Silvio E' Inzucchi MD

1
Management of Diabetes Heart
FailureGlucose Control Strategies

• Silvio E. Inzucchi MD
• Section of Endocrinology
• Yale University
• New Haven, CT

2
Diabetes Therapy and Heart Failure

• Epidemiological points

2. The link between DM and HF
3. Current landscape of antihyperglycemic
therapy
4. Metformin HF 5. Thiazolidinedione HF
6. Summary Recommendations
3
DM HF Epidemiology

• Prevalence of DM in HF clinical trials is 25
• (20-40 in community surveys)
• RR of incident HF in DM 2.5
• Increased hospitalization rates
• Increased mortality

Nichols GA, Diabetes Care 27879, 2004
Gustafsson I et al. J Am Coll Cardiol 43771,
2004 Nichols GA, et al. Diabetes Care 241614,
2001 Shindler DM et al. Am J Cardiol 1771017,
1996
4
Glycemic Control HF Incidence
Men n25,958
Women n22,900
10.2
9.2
plt0.001
plt0.01
8.0
7.2
6.6
6.0
CHF Rate /year per 1000
5.9
5.6
4.5
4.5
lt7
7-8
8-9
9-10
gt10
7-8
8-9
9-10
gt10
lt7
Hemoglobin A1c ()
Iribarren et al. Circulation 20011032668-2673
5
High Prevalence of HF in DiabetesMechanisms

• Associated Comorbidities
• CAD
• Hypertension
• Obesity

• Diabetic Cardiomyopathy
• Microvascular disease
• Oxidative stress
• Inflammatory mediators
• Endothelial dysfunction
• RAS activation
• Myocardial fibrosis
• Metabolic dysfunction

6
Insulin Resistance HF Uppsala Study

• 1187 men (126 DM) gt70 yrs old, no CHF at
  baseline
• Insulin sensitivity measured by insulin-glucose
  clamp, HOMA-IR, OGTT, proinsulin levels.
• Anthropometrics, HF risk factors
• Mean follow-up, 8.9 yrs.
• 8.7 (N104) developed HF
• Incidence 10.5 / 1000 person-years

Ingelsson E et al. JAMA 294334, 2005
7
Insulin Resistance HF Uppsala Study

• Unadjusted analysis predictors of incident HF
  clamp glucose disposal rates (GDR), 2-h OGTT
  glucose, proinsulin level, BMI, and waist
  circumference.
• Adjusted analysis risk of HF associated with 1
  SD increase in
• HR 95 C.I.
• - 2-h OGTT glucose 1.44 1.08-1.93
• - proinsulin level 1.29 1.02-1.64
• BMI 1.35 1.11-1.65
• Waist circumference 1.36 1.10-1.69
• Clamp GDR 0.66 0.51-0.86

Ingelsson E et al. JAMA 294334, 2005
8
CAD, HTN Inflammation ? Endothelial Function ?
Glucose Utilization ? FFAs
Insulin Resistance
Heart Failure
SNS Activation ? Muscle Mass ? Tissue
Perfusion Physical Inactivity
9
(No Transcript)
10
(No Transcript)
11
Insulin Sensitizers Mechanism of Action
60
0
Plt0.006
50
NS
-5
40
? Hepatic Glucose Output()
-10
? Peripheral Glucose Disposal ()
30
-15
20
Plt0.03
-20
10
13
Plt0.001
-25
0
Between groups p lt 0.04
Between groups p lt 0.03
Metformin Troglitazone
Inzucchi SE et al. N Engl J Med.1998338867
12
Potential CV Benefits of Metformin

• ? weight
• ? hyperinsulinemia
• ? TGs, ? LDL, ? FFAs
• ? vascular reactivity / endothelial function
• ? AMP-activated protein kinase (AMPK)
• ? PAI-1, ? platelet aggregation
• ? oxidative stress
• ? atherosclerosis in animal models
• ? CVD event rates

