[PDF] - 12/17/16 Cardiac CT, Nuclear Cardiology, and Cardiac MRI DISCLOSURE PDF Document

SLIDE 1

12/17/16 1

Daniel S. Berman, MD Director, Cardiac Imaging Cedars-Sinai Heart Institute Professor of Medicine Cedars-Sinai Medical Center

UCSF December 2016

Cardiac CT, Nuclear Cardiology, and Cardiac MRI Roles in the Era of Value-based Imaging

DISCLOSURE

Daniel S. Berman, M.D. declares the following relationships:

Consultant: Molecular-Dynamics Royalties: Cedars-Sinai Medical Center Research grant: HeartFlow

The exciting:

–Technology: always improving

The realistic:

– Someone is paying for it all – Proven value will be required for its use

Value-based Cardiac Imaging in CAD

SLIDE 2

12/17/16 2

The exciting:

–Technology: always improving

The realistic:

– Someone is paying for it all – Proven value will be required for its use

Value-based Cardiac Imaging in CAD

Does the test PREDICT risk? Does the test result in improved

utcomes or

reduce costs?

Value-based Cardiac Imaging in CAD Value-based Cardiac Imaging in CAD

Technologic developments
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure

Challenges

Automatic Quantitative Analysis of SPECT/PET Emory Cardiac Tool Box 4D M-SPECT Cedars-Sinai QGS/QPS/QPET

SLIDE 3

12/17/16 3

Prediction of MACE with SPECT-MPI Machine Learning with Nuclear and Clinical Variables

N=2818 Mean f/u: 3.0±0.8 years MACE: 12% Otaki…Slomka, et al ASNC 2016

Otaki…Slomka, et al ASNC 2016

Prediction of MACE with SPECT-MPI Machine Learning with Nuclear and Clinical Variables

N=2818 Mean f/u: 3.0±0.8 years MACE: 12% Otaki…Slomka, et al ASNC 2016

Otaki…Slomka, et al ASNC 2016

Automated quantitative assessment with machine learning: Superior to expert reading alone and will become routine

CZT Cardiac SPECT Systems

CZT Detectors No PMT sensitivity spatial resolution energy resolution smaller size Einstein JNM 2014 <1 mSV

GE

SPECT/CT

GE Siemens Philips

Attenuation correction
Coronary artery calcium scoring

SLIDE 4

12/17/16 4

GE

SPECT/CT

GE Siemens Philips

Attenuation correction
Coronary artery calcium scoring

PET/CT Cameras

GE Siemens Philips

STRESS REST

CAC 0

PET/CT and SPECT/CT Combining anatomy with function: CAC+ MPI

V GFADS REST

PET/CT and SPECT/CT Combining anatomy with function: CAC+ MPI

LAD + LCX stenosis: CABG

SLIDE 5

12/17/16 5

V GFADS REST

PET/CT and SPECT/CT Combining anatomy with function: CAC+ MPI

Adding CAC to PET/SPECT: Increases diagnostic certainty Detects subclinical atherosclerosis

LAD + LCX stenosis: CABG

Absolute MBF Assessment With Rb-82 PET Prognostic Value Over Regional Perfusion

10 20 30 40 50 20 40 60 80 100 120 seconds

LV GFADS LV VOI RV GFADS RV VOI MYO GFADS MYO VOI

El Fakhri, Sitek, et al. J. Nucl. Med 2005, Brigham & Women’s Hospital; Ziadi et al. JACC 2011

Upper Tertile Middle Tertile Lower Tertile 0% 2% 4% 6% 8% 10% 12% ≥10% 1-9% 0% Annualized Mortality ≥10% 1-9% 0% Upper Tertile 2.4% 0.3% 0.1% Middle Tertile 4.4% 4.0% 1.1% Lower Tertile 10.2% 6.0% 3.6%

P<0.0001

(416) (217) (119) (n) (190) (202) (195) (n) (321) (509) (614) (n)

Murthy VL, et al. Circulation 2011 N= 2,783 CD= 137

Coronary Flow Reserve (CFR) Predicts Mortality Independent of Perfusion Defects

Upper Tertile Middle Tertile Lower Tertile 0% 2% 4% 6% 8% 10% 12% ≥10% 1-9% 0% Annualized Mortality ≥10% 1-9% 0% Upper Tertile 2.4% 0.3% 0.1% Middle Tertile 4.4% 4.0% 1.1% Lower Tertile 10.2% 6.0% 3.6%

