Optimising the Windkessel Model for Cardiac Output Monitoring During Changes in Vascular Tone

Peter Charlton; John Smith; Luigi Camporota; Richard Beale; Jordi Alastruey

Optimising the Windkessel Model for Cardiac Output Monitoring During Changes in Vascular Tone

Peter Charlton, John Smith, Luigi Camporota, Richard Beale, Jordi Alastruey

Research output: Chapter in Book/Report/Conference proceeding › Poster abstract

156 Downloads (Pure)

Abstract

INTRODUCTION. Algorithms for estimating cardiac output (CO) from the arterial blood pressure wave have been observed to be inaccurate during changes in vascular tone. Many such algorithms are based on the Windkessel model of the circulation. We investigated the optimal analytical approaches and assumptions to apply this model for CO estimation during changes in vascular tone.

METHODS. Equations relating CO to the measured arterial blood pressure were derived by applying conservation of mass and Pouiseille’s law to the Windkessel model. Analytical approaches and assumptions used in the literature were applied to these equations to construct CO algorithms. We evaluated these algorithms on data collected from 15 critically ill patients. The algorithms were calibrated with reference transpulmonary thermodilution CO measurements at baseline, and their accuracy was determined by comparison with reference measurements during a pharmacologically controlled change in vascular tone.

RESULTS. We found that the most accurate algorithms: (i) maintained the compliance and outflow terms in the Windkessel model; and (ii) re-calculated compliance on a beat-by-beat basis. They produced a percentage error of ±31%. In addition, algorithms had higher accuracy when the end of systole was identified using the zero crossing of the second derivative of pressure, and when the outflow pressure was assumed to be zero. The choice of methods for estimating compliance and the systemic time constant had little effect on accuracy.

CONCLUSION. We found large differences in the accuracy of CO algorithms based on the Windkessel model of the circulation during changes in vascular tone. Optimal analytical approaches and assumptions were identified from those tested. However, none of the tested algorithms was within the clinically acceptable accuracy of ±30%.

Original language	English
Title of host publication	MEC Annual Meeting and Bioengineering14 Programme and Abstracts
Place of Publication	London
Publisher	MECbioeng14, Imperial College London
Pages	95-95
Number of pages	1
ISBN (Print)	9780993039003
Publication status	Published - 2014

Access to Document

Poster PresentationSubmitted manuscript, 844 KB

http://www.mecbioeng.org/2014/wp-content/uploads/MECbioeng14-brochure.pdf

Cite this

Charlton, P., Smith, J., Camporota, L., Beale, R., & Alastruey, J. (2014). Optimising the Windkessel Model for Cardiac Output Monitoring During Changes in Vascular Tone. In MEC Annual Meeting and Bioengineering14 Programme and Abstracts (pp. 95-95). Article 88 MECbioeng14, Imperial College London. http://www.mecbioeng.org/2014/wp-content/uploads/MECbioeng14-brochure.pdf

@inbook{663c65aab80e4f748c502d064220ed5d,

title = "Optimising the Windkessel Model for Cardiac Output Monitoring During Changes in Vascular Tone",

abstract = "INTRODUCTION. Algorithms for estimating cardiac output (CO) from the arterial blood pressure wave have been observed to be inaccurate during changes in vascular tone. Many such algorithms are based on the Windkessel model of the circulation. We investigated the optimal analytical approaches and assumptions to apply this model for CO estimation during changes in vascular tone.METHODS. Equations relating CO to the measured arterial blood pressure were derived by applying conservation of mass and Pouiseille{\textquoteright}s law to the Windkessel model. Analytical approaches and assumptions used in the literature were applied to these equations to construct CO algorithms. We evaluated these algorithms on data collected from 15 critically ill patients. The algorithms were calibrated with reference transpulmonary thermodilution CO measurements at baseline, and their accuracy was determined by comparison with reference measurements during a pharmacologically controlled change in vascular tone.RESULTS. We found that the most accurate algorithms: (i) maintained the compliance and outflow terms in the Windkessel model; and (ii) re-calculated compliance on a beat-by-beat basis. They produced a percentage error of ±31%. In addition, algorithms had higher accuracy when the end of systole was identified using the zero crossing of the second derivative of pressure, and when the outflow pressure was assumed to be zero. The choice of methods for estimating compliance and the systemic time constant had little effect on accuracy.CONCLUSION. We found large differences in the accuracy of CO algorithms based on the Windkessel model of the circulation during changes in vascular tone. Optimal analytical approaches and assumptions were identified from those tested. However, none of the tested algorithms was within the clinically acceptable accuracy of ±30%.",

