Todays Outline 8 articles to discuss SHOULD IT CHANGE YOUR - - PowerPoint PPT Presentation

8/9/2019 1

Michael G. Shlipak, MD MPH SHOULD IT CHANGE YOUR PRACTICE? A DEEPER LOOK AT SOME OF THE PAST YEAR’S MOST IMPORTANT PAPERS

AUGUST 8, 2019

Today’s Outline

• 8 articles to discuss

 10 minutes per article

• Audience participation is essential
• Questions and comments can relate to:

 Overall topic  Methods  Clinical implications  Practical experience

Keeping up with the Literature

• Impossible task for busy clinicians
• Nearly impossible task for academician
• My sources:

 Journal table of contents  Email newsletters (Journal Watch, specialty newsletters, AMA, etc.)  Popular press

• If the topic is interesting, then I go to the manuscript

Manuscript Review:

Questions I Consider

• Does this study address an important question?
• Can the study design answer the question?
• What were the results?

 Overall conclusion  Strength of the findings  Generalizability of population  Ethical or cost considerations

• Does this change my practice?

8/9/2019 2

A year of Surprising Trials

• 1. Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease
• 2. Marine n−3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer
• 3. Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia
• 4. Acute Illness Associated With Cannabis Use, by Route of Exposure: An

Observational Study

• 5. Canagliflozin and Renal Outcomes in Type 2 Diabetes and Nephropathy
• 6. Use of aspirin to reduce risk of initial vascular events in patients at moderate risk of

cardiovascular disease (ARRIVE): a randomised, double-blind, placebo-controlled trial

7.

Effects of Aspirin for Primary Prevention in Persons with Diabetes Mellitus

• 8. Effect of Aspirin on All-Cause Mortality in the Healthy Elderly

Vitamin D Supplements

SMG1

Why do we need RCTs of Vitamins?

• Supplements heavily used
• Nutritional epidemiology – observational studies

seem to be overly optimistic

very controversial, contradictory studies substantial confounding (activity, nutrition, socioeconomic status)

• Large trials required to determine effect on

clinically meaningful outcomes

• 3 major supplements trials to discuss today

Slide 7 SMG1 Add the date, please

Shlipak, Michael G., 7/22/2019

8/9/2019 3

Why did VITAL study Vitamin D?

• Vitamin D is believed to have many beneficial properties
• Enormous literature from cohort studies suggesting

beneficial effects

• Anti-inflammation, anti-aging, prevention of atherosclerosis,

prevention of cancer

• USPTF guidance on vitamin D (before VITAL)

 Screening: “I” indeterminant evidence  Fracture prevention in men and women: “I”

Design: The VITAL Trial

• Setting: United States
• Age: Men ≥ 50; Women ≥ 55
• Goal: 5,000 African Americans; 25,871 total
• Intervention: 2000 IU vs. placebo
• Exclusions: cancer, CVD, ESRD, cirrhosis, hypercalcemia
• Co-treatment: max of 800 IU of vitamin D
• Outcomes: invasive cancer, CVD events
• Funding: NIH

Effect of the Intervention on 25(OH) D levels

• Vitamin D:

↑40% (3042)

• Placebo:

3030 (no change) Is that enough to change outcomes?

Participants in VITAL

Characteristic Total (N= 25,871) Mean Age

67±7

Female sex

51%

Race Non-Hispanic White

71%

Black

20%

Nonblack Hispanic

4%

Asian or Pacific Islander

1%

Body Mass Index

28±6

Current Smoking

7%

Hypertension

50%

Diabetes

14%

Manson JE et al., N Eng J Med, 2019

8/9/2019 4

Vitamin D: No effect on Cancer

Manson JE et al., N Eng J Med, 2019

Vitamin D: No effect on CVD Risk

Manson JE et al., N Eng J Med, 2019

Vitamin D and Specific Cancers Vitamin D and Specific CVD Outcomes

8/9/2019 5

Subgroup

• No. of

Participants Invasive Cancer Cardiovascular Events Baseline serum 25- hydroxyvitamin D category 15,787 Vitamin D Placebo Hazard Ratio (95% CI) Vitamin D Placebo Hazard Ratio (95% CI) <Median of 31 ng/ml 7,812 128 139 0.94 (0.74–1.20) 251 252 1.02 (0.86–1.21) ≥Median of 31 ng/ml 7,975 124 111 1.09 (0.84–1.41) 266 275 0.95 (0.80–1.12) Manson JE et al., N Eng J Med, 2019

