Coronary artery disease: Implications of Preoperative Imaging and functional analysis of areas stenosis in planning CABG/PCI Clinical implication of genomics in the use of bilateral internal mammary artery for coronary artery bypass graft

1
Coronary artery disease Implications of
Preoperative Imaging and functional analysis of
areas stenosis in planning CABG/PCI Clinical
implication of genomics in the use of bilateral
internal mammary artery for coronary artery
bypass graft

• Giovanni Ferrari PhD Assistant Professor of
  Surgery
• University of Pennsylvania School of Medicine
  Penn Cardiovascular Institute Institute for
  Translational Medicine and Therapeutics
• Children's Hospital of Philadelphia, Abramson
  Research Building
• E Giovanni.Ferrari_at_uphs.upenn.edu

2

• Disclosure
• Scientific consultant for The Valley Hospital,
  Ridgewood, NJ - USA

3

• a biomarker is a measurable indicator of the
  severity or presence of some disease state.
• Imaging biomarkers have many advantages
• They are usually noninvasive, and they produce
  intuitive, multidimensional results.
• Yielding both qualitative and quantitative data,
  they are usually relatively comfortable for
  patients.
• When combined with other sources of information,
  they can be very useful to clinicians seeking to
  make a diagnosis.

4
Objectives
Translational Research is a transformative
opportunity for CT Surgery Departments
Work with surgeons and cardiologists to
understand the fundamental processes of specific
CT diseases IP and Patents NIH Extramural
Support Visibility (Meetings and
Publications) Increase enrollment and retain pts
5

• LIMA widely accepted as a superior conduit due
  to his exceptional long-term patency compared
  to the SV and RA grafts6
• Studies have demonstrated RIMA grafts having
  similar patency rates to LIMA6,7
• Despite proven clinical successes, BIMA use
  remains low among CABG patients (only 4.0 STS)
  in the US and around 12 in Europe 12

Figure from the Cleveland Clinic5
6

• Retrospective analysis of two propensity-matched
  groups of 928 BIMA and 928 LIMA-SVG patients from
  a single institution from 1994-2010 selected from
  6.000 CABG cases.
• CABG X1 including MIDCABG, Reoperations, combined
  valve CABG and or use of radial artery were
  excluded from the analysis
• Assessment of in-hospital to 30-day operative
  morbidity and mortality
• Follow-up of long-term survival up to 17 years by
  assessing all-cause mortality using the SSDI

7
    BIMA 928 (50.0) LIMA-SVG 928 (50.0) p Value
Age (mean SD) 60.9 9 62.1 9 0.004
Body Mass Index (mean SD) 27.6 4 28.9 5 lt0.001
Gender (male) 828 (89.2) 829 (89.3) 0.940
History of Smoking 598 (64.4) 635 (68.4) 0.069
History of Diabetes Mellitus 100 (10.8) 101 (10.9) 0.940
History of Renal Failure 13 (1.4) 17 (1.8) 0.482
Elevated Cholesterol 625 (67.3) 605 (65.2) 0.340
History of Hypertension 631 (68.0) 641 (69.1) 0.617
Prior Myocardial Infarction 416 (44.8) 420 (45.3) 0.852
Prior Cerebrovascular Accident 35 (3.7) 23 (2.5) 0.109
History of Cardiovascular Disease 39 (4.2) 29 (3.1) 0.217
History of Peripheral Vascular Disease 84 (9.1) 92 (9.9) 0.526
Chronic Lung Disease 17 (1.8) 20 (2.1) 0.962
History of Arrhythmias 56 (6.0) 69 (7.4) 0.229
Hematocrit 41.1 4 40.8 5 0.737
Left Main Disease 304 (32.8) 295 (31.8) 0.655
Triple Vessel Disease 752 (81.0) 761 (82.0) 0.860
Left Ventricular Ejection Fraction 52 11 51 12 0.091
8
Perioperative Characteristics and Outcomes
BIMA 928 (50.0) LIMA-SVG 928 (50.0) p Value
Surgery Acuity      
Elective 424 (45.7) 405 (43.6) 0.772
Urgent 482 (51.9) 498 (53.7)  
Emergent 21 (2.3) 23 (2.5)  
Emergent/Salvage 1 (0.1) 2 (0.2)
Arteries Bypassed 3.6 3.4 lt0.0001
Off-Pump 454 (48.9) 476 (51.3) 0.307
Intra-Aortic Balloon Pump Used 89 (9.6) 96 (10.3) 0.588
STS Predicted Mortality Rate 1.1 1.2 0.136

