Wwwicsiorg

Work Group's Conclusion: For patients with type 2 diabetes mellitus, the A1C goal is less than 7%. Conclusion Grade: II

Author/Y ear

UK Prospective Diabetes Study Group (UKPDS 33), 1998

Ohkubo et al., 1995

Design Type

Class

Population Studied/Sample Size

-3867 newly diagnosed type 2 diabetes patients with mean glucose concentrations of 6.1-15.0 mmol/1 after 3 months of diet treatment

-all patients age 48-60 years of age -patients randomly assigned to receive treatment with a sulfonylurea (chlorpropamide, glibenclamide, or glipizide), insulin, or continue with the diet -10-year follow-up

-110 patients with non-insulin-dependent diabetes mellitus (NIDDM) -patients randomly assigned multiple insulin injection treatment (MIT) or conventional insulin injection treatment (CIT)

-patients split into primary prevention cohort (no retinopathy and urinary excre-tions<30mg/24hours) and secondary prevention cohort (simple retinopathy and urinary excretions<300mg/24hours) -all patients <70 years of age -6-year follow-up

-ETDRS scale: worsening = increase 2+ steps in 19 stage scale for retinopathy or increase 1+ in 3-stage nephropathy scale

Primary Outcome Measure(s)/Results (e.g., p-value, confidence interval, relative risk, odds ratio, likelihood ratio, number needed to treat)

â– Over 10 years, HbAlC was 11% lower in the 2 treatment groups as compared to the diet alone group (7% vs 7.9%)

-treatment group had a 25% risk reduction (p=0.0099) in microvascular endpoints and 12% (p=0.029) reduction in any diabetes related event as compared to the diet group -non-significant reductions for the treatment group were 10% (p=0.34) for any diabetes related death and 6% (0.44) for all cause mortality -weight gain and hypoglycemic events were significantly higher in the treatment group (p<0.001 andpcQ.OOOl)

-near normoglycemia was obtained by month 3 in the MIT group

-mean values of FBG, HbAlC, MBG, M-value, and MAGE were significantly lower in NIT group as compared to CIT group (p<0.001) -retinopathy was significantly higher in MIT groups (7.7% primary and 19.2% secondary) as compared to CIT groups (32% primary [p=0.039] and 44% secondary [0.049]) after 6 years -nephropathy was significantly higher in MIT groups (7.7% primary and 11.5% secondary) as compared to CIT groups (28% primary [p=0.032] and 32% secondary [0.044]) after 6 years -MIT group showed significant improvement in nerve conduction velocities while CIT group showed deteriorated velocities and vibration thresholds

Authors' Conclusions/

Work Group's Comments (italicized)

-Intensive blood-glucose control by either sulphonylureas or insulin substantially decreases the risk of microvascular complications, but not macrovascular disease, in patients with type 2 diabetes.

-Intensive glycemic control by multiple insulin injection therapy can delay the onset and the progression of diabetic retinopathy, neuropathy, and nephropathy in Japanese patients with NIDDM. From this study, the glycemic threshold to prevent the onset and the progression of diabetic microangiopathy is indicated by HbAlC<6.5%, FBG<110mg/dl, and 2-h postprandial blood glucose concentra-tion<180mg/dl.

O 4b

CD O

CSs S

cn cn

Author/Y ear

Design Type

Class

Quality +,-,0

Population Studied/Sample Size

Primary Outcome Measure(s)/Results (e.g., p-value, confidence interval, relative risk, odds ratio, likelihood ratio, number needed to treat)

Authors' Conclusions/

Work Group's Comments (italicized)

UK Prospective Diabetes Study Group (UKPDS 35), 2000

Cohort study

B

0

-4585 patients with HbAlC measured 3 months after diagnoses of type 2 diabetes included in analysis of incidence (3867 with fasting plasma glucose levels of 6.1-15.0 mmol/1 and no symptoms of hyperglycemia included in analysis of RR) -mean age 53 years -mean 10-year follow-up

-each 1% reduction in updated mean HbAlC was associated with reductions of risk of: 21% for any end point related to diabetes (95CI 17%-24%, p<0.0001), 21% for any end point related to diabetes (95CI 15%-27%, pcO.OOOl), 14% for MI (95CI 8%-21%, p<0.0001), 37% for microvascular complications (95CI 33%-41%, p<0.0001) -no threshold of risk was observed for any end point

-In patients with type 2 diabetes the risk of diabetic complications was strongly associated with previous hyperglycemia. Any reduction in HbAlC is likely to reduce the risk of complications, with the lowest risk being in those with HbAlC values in the normal range (<6.0%).

