Transcript Engineering in Medical Research - CEProfs

Advanced Physiologic Monitoring Laboratory
Texas Heart Institute
TAMU
ENGR 281
ESP
A Brief Look at
Engineering in Medical Research
April 5, 2006
Alan Brewer, BSBE, MBA
Advanced Physiologic Monitoring (APM)
Disclosures
Major portions of this research were supported by a
grant from TATRC, a department of the U.S. Army, under
a sub-contact from UT-HSC-H. Other smaller portions of
this research were supported by various other grants
including the MacDonald Foundation.
The APM department expresses its gratitude and
appreciation to all supporting parties.
Advanced Physiologic Monitoring (APM)
Team Members







Dr. S. Ward Casscells, III, MD – PI
Dr. Amany Ahmed, MD
Dr. Muhammad S. Munir, MD
Dr. K. J. Shankar, MD
Dr. Igor Stupin, MD
Alan Brewer, BSBio Engr, MBA
Bioengineer (F/T) – to be hired
Texas Medical Center – Houston, TX
Texas Heart Institute
Texas Medical Center – Houston, TX








Largest Medical Center in the World
Direct Employment - over 65,000
Approximately 45 Institutions including over 12
separate hospitals
Two Medical schools, Dental, Pharmacy, four Nursing
schools, School of Public Health, Graduate School of
Biological sciences
Affiliations with UT, Texas A&M U, Rice U and U of H
Immense amounts of research continuously underway,
$3.5B for 2000-2004 alone
Over 5M patient visits per year, 6600+ beds
22,000+ students
Texas Medical Center – Houston, TX
Academic and Research Institutions