Conclusions of Vitamin D Trial

• What are the major limitations?
• Does it change your practice?

Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease and Cancer

8/9/2019 6 Why did VITAL study Fish Oil/Omega 3 fatty acids?

• Marine derived long-chain n-3 (omega-3) fatty

acids may prevent CVD

• Mostly observational studies, small RCTs
• Not definitive in the literature
• Effect on cancer unknown
• What is a 2x2 factorial design? Why did the

investigators choose this design for VITAL?

Design of VITAL – Omega-3

• Design: RCT, double-blind; 2x2 factorial design

in combination with Vitamin D/placebo trial

• Intervention: fish oil capsule vs. placebo

840mg of n-3 fatty acids 460 EPA 380 DHA

• The chosen dose was AHA recommended
• Donated by industry (BASF)

Omega 3’s and CVD Risk Omega 3’s and Cancer Risk

8/9/2019 7 Omega 3’s and Specific CVD Outcomes

Manson JE et al. New Eng J Med, 2019

Omega 3’s and More CVD Outcomes

§ These events were not prespecified as primary or secondary end points. ¶ This end point was a composite of myocardial infarction, coronary revascularization (PCI or CABG), and death from coronary heart disease.

Manson JE et al. New Eng J Med, 2019

Omega-3’s and Cancer Risk

Manson JE et al. New Eng J Med, 2019

Subgroups by Race and Fish Consumption on Risk of Primary CVD outcome

8/9/2019 8

Discussion

• How do we reconcile the positive outcomes

with all the negative outcomes?

• Does it change your practice?

Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia

Why was this Fish Oil study done?

• Is TG a modifiable risk factor for CVD?

Most trials that lower TG’s have been null

• Japan EPA Lipid Intervention Study (JELIS)- 19%

lower CVD death

• Icosapent-ethyl: highly purified and stable EPA

ethyl ether

• Hypothesized effects: anti-inflammatory, anti-
• xidative, plaque-stabilizing
• Can it decrease CVD events?

8/9/2019 9

REDUCE-IT

• Icosapent-Ethyl: highly purified EPA

ethyl ester vs. placebo (mineral oil) 2g BID (>4x the VITAL dose)

• Is increased TG really a CVD risk

factor?

Design of REDUCE-IT Trial

• Icosapent-Ethyl: highly purified EPA ethyl

ester vs. placebo (mineral oil)

• Intervention: 2g BID vs. placebo (>4x the

VITAL dose)

• Design: RCT, double-blind, phase 3b
• Participants (High CVD Risk):

≥ 45 years old with cardiovascular disease ≥ 50 years old with diabetes

Manson JE et al. New Eng J Med, 2019

Design of REDUCE-IT Trial

• TG fasting range: 150-499
• LDL range: 41-100
• On statins for 4 weeks
• Outcomes: broad cardiovascular disease-

MI, CVA, unstable angina, cardiovascular death

• Funding: Amarin Corporation

11 countries, 473 sites

REDUCE-IT

• N= 8,179; Mean age: 64
• 71% CVD
• 29% diabetes without CVD
• 39% US
• 29% female
• Mean LDL: 75 mg/dl
• Mean HDL: 40 mg/dl
• TG: 216 mg/dl

Bhatt DL et al., N Eng J Med, 2019

8/9/2019 10

Primary Outcome

Bhatt DL et al., N Eng J Med, 2019

17% vs. 22% at 5 years NNT= ??

