Pressure transducers and cardiac output measurement
How pressure is measured non-invasively and invasively, the components of an arterial line, the Wheatstone bridge, zeroing and levelling, natural frequency, damping and the fast flush test, Fourier analysis of the waveform, and the principles of measuring cardiac output by the Fick principle, indicator dilution, thermodilution, pulse contour analysis and Doppler.
By the end of this page you should be able to
- Describe how an oscillometric non-invasive blood pressure monitor works and where it errs.
- Describe the components of an invasive pressure system, the strain gauge and Wheatstone bridge, and the difference between zeroing and levelling.
- Explain natural frequency, resonance and damping, and interpret a fast flush test.
- Explain the Fick principle and apply it to cardiac output measurement, including indicator and thermodilution.
- Outline the principles and limitations of pulse contour and Doppler methods.
Non-invasive blood pressure
Pressure is force per unit area; physiological pressures are quoted in mmHg or cmH2O, and gauge pressures are relative to atmospheric. Oscillometric monitors inflate a cuff above systolic pressure and deflate it in steps, sensing the small pressure oscillations transmitted from the pulsating artery. The point of maximal oscillation is the mean arterial pressure, which is the most accurate value; systolic and diastolic are then calculated by manufacturer algorithms from the rate of rise and fall of the oscillations. Errors follow: a cuff that is too small over-reads and one that is too large under-reads (the bladder should be about 40% of arm circumference in width and encircle at least 80% of it); arrhythmia, shivering and movement confuse the algorithm; and readings are inaccurate at extremes of pressure and in severe vasoconstriction. Frequent cycling can injure nerves and skin. Auscultatory measurement uses Korotkoff sounds; the fifth phase (disappearance) is diastolic.
The invasive pressure system
An arterial line consists of a cannula, non-compliant tubing filled with saline, a flush system (a pressurised bag delivering a few mL/h to keep the line patent, with a valve for a fast flush), a three-way tap and a transducer, connected to an amplifier and display.
The transducer converts the pressure wave into an electrical signal. A diaphragm moves with pressure and stretches a strain gauge, a resistor whose resistance changes when it is stretched. Small resistance changes are measured accurately with a Wheatstone bridge: four resistors in a diamond with a voltage applied across it. When the bridge is balanced, no current flows across the middle; when the strain gauge resistance changes, the imbalance produces a voltage proportional to pressure. Using strain gauges as several arms of the bridge increases sensitivity and compensates for temperature.
Zeroing opens the transducer to air so that atmospheric pressure is read as zero; it corrects for drift and offset. Levelling places the transducer at the reference point, usually the phlebostatic axis (fourth intercostal space, mid-axillary line, roughly the right atrium). Every 10 cm that the transducer sits below the heart adds about 7.4 mmHg (a column of water 10 cm high) to the reading, and every 10 cm above subtracts it. For cerebral perfusion pressure, the arterial transducer may instead be levelled at the external auditory meatus.
Natural frequency, damping and the fast flush test
The arterial waveform is complex, but by Fourier analysis it can be described as the sum of a fundamental sine wave (at the heart rate) and its harmonics. To reproduce it faithfully, a system must respond evenly to frequencies up to about the eighth to tenth harmonic, which at a heart rate of 120/min (2 Hz) means up to roughly 16 to 20 Hz.
Every system has a natural (resonant) frequency at which it oscillates when disturbed. If the natural frequency is close to frequencies within the waveform, those components are amplified: resonance. The natural frequency is higher with short, wide, stiff tubing and a small, stiff diaphragm, and lower with long, narrow, compliant tubing and air bubbles. A good system has a natural frequency well above the frequencies it must record.
Damping is the tendency of a system to resist oscillation, from friction and compliance. It is quantified by the damping coefficient: 0 means undamped; 1 is critical damping (the fastest return to baseline without overshoot, but slow to respond); about 0.64 is optimal for pressure measurement, giving the best compromise of speed and accuracy with a small overshoot. An underdamped system (coefficient low) oscillates and overestimates systolic and underestimates diastolic; an overdamped system (high coefficient) is sluggish, underestimates systolic and overestimates diastolic. Mean pressure is relatively preserved in both, which is why it is the most reliable value when the trace is doubtful.
