Congestive heart failure (CHF) is a condition caused by the heart’s inability to correctly circulate blood to peripheral tissues, which results in a persistent state of increased plasma volume, due to the overactivation of neurohumoral complexes that induce the body to retain renal sodium and water. The most common therapy for decongestion is the administration of diuretics, in particular loop diuretics such as furosemide. Since the effects of diuretic drugs in case of CHF can largely vary between patients, the actual course of the treatment is not standardised, and its efficacy relies on the physician’s expertise. This work aimed to implement a control system, using a model predictive control (MPC) strategy, that could automatically and dynamically adapt the administration of furosemide by predicting the patient’s response to the treatment based on the patient’s characteristics, in order to achieve decongestion more efficiently than with standard therapy. As part of the simulation set up, literature was extensively consulted in the search for a pharmacokinetic and pharmacodynamic (PKPD) model of furosemide, finally settling on a three-compartment representation of the pharmacokinetics and a four-parameters sigmoidal pharmacodynamic response that describes the urine production in response to the drug. In addition to that, a mathematical model of renal activity was needed. Two models were shortlisted. One, more articulated and capable of describing several physiological events, was discarded because it was too complex to integrate with the PKPD model. The second model was simpler and described a first-order, negative-feedback response of the kidneys to changes in plasma volume from its normal physiological level, which could be easily integrated with the furosemide PKPD model. Once the control simulation was set up, results showed that an automated MPC controlled diuretic administration can outperform standard diuretic treatment in case of congestive heart failure by achieving comparatively higher level of decongestion in less time.

Congestive heart failure (CHF) is a condition caused by the heart’s inability to correctly circulate blood to peripheral tissues, which results in a persistent state of increased plasma volume, due to the overactivation of neurohumoral complexes that induce the body to retain renal sodium and water. The most common therapy for decongestion is the administration of diuretics, in particular loop diuretics such as furosemide. Since the effects of diuretic drugs in case of CHF can largely vary between patients, the actual course of the treatment is not standardised, and its efficacy relies on the physician’s expertise. This work aimed to implement a control system, using a model predictive control (MPC) strategy, that could automatically and dynamically adapt the administration of furosemide by predicting the patient’s response to the treatment based on the patient’s characteristics, in order to achieve decongestion more efficiently than with standard therapy. As part of the simulation set up, literature was extensively consulted in the search for a pharmacokinetic and pharmacodynamic (PKPD) model of furosemide, finally settling on a three-compartment representation of the pharmacokinetics and a four-parameters sigmoidal pharmacodynamic response that describes the urine production in response to the drug. In addition to that, a mathematical model of renal activity was needed. Two models were shortlisted. One, more articulated and capable of describing several physiological events, was discarded because it was too complex to integrate with the PKPD model. The second model was simpler and described a first-order, negative-feedback response of the kidneys to changes in plasma volume from its normal physiological level, which could be easily integrated with the furosemide PKPD model. Once the control simulation was set up, results showed that an automated MPC controlled diuretic administration can outperform standard diuretic treatment in case of congestive heart failure by achieving comparatively higher level of decongestion in less time.

Modeling and Control of Loop Diuretic Therapy in Congestive Heart Failure: A Feasibility Study

SEMENZATO, CORINNA
2022/2023

Abstract

Congestive heart failure (CHF) is a condition caused by the heart’s inability to correctly circulate blood to peripheral tissues, which results in a persistent state of increased plasma volume, due to the overactivation of neurohumoral complexes that induce the body to retain renal sodium and water. The most common therapy for decongestion is the administration of diuretics, in particular loop diuretics such as furosemide. Since the effects of diuretic drugs in case of CHF can largely vary between patients, the actual course of the treatment is not standardised, and its efficacy relies on the physician’s expertise. This work aimed to implement a control system, using a model predictive control (MPC) strategy, that could automatically and dynamically adapt the administration of furosemide by predicting the patient’s response to the treatment based on the patient’s characteristics, in order to achieve decongestion more efficiently than with standard therapy. As part of the simulation set up, literature was extensively consulted in the search for a pharmacokinetic and pharmacodynamic (PKPD) model of furosemide, finally settling on a three-compartment representation of the pharmacokinetics and a four-parameters sigmoidal pharmacodynamic response that describes the urine production in response to the drug. In addition to that, a mathematical model of renal activity was needed. Two models were shortlisted. One, more articulated and capable of describing several physiological events, was discarded because it was too complex to integrate with the PKPD model. The second model was simpler and described a first-order, negative-feedback response of the kidneys to changes in plasma volume from its normal physiological level, which could be easily integrated with the furosemide PKPD model. Once the control simulation was set up, results showed that an automated MPC controlled diuretic administration can outperform standard diuretic treatment in case of congestive heart failure by achieving comparatively higher level of decongestion in less time.
2022
Modeling and Control of Loop Diuretic Therapy in Congestive Heart Failure: A Feasibility Study
Congestive heart failure (CHF) is a condition caused by the heart’s inability to correctly circulate blood to peripheral tissues, which results in a persistent state of increased plasma volume, due to the overactivation of neurohumoral complexes that induce the body to retain renal sodium and water. The most common therapy for decongestion is the administration of diuretics, in particular loop diuretics such as furosemide. Since the effects of diuretic drugs in case of CHF can largely vary between patients, the actual course of the treatment is not standardised, and its efficacy relies on the physician’s expertise. This work aimed to implement a control system, using a model predictive control (MPC) strategy, that could automatically and dynamically adapt the administration of furosemide by predicting the patient’s response to the treatment based on the patient’s characteristics, in order to achieve decongestion more efficiently than with standard therapy. As part of the simulation set up, literature was extensively consulted in the search for a pharmacokinetic and pharmacodynamic (PKPD) model of furosemide, finally settling on a three-compartment representation of the pharmacokinetics and a four-parameters sigmoidal pharmacodynamic response that describes the urine production in response to the drug. In addition to that, a mathematical model of renal activity was needed. Two models were shortlisted. One, more articulated and capable of describing several physiological events, was discarded because it was too complex to integrate with the PKPD model. The second model was simpler and described a first-order, negative-feedback response of the kidneys to changes in plasma volume from its normal physiological level, which could be easily integrated with the furosemide PKPD model. Once the control simulation was set up, results showed that an automated MPC controlled diuretic administration can outperform standard diuretic treatment in case of congestive heart failure by achieving comparatively higher level of decongestion in less time.
Loop diuretic
Heart failure
Modeling
MPC
Kidney modeling
File in questo prodotto:
File Dimensione Formato  
Semenzato_Corinna.pdf

accesso aperto

Dimensione 2.99 MB
Formato Adobe PDF
2.99 MB Adobe PDF Visualizza/Apri

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/46249