Mortality (In-hospital to 30 days) 7 (0.8) 10 (1.1) 0.465
Post-Operative Permanent Stroke 5 (0.5) 5 (0.5) 1.000
Post-Operative Renal Failure 6 (0.6) 7 (0.8) 0.781
Post-Operative MI 18 (1.9) 22 (2.4) 0.523
Re-Operation for Bleeding 10 (1.1) 16 (1.7) 0.236
Deep Sternal Wound Infection 3 (0.3) 3 (0.3) 1.000
Sepsis 8 (0.9) 4 (0.4) 0.247
Post-Operative Atrial Fibrillation 185 (19.9) 178 (19.2) 0.682
Required Blood Transfusion 332 (35.8) 292 (31.5) lt0.05
Post-Surgery Length of Stay (Days) 6.9 4 6.9 4 0.978
9
    BIMA Off-pump 454 (48.9) BIMA On-pump 474 (51.1) LIMA-SVG Off-pump 476 (51.3) LIMA-SVG On-pump 452 (48.7) p Value
Mortality (In-Hospital to 30 Days) 5 (1.1) 2 (0.4) 5 (1.1) 5 (1.1) 0.632
Postoperative Permanent Stroke 0 (0.0) 5 (1.1) 0 (0.0) 5 (1.1) 0.022
Postoperative Renal Failure 5 (1.1) 1 (0.2) 3 (0.6) 4 (0.9) 0.401
Postoperative Myocardial Infarction 6 (1.3) 12 (2.5) 15 (3.2) 7 (1.5) 0.184
Reoperation for Bleeding 5 (1.1) 5 (1.1) 10 (2.1) 6 (1.3) 0.491
Deep Sternal Wound Infection 0 (0.0) 3 (0.6) 0 (0.0) 3 (0.7) 0.109
Sepsis 2 (0.4) 6 (1.3) 2 (0.4) 2 (0.4) 0.284
Postoperative Atrial Fibrillation 83 (18.3) 102 (21.5) 81 (17.0) 97 (21.5) 0.202
Required Blood Transfusion 152 (33.5) 180 (38.0) 136 (28.6) 156 (34.5) 0.021
Post-Surgery Length of Stay (Days) 6.3 3 7.6 4 6.1 3 7.8 6 lt0.0001
Surgical Results Based on Pump Status
10
Overall Long-Term Survival
BIMA over LIMA use in CABG yielded a survival
advantage of 10 at 10 years and 18 at 15 years.
Year(s) of Follow-Up 1 5 10 15
BIMA 99 96 89 79
LIMA-SVG 99 91 79 61
11
Long-Term Survival A Sub-Analysis
Off-pump BIMA patients benefited from a 22
survival advantage over on-pump LIMA-SVG patients
at 15 years of follow-up.
Year(s) of Follow-Up 1 5 10 15
Off-Pump BIMA 99 96 89 81
On-Pump BIMA 99 95 88 78
Off-Pump LIMA-SVG 99 92 80 61
On-Pump LIMA-SVG 99 90 77 59
12

• Overall rates of morbidity and mortality for CABG
  were very low and correlated to STS predictions.
• The risk of stroke was low overall (1), the off
  pump groups had the lowest risk in the propensity
  matched groups.
• The use of BIMA provided superior survival
  advantage at 10 and 15 years.
• Subgroup analysis of the pump status/IMA grouping
  confirms the independent increase in risk of
  late mortality associated with patients having
  on-pump LIMA-SVG.

13

• Cardiac Surgeons have gotten quite good at doing
  this operation (CABG) with continuously
  decreasing perioperative morbidity mortality over
  the past decades.
• LIMA to LAD is a good strategy to improve
  survival in patients with CAD treated surgically.
  Use of BIMAs adds an incremental benefit that
  seems to further grow as follow up lengthens.
• In our series, use of BIMA did not incur into any
  additional risks or complications.