Baylor College of Medicine
The University of Texas M. D. Anderson Cancer
Houston Academy of Medicine-Texas Medical Center Library
Houston Community College System, Health Science Programs
Prairie View A&M University, College of Nursing
Rice University
Texas A&M University Health Science Center - Institute of Biosciences and
Technology
Texas Heart Institute
Texas Southern University College of Pharmacy & Health Sciences
Texas Woman's University Institute of Health Sciences-Houston
University of Houston College of Pharmacy
The University of Texas Health Science Center at Houston - Dental Branch,
Graduate School of Biomedical Sciences, Harris County Psychiatric Center,
Medical School, School of Health Information Sciences, School of Nursing,
School of Public Health
Michael E. DeBakey High School for Health Professions
Potential Roles for Engineers in
Medical Research
 Experiment Design
 Good hypothesis development
 Insure endpoints are “testable”
 Control for extraneous or confounding influences
 Prototype Device Design and Construction
 Materials
 Software
 Safety
 Data Collection, Analysis and Interpretation
 DAQ Implementation
 Optics
 Telemetry
 Complex Imaging
 Project management
Why is a modern pacemaker
(or many other medical devices)
similar to a space vehicle?
or a down hole instrument package? or an undersea probe? etc?
 Operate in a hostile environment
 Measure environment continuously and react independently
 Impractical to service more than infrequently, thus high
reliability is paramount
 Must telemeter real-time and stored data to “mission control”
upon command
 Highly advantageous to be remotely programmable as
desired
 Data up- and down-links must be secure from interference
 Very energy efficient to maximize battery life
 Small in size and weight
 Able to compress and store data efficiently
 Small program footprint
Past APM accomplishments
Manuscript regarding CHF and Core Temperature
published in Am Heart Journal in May 2005
American Heart Journal, Vol 149, no 5, May 2005
APM Priority Projects
 Swine Study – Local Myocardial Temperature
and Ischemia Detection DAQ was developed
for this study
 Hamster Study – CHF and Hypothermia
 Prospective Clinical Study – CHF and
Hypothermia
APM Swine Study Engineering
Local Myocardial Temperature and Ischemia Detection
 Experiment design – Understanding heat
transfer
 Sensor selection and modification, device
prototyping
 DAQ development, LabVIEW
 Data analysis
APM Hamster Study Engineering
Body Temperature as a Predictor of Impending Heart Failure/CHF
 Experiment design
 Biotelemetry sensor selection
 Data analysis
Swine Study Project Status
April 3, 2006
 Pilot study - Four experiments performed
 Six more subjects planned and requested
 Sensor improvements made and validated –
more minor modifications desired
 Results from pig #1 and pig #4 were promising
Results from pig #2 and pig #3 were equivocal
 Surgery for epicardial temperature sensor
placement is non-complicated, but proper
location is still a possible issue
 Issues with trans-venous approach to RADI wire
in CS and balloon placements (esp. in RCA) for
swine subjects
Clinical Need for Better Ischemia
Detection
Driven by  12 million in USA with coronary artery
disease (CAD)
 One-half suffer angina pectoris
 Thus silent myocardial ischemia (SMI) is
signifcant problem, estimated that 2 to 4 %
of USA population have SMI.
 Pain is a poor predictor
 Patients with angina also have intermittent
SMI
 Anesthetized patients cannot register pain
symptoms, also some disabled patients
Clinical Need
Clinical Need driven by –
 Episodic diagnostic testing is
inconvenient and costly
 Exercise treadmill testing
 Low sensitivity, high false-positive rate
 Not tolerated by many patients
 Must be confirmed by radionuclide imaging techniques
(perfusion scintigraphy or exercise ventriculography) or
stress echocardiography
 Holter or 30-day ambulatory monitoring not highly specific
 e.g. LBBB, etc. can mask SMI in H-ECG
 Serial testing as method of monitoring for SMI is even
more problematic
Clinical Need
Clinical Need driven by –
 Currently, real-time diagnosis of SMI by ECG is
problematic
 Electrolyte imbalance, Rx side-effects, and noncardiac events can manifest with S-T segment
shifts in ECG
 Sensor positioning relative to local ischemia,
and near-field versus far-field signal amplitudes
complicate using intra-cardiac ECG for SMI