13
UKPDS Myocardial Infarction
Myocardial Infarction
Coronary Deaths
20
10
p0.02
p0.01
NS
8
39
Incidence per 1000 patient years
15
Reduction
50
Reduction
6
Incidence per 1000 patient years
10
4
5
2
0
0
Insulin
Metformin
Conventional
Conventional

Metformin


or
Diet
Diet
SUs
UKPDS 34, Lancet 352 854, 1998
14
Metformin Package Insert
WARNING
15
Potential CV Benefits of TZDs

• ? insulin resistance, ? hyperinsulinemia
• ? TGs, ? HDL, ? LDL particle size, oxidation
  rates, ? FFAs
• ? blood pressure
• ? vascular reactivity / endothelial function
• improved ventricular remodelling
• ? cardiac metabolism (? glucose utilization)
• ? PAI-1, ? platelet aggregation
• ? vascular smooth muscle cell proliferation
• ? neointimal proliferation post-vascular injury
• ? expression of adhesion molecules,
  metalloproteinases
• ? CRP, other inflammatory mediators,
  adipocytokines
• ? oxidative stress
• ? atherosclerosis in animal models
• ? carotid IMT, ? aortic pulse wave velocity

16
TZDs Package Insert
WARNINGS
17
Insulin Sensitizers HFA Therapeutic Conundrum!

• Both metformin TZDs are contraindicated in
  patients with advanced heart failure
• Both metformin TZDs have vascular and
  metabolic effects that may potentially benefit
  the failing heart.

18
Management of Diabetes Heart Failure Glucose
Control Strategies
Metformin
19
Contraindications to Metformin

• Absolutely contraindicated in conditions that
  predispose to tissue anoxia, increase circulating
  lactate levels, lower pH, or decrease metformin
  clearance
• Renal dysfunction (SCr gt1.5 men, gt1.4 women)
• Radiocontrast studies
• Age gt80 (unless GFR normal)
• Liver disease
• Hemodynamic impairment
• Dehydration
• Hypoxia
• Metabolic acidosis
• Heart Failure

20
The Phantom of Metformin-Associated Lactic
Acidosis (MALA)

• Concerns date from phenformin experience
  (1960-70s)
• Risk at most is 1 per 30,000 patient-years -
  typically occurs when prescribed inappropriately
  (i.e., CKD)
• Cochrane meta-analysis of 194 studies (36,893
  patient-years metformin) no increased risk of
  lactic acidosis
• Surveys show 20-30 of metformin-treated patient
  have an active contraindication (no MALA
  reported!)

Inzucchi, Diabetes Care 282585, 2005
Emslie-Smith et al., Diabet Med 18483, 2003
Salpeter et al. Arch Intern Med 1632594, 2003
Holstein et al. Diabet Med 16692, 1999 Sulkin
TV et al. Diabetes Care 20925, 1997.
21
Saskatchewan Health Study
12,272 New Users of Metformin/Sulfonylureas
2,793 (23) HF

• Exclusions
• 625 subjects with prevalent HF
• 335 subjects treated with insulin

1,883 Incident HF
773 (42) Sulfonylurea Monotherapy
208 (11) Metformin Monotherapy
852 (47) Combination Therapy
Eurich DT et al., Diabetes Care 2005
22
Saskatchewan Health Study Baseline
Characteristics
plt0.05
Eurich DT et al., Diabetes Care 2005
23
Saskatchewan Health Study Baseline Medications
Eurich DT et al., Diabetes Care 2005
24
Saskatchewan Health Study Crude Outcomes
Plt0.05
Pgt0.05
Plt0.01
Eurich DT et al., Diabetes Care 2005
25
(No Transcript)
26
Management of Diabetes Heart Failure Glucose
Control Strategies
Thiazolidinediones
27
TZDs Edema
TZD
TZD Rosiglitazone No insulin 1-2
4-5 Insulin
4-5 13-16 Pioglitazone No insulin
1-2 4-5 Insulin
7 15-16
28
TZDs Edema Proposed Mechanisms

