In vitro• Total control of confounding variables

– Vasomotion, temperature changes, autoregulation, mean BP• Most accurate because vessel examined directly• Best for detailed information about mechanical properties of vessel material

In vivo (invasive)• Realistic clinical information• Limited by technical problems

– Measurement errors, transitory changes in diam. BP etc

In vivo (non-invasive)• Further technical problems

– Especially pressure

Measurement of blood pressure

• Invasive– Pressure catheter and transducer

• Non invasive– Sphygmomanometry

• Auscultation (by ear or automatically by microphone)• Oscillometry

– Volume clamp– Tonometry

• Invasive– Accurate reproduction of central pressure waveforms– Risk of thrombosis and arrhythmias

• Non-invasive– Quick, cheap, widely used– Lack of central pressure measurement– Requires skilled and experienced operators

Advantages/ drawbacks

Sphygmomanometry

www.fmshk.com.hk/sahk/lecture_noninvasive.pp

Pulse detector(stethoscope or microphone)

Manometer(mercury or capsule type)

Manometer(mercury or capsule type)

Sphygmomanometry

• 1896 Blood pressure cuff (Riva Rocci)

• 1905 First report of audible detection of heart sounds used with cuff (Korotkov)

• 1968 Microphone used for automatic pressure measurement (Stegall)

Sphygmomanometry

Capsule manometerReplacing mercury spymomanometerMercury sphygmomanometer

Korotkov Soundscaused by vibration collapse of the arterial wall??

www.fmshk.com.hk/sahk/lecture_noninvasive.pp

Cuf

f pr

essu

re

Systolic

Diastolic

– Korotkoff IV is a better indication of diastolic pressure according to theory

– However Korotkoff V is the commonly recommended measuring point except in pregnant patients because

• It is associated with less inter-observer variations

• It is easier to detect by most observers

Errors

• Korotkoff sounds compared to invasive blood pressure measurement– Korotkoff IV is on average 8mm Hg above the invasively

measured diastolic blood pressure– Korotkoff V is on average 2mm Hg above the invasively

measured diastolic blood pressure

Oscillometry

• Cuff round the arm

• Pressurise cuff (> systolic)

• Allow pressure to drop slowly to zero

• Measure pressure in the cuff during deflation

Oscillometry: set up

Pressure transducerPressure transducerAir pumpAir pump Bleed valveBleed valve

Micro-processor

Micro-processor DisplayDisplay

Principle of oscillometry

Variation of cuff pressure as cuff is deflated

Filt

ered

sig

nal

Of

cuff

pre

ssur

e

Limitations• Inaccurate / unreliable in shock patients• Inaccurate / unreliable in patients with arrhythmias

– The algorithm of measurement assumes a regular pulse, so the reading is unreliable in patients with irregular pulse

Advantages• No skill required• No subjective errors

Volume clamp

Air

Air

Infra red emitter

Detector

Artery

FingerPressure

Detectedsignal

Change cuffpressure

Measure cuffpressure

To pump

Diameter

Applanation tonometryDetects pressure of arterial pulsations through the skin

Problem:

AorticRadial

• Aortic and peripheral pressures are different.• The heart doesn’t care what the pressure is in the radial artery.• It only “sees” aortic pressure.• Aortic pressure is difficult (impossible?) to measure non-

invasively• Can we reconstruct the aortic waveform from the radial?

80

100

120

Systolic

Diastolic

Mean

Yes we can. At least in principle

• Record radial waveform with tonometry

• Apply inverse transfer function

• “Reconstruct” aortic waveform– What is an inverse transfer function?– How do we reconstruct the waveform?

Fourier analysis

36027018090

-2

-1

0

1

2

36027018090

-2

-1

0

1

2 H1 + H2H3

36027018090

-2

-1

0

1

2 H1 + H2 + H3H4

36027018090

-2

-1

0

1

2

Mean

H1H2

Measured

H1+H2+H3+H4

Pa(t) = pa0

+ pa1Cos(t - a1)+ pa2Cos(t - a2)+ pa3Cos(t - a3)+ ...

Pb(t) = pr0

+ pr1Cos(t - r1)+ pr2Cos(t - r 2)+ pr3Cos(t - r 3)+ ...

For each harmonic (n)

Transfer function phase = an - rn

Transfer function amplitude = pan / prn

aortic pressure radial artery pressure

Amplification of the pulse

AA - CA

CA - RA

AA - RA

How to derive the central pressure from peripheral measurements

• Compare Fourier series of “typical” aortic pressure waves with Fourier series of the radial pressure computed from tonometric measurements.