Diabetes Control and Complications Trial Research Group (DCCT), 1996

RCT

A

0

-1441 patients with insulin-dependent diabetes mellitus (IDDM) -patients randomly assigned to receive intensive treatment of 3-4 daily insulin injections or use of external insulin pump (711 patients) or conventional therapy consisting of 1-2 daily insulin injections (730 patients)

-intensive treatment patients had a goal of achieving glycemic control as close to nondiabetic range as safely possible -all patients age 48-60 years of age -mean 6.5-year follow-up

-risks of retinopathy progression, developing microalbuminuria, and neuropathy were continuous but nonlinear over entire range of glycosylated hemoglobin values in both groups and in the two groups combined

-no HbAlC threshold was identified, short of normal glycemia, below which there was no risk of the development or progression of complications

-as HbAlC was reduced proportionately, proportional rate of decline in RR for each complication was similar for HbAlC levels 8.0% or less and greater than 8.0%

-absolute risk of severe hypoglycemia in intensive group increased as HbAlC decreased but RR gradients were significantly less for HbAlC levels 8.0% or less than for levels greater than 8.0%

-DCCT data do not support the conjecture that a glycemic threshold for the development of complications exists at a HbAlC of 8.0% or that an HbAlC goal of 8% is maximally beneficial. In the DCCT, as HbAlC was reduced below 8% there were continuing relative reductions in the risk of complications, whereas there was a slower rate of increase in the risk of hypoglycemia. -The DCCT continues to recommend implementation of intensive therapy with the goal of achieving normal glycemia as early as possible in as many IDDM patients as is safely possible.

5"

n"

CT S Ci oo

to S"

01 Co

Conclusion Grading Worksheet - Appendix B -Annotation #11B (Statin Use)

00 2

tf P

"S

R tu

S CvS

1 S3

O^OOH

f^oo

en cd

Author/Y ear

Design Type

Class

Quality +,-,0

Population Studied/Sample Size

Primary Outcome Measure(s)/Results (e.g., p-value, confidence interval, relative risk, odds ratio, likelihood ratio, number needed to treat)

Authors' Conclusions/

Work Group's Comments (italicized)

Colhoun et al, CARDS 2004

RCT

A

+

2838 patients (age 40-75 years, 94% caucasion and 68% male), in 132 centers in the UK/Ireland

Atorvastatin 10 mg vs Placebo

Acute coronary event HR 0.63(0.48-0.83)

Death from any cause HR 0.73(0.52-0.85)

Randomization with equal groups at baseline and 1% lost to follow-up after a mean follow-up of 4 years. Analysis was with intention to treat, and during the course of study 9% of placebo group was known to take a statin and 85% of the intervention (either atorvastatin or another statin). Overall frequency of adverse events or serious adverse events did not differ between treatments. In each group 1.1% of patients randomized had one or more serious adverse events. Based on pre- and post-LDL values in intervention and control group there did not appear to be a particular threshold level of LDL-cholesterol to reduce cardiovascular events.

Robins et al, 2001

RCT

A

+

2531 men with coronary heart disease and low HDL-C levels (avg 32 mg/dl). 620 patients had diabetes.

Gemfibrizol 1200 mgm/day vs Placebo

RRR 95% CI (4-46%)

Patients were randomized with concealed allocation; they were similar at baseline and treated relatively similarly throughout the trial; patients, study personnel, health care providers, and outcomes assessors were blinded; intention-to-treat analysis was conducted; there was trivial loss to fol-low-up. No validity concerns.