• Increased plasma volume
• Increased distal nephron Na retention (PPAR-
  ? stimulation of amiloride sensitive ENaC in
  collecting duct)
• Vasodilatation
• Calcium blocker effect
• PPAR-? mediated increase in vascular permeability
  (?VEGF)

Guan et al. Nature Medicine 11861, 2005
29
TZDs Cardiac Performance
15
Glyburide
10
Troglitazone
5
0
-5
-10
-15
N154, all with normal LV systolic
function Randomized to treatment for 48 weeks
Ghazzi MN et al. Diabetes 1997 46 433
30
TZDs Heart Failure

• In clinical trials, HF in lt1 of patients.
• More likely in those
• - receiving insulin
• - receiving higher TZD doses
• - have gt1 HF risk factors

31
TZDs Edema ? HF
Patients with chronic systolic dysfunction, 50
NYHA III-IV
Tang WHW et al. JACC 2003411394-1398
32
AHA/ADA Consensus Statement
Thiazolidinedione Use, Fluid Retention, and
Congestive Heart Failure A Consensus
Statement From the American Heart Association
and American Diabetes Association Richard W
Nesto, MD David Bell, MD Robert O Bonow, MD
Vivian Fonseca, MD Scott M Grundy, MD, PhD
Edward S Horton, MD Martin Le Winter, MD Daniel
Porte, MD Clay F Semenkovich, MD Sidney Smith,
MD Lawrence H Young, MD Richard Kahn,
PhD Circulation 20031082941-2948 Diabetes
Care 200427256-263
www.americanheart.org www.diabetes.org
33
TZDs HF AHA-ADA Consensus Statement
NYHA Class I or II symptoms TZDs are not
contraindicated Start with low dose, increase
dose very gradually to optimize glycemic
control, observe for excessive weight gain,
edema, or CHF Patients with NYHA Class III or IV
symptoms TZDs not recommended at this time
34
Use of Metformin or TZDs in HF
13.5
1998-99
Masoudi FA et al. JAMA 2003
35
Use of Metformin or TZDs in HF
24.4
What are the outcomes associated with off-label
insulin sensitizer use in diabetic patients with
HF?
2000-01
Masoudi FA et al. JAMA 2003
36
The National Heart Care Project

• A CMS-sponsored program to improve the quality
  of heart failure care
• 2 cross-sectional databases of Medicare
  fee-for-service patients hospitalized with
  primary diagnosis of HF. Up to 800 discharges
  sampled from each state, DC Puerto Rico during
  2 sampling periods 3/98?4/99, 7/00?6/01
• Records abstracted by trained reviewers
  (validated).

ICD-9 Codes 402.01, 402.11, 402.91, 404.01,
404.91, 428
37
Study Objectives

What are the outcomes associated with the
prescription of an insulin sensitizing drug in
elderly diabetic patients with heart failure
on mortality?
readmisssion?
Masoudi FA et al. Circulation 2004
38
78,882 Medicare beneficiaries discharges with
principal diagnosis of heart failure
261 Met TZD
39
NHCP HF
Variables (78)

• Demographics (age, gender, race)
• Patient medical history, diabetes comorbidities,
  clinical
• presentation, laboratory tests, imaging,
  including LVEF
• Medications at discharge
• Physician characteristics (AMA Masterfile)
• - Board certification, specialty
• Hospital characteristics (AHA Surveys)
• - Teaching vs. non-teaching, ownership, cardiac
  facilities, location