The fast flush test applies a square wave and watches the response. The time between oscillation peaks gives the natural frequency (paper speed ÷ distance between peaks), and the ratio of successive peak amplitudes gives the damping coefficient. One or two oscillations before settling indicates near-optimal damping. Overdamping is caused by air bubbles, clots, kinks, soft tubing and loose connections; underdamping by long or stiff tubing with a low natural frequency, and is worsened by tachycardia.
Central venous and pulmonary artery pressure
Central venous pressure is measured with the same transducer system from a catheter whose tip lies at the junction of the superior vena cava and right atrium. The waveform has a, c and v waves (atrial contraction, tricuspid bulging in isovolumetric contraction, venous filling against a closed valve) and x and y descents. Cannon a waves appear when the atrium contracts against a closed tricuspid valve (complete heart block, junctional rhythm); large v waves with tricuspid regurgitation; absent a waves in atrial fibrillation. It is read at end-expiration, when intrathoracic pressure is closest to atmospheric. A single value is a poor guide to fluid responsiveness. A pulmonary artery catheter adds pulmonary artery and occlusion (wedge) pressures, the latter approximating left atrial pressure when the catheter tip lies in a zone where pulmonary venous pressure exceeds alveolar pressure.
Measuring cardiac output
- Fick principle: the amount of a substance taken up by an organ per minute equals the blood flow multiplied by the arteriovenous concentration difference. For oxygen, cardiac output = oxygen consumption ÷ (arterial oxygen content minus mixed venous oxygen content). It is the reference method but needs mixed venous blood and a measured oxygen consumption at steady state. Partial carbon dioxide rebreathing applies the Fick principle to carbon dioxide non-invasively.
- Indicator dilution (Stewart-Hamilton): a known amount of indicator (a dye, lithium) is injected centrally and its concentration measured downstream; cardiac output = amount injected ÷ area under the concentration-time curve. Recirculation distorts the tail, so the downslope is extrapolated exponentially.
- Thermodilution: cold saline is the indicator; a thermistor at the catheter tip measures the temperature change, and cardiac output is inversely proportional to the area under the temperature-time curve (a small, brief change means high output). Errors come from tricuspid regurgitation, intracardiac shunts, the wrong injectate volume or temperature, and timing with the respiratory cycle. Continuous methods warm blood with a heating filament instead. Transpulmonary thermodilution injects centrally and measures in a femoral artery, and also estimates extravascular lung water and global end-diastolic volume.
- Pulse contour and pulse power analysis: stroke volume is estimated from the area under the systolic part of the arterial waveform or from its power, using a model of aortic impedance that is calibrated against a dilution measurement or estimated from demographics. Accuracy falls with changes in vascular tone, arrhythmia, aortic regurgitation, intra-aortic balloon pumps and a poorly damped trace. Respiratory variation in pulse pressure or stroke volume predicts fluid responsiveness only in sinus rhythm with controlled ventilation at adequate tidal volume and a closed chest.
- Doppler: the frequency shift of ultrasound reflected from moving red cells gives velocity (shift = 2 × transmitted frequency × velocity × cos θ ÷ speed of sound). Integrating velocity over systole gives the velocity-time integral (stroke distance), which multiplied by the cross-sectional area of the vessel or outflow tract gives stroke volume. Oesophageal Doppler and echocardiography use this; errors come from the angle of insonation (cos θ) and estimating the area.
What gets marks
- Draw the system from cannula to screen, then explain each component's role in accuracy.
- Separate zeroing (atmospheric reference) from levelling (hydrostatic reference); give the 7.4 mmHg per 10 cm figure.
- Explain natural frequency and damping before interpreting a flush test, and say what happens to systolic, diastolic and mean.
- For cardiac output methods, state the principle as an equation, then the assumptions, then the errors.
Common traps
- Saying zeroing corrects for the height of the transducer; that is levelling.
- Saying critical damping is ideal; it is too slow, and about 0.64 is optimal.
- Saying oscillometry measures systolic and diastolic directly; it measures mean, the rest is calculated.
- Saying a larger area under the thermodilution curve means higher output; it means lower.
- Using respiratory pulse pressure variation in a spontaneously breathing patient or in atrial fibrillation.
Where this meets intensive care
- Check damping with a fast flush before acting on a high systolic pressure; an underdamped line overreads systolic, and mean pressure is the most reliable number.
- Releveling the transducer after moving the bed or patient matters: each 10 cm error changes the reading by about 7 mmHg.