• Calculate the amplitude ratio and phase difference for each harmonic

• Apply this ratio and phase difference to each harmonic of the measured radial wave and reconstruct aortic wave that would when transmitted down the arm, produce the measured radial wave

Question• How well does the typical transfer function

apply to people of different ages and disease states

Answer• Surprisingly well considering the changes that

occur in the arterial system with age and vascular disease

• However, most believe that more work is needed to validate the method

Pressure transducers(for invasive measurement)

Fluid filled chamber

Stiff diaphragmMeasure its movementelectronically

To pressure to be measured,(via an intra arterial cannula)

Diaphragm manometer

Advantages• Cheap, disposable• easy to use• Accurate mean pressure

Disadvantages• Clotting in cannula, air

bubbles• Therefore errors in pulse

pressure

Pressure transducers(for invasive measurement - 2)

Cannula tip manometer Semi conducting strain gauge

Diameter may be as small as 0.67 mm

Advantages• High accuracy• Especially in very small

vessels

Disadvantages• No calibration possible when

in position• Expensive• Fragile

Pressure: comparison of methods

Method Sensitivity Invasive Advantages/ disadvantages

Auscultation + cuff

OK No Subjective, limited to arm or leg. Good in skilled hands

Oscillometry + cuff

OK No As above but less subjective. No mean pressure.

Catheter Good Yes Only direct way to measure in central vessels

Volume clamp Good No Limited to peripheral arteries but can do small ones

Tonometry V. Good No Superficial vessels only, sensitive to movement, good for carotid. No absolute P values. Can be calibrated against cuff methods

PPG V. Good No Superficial vessels only. Used as a pulse detector in conjunction with cuff. PROMISING

Flow Measurement

• Invasive– Electromagnetic flow velocimetry– Ultrasonic transit time

• Non invasive– Doppler ultrasound– Ultrasonic transit time – Optical (small superficial vessels only)– MRI

Flow measurement• 1870 Fick principle described

Flow in a given period of time = Amount of substance injected in that time/concentration difference before and after point of entry

• 1886 Fick method first used by Grehart & Quinquardt

• Modern instruments– Optical– Electromagnetic 1936-1937 Kolin– Ultrasonic transit time 1959– Ultrasonic Doppler 1961– MRI 1990’s (not commercial)

i.d. 0.5 - 26 mm

E = H.d.VInduced voltage

Magnetic field strength

Vessel diameter

Mean blood velocity

Electrode

Principle of Doppler flow velocimetry

v

c fc

cf

c vf'

cf' (c v)f

cf' cf vf

f' ff

vc

fvfc

c f'

Flow: comparison of methods

Method Sensitivity Invasive Advantages/disadvantages

Dilution Adequate No Cumbersome, slow,mean values only

Optical Good No Small superficial vessels only

Doppler OK No Absolute flow values difficultto measure

Transit time Good Yes None apart from expense andinvasiveness

E.M OK Yes Electrical noise, hard tocalibrate accurately

Diameter Measurement

• Mechanical• Optical• Ultrasonic

– Implanted crystals– Pulse echo

• Cine-angiography • MRI

Invasive Diameter Measurement

• Ultrasound (external transducers)• IVAS• TV• Mechanical• Cine angiography

Non-invasive diameter measurement

• Pulse echo ultrasound (direct)• PWV (indirect)

– Diameter wave– Flow wave– Pressure wave

Artery

Differential transformer

Springy stainless steel

Ultrasonic crystals(glued or sutured)Measure time delay

TV camera

Other diameter methods

Transmitter

Receiver

Measure time delay between transmitted and received pulse

Principle of pulse echo ultrasound

Diameter: comparison of methods

Method Sensitivity Invasive Advantages/disadvantages

Mechanical OK Yes Cumbersome, but insensitiveto wall movement

Optical Good Yes (no) Non contact but sensitive towall movement. N.I. methodonly measures rel. diam.

Ultrasound(crystals)

Very good Yes Difficult to set up, insensitiveto wall movement

Ultrasound(echo)

Good No Sensitive to wall/patientmovement, but only absolutenon invasive method

Elasticity measurement

• Direct– Stress

•pressure, tension, area, wall thickness

– Strain•length, diameter

• Indirect– Pulse wave velocity

•detect pressure, diameter or flow pulse

PWV Methods

• Pressure pulse– Tonometry

• Flow pulse– Doppler

• Diameter Pulse– PPG

Nature of the PPG Signal

• Commonly regarded as a measure of changes in tissue volume due to arteriolar and capillary blood flow time varying absorption of light or i.r.