Major CVD Events

(CVD death, MI, stroke)

Bhatt DL et al., N Eng J Med, 2019

11% vs. 15% at 5 years NNT= 25

Effect by Key Subgroups on Major CVD Events

Bhatt DL et al., N Eng J Med, 2019

Effect on Specific CVD Outcomes

Bhatt DL et al., N Eng J Med, 2019 Hazard Ratio 95% CI)

8/9/2019 11

Discussion of REDUCE-IT

• Do you believe the results?
• What might the mechanism be?
• Could mineral oil block statin absorption?
• Multiple trials in progress

Acute Illness Associated with Cannabis Use, by Route of Exposure

Background

• Colorado: recreational cannabis approved in 2014
• Increase in ED visits
• 40% of marijuana users (smoke only)
• Only deaths in CO from cannabis were from edibles
• Anecdotally, more toxic
• Does the mode of cannabis use affect the risk of

adverse events?

8/9/2019 12

Design

• Design: Retrospective observational; cases only
• Setting: Academic ED; University of Colorado Hospital

Emergency Department

• Case ID: ICD-9 codes with screen for cannabis toxicity

 Attributable to cannabis: 2,567  238 (9.3%) edible

• Compared with cannabis sales in CO
• Funding: Colorado Department of Public Health and

Environment

Monte AA, Annals Int Med, 2019

Methods

• Urban hospital, ED

~100,000 visits/year

• 2012-2016
• Definition of “cannabis related”:

Identified cannabis as likely causing or contributing Toxicity screen + for cannabinoids Plausible association of symptoms with cannabis use Reviewed by toxicologists

Cannabis-related ED Visits

N= 2,567

ED Visits With Cannabis-Related ICD Code

Age 30 years Men 65% White 43% Black 37% Hispanic 13% Admitted (no deaths) 32%

Most Common Clinical Conditions Associated With Cannabis-Attributable Visits

Condition Edible Exposure (n=238; 9%) Inhalable Exposure (n=2329; 91%) Gastrointestinal symptoms 15% 32% Cannabinoid hyperemesis syndrome 8% 18% Intoxication 48% 28% Psychiatric symptoms 26% 25% Acute psychiatric symptoms 18% 11% Acute exacerbation of chronic disease 0.4% 4% Chronic psychiatric condition 0.4% 4% Cardiovascular symptoms 8% 3%

Monte AA, Annals Int Med, 2019

8/9/2019 13 Inhalable and Edible Sales 309 ratio of inhalable/edible, means 30-fold risk of edible/inhaled

Discussion: Do you believe that edible cannabis is really more harmful?

• Edibles known to be hazards for children
• In adults, relative safety had been unclear; amateur users?
• Do you believe the results? What would you tell your patients?

Kinetics 1st Effects Peak in Blood Clearance complete Inhaled 10 min 30-90 min 4 hours Edible 30 min 3 hours 12 hours

Canagliflozin and Renal Outcomes in Type 2 Diabetes and Nephropathy

The CREDENCE TRIAL

8/9/2019 14

CREDENCE Trial

• Diabetic nephropathy

Top cause of kidney failure worldwide Treatments: glycemic control, ACE/ARBs Will new drugs have beneficial effects for the kidney?

CREDENCE Trial

• SGLT2-Inhibitors

 Inhibit sodium-glucose transporter 2 (proximal tubule)  Glucose lowered by urine elimination  Small change in Hb A1C  Surprisingly large decrease in cardiovascular events, especially heart failure  Decrease in urine albumin  Slowing of eGFR decline  Can they lower the risk of ESKD?