Masoudi FA et al. Circulation 2004
40
NHCP HF
Baseline Characteristics
41
NHCP HF
Baseline Characteristics
Masoudi FA et al. Circulation 2004
42
NHCP HF
Results Crude Mortality Rates at 1 Yr
Plt.0001
Plt.0001
12,069
2,487
2,122
Masoudi FA et al. Circulation 2004
43
NHCP HF
Adjusted K-M Curve Mortality
Proportion Alive
Masoudi FA et al. Circulation 2004
44
NHCP HF
45
NHCP HF
46
NHCP HF
47
NHCP HF
48
NHCP HF
Readmission Rates
Masoudi FA et al. Circulation 2004
49
71,120 Medicare beneficiaries discharged with
diagnosis of AMI
50
NHCP AMI
Adjusted Mortality Readmissions
Inzucchi SE et al. Diabetes Care 2004
51
NHCP AMI
Adjusted Mortality Readmissions
Inzucchi SE et al. Diabetes Care 2004
52
PROactive Study Design

• Prospective, multi-center randomized,
  double-blind, placebo-controlled, parallel-group
  study
• Planned recruitment of 5000 patients with
  minimum 2.5 years of exposure to treatment
• Pioglitazone vs. Placebo added to existing
  anti-DM therapy
• Major inclusion criteria T2DM, age 35-75,
  previous macrovascular disease, HbA1c gt6.5
• Major Exclusions insulin alone, heart failure
  (NYHA II-IV)

MI, stroke, CABG, PCI, CAD, PAD
Dormandy JA, Charbonnel B, Ekland D, et al.
Lancet 3661279, 2005
53
PROactive Baseline Characteristics
Pioglitazone (2605) Placebo (2633)
Male 67 66 Caucasian 98 99 Age
62 yrs 62 yrs DM duration 8 yrs 8
yrs BMI (kg/m2) 30.7 31.0 HbA1c
() 7.8 7.8 LDL-C 112 112 HDL-C 43 4
3 TG 159 159 ACEI/ARB 69
70 b-Blockers 55 55 Statins 43 43
ASA 85 83
Dormandy JA, Charbonnel B, Ekland D, et al.
Lancet 3661279, 2005
54
PROactive Macrovascular Disease Criteria
Prior Stroke 984
Prior MI 2445
662
82
763
95
145
1505
1904
Other criteria 3649
Dormandy JA, Charbonnel B, Ekland D, et al.
Lancet 3661279, 2005
1043 with PAD
55
PROactive Principal Secondary Endpoint

• Time from randomization to first occurrence of
  any of the following events
• All-cause mortality
• Non-fatal MI (excluding silent MI)
• Stroke
• Major leg amputation (above the ankle)
• Acute coronary syndrome
• CABG or PCI
• Leg revascularization

Dormandy JA, Charbonnel B, Ekland D, et al.
Lancet 3661279, 2005
56
(No Transcript)
57
PROactive Adverse Events
Pioglitazone Placebo
Reported Heart Failure 281 (10.8) 198
(7.5) (non-adjudicated) Heart Failure
Hospitalization 149 (5.7) 108 (4.1) Heart
Failure Mortality 25 (0.96) 22 (0.84) Edema
(in absence of HF) 562 (21.6) 341(13.0) Cessa
tion due to Wt Gain 21 (0.8) 6 (0.2)
Dormandy JA, Charbonnel B, Ekland D, et al.
Lancet 3661279, 2005
58
SUMMARY

• Diabetes and heart failure are frequently
  coexisting conditions, especially in elderly
  patients.

• There are significant safety concerns with
  currently available insulin sensitizers,
  metformin and TZDs, rendering
  anti-hyperglycemic treatment options more
  limited in patients with heart failure.

• Yet, both these drugs may have unique benefits
  in DM patients, especially those with insulin
  resistance.

59
SUMMARY

• Several observational studies now demonstrate
  that metformin is probably safe - and potentially
  effective - in HF patients. Randomized studies
  are needed to confirm these findings. If used,
  should be only in the most stable patients
  without other contraindications (i.e. CKD).

• TZDs present a more complex issue - they
  modestly improve CVD outcomes in high-risk
  patients, yet appear to increase the HF
  diagnosis and hospitalization rates. Further
  study is warranted. If used, should be only in
  stable, compensated patients, at the lowest
  doses, probably never with insulin, and with
  very cautious monitoring of fluid status.