14
Limitations
Limitations

• Retrospective analysis, non-randomization of
  techniques or conduits. We did not account for
  coronary anatomy. Multiple surgeons with
  different philosophies and/or technical skills.
• No information is provided regarding cause of
  death, incidence of MACE, Need for reoperation or
  PCI during follow up period, patency etc.
• Many other factors other than surgical grafting
  technique may affect long-term survival and may
  have not be accounted for in our series

15

• Comparative genome-wide transcriptional analysis
  of human left and right internal mammary
  arteries. Genomics 104, 3644 (2014).

• The complete genome sequence demands a holistic
  approach to understanding human disease.
• Genes cannot be considered in isolation, but
  rather in the context of the other genes encoded
  within the genomes working in the context of
  networks and pathways.
• A fundamental question we want to ask is what
  genesand what processesdiffer between
  biological states.

16

• DNA microarrays allow gene expression levels to
  be measured by assaying RNA abundance for each
  gene.
• RNA is extracted, labeled, and hybridized to an
  array of DNA fragments representing each gene in
  the genome.

Quackenbush J. Microarray analysis and tumor
classification. NEJM, 2006.
17
Transcriptional Comparison of IMAs and Aorta

• The objective of this project was to explore and
  compare gene expression profiles of the aorta and
  the left mammary and right mammary arteries. The
  motivation was that the mammary arteries are
  generally immune from atherosclerosis while the
  aorta and coronary arteries are not.
• Primary hypothesis LIMA and RIMA are similar in
  their expression profiles.
• Secondary hypothesis aorta is different in
  expression from internal mammary artery, which
  may explain the differences in performance and
  outcome

18

• Selected a cohort of 32 patients from our CV
  biobank resulting in 32 paired LIMA/RIMA/Aorta
  for further molecular analysis
• Mammary and aortic tissues were harvested at the
  time of CABG 1 cm sample of the distal aspect
  of LIMA and RIMA. Aortic button taken from the
  punch aortotomy
• All samples stored using standard OCT gel or RNA
  later as a buffer in a secure -80C freezer

19
Approach
RIMA
Extract and LabelRNA
LIMA
Extract and LabelRNA
20
LIMA vs RIMA
Using Fishers Exact Test and looking at gene
functional classes, we see small, slightly
significant differences in gene expression
between LIMA and RIMA for a small number of genes
associated with T-cell properties.
21
BIMA
Extract and LabelRNA
Aorta
Extract and LabelRNA
22
BIMA, Lipid Metabolism, and Cellular Movement

• Red Up-regulation in IMA
• Green Down-regulation in IMA
• Lipid metabolism cellular movement
  atherosclerosis
• Again, analysis presents a central role for NF-?B
  

23
BIMAs and the Inflammatory Response

• Red Up-regulation in IMA
• Green Down-regulation in IMA
• Analysis shows lower levels of inflammatory
  proteins, highlight a central role for NF-?B in
  BIMA

24
BIMA vs AortaDifferentially Expressed Pathways
25
Differentially Expressed Pathways
Differential Expression in BIMAS vs Aorta
Atherosclerosis Signaling
Eicosanoid Signaling
Cytokine Macrophage Migration Inhibitory
Hepatic Fibrosis/Hepatic Stellate Cell Activation
Macrophage Migratory Inhibitory Factor
Arachidonic Acid Metabolism
Liver X Receptors/Retinoid X Receptor Activation Signaling
Peroxisome Proliferator activated Receptor Signaling
26
BIMA vs Aorta

• The genes in the Atherosclerosis Signaling
  pathway and Inflammatory Processes are
  significantly down-regulated in the BIMAs
  relative to aorta.
• Eicosanoid Signaling pathway genes and
  Arachidonic Acid Metabolism among others are also
  expressed at lower levels in the BIMA tissues.
• These protective factors appear to be shared by
  the mammary conduits from different patients and
  as a group when compared to the aortic buttons

27
Summary

• The supporting transcriptional evidence presented
  here provides additional evidence necessary to
  persuade surgeons to favorably reconsider the use
  of BIMAs as the primary conduits for CABG.
• The BIMAs should be viewed as complex living
  metabolic biological units with the additional
  ability to protect the coronary artery target
  from atherosclerosis progression.

28
(No Transcript)
29
This project was supported by the Marjorie G
Bunnell Charitable Fund of The Valley Hospital
Foundation