Design

Perkovic V et al. N Eng J Med, 2019

• Design: RCT; 1:1 (Canagliflozin 100 vs. placebo)
• Inclusion:
• Targeted ACR >300
• eGFR 30-90 (2/3 with 30-60 ml/min/1.73m2 by design)
• All using ACE/ARB at stable dose
• Outcomes:

 ESRD (dialysis or treatment)  < 15 eGFR for 3 months; 2x creatinine  “death from kidney or CVD”

• Trial stopped early (harm or benefit?)
• Funding: Janssen Research and Development

Demographic and Clinical Characteristics

Characteristic All Patients (N=4,401)

Age— year 63 ± 9 Female sex 34% Race or ethnic group White 67% Black 5% Asian 20% Other 8% Duration of Diabetes— yr 16 ± 9 Glycated hemoglobin— % 8.3 ± 1.3 Estimated GFR— ml/min/1.73m2 56 ±18 Median urinary ACR 927 mg/g

Perkovic V et al. N Eng J Med, 2019

8/9/2019 15 Canagliflozin and Kidney Outcomes

Perkovic V et al. N Eng J Med, 2019

Canagliflozin: Efficacy and Safety

Efficacy Canagliflozin Placebo Hazard Ratio (95% CI) %/year Primary composite outcome 4.3% 6.1% 0.70 (0.59–0.82) End-stage kidney disease 2.0% 2.9% 0.68 (0.54–0.86) Cardiovascular death 1.9% 2.4% 0.78 (0.61–1.00) Hospitalization for heart failure 1.8% 2.5% 0.61 (0.47–0.80) Death from any cause 2.9% 3.5% 0.83 (0.68–1.02) Hyperkalemia 3.0% 3.7% 0.80 (0.65–1.00) Diabetic ketoacidosis 0.2% 0.02% 10.80 (1.39–83.65)

Perkovic V et al. N Eng J Med, 2019

Mechanism for the Benefit?

Change in A1C ↓ 0.3% Change in SBP ↓ 3.3 mmHg Change in weight ↓ 0.8 kg Change in eGFR ↓ 3 ml/min (3 weeks)

Effects on eGFR and ACR during follow-up

Perkovic V et al. N Eng J Med, 2019

8/9/2019 16

Discussion of the CREDENCE trial

• High risk population
• Changes in ACR and eGFR seem in earlier

trials

• Mechanisms appear more like ACE-

inhibitors than a diabetes medication

• Trials will evaluate non-diabetic CKD next
• Several agents in this class
• Will this change your practice?

Aspirin for Primary Prevention

ASA and CVD Prevention

• Secondary Prevention: ASA clearly beneficial
• ASA use for primary prevention clearly requires

balance of risks and benefits

• In the era of statins and CVD prevention,

should ASA be used for primary prevention?

• 3 new major trials for primary prevention

Meta of 16 Secondary ASA prevention trials

• Lancet. 2009 May 30; 373(9678): 1849–1860.

8/9/2019 17 3 New Trials on ASA in Primary Prevention of CVD Risk

1.Aspirin to Reduce Risk of Initial Vascular

Events (ARRIVE) Trial

2.A Study of Cardiovascular Events in

Diabetes (ASCEND) Trial

3.Aspirin in Reducing Events in the Elderly

(ASPREE) Trial

ARRIVE Trial – N=12,500, No Diabetes

• RCT: ASA 100mg vs. Placebo
• Exclude: CVD, HF, diabetes

 55 (60) years old  ACC/AHA CV =17% 10 year risk

• Setting: 7 Western countries

 40% UK, 25% Poland, 24% Germany, 4% US, 3% Italy, 3% Spain, 1% Ireland

• 30% women, 97% white
• 5 year follow-up
• CV Outcomes

 Composite CV death, MI, unstable angina, stroke/TIA

• Safety Outcomes

 Hemorrhage

Gaziano JM et al. The Lancet, September 2018

Effect of ASA on CVD and Death in ARRIVE

Number of events in the intention-to-treat population

Aspirin (n=6,270) Placebo (n=6,276) Hazard Ratio (95% CI) MI, stroke, CVD death, UA, TIA 269 (4.3%) 281 (4.5%) 0.96 (0.81–1.13) MI, stroke, or CVD death 208 (3.3%) 218 (3.5%) 0.95 (0.79–1.15) Stroke 75 (1.2%) 67 (1.1%) 1.12 (0.80–1.55) Any death 160 (2.6%) 161 (2.6%) 0.99 (0.80–1.24)