Viva questions
Answer each one aloud, as you would to an examiner, before you open the model points.
- 1.
How does an oscillometric blood pressure monitor work?
Show answer
- Cuff inflated above systolic and deflated in steps, sensing oscillations transmitted from the artery.
- Maximal oscillation corresponds to mean arterial pressure, the most accurate value.
- Systolic and diastolic are derived by algorithm.
- Errors: wrong cuff size, arrhythmia, movement, extremes of pressure, vasoconstriction; complications of frequent cycling.
- 2.
Describe the components of an arterial line and the function of each.
Show answer
- Cannula, stiff saline-filled tubing, pressurised flush system with fast-flush valve, three-way tap, transducer, amplifier, display.
- Diaphragm and strain gauge convert pressure to a resistance change.
- Wheatstone bridge converts resistance change to a voltage proportional to pressure.
- Short, stiff, wide tubing keeps natural frequency high.
- 3.
What is the difference between zeroing and levelling?
Show answer
- Zeroing exposes the transducer to atmosphere and sets that as zero, correcting offset and drift.
- Levelling places the transducer at the reference point, usually the phlebostatic axis.
- Each 10 cm below the heart adds about 7.4 mmHg; above subtracts it.
- For cerebral perfusion pressure, the external auditory meatus may be used as the reference.
- 4.
Explain natural frequency and resonance in a pressure system.
Show answer
- The waveform is a sum of harmonics of the heart rate (Fourier analysis); faithful recording needs response up to about the tenth harmonic.
- Natural frequency is the frequency at which the system oscillates when disturbed.
- If it lies close to waveform frequencies, those are amplified: resonance, with systolic overshoot.
- Raised by short, wide, stiff tubing; lowered by long, narrow, compliant tubing and bubbles.
- 5.
Interpret a fast flush test.
Show answer
- Square wave applied and released; watch the oscillations.
- Natural frequency from the distance between peaks and the paper speed; damping coefficient from the ratio of successive amplitudes.
- Optimal: one or two oscillations then settling; coefficient around 0.64.
- Underdamped: many oscillations, systolic overread; overdamped: no oscillation, systolic underread; mean relatively preserved.
- 6.
State the Fick principle and how it is used to measure cardiac output.
Show answer
- Uptake of a substance equals flow × arteriovenous concentration difference.
- Cardiac output = oxygen consumption ÷ (arterial minus mixed venous oxygen content).
- Requires steady state, measured oxygen consumption and mixed venous sampling from the pulmonary artery.
- Carbon dioxide rebreathing methods use the same principle.
- 7.
How does thermodilution measure cardiac output, and what causes errors?
Show answer
- Cold injectate into the right atrium; thermistor in the pulmonary artery records the temperature-time curve.
- Stewart-Hamilton: output inversely proportional to the area under the curve.
- Errors: tricuspid regurgitation, shunts, wrong injectate volume or temperature, slow injection, respiratory timing.
- Average several readings; continuous and transpulmonary variants exist.
- 8.
How does pulse contour analysis estimate stroke volume, and when is it unreliable?
Show answer
- Stroke volume derived from the systolic area or power of the arterial waveform via a model of aortic impedance.
- Calibrated by dilution or estimated from demographic data.
- Unreliable with changing vascular tone, arrhythmia, aortic regurgitation, balloon pumps and poor damping.
- Dynamic indices need controlled ventilation, sinus rhythm and adequate tidal volume.
- 9.
Describe the central venous pressure waveform.
Show answer
- a wave atrial contraction; c wave tricuspid bulging; x descent atrial relaxation; v wave filling against closed valve; y descent valve opening.
- Cannon a waves in atrioventricular dissociation; large v waves in tricuspid regurgitation; absent a waves in atrial fibrillation.
- Read at end-expiration, levelled at the phlebostatic axis.
- A single value is a poor predictor of fluid responsiveness.
Practise this week
- Physiology Bench: move the graphs yourself
- First Part MCQs: measurement and equipment
- First Part SAQs, marked part by part
- Anatomy Scroller: procedural anatomy on a real CT
Week 9 in the other subjects
- Physiology: Cardiac output and its control
- Pharmacology: Fluids, electrolytes and blood products
Not affiliated with the CICM. Notes are a revision aid, not a substitute for the curriculum or for clinical guidelines.