author = "Peter Charlton and John Smith and Luigi Camporota and Richard Beale and Jordi Alastruey",

year = "2014",

language = "English",

isbn = "9780993039003",

pages = "95--95",

booktitle = "MEC Annual Meeting and Bioengineering14 Programme and Abstracts",

publisher = "MECbioeng14, Imperial College London",

}

TY - CHAP

T1 - Optimising the Windkessel Model for Cardiac Output Monitoring During Changes in Vascular Tone

AU - Charlton, Peter

AU - Smith, John

AU - Camporota, Luigi

AU - Beale, Richard

AU - Alastruey, Jordi

PY - 2014

Y1 - 2014

N2 - INTRODUCTION. Algorithms for estimating cardiac output (CO) from the arterial blood pressure wave have been observed to be inaccurate during changes in vascular tone. Many such algorithms are based on the Windkessel model of the circulation. We investigated the optimal analytical approaches and assumptions to apply this model for CO estimation during changes in vascular tone.METHODS. Equations relating CO to the measured arterial blood pressure were derived by applying conservation of mass and Pouiseille’s law to the Windkessel model. Analytical approaches and assumptions used in the literature were applied to these equations to construct CO algorithms. We evaluated these algorithms on data collected from 15 critically ill patients. The algorithms were calibrated with reference transpulmonary thermodilution CO measurements at baseline, and their accuracy was determined by comparison with reference measurements during a pharmacologically controlled change in vascular tone.RESULTS. We found that the most accurate algorithms: (i) maintained the compliance and outflow terms in the Windkessel model; and (ii) re-calculated compliance on a beat-by-beat basis. They produced a percentage error of ±31%. In addition, algorithms had higher accuracy when the end of systole was identified using the zero crossing of the second derivative of pressure, and when the outflow pressure was assumed to be zero. The choice of methods for estimating compliance and the systemic time constant had little effect on accuracy.CONCLUSION. We found large differences in the accuracy of CO algorithms based on the Windkessel model of the circulation during changes in vascular tone. Optimal analytical approaches and assumptions were identified from those tested. However, none of the tested algorithms was within the clinically acceptable accuracy of ±30%.

AB - INTRODUCTION. Algorithms for estimating cardiac output (CO) from the arterial blood pressure wave have been observed to be inaccurate during changes in vascular tone. Many such algorithms are based on the Windkessel model of the circulation. We investigated the optimal analytical approaches and assumptions to apply this model for CO estimation during changes in vascular tone.METHODS. Equations relating CO to the measured arterial blood pressure were derived by applying conservation of mass and Pouiseille’s law to the Windkessel model. Analytical approaches and assumptions used in the literature were applied to these equations to construct CO algorithms. We evaluated these algorithms on data collected from 15 critically ill patients. The algorithms were calibrated with reference transpulmonary thermodilution CO measurements at baseline, and their accuracy was determined by comparison with reference measurements during a pharmacologically controlled change in vascular tone.RESULTS. We found that the most accurate algorithms: (i) maintained the compliance and outflow terms in the Windkessel model; and (ii) re-calculated compliance on a beat-by-beat basis. They produced a percentage error of ±31%. In addition, algorithms had higher accuracy when the end of systole was identified using the zero crossing of the second derivative of pressure, and when the outflow pressure was assumed to be zero. The choice of methods for estimating compliance and the systemic time constant had little effect on accuracy.CONCLUSION. We found large differences in the accuracy of CO algorithms based on the Windkessel model of the circulation during changes in vascular tone. Optimal analytical approaches and assumptions were identified from those tested. However, none of the tested algorithms was within the clinically acceptable accuracy of ±30%.

M3 - Poster abstract

SN - 9780993039003

SP - 95

EP - 95

BT - MEC Annual Meeting and Bioengineering14 Programme and Abstracts

PB - MECbioeng14, Imperial College London

CY - London

ER -

Optimising the Windkessel Model for Cardiac Output Monitoring During Changes in Vascular Tone

Abstract

Access to Document

Fingerprint

Cite this