P<0.0001

(416) (217) (119) (n) (190) (202) (195) (n) (321) (509) (614) (n)

Murthy VL, et al. Circulation 2011 N= 2,783 CD= 137

Coronary Flow Reserve (CFR) Predicts Mortality Independent of Perfusion Defects

CFR: Adds prognostic information to perfusion defect Increases certainty; identifies diffuse disease

SLIDE 6

12/17/16 6

F-18 Sodium Fluoride PET Identifies Ruptured and High-Risk Coronary Plaques

Joshi, Dweck…Newby Lancet 2013

40 AMI
93% uptake in

culprit plaque at ICA

40 Stable angina
45% uptake in

plaques with high risk features (IVUS)

Nuclear Imaging Targets for Vulnerable Plaque

Courtesy: Andreas Kjaer

Beautiful Images

Who needs them? Will the results affect managment/outcome?

S. Achenbach, MDbb
Instrumentation/software

– Full coverage single beat – Higher temporal resolution 66 ms – Lower radiation <1 mSv – Model based iterative reconstruction

Value-based Cardiac Imaging in CAD Cardiac CT

SLIDE 7

12/17/16 7

Instrumentation/software

– Full coverage single beat – Higher temporal resolution 66 ms – Lower radiation <1 mSv – Model based iterative reconstruction

Assessments

– Plaque – Perfusion – Flow

Value-based Cardiac Imaging in CAD Cardiac CT

ACS LAD Positive remodeling (+), Soft plaque (+), Fibrous plaque (+), Calcification (-)

Motoyama et al. JACC 2007;50:319-26

ACS LAD Positive remodeling (+), Soft plaque (+), Fibrous plaque (+), Calcification (-)

Motoyama et al. JACC 2007;50:319-26 Adverse Plaque Features: positive remodeling, low attenuation plaque

5 10 15 20 25 nl (0/167) F(4/20) 1 F (1/27) 2 F (10/45) % with events

% of Patients Subsequently Having ACS Adverse Plaque Features on CCTA

1,059 pts with CCTA followed up for 27 ± 10 mo ACS developed in 15 patients. None had >75% stenosis in the culprit lesion at time of CCTA

Adverse features (F): positive remodeling low-attenuation plaques

Motoyama et al. JACC 2009;54:49-57

SLIDE 8

12/17/16 8

Motoyama, et al JACC 2015 N=3,158; 88 ACS in mean f/u 3.9 ± 2.1 years HRP: positive remodeling or low attenuation plaque; SS: ≥70% stenosis

High risk plaque Features: Predict ACS HRP and Significant Stenosis: Complementary

81% stenosis 70% stenosis

Shmilovich, Cheng, et al., Atherosclerosis 2011

APFs on CCTA Predict Ischemia

51 patients with coronary CTA and rest-stress 13N-ammonia PET
Ischemia automatically derived from PET. Plaque analysis by Autoplaq

CTA plaque assessment predicts regional PET flow

Dey et al Circulation Cardiovascular Imaging, 2015 Collaboration with Dr. Erick Alexanderson, Mexico City

Plaque Features on CCTA Add to Stenosis in Prediction of Impaired PET MFR True positive rate False positive rate

Composite Score 0.83 (0.79-0.91) Stenosis 0.66 (0.57-0.76)

p = 0.005* Dey et al Circulation Cardiovascular Imaging, 2015

153 vessels; 51 patients
Noncalcified Plaque Burden

Strongest Plaque Variable Dey, et al Circ CV Imaging 2015 ML of Clinical, stenosis,and plaque variables

SLIDE 9

12/17/16 9

Dey et al JCCT 2009, Dey et al JCCT 2014, Diaz Zamudio et al Radiology 2015

Autoplaq: Automated method for quantitative plaque characterization

% Diameter Stenosis
% Area Stenosis
NCP, CP, total plaque volume/burden
Low-density NCP plaque volume/burden
% NCP/Total plaque Volume
% Aggregate plaque volume
Remodeling index
Contrast density difference
Minimum luminal area, lesion length
Reproducible, quantitative of global plaque burden
Assessing response to therapy
Patient managment and clinical trials
CAC not useful for this purpose
Could extend application of CCTA to

asymptomatic patients

Serial Quantitative Coronary Plaque Assessment

CT Perfusion Predicts Ischemia: CTP + CTA CORE320: 64-year-old male with chest pain

Rochitte C E et al. Eur Heart J 2014

CT Perfusion Predicts Ischemia: CTP + CTA CORE320: 64-year-old male with chest pain