• When detected in the vicinity of a large superficial artery, the signal is dominated by changes in the diameter (volume) of the artery.

SKIN

ARTERYFLOW

Downstreamprobe

MUSCLE/BONE

Upstreamprobe

Infra red emitter Detector

Loukogeorgakis, et al. (2002). Physiological Measurement 23: 581-96.

Optical detection of the diameter wave

PhotoPlethysmoGraphy (PPG)for pulse wave velocity measurement.How does it work?

• Infra red probes detect transitory change in conduit artery volume due to the passage of the pulse wave

• Measure time delay and distance between the probes

• Pulse wave velocity = d/t

• Pulse wave velocity (compliance)-1/2

t

d

LED (emitter)Photo-transistor (detector)

20 mm

20 mm

Validation experiments.

Comparison of PPG with• Echo Tracking.

– Does PPG method really measure large artery diameter?

• Doppler.– How well do PPG derived pulse wave transit times

compare to measurements using an established method?

• Intra-arterial pressure wave.– Do transcutaneous transit time measurements compare

with intra-arterial ones?

PPG/Echo tracking methods

PPG

Probes on the posterior tibial artery

Probes on the radial artery

PPG

NIUS ultrasoundprobe

NIUS ultrasoundprobe

PPG

Ultrasound

0 2 4 6 8 10

Frequency (Hz)

Ph

ase

0

100

200

300

400

Ultrasound

Rel

ativ

e am

plit

ud

e

0.01

0.1

1PPG

PPG/Echo Tracking - Conclusions.

• PPG reproduces the diameter wave with reasonable fidelity, when compared to high precision echo tracking system.

• Timing of the foot is close

Validation experiments.

Comparison of PPG with• Echo Tracking.

– Does PPG method really measure diameter?

• Doppler.– How well do PPG derived pulse wave transit times

compare to measurements using an established method?

PPG/Doppler methods

PPGDoppler

Probes on the posterior tibial artery

Probes on the radial artery

PPG

Doppler

ECG

PPG

ECG used as time reference

Doppler

100

150

200

250

300

350

TT PPG [ms]

100 150 200 250 300 350

TT Doppler [ms]

y = 0.90x + 12.8 r = 0.95

Comparison of PPG and Doppler transit times

Leg

Arm

Comparison of PPG and Doppler.Difference v mean

-50

-25

0.0

25

50

Doppler - PPG [ms]

0 100 200 300 400

Average [ms]

+ 2SD

- 2SD

Leg

Arm

Mean difference = 8.6 ms

PPG/Doppler - Conclusions.

• PPG transit times agree satisfactorily with Doppler values recorded at the ‘same’ site.

• The difference plot shows– the transit time estimated by the Doppler instrument is

consistently greater than that derived from the PPG signals (mean difference 8.6 ms)

• The discrepancy is due to the Doppler signal processing

Validation experiments.

Comparison of PPG with• Echo Tracking.

– Does PPG method really measure diameter?

• Doppler.– How well do PPG derived pulse wave transit times compare to

measurements using an established method?

• Intra-arterial pressure wave.– How well do transcutaneous transit time measurements

compare with intra-arterial ones?

Subjects

Measurements on 21 volunteers (8 female, age

range 33 to 78 years, mean 57) after elective

coronary angiography, under the approval of

the regional research ethics committee.

Inguinal ligament

ECG

TP1 Pressuremeasurement pos. 1

Femoral arteriotomy

Inguinal ligament

Pressuremeasurement pos. 1

ECG

TP1

Pressuremeasurement pos. 2

TP2

PPG measurementpos.

TPPG

TP = TP2-TP1

TPPG = TPPG-TP1

TC

+TC

Femoral arteriotomy

40

50

60

70

80

90

100

110

PPG transit time [ms]

40 50 60 70 80 90 100 110

Intra arterial transit time [ms]

y = 0.68x + 22, r = 0.66, P < 0.005

Comparison of PPG and intra-arterial transit times

-30

-20

-10

0.0

10

20

30

I.A. - PPG [ms]

40 50 60 70 80 90 100 110

Mean transit time [ms]

+ 2SD

- 2SD

Comparison of PPG and intra-arterial transit times.Difference v mean

Mean difference = 0.0 ms

Some limitations.

• Non simultaneous measurement of proximal and distal signals

– Ethical constraint of one catheter

• Proximal signal not transcutaneous

– ‘Hybrid’ measurements will avoid this. i.e. aortic signal from Doppler, distal signal

from PPG.

