VVP (2)

Ralf Schüler

IMPROVING THE ACCURACY OF VENOUS OCCLUSION PLETHYSMOGRAPHY WITH IMPEDANCE METHOD

Proceedings of the IX. International Conference on electrical Bio-Impedance,
Heidelberg 1995, pp. 295

In the tracings of venous occlusion plethysmograhy (VOP) there is a spike immediatelly after releasing the pneumatic occlusion cuffs. This spike only occures if the measurment of the blood volume is done via bio-impedance not via other measurment methods. This study investigates the origin of this spike and the error it produces. This was done by observing practical examinations with the measurement system rheoscreen® as well as by mathematical modelling. It could be found that there are large interindividual differences in the occurence of this spike. Often it is very small but sometimes it produces significant errors. To solve this problem we developed a method to eliminate the effects of the spike and integrated the method in the measuring system rheoscreen® to reach better diagnostic validity in such cases.

Introduction

The venous occlusion plethysmography (VOP) is a widespread method for diagnosing deep vein thrombosis. The method is to obstruct the venous outflow by a pneumatic cuff while arterial inflow continues because of the higher arterial pressure. The veins fills with blood. After some minutes the cuffs were released quickly. The speed of outflowing blood is significantly reduced in case of an outflow obstruction like a thrombosis. One method to measure limb blood volume changes under VOP is the use of electrical bio-impedance. Fig. 1 shows a typical VOP-tracing. Please note, that the impedance is indirect proportional to the blood volume (decrease blood volume means increase impedance). In case of modern equipment the bio-impedance method detecting blood volume changes is highly sensitive and easy to handle.


Fig1: Typical impedance tracing during VOP

But in contrast to other methods like strain gauge there is a spike in the impedance signal immediately after opening the occlusion cuff. Mohapatra (1) described this as the effect of changing blood resistivity by changing the blood velocity. Any authors don´t notify the presence of this effect, other (3) describe this as a source of a large error.

Our study analyses the effect of the mentioned spike on the accuracy of the impedance method.

Materials and Methods

Practical Investigations

We determined the size of the spike by analysing the data of about 200 medical routine examinations with the angiologic measuring system rheoscreen®. We evaluate the height of the spike in relation of the total amplitude (also called „venous capacitance“) of the VOP-curve.

Mathematical Modelling

A mathematical model of the blood outflow in the limbs after venous occlusion was used to confirm the genesis of the spike by changes of blood resistivity and to show the kind of errors produced by this effect. A mathematical model is necessary to evaluate this because of practical measurements with different measuring methods are not usable. The error caused by a comparision of different measurement methods is normaly higher then the error caused by the spike.

We used a blood flow model of the limb that was originate used to evaluate the influence of various factors like measurement conditions and various physiological changes of the shape of the outflow curve obtained by the VOP. To consider the blood resistivity we expand the model with the dependency of the blood resistivity on the blood velocity (2).

The model considers the following factors:

  • the resting volume of the veins
  • the nonlinear dependency of the venous compliance on the venous pressure
  • the different hydrostatic pressure in different sections
  • the arterial inflow
  • the flow resistance of the venous system

The physiological constants used in the model can be found in (4, 5). The model requires an iterative computation to solve the equation system. The calculations of the simulations were done at a Pentium-PC.

The model doesn´t consider the phase in which the cuff is just opening. That´s why it isn´t possible to simulate the shape of the spike.

Results

Practical investigations

There are large interindividual differences in the spike amplitude. The spike could be found in nearly all observed examinations. In most cases (approx. 80%) the spike was very little (below 5%). Only in approx. 5% the spike has an amplitude of greater then 25% of the venous capacitance. It was found that the spikes in patients with any kind of outflow obstructions was more often higher. This is caused by the low venous capacitance in this group as well as by an increase of the total amplitude. Because of the very large interindividual differences there is no rule to consider this by the evaluation of the measuring results.

Model


Fig 2: Graphical results of the mathematical simulation

The mathematical model corresponds good with the practical results. The simulated outflow curve shows a high conformity with real obtained curves. The results show clearly the difference between simulating the volume changes after opening the venous occlusion with and without the dependence of blood resistivity on blood velocity. The variation of the blood resistivity cause a higher curve level. The error is as higher as faster the blood flows out. This means that the highest error occured immediately after opening the cuff.


Fig 3: Two possibilities to measure outflow volume

Effect of the calculation of diagnostically relevant parameters

To diagnose venous outflow obstructions such as thrombosis there are used the amount of blood flowing out two, three or five seconds after opening the venous occlusion. Figure 3 illustrates two possible ways to determine this levels from the top of the spike or from the level before opening. For the first method the error is small or nearly zero immediately after opening the cuffs and largest after reaching the resting level. The error trend of the second method is opposite. That´s why the first method should be preffered to calculate the blood outflow in the first time and the second method to calculate the total amount of blood flowing out. In most cases this should reduce the error to a low level.

But in cases of low venous capacitance and relative high spikes the error reaches a level that can cause poor interpretations of the measuring results. Data from thrombosis can be interpreted as borderline values. To avoid this it is necessary to eliminate the effect of varying blood resistivity by the evaluation software. This can be done by measuring the outflow velocity to correct the curve in dependence on the spike height on basis of the relations we used in the model.

Conclusions

The measurement of blood volume changes under VOP by bio-impedance is a highly sensitive method. But the dependence of the blood resistivity on the blood velocity may sometimes cause a high error level especially in cases of outflow obstructions. Therefore this effect has to be eliminated to calculate precise parameters. We developed a method to do this and implemented it in the angiologic diagnostic system rheoscreen®. So the bio-impedance becomes a highly sensitive mesuring method with no loss of accuracy by impedance specific effects.

References

1. Surya N. Mohapatra
Non-invasive Cardiovascular Monitoring by Electrical Impedance Technique
Pitman Medical 1981

2. Lamberts, R.
Impedance Cardiography
Diss. Uni Gorcum 1984

3. G. Rudofsky, J. Gutmann, M. Althoff
Apparative Gefäßdiagnostik mit Doppler, Duplex, Verschluß-Plethysmograf und Lichtreflex-Plethysmograf
Eurasburger Berichte zu Themen der Angiologie und Kardiologie, Heft 7, 1993

4. Rudi Busse
Kreislaufphysiologie
Georg Thieme Verlag Stuttgart, New York, 1982

5. D. H. Bergel
The static elastic properties of the of the arterial wall
J. Physiol. 156 (1961), S. 455-457

6. T. M. Ravi Shankar, John G. Webster, Shu-Yong Shao
The Contribution of Vessel Volume Change and Blood Resistivity Change to the Electrical Impedance Pulse
IEEE Transactions on Biomedical Engineering, Vol. 33, 3/1985

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert.

*