Rochitte C E et al. Eur Heart J 2014

SLIDE 10

12/17/16 10 Non-Invasive FFRCT

Source: Min JK et al. J Cardiovasc Comput Tomogr 2011, Min JK et al. Am J Cardiol 2012; Min JK et al. J Cardiovasc Comput Tomogr. 2012; Grunau GL et al. Curr Cardiol Report; Min JK et al. JAMA 2012; Koo et al. J Am Coll Cardiol 2012

From typically acquired CCTA
Computational fluid dynamics
Stenosis
Vessel volume after lesion
Myocardial mass distal to

lesion

No additional acquisition, radiation
No modification to imaging

protocols

No administration of medications

Non-Invasive FFRCT

Source: Min JK et al. J Cardiovasc Comput Tomogr 2011, Min JK et al. Am J Cardiol 2012; Min JK et al. J Cardiovasc Comput Tomogr. 2012; Grunau GL et al. Curr Cardiol Report; Min JK et al. JAMA 2012; Koo et al. J Am Coll Cardiol 2012

From typically acquired CCTA
Computational fluid dynamics
Stenosis
Vessel volume after lesion
Myocardial mass distal to

lesion

No additional acquisition, radiation
No modification to imaging

protocols

No administration of medications
Limitations:
Must send to HeartFlow
Significant added expense

FFR 0.65 = Lesion-specific ischemia FFRCT 0.62 = Lesion-specific ischemia LAD stenosis FFRCT 0.87 = No ischemia RCA stenosis FFR 0.86 = No ischemia

FFRCT for Lesion-Specific Ischemia

Case 1 Case 2

CT ICA and FFR FFRCT CT FFRCT ICA and FFR FFRCT <.80 CT >50

NXT Per-Vessel: FFRCT vs FFR

N=484

FFRCT diagnostic accuracy superior to both CT and ICA stenosis

Norgaard et al JACC 201 ICA ≥50

SLIDE 11

12/17/16 11

FFRCT <.80 CT >50

NXT Per-Vessel: FFRCT vs FFR

N=484

FFRCT diagnostic accuracy superior to both CT and ICA stenosis

Norgaard et al JACC 201 ICA ≥50

Stenosis Severity, Plaque Characteristics*, and FFRCT Predict Lesion-Specific Ischemia (FFR)

Gaur, et al EHJ 2016 484 vessels in 254 patients (NXT) * Autoplaq

LAD Diag 1 Diag 2 RCA LCX

N=10,030; Duration: 5 years; 44 ACD / 25 clinical parameters

Motwani…Slomka, et al: EHJ 2015

Prediction of MACE with Coronary CTA Machine Learning with CCTA and Clinical Variables

Cardiac MRI in CAD

anatomy & morphology function & wall motion perfusion coronary plaque coronary MRA Viability

SLIDE 12

12/17/16 12

CMR Push-button cine T1 mapping

No ECG Free-breathing Push-button Standard (MOLLI) ECG-triggered Breath-held

2000 1500 1000 500 T1 (ms)

Christodoulou AG, Shaw JL, Sharif B, Li D ISMRM 2016 Shaw JL, Christodoulou AG, Sharif B, Li D SCMR 2017

Single slice: 1 minute 3D whole heart: 10 minutes

End-diastolic myocardial T1

Push-button MOLLI

1200 1250 1300 1350 1150

p = 0.83, n = 10

An entire CMR exam in one 30-minute scan:

Cine
Native T1
Rest Perfusion
Stress Perfusion
Late Gadolinium Enhancement
Extracellular Volume Fraction

No ECG leads No breath holds No slice orientation scouts

CMR: The Future within Reach

Christodoulou AG, Shaw JL, Sharif B, Li D ISMRM 2016 Shaw JL, Christodoulou AG, Sharif B, Li D SCMR 2017

Diagnostic accuracy of CMR perfusion for ischemic stenosis with FFR as reference: meta-analysis

Per patient (n=650):90% sensitivity; 87% specificity Per vessel (n=1073): 89% sensitivity; 86% specificity

Li M, et al JACCi 2014

Biograph mMR Simultaneous, whole-body molecular MR (PET/MR)

SLIDE 13

12/17/16 13

Value-based Cardiac Imaging in CAD

Technologic improvements
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure

Challenges

Value-based Cardiac Imaging

Value = quality/cost

– Quality ≈ outcomes – Cost: total costs related to episode of care

New paradigm:

– Only tests/procedures that add value will be “purchased”

