
+
++
+
0
+++
+
NAFLD
+
0
+
0
0
++
0
0
0
0
Abbreviations: FPG, fasting plasma glucose; NAFLD, nonalcoholic fatty liver disease; PPG, postprandial
glucose.
Rodbard HW, et al. Endocrine Pract. 2009;15:541-559.
Risks of T2DM Medications
AACE/ACE Algorithm
Hypoglycemia
0
0
0
++
+
0
0
0
+++
0
GI symptoms
++
0
++
0
0
0
++
++
0
++
Risks in renal
insufficiency
+++
 dose
++
++
0
+
0
0
++
?
Lactic
acidosis/risk in
liver failure
+++
0
0
++
++
++
0
0
0
0
Heart
failure/edema
Caution
0
0
0
0
++*
0
0
0†
0
Weight gain
–
0
–
+
+
++
0
0
++
–
Fractures
0
0
0
0
0
++
0
0
0
0
Drug-drug
interactions
0
0
0
++
++
0
0
0
0
0
*TZD contraindicated in class 3 or 4 CHF. †Unless used with TZD.
Rodbard HW, et al. Endocrine Pract. 2009;15:541-559.
Risk of Hypoglycemia
Bodmer M, et al. Diabetes Care. 2008;31:2086-2091.
Pooled Risk of Hypoglycemia
Drug Comparison, Hypoglycemia Risk
Weighted Absolute Risk
Difference (95% CI)
Metformin = metformin + thiazolidinedione
0.00 (-0.01–0.01)
Sulfonylurea > repaglinide
0.02 (-0.02–0.05)
Glyburide > other sulfonylurea
0.03 (0.00–0.05)
Sulfonylurea > metformin
0.04 (0.00–0.09)
Sulfonylurea + thiazolidinedione > sulfonylurea
0.08 (0.00–0.16)
Sulfonylurea > thiazolidinedione
0.09 (0.03–0.15)
Sulfonylurea + metformin > sulfonylurea
0.11 (0.07–0.14)
Sulfonylurea + metformin > metformin
0.14 0.07–0.21)
Bolen S, et al. Ann Intern Med. 2007;147:386-399.
Pooled Risk of Weight Gain
Drug Comparison, Risk of Weight Gain
Weighted Absolute Risk
Difference (95% CI)
Sulfonylurea ≈ repaglinide
0.03 (-1.0–1.0)
Sulfonylurea ≈ metformin + sulfonylurea
0.05 (-0.05–0.6)
Thiazolidinedione > sulfonylurea
1.1 (-0.9–3.1)
Sulfonylurea > acarbose
1.9 (0.2–4.0)
Thiazolidinedione > metformin
1.9 (0.5–3.3)
Sulfonylurea > metformin
Metformin + sulfonylurea > metformin
Bolen S, et al. Ann Intern Med. 2007;147:386-399.
1.9 (1.4–2.4) for trials <24 weeks
3.5 (3.0–4.0) for trials ≥24 weeks
2.4 (1.1–3.6)
What are the benefits and risks of
intensive glycemic control?
Intensive Glucose Control
DCCT1,2
UKPDS3,4
1441
4209
HbA1c with intensive vs
standard treatment
7.4% vs 9.1%
7.0% vs 7.9%
Effect on microvascular
risk
Significant risk reduction
Significant risk reduction
Effect on cardiovascular
risk
Benefit seen in follow-up
Benefit seen in follow-up
Number of patients
1. DCCT Research Group. N Engl J Med. 1993;329:977-986. 2. DCCT/EDIC Study Research Group. N
Engl J Med. 2005;353:2643-2653. 3. UKPDS Group. Lancet. 352:837-853. 4. Holman RR, et al. N Engl J
Med. 2008;359:1577-1589.
Intensive Glucose Control
ACCORD1
ADVANCE3
VADT4
10,251
11,140
1791
6.4% vs 7.5%
6.4% vs 7.0%
6.9% vs 8.4%
Effect on
microvascular risk
Some reductions in
albuminuria, eye
complications,
neuropathy2
Reduced risk of
nephropathy
Minimal benefit
(albumin
excretion)
Effect on
cardiovascular risk
Decreased risk of nonfatal
MI; increased risk of death
from CVD
No benefit
No benefit
Effect on mortality
Increased*
No difference
No difference
More hypoglycemia,
weight gain, fluid retention
More
hypoglycemia,
weight gain
More
hypoglycemia,
weight gain
Number of patients
HbA1c with intensive
vs standard treatment
Adverse effects
*Intensive control arm discontinued February 2008 due to increased mortality risk.
1. ACCORD Study Group. N Engl J Med. 2008;358:2545-2559. 2. Ismail-Beigi F, et al. Lancet.
2010;376:419-430. 3. ADVANCE Collaborative Group. N Engl J Med. 2008;358:2560-2572. 4. Duckworth
W, et al. N Engl J Med. 2008;360:129-139.
HbA1c Concentration (%) at Follow-up
Meta-analysis of RCTs of Intensive
Glycemic Control
n
4620
5238
11,140
1791
10,251
33,040
Years
10.1
2.9
5.0
5.6
3.5
4.95
46,237
15,059
55,700
10,030
35,879
162,905
Patient-Years
Abbreviation: RCTs, randomized controlled trials.
Ray KK, et al. Lancet. 2009;373:1765–1772.
Meta-analysis of RCTs of Intensive
Glycemic Control
Risk of Nonfatal MI
Odds Ratio
95% CI
UKPDS
0.78
0.62–0.98
PROactive
0.83
0.64–1.06
ADVANCE
0.98
0.78–1.23
VADT
0.81
0.58–1.15
ACCORD
0.78
0.64–0.95
Overall
0.83
0.75–0.93
OR <1 favors intensive treatment
OR >1 favor standard treatment
Ray KK, et al. Lancet. 2009;373:1765-1772.
Meta-analysis of RCTs of Intensive
Glycemic Control
Risk of Mortality (All-Cause)
Odds Ratio
95% CI
UKPDS
0.79
0.53–1.20
PROactive
0.96
0.77–1.19
ADVANCE
0.93
0.82–1.05
VADT
1.09
0.81–1.47
ACCORD
1.28
1.06–1.54
Overall
1.02
0.87–1.19
OR <1 favors intensive treatment
OR >1 favor standard treatment
Ray KK, et al. Lancet. 2009;373:1765-1772.
Mortality Associated with
Severe Hypoglycemia
ACCORD
Bonds DE, et al. BMJ. 2010;340:b4909.
Cardiovascular Deaths in ACCORD
Incidence of Mortality (%)
Sudden death accounted for nearly two thirds of cardiovascular
deaths! 86/135 with intensive therapy and 67/94 with standard therapy
ACCORD. N Engl J Med. 2008;358:2545-2559.
Should treatment goals be
standardized or individualized?
Glycemic Management
AACE Guideline

