Transcript DIABETES AND THE ELDERLY

Diabetes Mellitus
James W. Mold, M.D., M.P.H.
OU-HSC
Department of Family and Preventive
Medicine
Objective
Attendees should be able to:
 Construct a rational management plan for
an older patient with type 2 DM, taking
into account life expectancy, risks for
adverse events, and patient goals,
preferences, and resources
Format
 Lecture (30 minutes)
 Small group case-discussions (30
minutes)
 Presentations of case discussions in
large group (30 minutes)
Type 2 Diabetes Mellitus
Insulin resistance plus beta cell failure
 Insulin resistance (metabolic syndrome)
usually precedes beta cell failure by 10-20
years
 Connection between the two is still unclear
Beta cell fatigue? Genetically linked?
Inflammation?
 DM diagnosed when beta cell function
insufficient to control blood glucose levels
First manifestation is postprandial hyperglycemia
Insulin Resistance/Metabolic
Syndrome





Insulin resistance
Hypertension
Lipid abnormalities
Endovascular inflammation
Hypercoagulability
Doubles the risk of:
Macrovascular events (MI, CVA)
Metabolic Syndrome
NCEP ATP III Criteria (3 or more criteria)
Criteria
Defining Level
Waist circumference
Men
Women
>40 inches
>35 inches
Triglycerides
>150 mg/dL
HDL Cholesterol
Men
Women
<40 mg/dL
<50 mg/dL
Blood Pressure
>130/>85 mm Hg
Fasting Glucose
>110 mg/dL
NCEP ATP III. JAMA. 2001;285:2486–2497.
Metabolic Syndrome: Prevalence
by ATP III Criteria — NHANES III
Population
Overall 22% for age 20 and older
Prevalence (%)
45
40
35
30
25
Men
Women
20
15
10
5
0
30-39
40-49
50-59
60-69

70
Age (yr)
Adapted from: Ford ES et al. JAMA. 2002;287:356–359.
Why Worry About Insulin
Resistance
 Twice the risk of MI, CVA, PAD
 Eight times the risk of development of
type 2 DM
 Substantially reduced life expectancy
(by approximately 8 years)*
*Franco et al. Arch Intern Med 2007;167:11451151
Beta Cell Failure
Hyperglycemia increases risk for:
 Microvascular disease (retinopathy,
nephropathy)
 Neuropathy
 Infection
Some small additional risk for:
 Macro-vascular disease/events
 Hypercoagulability
 Endothelial cell dysfunction
UKDPS: Benefits of Glycemic Control
vs BP Control With ACEIs or -Blockers
20
Heart Failure
Relative Risk
Reduction %
0
Stroke
Diabetic Death
+7
-9
-8
-12
-20
-21
-32
-40
-44
-60
MI
-56
UKPDS Group. BMJ. 1998;317:703–713. Lancet. 1998;352:837–853.
Glycemic Control
ACEI or BB
Early Detection/Screening
 USPSTF: Screen all adults with
hypertension or hyperlipidemia for type
2 diabetes mellitus.
Prevention
In at risk patients:
 The onset of beta cell failure (diabetes) can
be delayed or prevented with exercise, diet,
metformin, glitazones, acarbose, an ACEI
or an ARB.
 Exercise and diet are much more efficacious
than medications.
However,
 The cost effectiveness of delaying the onset
of diabetes has not yet been established.
American Geriatrics Society
 Offer individualized therapy that
considers