Value-based Cardiac Imaging

Value = quality/cost

– Quality ≈ outcomes – Cost: total costs related to episode of care

New paradigm:

– Only tests/procedures that add value will be “purchased”

Value-based Cardiac Imaging

Value = quality/cost

– Quality ≈ outcomes – Cost: total costs related to episode of care

New paradigm:

– Only tests/procedures that add value will be “purchased” – Evidence will be required

SLIDE 14

12/17/16 14

Demonstrating Value of Imaging

Mosts meaningful outcomes:

– All-cause mortality – MACE: Cardiac death; + MI; + ACS – Quality of life

Value-based Cardiac Imaging in CAD

Technologic improvements
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure

Challenges

CT Coronary Artery Calcium (CAC)

CAC: a marker of CAD

– Burden of coronary atherosclerosis

Integrated lifetime effect of all risk factors

– Overcomes the limitations of FRS – Consistent evidence: incremental prognostic value

<1 mSv (~ mammogram) Single breath No contrast

MG 72 M 09/98

70 y/o male physician tennis player
Asymptomatic
Total cholesterol: 220 mg/dL; LDL 152
Recent exercise thallium scan was normal

SLIDE 15

12/17/16 15

MG 72 M 09/98

After normal thallium:

–No changes

MG 72 M 09/98

After normal thallium:

–No changes

Offered a coronary calcium scan

MG 72 M 09/98

CAC Results

Location # Calcified Lesions Calcified Plaque Volume (mm3) Calcium Score LAD 2 806 1063 LCX 4 645 782 RCA 2 830 1101 Total 8 2282 2946*

* 97th percentile

MG 72 M 09/98

After CAC scan:

– Started statin and aspirin – Changed eating habits – Lost 10-15 lbs within one month – Increased exercise

SLIDE 16

12/17/16 16

Time to Follow-up (Years) 0 (n=11,044) 1-10 (n=3,567) 11-100 (n=5,032) 101-299 (n=2,616) 300-399 (n=561) 400-699 (n=955) 700-999 (n=514) 1,000+ (n=964) c2=1363, p<0.0001 for variable overall and for each category subset.

Cumulative Survival

0.0 2.0 4.0 6.0 8.0 10.0 12.0 0.70 0.75 0.80 0.85 0.90 0.95 1.00

All Cause Mortality and CAC Scores: Long Term Prognosis in 25,253 patients

Budoff, et al. JACC 2007; 49: 1860-70

10.4 Fold

Increased Risk

Yeboah JAMA 2012 - MESA

CAC Superior to Novel Risk Markers for Prognosis

Yeboah JAMA 2012 - MESA

= FRS + CAC

CAC Superior to Novel Risk Markers for Prognosis

Yeboah JAMA 2012 - MESA

= FRS + CAC

Consistent findings in multiple population-based studies CAC Superior to Novel Risk Markers for Prognosis

SLIDE 17

12/17/16 17

Yeboah JAMA 2012 - MESA

= FRS + CAC

Consistent findings in multiple population-based studies CAC Superior to Novel Risk Markers for Prognosis

CAC 0 reclassifies ~ ½ of candidates as not eligible for statins

More targeted preventive treatment
Upscale or downscale
Improvement in risk factor profile1
Intensification of Rx2
Better adherence to Rx3,5
Dietary modifications4
Increased exercise4

CAC leads to better treatment / lifestyle

1 Rozanski et al, JACC 2011 (EISNER Study) 2 Nasir K et al, Circ Cardiovasc Qual Outcomes 2010 (MESA) 3 Kalia NK et al, Atherosclerosis. 2006 4 Orakzai RH et al, Am J Cardiol 2008 5 Taylor A t al, JACC 2008

Screening for Subclinical Atherosclerosis

CT coronary calcium: effective
No primary role for other imaging
Stress imaging useful for guiding management

when CCS is high (~10%)

Value-based Imaging in CAD Imaging for Prevention Coronary Calcium Screening

Outcomes: ↑↑
Costs: no Δ
Value: ↑

SLIDE 18

12/17/16 18

Value-based Cardiac Imaging in CAD

Technologic improvements
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure

Challenges

Coronary CTA

Sensitivity and specificity for CAD: ~ 95%/90%
Higher than all other modalities

– Per patient – Per vessel – Per segment

Coronary CTA

Sensitivity and specificity for CAD: ~ 95%/90%
Higher than all other modalities

– Per patient – Per vessel – Per segment

Very high negative predictive value for events

Coronary CTA

Sensitivity and specificity for CAD: ~ 95%/90%
Higher than all other modalities

– Per patient – Per vessel – Per segment

Very high negative predictive value for events
Very unlikely to miss high risk CAD