Encourage all patients with T2DM to achieve
glycemic levels as near normal as possible without
inducing clinically significant hypoglycemia
(grade A evidence)

Glycemic targets include
– HbA1c ≤6.5% (grade B)
– FPG <110 mg/dL (grade B)
– 2-hour postprandial glucose <140 mg/dL
(grade B evidence)
AACE Diabetes Mellitus Clinical Practice Guidelines Task Force. Endocrine Pract.
2007;13(suppl 1):3-68.
Cardiovascular Outcome Subgroup
Analyses in ACCORD

Heterogeneity in prespecified subgroups

Intensive therapy was associated with fewer
cardiovascular events compared with
standard therapy in
– Patients with no history of cardiovascular events
before randomization (P = .04)
– Patients with baseline HbA1c ≤8.0% (P = .03)
ACCORD. N Engl J Med. 2008;358:2545-2559.
Coronary Calcium Subgroup
Analysis in VADT
Coronary Artery
Calcium Score
Hazard Ratio for
Intensive vs
Standard
95% CI
P-Value
0
0.07
0.01–0.55
.01
10
0.16
0.04–0.61
<.01
100
0.34
0.16–0.73
<.01
1000
0.75
0.47–1.19
.22
Low (≤100)
0.08
0.008–0.77
.03
High (>100)
0.74
0.46–1.18
0.21
HR <1 favors intensive therapy
Only those with low coronary artery calcium scores (<100)
benefited from intensive therapy.
Reaven PD, et al. Diabetes. 2009;58:2642-2648.
Mortality Associated with Baseline
HbA1c in ACCORD
•
Mortality in the intensive care arm correlated with baseline
HbA1c
−
•
No such correlation seen in the standard treatment arm
−
•
Mortality was lower with low HbA1c but rose with HbA1c
values >8.2%
Optimal baseline HbA1c was about 9.0%
Other factors associated with higher mortality in the intensive
treatment arm
−
History of neuropathy
−
History of aspirin use
−
≥35 kg/m2 (trend)
Calles-Escandon J, et al. Diabetes Care 2010;33:721–727.
Impact of Treatment Intensification
by Switch from Oral Monotherapy
to Insulin
Reference
HR for All-Cause Mortality
n = 20,005
Mean On-Treatment HbA1c (%)
With permission from Currie CJ, et al. Lancet. 2010;375:481-489.
Mortality Associated with
On-Trial HbA1c in ACCORD

Mortality in standard therapy arm of ACCORD was
increased with either HbA1c >8% or HbA1c <7%
(nonlinear U-shaped curve)
– Lowest risk was HbA1c between 7% and 8%

Mortality in intensive therapy arm steadily increased
with higher average HbA1c of 6% to 9%
– Those with lesser 1-year decrease in HbA1c also
had higher mortality risk
Riddle MC, et al. Diabetes Care. 2010;33:983-990.
Individualizing Glycemic
Goal Setting
Favors Intensive Therapy
Favors Standard Therapy

No CVD, few
comorbidities

CVD, neuropathy,
other comorbidities

HbA1c ≤8%


Glycemic control more
readily attained
Glycemic control
difficult

History of
hypoglycemia

Older age/shorter life
expectancy

Younger age

Shorter duration of
diabetes

Lower hypoglycemia
Conclusions

Diabetes is associated with adverse outcome,
mediated in part by hyperglycemia

There is evidence that glycemic interventions provide
benefit, but glycemic control is often lost over time

There is heterogeneity in the relationship between
HbA1c and glycemia

Risks and benefits of intensive glycemic control vary
for different subgroups
– Must take into account difficulties of attaining
control, hypoglycemia risk, and prior CVD
Thank you for your participation.
To earn CME/CE credit, please complete the posttest and
evaluation. (Click link in the navigation bar above or to
the left of the slide presentation.)
Your feedback is appreciated and will help us continue to
provide you with clinically relevant educational activities
that meet your specific needs.

similar documents