Gaziano JM et al. The Lancet, September 2018

8/9/2019 18 Kaplan-Meier cumulative incidence

• f Gastrointestinal bleeding

0.97%- ASA 0.46%- Placebo HR 2.11; p= 0.0007

Gaziano JM et al. The Lancet, September 2018

Discussion ASA Part 1: Primary Prevention, Non-Diabetic

• No benefit on CVD, and increased GI bleed risk
• 5-year event rates = 4.3%
• Estimated CVD risk= 17% at 10 years
• Will this study change your practice?

ASCEND- A Study of CV Events in Diabetes

• RCT: ASA 100mg vs. Placebo
• Participants: 15,500 British persons with diabetes
• Age: 63 yrs
• Sex: 37% women
• Race: 97% white

 Vasc risk score- 40% low, 42% mod, 18% high  7.4 years follow up

• Outcomes

 Composite CV death, MI, USA, stroke/TIA

N Engl J Med. 2018 Oct 18;379(16):1529-1539

8/9/2019 19

CVD Benefit: HR = 0.88 (0.79-0.97)

N Engl J Med. 2018 Oct 18;379(16):1529-1539

ASCEND- Outcomes

• Composite outcome: CV death, MI, USA, TIA/CVA
• ASA events: 8.5% in 7 years
• Placebo events: 9.6%
• Rate Ratio = 0.88
• 12% reduction
• NNT=100 over 7 years
• Major bleeding: ASA 4.1%, Placebo 3.2%
• Bleeding RR: 1.29
• NNH= 100
• Benefit = Risk

N Engl J Med. 2018 Oct 18;379(16):1529-1539

Discussion ASA Part 2: Primary Prevention in Persons with Diabetes

• Benefit on CVD Risk but Harm for GI Bleeding
• Balanced benefit and harms
• No effect on cancer risk
• Will this study change your practice?

8/9/2019 20

ASPREE- Aspirin in Reducing Events in Elderly

• RCT- 19,000 patients from Australia/U.S.

 100mg ASA vs. Placebo-  5 years follow-up  Mean age: 74 yrs  56% women  85% White, 5% Black, 2.5% Hispanic

• Diabetes: 11%
• Hypertension: 74%
• Current Smoking: 4%
• Mean BMI: 28

McNeil JJ et al. N Engl J Med., 2018

ASPREE- Aspirin in Reducing Events in Elderly

• Median 4.7 years
• Outcomes:

 Death, dementia, persistent physical disability  All cause mortality  Cardiovascular events and bleeding

McNeil JJ et al. N Engl J Med., 2018

Mortality According to the Underlying Cause of Death

Aspirin (N = 9525) Placebo (N = 9589) Hazard Ratio (95% CI)

All cause mortality 5.9% 5.2% 1.14 (1.01–1.29) Cancer death 3.1% 2.3% 1.31 (1.10–1.56) CVD death 1.0% 1.2% 0.82 (0.62–1.08) Major hemorrhage 3.8% 2.8% 1.38 (1.18-1.62) Dementia 6.7% 6.9% 0.98 (0.83-1.15)

McNeil JJ et al. N Engl J Med., 2018

Aspirin and Cancer Death - ASPREE

McNeil JJ et al. N Engl J Med., 2018

Colorectal cancer: 35 vs. 20 fatal events 1.77 (1.02-3.06)

8/9/2019 21

Discussion ASA Part 3: Primary Prevention in Elderly

• No benefit on CVD risk
• Increased risk of Death!
• Appears due to higher Cancer risk, particularly GI
• Will this study change your practice?

Thank you!