SLIDE 19

12/17/16 19

Primary Outcome - Length of Hospital Stay

8.6 hours 26.7 hours 62% 21%

Hoffman, et al ACC 2012

– Reduces length of stay and time to diagnosis* – Safely increases direct ED discharge rates – No increase in costs of care**

Consistent results in three large RCTs:

Goldstein, et al JACC 2011 Litt, et. al. NEJM, 2012 Hoffman, et al ACC 2012

Coronary CTA in Suspected ACS

Imaging for Acute Chest Pain Coronary CTA

Outcomes: no Δ or ↑*
Costs: ↓**
Value: ↑

* Length of stay and time to diagnosis: quality measures ** Large proportion of ED CCTAs: completely normal; Potential to reduce subsequent ED visits

Value-based Cardiac Imaging in CAD

Technologic improvements
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure

Challenges

SLIDE 20

12/17/16 20

Pre-test likelihood of CAD

Stress testing Low-to-Intermediate (15-85%)

Noninvasive Imaging for CAD Suspected CAD: Stable

Pre-test likelihood of CAD

Stress testing Low-to-Intermediate (15-85%)

Noninvasive Imaging for CAD Suspected CAD: Stable

Combining with CAC scanning adds strength

Pre-test likelihood of CAD

Stress testing Low-to-Intermediate (15-85%)

Noninvasive Imaging for CAD Suspected CAD: Stable

Could be performed without imaging (ECG alone)
Combining with CAC scanning adds strength

JONREB (61 F) 11-18-2013

Substernal chest discomfort; at times exertional Multiple prior normal SPECT MPI studies

SLIDE 21

12/17/16 21

JONREB (61 F) 11-18-2013

Substernal chest discomfort; at times exertional Multiple prior normal SPECT MPI studies

JONREB (61 F) 11-18-2013

JONREB (61F) CCTA 11/18/2013

LAD LCX RCA

JONREB (61F) CCTA 11/18/2013

LAD LCX RCA 3 year follow-up: occasional chest discomfort persists; no further testing

SLIDE 22

12/17/16 22

Pre-test likelihood of CAD

Coronary CTA Low-to-Intermediate (15-85%)

Noninvasive Imaging for CAD Suspected CAD: Stable

Growing as initial test

Patel et al. NEJM 2010;362:886-95

ACC National Cardiovascular Data Registry
397,954 patients with elective cath (663 hospitals)
only 37.6% had obstructive CAD (>70% stenosis)

Normal Non-Obstructive

p<0.0001

1-Vessel CAD

p<0.0001

2-Vessel CAD

p<0.001

3-Vessel/Left Main

p<0.0001

Survival Probability Survival Time (Years)

Source: CONFIRM Min et al. J Am Coll Cardiol 2011

Prognostic Value of CCTA CAD Extent / Severity

23,854 patients w/o known CAD (57+13 years), 2.3 year f/u

Normal Non-Obstructive

p<0.0001

1-Vessel CAD

p<0.0001

2-Vessel CAD

p<0.001

3-Vessel/Left Main

p<0.0001

Survival Probability Survival Time (Years)

Source: CONFIRM Min et al. J Am Coll Cardiol 2011

Prognostic Value of CCTA CAD Extent / Severity

23,854 patients w/o known CAD (57+13 years), 2.3 year f/u

Consistent findings in all populations studied to date

SLIDE 23

12/17/16 23

Normal Non-Obstructive

p<0.0001

1-Vessel CAD

p<0.0001

2-Vessel CAD

p<0.001

3-Vessel/Left Main

p<0.0001

Survival Probability Survival Time (Years)

Source: CONFIRM Min et al. J Am Coll Cardiol 2011

Prognostic Value of CCTA CAD Extent / Severity

23,854 patients w/o known CAD (57+13 years), 2.3 year f/u

“Warranty period” of a normal scan: likely 7 years

Prognostic Utility of CT Findings

Does risk increase as a function of CAD by CT?