GO I

ns NJ

Work Group's Conclusion: For patients with type 2 diabetes mellitus, the systolic blood pressure (BP) goal is less than 130 and the diastolic blood pressure (BP) goal is less than 80.

Conclusion Grade: II

Author/Year

Design Type

Class

Quality

Population Studied/Sample Size

Primary Outcome Measure(s)/Results (e.g., p-value, confidence interval, relative risk, odds ratio, likelihood ratio, number needed to treat)

Authors' Conclusions/ Work Group's Comments (italicized)

UK Prospective Diabetes Study Group (UKPDS 39), 1998

ROT

A

0

-758 patients allocated to tight control of BP among 1148 hypertensive patients with type 2 diabetes

-400 patients treated with captopril (25-50 mg twice daily), 358 with atenolol (50-100 mg twice daily)

-all patients age 48-60 years of age (mean age of treatment groups 56 years) -9-year follow-up -goal of BP <150/85 mm Hg

-captopril and atenolol euallele effective in mean BP reduction (144/84 and 143/81 mm Hg, respectively) -reduction of risk of macrovascular endpoints were similar in the two groups (31% and 37% showed deterioration in retinopathy by 2 grades; 5% and 9% developed clinical grade albuminuria greater or equal to 300 mg/1) -similar percent of patients required 3 or more antihypertensive treatments (27% and 31%) or developed hypoglycemic attacks but mean wt gain was greater in the atenolol group (1.6 kg vs 3.4kg) -78% captopril and 65% atenolol patients taking treatment at last visit (p<0.0001)

-Blood pressure lowering with captopril or atenolol was similarly effective in reducing the incidence of diabetic complications. This stody suggests that blood pressure reduction in itself may be more important than the treatment used.

UK Prospective Diabetes Study Group (UKPDS 38), 1998

ROT

A

+

-1148 hypertensive patients with type 2 diabetes

-758 patients allocated to tight control of BP with goal of <150/85 mm Hg (400 patients treated with captopril [25-50 mg twice daily], 358 with atenolol [50-100 mg twice daily]) and 390 patients allocated to less tight control of BP with goal of <180/105 mm Hg -all patients age 48-60 years of age (mean age of treatment groups 56 years) -8.4-year follow-up

-mean BP was significantly reduced in the tight BP group (144/82 Hg mm) as compared to the less tight BP group (and 154/87 mm Hg, p<0.0001) -reduction of risk in the tight BP group as compared to the less tight BP group were: 24% in diabetes related endpoints (95CI 8% to 38%, p=0.0046), 32% in deaths related to diabetes (95CI 6% to 51%, p=0.019), 44% in strokes (95CI 11% to 65%, p=0.013), 37% in microvascular endpoints (95CI 11% to 56%, p=0.0092) -tight BP group had a 34% reduction in risk of proportion with deterioration in retinopathy by 2 grades (99CI 11% to 50%, p=0.0004), and a 47% reduced risk for deterioration in visual acuity (99CI 7% to 70%, p=0.004)

-Tight blood pressure control in patients with hypertension and type 2 diabetes achieves a clinically important reduction in the risk of deaths related to diabetes, complications related to diabetes, progression of diabetic retinopathy, and deterioration in visual acuity.

Hansson et al., 1998

Hypertension Optimal Treatment (HOT) Trial

ROT

A

0

-1510 patients with diabetes (among 18,790 total patients with hypertension and diastolic BP 100-115 mm HG in trial) -all patients age 50-80 years of age -patients randomly assigned a target diastolic BP of less than or equal to 90 mm Hg, 85 mm Hg, or 80 mm Hg -all patients received felodipine for hypertension

-ACE inhibitors or B-blockers were used to treat to given target diastolic BP -3-8-year follow-up

-for patients with diabetes, the blood pressure intervention led to a significant reduction (51%) in number of major cardiovascular events (45 events in 90 mm HG group, 34 in 85 mm HG group, and 22 in 80 mm HG group; p=0.005 for trend) and cardiovascular mortalities (21, 21, and 7;p=0.016)

-for patients with diabetes, the blood pressure intervention reduced total mortality (30, 29, 17 events), Mis (14, 8, 7), and stroke (17, 13, 12) but none were statistically significant

-Intensive lowering of BP in diabetes patients with hypertension was associated with a significantly 51% lower rate of cardiovascular events.