Life expectancy
Cognitive impairment
Patient preferences
Functional status
Social support
 Keep therapy as simple and
inexpensive as possible
Goals
Things you would want to make happen
for which it makes very little sense to
ask, “so that….?”
Examples:
A longer life, the ability to communicate
through writing, the ability to make
decisions for myself.
Goals of Health Care
Prevent premature death and disability (QALE)
 Increase life expectancy (LOL)
 Reduce the risk of disabling complications
(future QOL)
Improve or maintain current quality of life (QOL)
 Maximize ability to function in ways that
make life worth living
Clinical Decision-making
Strategies in individual cases should depend
upon:
1. Outcomes of importance to the patient
 Desire to continue to try to stay alive
 Ability to participate in valued life activities
2. Estimated impact of interventions on
those outcomes
3. Ability and willingness of the patient to
adhere to the interventions
DM and Length of Life
75% of Type 2 diabetics die of cardiovascular
events (MI, CVA)
 2-4 times more likely to have cardiovascular
events
 Risk of MI is as high for type 2 diabetics with no
prior MI as for non-diabetics with a prior MI
 When they have an MI, diabetics are
significantly more likely to die or to develop CHF
Lifestyle Modifications
Exercise***
 Aerobic: Substantial benefits for both LOL and
QOL
 Strengthening: Substantial benefits for future
QOL, ??LOL
 Balance: Reduced falls
Diet
 Weight reduction: Difficult.** Associated with
better QOL. Be more careful in the elderly.
 Mediterranean diet: Associated with reduced
macrovascular events. Be careful in the elderly.
Low-Dose Aspirin
 Reduced risk of MI; greater in men
 Reduced risk of CVA; greater in
women
USPSTF recommends low dose aspirin
for men 45 to 79 and women 55 to
79
DM and LOL
 ACE inhibitor (ramipril): 24% reduction in
overall mortality over 4.5 years (16% after
controlling for effects of BP reduction)
 Overall mortality reduced in patients with HTN
and LVH with ARBs. 42% reduced risk of
CVA even if little change in BP. ARBs reduce
cardiovascular events more than atenolol.
HOPE Study Investigators. Lancet 2000; 355 (9200): 253-259
LIFE Study Investigators. Lancet 2002; 359: 1004-1010
DM and LOL
Possible reasons for benefits of ACEIs:
 Anti-ischemic
 Stimulate endothelial nitric oxide
 Decrease myocardial O2 consumption
 Anti-atherogenic
 Reduce systemic vascular resistance and
BP
 ?Reduce cardiac remodeling
RRR and ARR
Base Risk X RRR = Absolute Risk Reduction
If base
ARR
If base
ARR
risk = 10%, and RRR = 50%, then
= 5%
risk = 50%, and RRR = 50%, then
= 25%
Therefore, the actual benefit of risk reduction
is often greater in the elderly assuming
equal RRR (because base risk is higher).
Effects of DM on QOL
RCT of glipizide XL vs. placebo for 12 weeks
594 patients; mean age 58.5 (range: 30-85)
Glipizide XL titrated upward as needed
Home glucose monitoring
Final mean A1c’s: 7.5% (glipizide) vs. 9.3%
(placebo)
 Final average fasting BS’s: 126mg/dl (glipizide)
vs. 168 mg/dl (placebo)




Effects of DM on QOL
Global QOL directly related to A1c level
Glipizide group had significantly (p<0.001) less:





Weakness; fatigue
Urinary frequency; nocturia
Thirst; polydipsia
Dryness of mouth, eyes, or nose
Sweet taste in mouth
Effects of DM on QOL (cont.)
Glipizide group also had significantly (p<0.01)
less:





Foot cramps; foot pain
Sweating
Numbness of lips or mouth
Blurred or double vision
Crabbiness; short-temperedness
Effects of DM on QOL (cont.)
Glipizide group also had (p<0.05) less:







Headaches
Tiredness, drowsiness
Muscle cramps
Vertigo (spinning sensation)
Lightheadedness when standing up
Chest pain with exertion
Confusion
Effects of DM on QOL (cont.)
 Glipizide group also had fewer:
 Days absent from work
 Days spent in bed
 Days of restricted activity
Testa MA, et al. JAMA 1998; 280 (17): 1490-1496
Micro-vascular Disease
 Greater benefit from reduction of A1c
from 9 to 8 than from 8 to 7
 It takes 8-10 yrs of glycemic control to
realize the benefits for micro-vascular
disease
ESRD
End Stage Renal Disease (ESRD) by age at
diagnosis of DM:
Age at Dx
55
55
A1c
7.0
9.0
Lifetime Risk_
0.9%
1.6%
65
65
7.0
9.0
0.3%
0.6%
Blindness
Blindness from DM Retinopathy by age at diagnosis
of DM:
Age at Dx
55
55
A1c
7.0
9.0
Lifetime Risk
0.1%
1.2%
65
65
7.0
9.0
<0.1%
0.5%
Diabetic Peripheral and Autonomic
Neuropathies
 Proposed mechanisms: sorbitol, myoinosital,
ischemia, glycosylation, osmotic
 Improved glucose control probably slows
progression, but size of effect is unknown
 Some evidence of minor benefits from: C-peptide,
Vitamin E, other antioxidants, nerve growth
factors
Why Not Control Everything?
 Law of diminishing returns
 Less benefit from successive interventions
 Diminishing ability to correctly adhere to more
complicated regimens
 Increased side effects and drug interactions
from more meds (exponential increase above
4-5)
 Impact of testing and interventions on
lifestyle
Math
ARR = RRR (baseline risk)
Most interventions that can reduce risk of
heart attack have RRR of about 25%.
If 10-yr base = 20%, then the first
intervention results in an ARR of 5%, the
second in an ARR of 3.75% (25% of 15%),
and so on (diminishing returns)
Diminishing Returns
Equal RRRs
Interventio
n
Prior risk
0.3000
Incrementa
RRR
l ARR
Resulting
risk
Cumulative
ARR
Cumulativ
e RRR
1
0.1929
-0.0579
0.2421
-0.0579
0.1929
2
0.1929
-0.0467
0.1954
-0.1046
0.3485
3
0.1929
-0.0377
0.1578
-0.1422
0.4742
4
0.1929
-0.0304
0.1273
-0.1727
0.5756
5
0.1929
-0.0246
0.1028
-0.1972
0.6574
6
0.1929
-0.0198
0.0830
-0.2170
0.7235
7
0.1929
-0.0160
0.0670
-0.2330
0.7768
RRR = Relative Risk Reduction; ARR = Absolute Risk Reduction
Diminishing Returns
Treatments Chosen from Best RRR to Worst
Interventio
n
Prior risk
0.3000
Incrementa
RRR
l ARR
Resulting
risk
Cumulative
ARR
Cumulativ
e RRR
1
0.35
-0.1050
0.1950 -0.105
0.35
2
0.30
-0.0585
0.1365 -0.1635
0.545
3
0.20
-0.0273
0.1092 -0.1908
0.636
4
0.20
-0.0218
0.0874 -0.21264
0.7088
5
0.15
-0.0131
0.0743 -0.225744
0.75248
6
0.10
-0.0074
0.0668 -0.2331696 0.777232
7
0.05
-0.0033
0.0635 -0.2365111 0.7883704
RRR = Relative Risk Reduction; ARR = Absolute Risk Reduction
Problems with Hypoglycemia
 Case-control study involving 111
community-dwelling adults >75 years
of age
 Strong correlation between A1c<7 and
increased risk of falls
 NHLBI-funded ACCORD study, a RCT
with 10,251 participants
 Intensive treatment group had an excess
number of deaths
Simulation
Fictitious 79 year-old woman with DM, COPD,
HTN, OA, and osteoporosis.
Researchers applied relevant clinical practice
guidelines using a conservative approach and
generics.




Required 12 medications
$406 per month (pre-Medicare D)
Taken at 5 different times per day
Multiple potential interactions
Boyd CM, et al. JAMA 2005; 294(6): 716-724.
Case
Mr. M is a 65 yo man with type 2 diabetes mellitus
diagnosed a yr. ago. Sedentary lifestyle; no
cigarettes or alcohol. Recent onset of fatigue,
decreased libido, polyuria and polydipsia, and
increasing pain in his joints. He currently takes no
medications. Past and family history are
unremarkable. His primary care physician finds:
body mass index (BMI) of 30.5, BP 200/100 mmHg,
osteoarthritis involving fingers and knees,
hemoglobin A1c 10%, LDL cholesterol 140 mg/dl,
HDL cholesterol 40 mg/dl, total cholesterol 260 mg/dl,
ALT and AST slightly elevated, and serum
testosterone low.
Diabetes Personal Health Decisions
(PHD) Engine
Archimedes program
• Attempts to model diabetes by including
>100 biological variables, symptoms, signs,
tests, treatments, and outcomes
• Uses differential equations and objectoriented programming to model the links
between variables
• Keeps all continuous variables continuous
Diabetes Personal Health Decisions
(PHD) Engine
• Addresses co-morbidities and treatments
with multiple effects
• Includes not only individual patients, but
also aspects of the helath care delivery
system (facilities, equipment, policies and
procedures, costs, and utilities)
• Data based upon knowledge of
pathophysiology, clinical trials, and data
from Kaiser Permanente
PHD Validation
• Subjected to a series of 74 validation
exercises involving 18 clinical trials, 10 of
which were not used in the construction of
the engine
• Correlation between results of PHD
simulations and clinical trials overall was
astounding (r=0.99)
• Correlation between absolute differences in
outcomes also amazing (r=0.97)
Mr. Waldman (Diabetes PHD
Risk Engine)
Life expectancy is about 7 years.
7-yr. Risk
MI
CVA
40%
17%
Aspirin
Moderate Exercise
BP to 130 with ACEI
Lower LDL to 100
Lower A1c to 6.5%
Reduce BMI to 26
www.diabetes.org
ARR
11%
10%
7%
5%
1%
9%
Sum
20%
24%
27%
28%
31%
ARR Sum
2%
5% 6%
3% 9%
0% 9%
0% 9%
3% 11%
Objective
Attendees should be able to:
 Construct a rational management plan for
an older patient with DM, taking into
account goals and preferences, life
expectancy, abilities, and resources