Extent/Severity of CAD Findings Risk* *Adjusted or unadjusted

Elderly
Men and Women
No Hx CAD
CAC=0
No modifiable risk

factors

Known CAD
Post-CABG
Post-PTCA
Ethnicity
Diabetes
Young
Family history
LV dysfunction
Known CAD
Chronic total
cclusion
Non-obstructive

Dx

Smoking

Source: CONFIRM: Min et al. J Am Coll Cardiol 2011; Chow et al. Circ CV Imaging 2011; Villines et al. J Am Coll Cardiol 2011; Cheng et al. Circulation 2011; Cho et al. Circulation 2012; Chow et al. J Am Coll Cardiol 2011; Villines et al. Am J Cardiol 2012; Shaw et al. J Am Coll Cardiol 2012; Min et al. Eur Heart J 2012; Nakazato et al. Atherosclerosis 2012; Rana et al. Diabetes Care 2012; Otaki et al. J Nucl Cardiol 2012; Nakanishi et al. EHJ 2015. CONFIRM Registry, preliminary data

J Min

Revascularization for High-Risk Anatomic CAD

Mortality for CAD Stratified by Medical Tx vs. Revasc

Source: Min et al. Eur Heart J 2012

15,223 patients w/o known CAD followed for 2.1 years (IQR 1.4-3.3 yrs)
Post-test medical therapy vs. revascularization adjusted via propensity

score in a manner similar to that of RCT

High risk anatomic CAD inclusive of LM, 3VD and 2VD plus pLAD
Hypothesis: A CT-based strategy will result in improved outcomes

(death, MI, hospitalization for UA, major complications)

RANDOMIZE Anatomic (CCTA) Functional (Stress Testing) Suspected angina from CHD (n-=10,003)

SLIDE 25

12/17/16 25

Stable Chest Pain: SCOT-HEART RCT

Hypothesis: A CT-based strategy will improve the

diagnosis of angina secondary to CHD at 6 weeks

RANDOMIZE Standard of Care Standard of Care plus CCTA Suspected angina from CHD (n-4146) Additional Endpoints

Diagnosis of CHD
Changes in planned

investigations

Changes in therapy
Prediction of CHD death
r MI

Source: SCOT-HEART Investigators, Lancet 2015

12 chest pain clinics; 3 imaging sites J Min

50 100 150 200 250 300 350 400 50 100 150 200 250 300 350 400

Frequency Frequency

Preventative Therapy Anti- Anginal Therapy All Therapies

Cancellations New Treatments

CTCA + Standard Care Standard Care Preventative Therapy Anti- Anginal Therapy All Therapies

CTCA and Medical Therapy Baseline Compared to 6 Weeks

Overall Changes in Treatments: 23% versus 5%, P<0.001

Newby et al Lancet 2015

Scot-Heart

Post-hoc 50-Day Landmark Analysis

2.5 2.0 1.5 1.0 0.5 0.0

1 3

2073 1571 323 2073 1550 316 CTCA Standard Care

Follow Up (years) Proportion of patients with an event (%)

853 837

2 CTCA Standard Care 0-50 Days

2055 2054

Implementation Delay

HR 0.50 [0.28-0.90] P=0.020

Impact of Alterations in Therapy

CTCA Performed
Result Reviewed
Management Changed
Invasive Angiography Arranged
Prescription issued

…Newby, et al 2016

Hypothesis: Value of Coronary CTA in SIHD Depends on Pre-test CAD Risk Hypothesis: Value of Coronary CTA in SIHD Depends on Pre-test CAD Risk

Outcomes: ↑
Costs: no Δ or ↓
Value: ↑

Intermediate

SLIDE 27

12/17/16 27

Hypothesis: Value of Coronary CTA in SIHD Depends on Pre-test CAD Risk

Outcomes: ↑
Costs: no Δ or ↓
Value: ↑

Intermediate Hypothesis: Value of Coronary CTA in SIHD Depends on Pre-test CAD Risk Too Low or Too High

Outcomes: ↑
Costs: no Δ or ↓
Value: ↑
Outcomes: no Δ
Costs: ↑
Value: ↓

Intermediate

“What should I do about it?”

Medium Very Low Low High Borderline Can’t tell Assure Prevent + Stress + Cath

Level of risk

+ Stress

r cath

“What should I do about it?”

Medium Very Low Low High Borderline Can’t tell Assure Prevent + Stress + Cath

Level of risk

+ Stress

r cath

Gatekeeper

SLIDE 28

12/17/16 28

“What should I do about it?”