01 T

S OQ

CSs s

Work Group's Conclusion: For patients with type 2 diabetes mellitus, initiate low-dose aspirin therapy (81-325 mg daily) in patients 40 and older unless there is a contraindication to aspirin therapy.

Conclusion Grade: I

Author/Year

Design Type

Class

Quality +,-,0

Population Studied/Sample Size

Primary Outcome Measure( s)/Results (e.g., p-value, confidence interval, relative risk, odds ratio, likelihood ratio, number needed to treat)

Authors' Conclusions/

Work Group's Comments (italicized)

Early

Treatment

Diabetic

Retinopathy

Study

(ETDRS)

Report 14,

1992

ROT

A

0

-3711 patients with diabetes mellitus (31% type I, 31% type II, and 39% type I or II) randomized to receive aspirin or placebo (650 mg twice daily)

-all patients age 18-70 years of age -5-year follow-up

-RR for total mortality was 0.91 (99CI 0.75-1.11, p=NS) overall and 0.92 in type II patients (99CI 0.69-1.23, p=NS) treated vs placebo patients

-myocardial infarction rates were 9.1% with aspirin and 12.3% with placebo (RR 0.83, p=0.04) overall

-the NNT to prevent one MI in 5 years with aspirin was 31 patients

-Aspirin use may reduce the risk of myocardial infarction in adults with diabetes, but did not reduce total mortality or CV mortality rates.

-There was no evidence of harmful effects of aspirin.

-The ETDRS results support use of aspirin in persons with diabetes at increased risk of cardiovascular disease.

Hansson et al., 1998

Hypertension Optimal Treatment (HOT) Trial

ROT

A

0

-1510 patients with diabetes (among 18,790

total patients with hypertension and diastolic

BP 100-115 mm HG in trial)

-all patients age 50-80 years of age

-patients randomly assigned a target diastolic

BP of less than or equal to 90 mm Hg, 85

-all study subjects were randomized to receive aspirin 75 mg/day or placebo

-3-8-year follow-up

-for all patients, aspirin use significantly reduced cardiovascular events 15% (p=0.03), and reduced MI rates 36% (p=0.002), but did not reduce mortality

-the relative benefit of aspirin to those with diabetes was "about the same" as in the whole trial population

-Use of aspirin in diabetes and in non-diabetes patients significantly reduced Mis (36%) and cardiovascular events (15%), but did not significantly reduce mortality. -Aspirin use (75 mg/day) appears to benefit diabetes patients with hypertension, even those in whom blood pressure is very well controlled.

Harpaz, et al., 1998

Cohort

B

+

-2,368 NIDDM adults with CHD and 8,586

non-NIDDM adults with CHD

-mean follow-up 5.1 years

-52% of NIDDM patients reported no ASA

use

-all cause mortality was 18.4% in NIDDM ASA users and 26.2% in NIDDM ASA non-users (p < 0.001)

-cardiac mortality was 10.9% in NIDDM ASA users and 15.9% in NIDDM ASA non-users (p < 0.001)

-both significant differences persisted after adjustment for possible confounders

-Treatment with ASA was associated with a significant reduction in cardiac and total mortality among NIDDM adults with CHD.

-Hie absolute benefit of aspirin was greater in diabetes versus non-diabetes adults.

Physician's Health Study Research Group, 1989

ROT

A

0

-primary prevention of MI in subgroup of 533 physicians with diabetes (among 22,071 total participants)

-patients randomized to either 325 mg ASA/day or placebo -mean follow-up 5 years

-overall, 44% reduction in MI (p<0.00001)

in those who took ASA

-in diabetes subgroup, 4.0% had MI in ASA

-relative risk of MI in ASA group was 0.60

in entire cohort, and 0.39 in diabetes

-Aspirin reduced MI rate in overall study. -Benefits in DM group appear to be at least as great as in non-DM group.