detection
Ischemia Project - Added Opportunities
Clinical opportunities –
 As feature for implantable cardiac
devices (e.g. pacemakers and ICD’s)
 There is a spectrum of possible
utilizations of this technology in such
devices
 (one end of the spectrum) - Stored
diagnostic for physician retrieval and
use at time of device follow-up
 (opposite end of the spectrum) Potential modulator/input to
automatically programmed pacing
parameters (e.g. upper-rate tracking
limit)
Project History – Prior Studies
 Literature states that one-half of
myocardial heat production is carried
away convectively by the coronary
blood flow
 Other significant heat fluxes are to the
lungs, airways, great vessels and the
cardiac blood flow
 Open-chest themography shows
distinctly cooler regions of the
epicardium result from coronary
ligation
 Closed-chest heat fluxes are more
complex and difficult to model
 Predictions were for small amplitude
thermal signatures
Why is the thermal signal small?
 Normally other
significant heat fluxes
are to the lungs,
airways, great vessels
and the cardiac intracavity blood flow
• Once a local region
becomes cooler, the
direction of heat flux
will reverse from the
normal direction
Project History – Canine Studies
 Team initially pursued a closed-chest
experiment using endovascularly placed
sensors in a canine model
 Right atrium blood flow
 Coronary sinus blood flow
 Right ventricular apex
myocardium
Project History – Canine Studies
But…
 Long procedure times
 Well-developed canine collateral myocardial circulation
 Sensor positioning issues – RV not
sensitive as hoped
 RADI device used for CS
temperatures
 Small amplitude signals
 This clarified need for accuracy,
resolution and drift
Swine Project - Hypotheses
Hypotheses generated that  Ischemic myocardium quickly will become colder
than surrounding non-ischemic myocardium
 Temperature drop with sudden ischemia will be
as or more rapid than changes in surface 12-lead
ECG morphologies.
 Temperature drop with sudden ischemia will be
as or more widespread than locally measured
changes in the intra-cardiac electrogram
morphologies.
 Temperature in the coronary sinus will show
distinct changes if significant volumes of
myocardium become ischemic.
Instrumentation Development
“DAQ” and
sensor system
 Hi-res (0.05 ° C),
reasonably rapid
response (tc ≈ 1.5
sec) and 1000
samples/sec
LabVIEW-based
 3 temperatures, plus
12-lead ECG and up
to 3 channels of intracardiac electrograms
DAQ Design Issues
for Pilot study
 Other user requirements
 Minimize custom components
 Simultaneous display of 15-lead ECG and 3 temperature
channels
 Automatic file saving and labeling
 Annotation markers and comments record capability
 Adjustable ECG filters to accommodate electrically noisy OR
Swine Study - Sensor Design Issues
Which is better –
Thermistor or
Thermocouple?
Adapted sensor
 Thermocouple sensor
 No intrinsic heat production
 Accurate to 0.05 °C goal if
properly calibrated
 Small yet rugged and reusable
 Rapid time constant, << 1 sec
when bare sensor
Bare sensor
Sensor Design Issues, cont
Thermocouple sensor
 Hard to place bare sensors in
myocardium for pig # 1
 Clinical requirements for swine
study allowed changes for improved
surgical placement
 Added sewing needle to sensor
 Protects thermocouple and
provides for easy method to
secure sensor in position
 These changes increased time
constant of thermal response to
approximately 1.0 to 1.5 sec
2-0 prolene suture
inserted into PTFE
tube pre-shrinking
#2 sewing
needle attached
to PTFE shrink
tubing
Adapted sensor
Bare sensor
Sensor Placement
Placement designed to match each sensor
with a “Target Ischemic Zone”
Sensor
Position
 Anterior LV
PosteriorLateral LV
Lateral RV
Target
Zone/Vessel
 LAD
Circumflex
RCA
Advanced Physiologic Monitoring (APM)
Warning!
The following two slides contain graphic
medical images of swine hearts ex-vivo
Swine Study
Epicardial Sensor Placements
Implantation is similar to
temporary epicardial
pacing wire
Approx
4 mm
depth
Clip
Swine Study
Epicardial Sensor Placements
Sensor placement
complications
Too shallow
Clip
Hypertrophic
heart with
grossly
thickened left
ventricle
Swine Project – Preliminary Results
In our closed-chest swine with PCI-induced ischemia model -
Ischemic myocardium becomes colder (by about 0.5 ° C)
than other non-ischemic myocardium in approx 2 minutes
or less.

