University of Twente Student Theses
Cardiac function monitoring during venoarterial extracorporeal membrane oxygenation in the intensive care unit
Wösten, M. (2023) Cardiac function monitoring during venoarterial extracorporeal membrane oxygenation in the intensive care unit.
|
PDF
8MB |
| Abstract: | Background: Venoarterial extracorporeal membrane oxygenation (VA ECMO) is a mechanical support modality that can be used to quickly restore the systemic circulation in patients with cardiogenic shock. However, the clinical care management of patients supported with VA ECMO is complex, mainly due to the interaction between the ECMO circuit and the patient’s native hemodynamics. To improve this clinical management, it is important to gain insight in the native function of the left ventricle (LV). This thesis therefore describes the application of a lumped parameter model to continuously monitor LV flow using arterial blood pressure during VA ECMO. Methods: Cardiovascular simulators and clinical data were used to evaluate the LV flow calculated using the two-element Windkessel model. For the cardiovascular simulators, LV flow was compared using the stroke volume (SV) and normalized root mean squared error (nRMSE). In the clinical data, the calculated SV from the left ventricle was compared to the echocardiographic measurement of SV. In addition, to compare the two-element Windkessel model to a model described by linearized one-dimensional Navier-Stokes equations, transfer functions were derived and compared using Bode plots. Results: In two of the three cardiovascular simulators the calculation of the LV flow resulted in an average error in SV below 12%. For the third simulator, the average error was larger with an average of 26%. The nRMSE showed large differences between the different simulators, varying from 0.1 to 2.6. In the clinical data, using current patient specific parameters, the average error in stroke volume was 0.88 mL. Using patient specific parameters that are continuously estimated, the error in SV becomes larger. Comparison of the two-element Windkessel model to linearized one-dimensional Navier-Stokes equations showed no large differences for low frequencies. Conclusion: Application of the two-element Windkessel model can be used to accurately determine SV during VA ECMO in two of the three cardiovascular simulators and in clinical data, using known patient specific parameters. We also showed that for low frequencies the linearized one-dimensional Navier-Stokes equations will not lead to large improvements in the flow calculations from arterial blood pressure. Further studies should show if more comprehensive models can improve the calculated LV flow waveform and should focus on improving the accuracy of the continuous estimation of patient specific parameters. |
| Item Type: | Essay (Master) |
| Clients: | UMC Utrecht, Utrecht, The Netherlands |
| Faculty: | TNW: Science and Technology |
| Subject: | 31 mathematics, 44 medicine |
| Programme: | Technical Medicine MSc (60033) |
| Link to this item: | https://purl.utwente.nl/essays/97686 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page