-The non-significant differences in DM group were likely due to small sample size and insufficient power.

Ci oq

to S"

01 Co

Conclusion Grading Worksheet - Appendix E - Annotations #24, 30A (Treatment with ACE Inhibitors or ARBs)

J3 O

m in

"OO

tf S

S T3

o ir

Ph m

ox II

IN J

CNcOCNm i m

TJ S

c o

o

T!

c <2

& PH

&

^ o

c o

>

Work Group's Conclusion: For patients with type 2 diabetes mellitus, thiazide diuretics in the treatment of hypertension can reduce cardiovascular events, particularly heart failure.

Conclusion Grade: I

Author/Year

Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial

(ALLHAT)

Officers and

Research

Group,

2002

ALLHAT trial

Wing et al., 2003

ANBP2 Trial

Design Type

Class

Quality

Population Studied/Sample Size

-12,063 patients with Type 2 Diabetes with hypertension as part of a large, multicenter (623 North American centers) including a total of 33,357 patients -mean age 67 years -53% Male; 47% White, 32% Black, and 15% Hispanic -mean follow-up 4.9 years

-6,083 patients (from 1594 family medical practices throughout Australia)

-only 7% with diabetes -95% Caucasian -mean age 72 years -patient groups were equal at randomization, followed for 4.1 years with intention to treat analysis (0.2% lost to f/u)

Primary Outcome Measure(s)/Results (e.g., p-value, confidence interval, relative risk, odds ratio, likelihood ratio, number needed to treat)_

Amlodipine 2.5-10 mgm vs Chlorthalidone 12.5-25 mgm/d -All cause mortality: relative risk (RR) 0.96 (95%CI 0.82-1.07)

Lisinopril 10-40 mgm vs Chlorthalidone 12.5-25 mgm/d -All cause mortality: RR 1.02 (95%CI 0.91-1.13) -Stroke: RR 1.07 (95%CI 0.9-1.28) -Combined CV disease: RR 1.08 (95%CI 1.0-1.17) -Any Heart Failure: RR 1.22 (95%CI 1.06-1.42)

-Enalapril (ACE inhibitor) vs Hydrochlorothiazide (diuretic) -All CV events or death from any cause: hazard ratio (HR) 0.89 (95%CI 0.79-1.00)

-First CV event or death from any cause: HR 0.89 (95%CI 0.79-1.01)

-58%-62% receiving treatment assigned at the end of study and equal BP response (systolic/diastolic) in both groups -in post hoc analysis, largest effect seen in male patients

Authors' Conclusions/

Work Group's Comments (italicized)

-For type 2 diabetic patients, lisino-pril appeared to have no special advantage (and amlodipine no special detrimental effect) for most CVD outcomes when compared with chlorthalidone .

-Because the main intent was to compare thiazide, calcium channel blocker, and ace inhibitor treatment, the available step-up for further management of hypertension for patients on ace inhibitors led to less than typical regimen (use of sympatholytics rather than diuretics and calcium channel blockers). Since a large proportion of diabetes patients require more than one-drug to control their BP, this study suggests that a diuretic should be in-cluded in all multidrug regimens.

-Initiation of antihypertensive treatment involving ACE inhibitors in older subjects, particularly men, appears to lead to better outcomes than treatment with diuretic agents, despite similar reductions of blood pressure. -There was a lower prevalence of diabetes than might have been expected [7%] mostly because the study population was overrepresented by elderly Caucasian patients. -Vascular outcomes and death were worse using hypertensive regimen emphasizing hydrochlorothiazide compared to ACE inhibition. -Also, insufficient information is provided to discern whether groups were treated equally._

S Ci oq

CSs s

CSs S

Institute for Clinical Systems Improvement

Support for Implementation:

0 0

Post a comment