The temperature drop due to acute ischemia occurs at the
same time or slightly before changes are observable in
surface 12-lead ECG morphologies.

The temperature of non-ischemic myocardium remains
stable and consistently tracks with minor body temperature
variations

The temperature of non-ischemic myocardium neither
increases nor decreases when other portions of the
myocardium suffer an ischemic insult

Detailed data analysis ongoing
Pig 2 CFX Occlusion – 2005-12-02 – 12:14 pm
Pig 2 CFX Occlusion – 2005-12-02 – 12:15 pm
Advanced Physiologic Monitoring (APM)
Future research opportunities
 Extend this study to a 30-day
chronic model using telemetry
 Look in detail at confounding
influences
 Environmental temperature
 Fever
 Activity and exercise
Project Status – Hamster Study
Expanded Study Design Plans
BIO-TO-2 Cardiomyopathic Syrian Hamsters
ETA-F20 or CTA-F40 transmitter (DSI) implanted
intraperitoneal or Sub-Q under anesthesia
CONTROL
No therapy
Room
Temperature
21°C
CONSTANT
INTERVENTION
Sustained
Thermotherapy
(TT) of 25 °C
SELECTIVE
INTERVENTION
Sustained TT of
25 °C in event of
hypothermia.
All animals followed and monitored via implanted device
and manually until symptoms exacerbate,
then euthanized @ ≤ 6 months
Project Status
Prospective CHF Study
“Low Body Temperature is a Marker of Poor
Prognosis in CHF Patients”
 Approved by SLEH/THI IRB for non-significant risk study
status
 CHF NYHA III and IV patients already followed by a
CHF clinic and optimally medically managed
 All patients record temp BID
 Follow for 6 months
 Compare hospitalization rates, mortality, other
endpoints; and track utilization
 22 patients now recruited, of 140 patients planned
Project Status
Prospective CHF Study
Hypothesis: In HF patients, low and/or
falling body temperature can predict a
poorer prognosis, compared to HF
patients whose body temperature
remains consistently normothermic
Prospective Study Proposal
Study Design
Population: Patients enrolled at THI HF Clinic
Recruit 140 patients per inclusion & exclusion criteria
1. Pts. receive ‘optimal medical management’
2. Pts. record oral temp (BID) within 1 hour after
waking up and within 1 hour prior to retiring
All Pts.’s that complete data reviewed against endpoints
1. Normothermic vs. hypothermic/falling temp
2. Negative for endpoints vs. positive for endpoints
Prospective Study Proposal
Primary Endpoints
1. Hospitalization for CHF
2. Mortality
3. Significant CHF decompensation episode not
requiring hospitalization, but necessitating
unscheduled physician visit(s) and adjustment to
CHF therapy
4. Assignment to “Status 1A” on heart transplantation
list
5. Implantation of LVAD device
Prospective Study Proposal
Inclusion Criteria
1. CHF (Tx for at least 30 days prior)
2. LVEF ≤ 40% (measured within last 1
year)
3. NYHA Class III or IV
4. Age ≥ 18 yrs and ≤ 70 yrs
5. Stable Rx’s for ≥ 7 days
Prospective Study Proposal
Exclusion Criteria
1.
2.
3.
4.
5.
6.
7.
8.
9.
Recent implantation of CRT ICD or CRT pacemaker (within 30
days)
Recent CVA (within 30 days)
Recent participant in another clinical trial (within 30 days)
Significant liver disease
Thyroid Dysfunction (TSH level outside of normal limits; or
current hyper- or hypo-thyroid abnormality)
Current alcohol or other drug abuse
Active infection or sepsis
Living conditions are without reasonable heat and AC
Unable or unwilling to provide consent
Prospective Study Proposal
Routine Follow-up
1. QOL survey – Minn. Living with Heart
Failure Survey (administered monthly)
2. In each clinic visit BNP, Bio-impedance
hemodynamic state measurement, Na+,
BUN, creatinine, uric acid etc. will be
checked
Alan Brewer
 Served on Bioengineering Curriculum Advisory Board since
inception
 BSBE ’77 TAMU, MBA ’84 HBS
 Texas Heart Institute as Bioengineer
 Instrumentation Laboratory, Inc. - Mass
 Hewlett-Packard Medical (now part of Philips) - Mass
 Intermedics, Inc. - Texas
 Alaris Medical Systems - California
 Texas Heart Institute – Back to Texas
 Prior Guest lecturer (’04 and 05) at TAMU for Entrepreneurial
Studies course
Venture Capital in Perspective
Sources of Equity Capital for Start-ups
Classical Venture Capital is
defined to consist of seed, early,
and expansion-stage financing.
VC$
28%
Debt Financing
Other
Sources
SBA loan guarantees only help 2-3% of all
startups
Grants
SBIR/STTR grants - from various agencies
72%
Equity Financing
72% from informal sources (founders, family,
friends, and foolhardy strangers (“4 F’s”),
corporate VC and professional VC
Source: Kauffman Center for Entrepreneurial
Leadership, Babson College, 2002.
Fewer than 2% of all new business
(per Inc. magazine) are started with
venture capital
Advanced Physiologic Monitoring (APM)
Thank you
and
Questions?