Medium Very Low Low High Borderline Can’t tell Assure Prevent + Stress + Cath

Level of risk

Gatekeeper

+ Stress

r cath

? Gateway

Non-Invasive FFRCT

Source: Min JK et al. J Cardiovasc Comput Tomogr 2011, Min JK et al. Am J Cardiol 2012; Min JK et al. J Cardiovasc Comput Tomogr. 2012; Grunau GL et al. Curr Cardiol Report; Min JK et al. JAMA 2012; Koo et al. J Am Coll Cardiol 2012

From typically acquired CCTA
Computational fluid dynamics
Stenosis
Vessel volume after lesion
Myocardial mass distal to

lesion

No additional acquisition, radiation
No modification to imaging

protocols

No administration of medications
Limitations:
Must send to HeartFlow
Significant added expense

FFR 0.71 BRADER (79M): Recent onset exertional CP/SOB Statin, ASA, ARB, BB

BRADER (79M) ICA 4/22/2016

Proximal to mid LAD :80% FFR 0.71 DES (3.25X18mm Xience Alpine) stented à 0%

SLIDE 29

12/17/16 29

BAUSTE (59M) Non-exertional CP/SOB DM/Insulin; Hypertension

Normal SPECT-MPI

2.5 5.3 11.4 47.6 82.4 20 40 60 80 100

1-24% 25-49% 50-69% 70-89% 90-99%

Frequency of Abnormal FFRct Across Lesion by CCTA Stenosis Category (Cedars-Sinai)

Frequency of FFRct across the lesion ≤0.8 (%) Maximal CT stenosis grade 810 vessels (393 patients)

N=132 N=381 N= 238 N=42 N=17

Unpublished 2016

2.5 5.3 11.4 47.6 82.4 20 40 60 80 100

1-24% 25-49% 50-69% 70-89% 90-99%

Frequency of Abnormal FFRct Across Lesion by CCTA Stenosis Category (Cedars-Sinai)

Frequency of FFRct across the lesion ≤0.8 (%) Maximal CT stenosis grade 810 vessels (393 patients)

N=132 N=381 N= 238 N=42 N=17

Unpublished 2016 Tamarappoo, Berman et al JNC 2010

Frequency of Abnormal SPECT-MPI by CCTA Stenosis Category

SLIDE 30

12/17/16 30

Tamarappoo, Berman et al JNC 2010

Frequency of Abnormal SPECT-MPI by CCTA Stenosis Category

Nearly Identical to FFRct Findings

Pre-test likelihood of CAD

Coronary CTA Low-to-Intermediate (15-85%)

Noninvasive Imaging for CAD Suspected CAD: Stable

Functional assessments (eg FFRct) may reduce unnecessary catheterization

Coronary CTA is less useful in several settings

CCTA MPI

High CAC
+
Small stents
+
Renal failure
+
Arrhythmia -

+

Morbidly obese
+
Can’t hold breath
+

Noninvasive Imaging for CAD Suspected CAD

Pre-test likelihood of CAD

Ischemia Testing Intermediate to High (50-100%)

SLIDE 31

12/17/16 31

Men and women
Sx and Asx
Diabetics
Renal Failure
Arrhythmia
High CAC Score
Obese
Known CAD

– MI – PCI – CABG

SPECT: Risk Increases as a Function of Stress Perfusion Abnormality

Extent/Severity of Perfusion Defects Risk* *Adjusted or unadjusted

Data from over 50,000 patients

BENLEO (78 M)

Asymptomatic

BENLEO (78 M)

QGS+QPS: Change Page

BENLEO (78 M)

SLIDE 32

12/17/16 32

log Hazard Ratio

1 2 3 4 5 6

% Myocardium Ischemic

12.5% 25% 32.5% 50%

Medical Rx Revasc * * *p<0.001

Post-SPECT Cardiac Mortality and Rx Given

Early Revascularization vs Medical Therapy

N=10,627 F/U: 1.9 ± 0.6 yrs Risk adjusted Hachamovitch, et al. (Cedars-Sinai) Circulation 2003

Ischemia on SPECT- MPI Predicts Benefit from Revascularization

No Known CAD Circulation 2003
No Known CAD + EF JNC 2006
Elderly

Circulation 2010

Prior revascularization

EHJ 2010

Prior MI

EHJ 2011

Hachamovitch, et al

Absolute MBF Assessment With Rb-82 PET Prognostic Value Over Regional Perfusion

10 20 30 40 50 20 40 60 80 100 120 seconds

LV GFADS LV VOI RV GFADS RV VOI MYO GFADS MYO VOI

El Fakhri, Sitek, et al. J. Nucl. Med 2005, Brigham & Women’s Hospital; Ziadi et al. JACC 2011