– Current hardware and software will allow this.

• Effect of errors in distance between measurement sites not investigated

– Careful comparison between I.A. and external distance measurements required.

PPG/Intra-arterial - Conclusions.• Reasonable correlation between intra-arterial and PPG

transit times and pulse wave velocities.

• Mean difference between the two methods close to zero

• Transcutaneous estimation of pulse wave transit time provides an acceptable estimate of its intra-arterial value.

– Errors in distance measurement must be carefully considered

Summary of validation results.

Comparison of PPG with• Echo Tracking.

– Does PPG faithfully measure large artery diameter changes and pulse wave timing?

Yes!

• Intra-arterial pressure wave.– Do transcutaneous transit time measurements compare with intra-arterial ones?

Reasonably

• Doppler.– How well do PPG derived pulse wave transit times compare to

measurements using an established method?

Reasonably

Examples of current usage

• Paediatric PWV studies

– Kawasaki disease

– Twin to twin transfusion syndrome

– Children of diabetic mothers

– Zambian schoolchildren of known birthweight and

nutritional status

Conclusions

• PPG measurements of PWV in superficial arteries compare well with other methods

• Although we don’t yet know quite what we’re measuring– Capillary and/or large artery volume changes?– More work needed

Assessment of endothelial function

• Endothelial function– The ability of the vascular endothelium to release vasodilators in

response to reduced mean shear stress• Nitric oxide• PGI2

• EDHF

• Endothelial function is a reliable indicator of vascular “health”– Continuous production of nitric oxide maintains a low basal level of

vascular tone and peripheral resistance– If NO production is impaired:

• Coronary arteries angina• Peripheral arteries mean BP increases • Peripheral endo function closely mirrors that in coronary artery

Assessment of endothelial function

Impaired endothelial function has prognosticand diagnostic value

• a strong predictor of cardiovascular morbidity and mortality

• associated with a wide range of CV pathology– Angina– Type II diabetes– Smoking– Essential hypertension

How to assessendothelial function• Direct

– Measure diameter of muscular artery in response to change in shear stress (flow)• Normally induced by reactive hypersaemia after a period of downstream

occlusion• B mode or echo tracking ultrasound (+ doppler)

– Expensive– Highly skilled operators needed– Not in routine clinical use

• Indirect– Venous occlusion plethysmography– Peripheral artery tonometry– Distal temperature changes– Change in arterial compliance

Principle

• Relaxation of vascular smooth muscle reduction in arterial stiffness

• Reduced stiffness reduced pulsewave velocity

Protocol

Experiment AEffect of exercise on brachio-radial PWV

• Base line PWV measurement• 5 minutes biceps curl• PWV measurements at 1, 2, 5 and 10 minutes• 47 healthy volunteers

Experiment BEffect of ischaemia on brachio-radial PWV

• Base line PWV measurement• 3 minutes forearm artery occlusion (BP cuff)• PWV measurements at 0.5, 1, 2 and 5 minutes• 36 healthy volunteers

ResultsExercise test

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0 1 2 3 4 5 6 7 8 9 10 11

Time after exercise [minutes]

Me

an

pu

lse

wa

ve

ve

loc

ity

ch

an

ge

re

lati

ve

to

ba

se

lin

e

ResultsForearm ischaemia

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

0.4

1 2 3 4 5 6

Time after cuff deflation [minutes]

Me

an

pu

lse

wa

ve

ve

loc

ity

ch

an

ge

re

lati

ve

to

ba

se

lin

e

Conclusions

• PPG is a reliable, repeatable low cost and robust alternative to the range of methods available for measuring PWV

• It is ideal for paediatric studies

• Changes in PWV may be a simple, low cost method for assesing endothelial function. – Very preliminary study

Flow-mediated changes in pulse wave velocity: a new clinical measure ofendothelial function.

Naka KK. Tweddel AC. Doshi SN. Goodfellow J. Henderson AH.European Heart Journal. 27:302-9, 2006 Feb.

Arm

Leg

Hyperaemia increased brachial artery diameter by 8% at this time.

GTN had similar effect

Hyperaemia had negligible effect on brachial artery diameter.

GTN reduced diameter by similar amount to controls

Sleeve: i.d 4 - 12.5 mm o.d 18 mm

Centralising lid

Ejection screw

Outer Container

Mandrel: diameter 4-10 mm

Artery

Frozen Artery Reaming Trunnion