Resonant Frequency:
In order to reproduce any waveform accurately, the resonant (natural) frequency of the measurement system must be higher than resonant frequency of the system which it is trying to measure (Figure 1 ). For the human cardiovascular system, it is generally accepted that pulse rates of up to 180 beats per minute should be able to be accurately reproduced. In practical terms there will be no noticeable distortion of the waveform if the system can track frequencies of ten times the fundamental frequency which implies a requirement to reproduce accurately 30 Hz waveforms. This, in turn, implies that the natural frequency of the measurement system should be at least 45 Hz.

In fact, because a modern transducer has such a low displacement volume (Figure 2), its natural frequency is often several hundred Hz and problems with resonance generally only start to arise when the transducer is coupled to a manometer system.

Accuracy: The low displacement volumes of modern transducers may theoretically compromise the accuracy of the device, but in practice this is not a major issue and zero-stable measurements which are accurate to +/- 1 mm Hg are the norm for a current disposable transducer.

Damping:
The catheter:transducer system should be appropriately damped (Figure 3 ). - Because the transducer is hydraulically connected to the artery by a fluid-containing system which is subject to frictional resistance, any change in pressure is not immediately transmitted to the transducer diaphragm. This delay in response is known as 'damping'. An ideally damped system should be about 70% critically damped.

An example of an under-damped PA pressure trace is shown in Figure 4. In this figure, it should be noted that some of the ringing which occurs during the transition from diastole to systole is produced by movement of the catheter ("catheter whip") and is not entirely due to under-damping. Re-positioning of the catheter in such cases can eliminate this phenomenon.

Over-damping can be produced by the insertion of additional three-way taps or the introduction of small air bubbles anywhere in the catheter:transducer system.

Phase shift:
The frictional resistance of the transducer system also causes a time delay or 'phase shift' between the occurrence of the pressure change and its registration by the transducer. For complex waveforms which incorporate a wide spectrum of frequencies, phase shift of the various frequencies is only constant when the system is about 70% critically damped.

Some manufacturers (eg Becton Dickinson) produce pressure measurement systems where the performance of the entire measurement path has been measured dynamically and is guaranteed. When using such systems, it is important that additional components not be inserted into the hydraulic 'chain' (Figure 5).

Zeroing:
The transducer should be zeroed by opening the transducer to atmosphere at its 'three way' tap. This same 'three way' tap should subsequently be levelled to the transducer reference point.

Reference point:
The transducer should be referenced to the level of the tricuspid valve. In the Intensive Care Unit it is common to mark the tricuspid position on the patients chest and to adjust the transducer to this reference point with a spirit level before any measurement is made. In the supine patient, the coordinates of the valve can be represented as: midline on the right-to-left axis, 80% of the distance between sternal notch and the tip of the xiphoid process and 40% of the 'depth' of the chest at this point ¹. (The corresponding reference point on the transducer should be the 'three way tap' which is used for zeroing the transducer.)

It is usually not possible to adopt this policy in the operating theatre but at least the transducers should be fixed to the operating table at about mid-thorax level, so that no bias-error will be introduced if the surgeon adjusts the height of the table.

Whenever possible, measurement of pulmonary arterial and occlusion pressures should be made in the supine position as reliable measurements cannot be obtained with patients in the 60 degrees lateral position ³.

The clinical practice of referencing the transducer to atmospheric pressure at the level of the tricuspid valve is not strictly correct in physiological terms. - As we are attempting to estimate transmural pressures when we make the measurement, the transducer should really be referenced to the intrapericardial pressure at the mid-atrial (tricuspid) point. This is obviously impractical, but for research purposes a valid estimate of this pressure can be obtained (in the erect position) by measuring the intra-oesophageal pressure at atrial level and referencing the intravascular pressures to this.

Finally, it should be remembered that infusions should not be administered through the CVP lumen of the catheter simultaneously with measurement as this will artifactually raise the reading ².

The section on 'Special purpose catheters' contains detailed information on the devices which are currently available.

1. Parkin WG, Bellomo R, Boyd H et al A device to automatically level vascular pressure transducers to the tricuspid valve. Anaesth Intensive Care 27:94, 1999.

2. Davis RF Yet another CVP artifact. Anesthesiology 60:262, 1984

3. Aitken LM Reliability of measurements of pulmonary artery pressure obtained with patients in the 60 degrees lateral position. Am J Crit Care 2000 Jan;9(1):43-51

Last edited on: 23/12/2000