Upper Tertile Middle Tertile Lower Tertile 0% 2% 4% 6% 8% 10% 12% ≥10% 1-9% 0% Annualized Mortality ≥10% 1-9% 0% Upper Tertile 2.4% 0.3% 0.1% Middle Tertile 4.4% 4.0% 1.1% Lower Tertile 10.2% 6.0% 3.6%

P<0.0001

(416) (217) (119) (n) (190) (202) (195) (n) (321) (509) (614) (n)

Murthy VL, et al. Circulation 2011 N= 2,783 CD= 137

Coronary Flow Reserve (CFR) Predicts Mortality Independent of Perfusion Defects

SLIDE 33

12/17/16 33

CFR is Associated with Cardiac Events Independently of Stenosis and Modifies the Effect of Early Revascularization

329 patients referred for ICA after PET; median f/u 3.1 year for CV death or HF CFR and CAD prognostic index (ICA): independent predictors Significant interaction (p=0.039) between CRF and early CABG but not PCI Taqueti et al, Circulation 2015

CFR + or - vs Revasc + or - CFR + or - vs Type of Revasc

CFR low, revasc - CFR low, revasc + CFR low, CABG CFR low, PCI

CFR is Associated with Cardiac Events Independently of Stenosis and Modifies the Effect of Early Revascularization

329 patients referred for ICA after PET; median f/u 3.1 year for CV death or HF CFR and CAD prognostic index (ICA): independent predictors Significant interaction (p=0.039) between CRF and early CABG but not PCI Taqueti et al, Circulation 2015

CFR + or - vs Revasc + or - CFR + or - vs Type of Revasc

CFR low, revasc - CFR low, revasc + CFR low, CABG CFR low, PCI

ISCHEMIA Trial

International Study of Comparative Health Effectiveness with Medical and Invasive Approaches

Study Chair: Judith Hochman Principal Investigator: David Maron

Sponsor: NHLBI

RANDOMIZE Cath (Revasc+ OMT) No Cath (OMT) >10% Ischemia* LVEF >35% Blinded CCTA Exclude LM/NCA

3-6 yr. F/U

ISCHEMIA Trial

8,000 stable CAD

patients

3-6 yr. F/U

*SPECT, PET, echo, CMR Core lab verification

SLIDE 34

12/17/16 34

ISCHEMIA Trial ISCHEMIA Trial Intermediate-to-high

Outcomes: ↑
Costs: ↓
Value: ↑

Suspected/Known Stable CAD Hypothesis: Value of Ischemia Testing Depends on Pre-test Risk

0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009

% Prevalence Year Yearly Prevalence of Abnormal and Ischemic SPECT Myocardial Perfusion Imaging Studies between 1991 and 2009

% Ischemia % Abnormal SPECT

N=39, 515 Rozanski, et al JACC 2012 (Cedars-Sinai)

SLIDE 35

12/17/16 35

Low Intermediate-to-high

Outcomes: No Δ
Costs: ↑
Value: ↓
Outcomes: ↑
Costs: ↓
Value: ↑

Suspected/Known Stable CAD Hypothesis: Value of Ischemia Testing Depends on Pre-test Risk

Value-based Cardiac Imaging in CAD

Technologic improvements
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure:

Challenges

Value-based Cardiac Imaging in CAD

Technologic improvements
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure: PET, CMR

Challenges
Technologic improvements
Applications

– Prevention – Acute coronary syndromes – Stable ischemic heart disease – Heart failure

Challenges

Value-based Cardiac Imaging in CAD

SLIDE 36

12/17/16 36

Must improve outcomes or reduce costs

–Select right patient for testing

Imaging tests cannot improve outcomes unless

they result in improved therapy Can Cardiac Imaging in CAD Provide Value? Challenges

Must improve outcomes or reduce costs

–Select right patient for testing

Imaging tests cannot improve outcomes unless

they result in improved therapy Can Cardiac Imaging in CAD Provide Value? Challenges

* ≥ 50% stenosis on CCTA

PROMISE: Prevalence of CAD*

Douglas, et al NEJM 2015

SLIDE 37

12/17/16 37

Must improve outcomes or reduce costs

–Select right patient for testing

Imaging tests cannot improve outcomes unless

they result in improved therapy Can Cardiac Imaging in CAD Provide Value? Challenges

Technology/assessments will improve across

modalities and clinical settings

Applications providing value will dominate
Value

– Will depend on the clinical setting – Requires linkage to therapeutic change – Evidence will be required

Value-based Cardiac Imaging in CAD

Thank you very much