This thesis work addresses the respiratory care of preterm infants suffering from Respiratory Distress Syndrome. Preterm birth is defined as a delivery at less than 37 completed weeks, and can be further divided in late preterm (33-37 weeks), moderate preterm (28-32 weeks) and severe preterm (20-27 weeks). Preterm infants, especially severe preterm, suffer from lack of surfactant, their chest wall is poorly ossified and their intercostal muscle are not fully functional. All this causes, but especially the deficiency of surfactant, may leads to respiratory failure and, in the next hours to the birth, to a particular disease called Infant Respiratory Distress Syndrome (IRDS), characterized by rapid respiratory rate, respiration difficulties, expiratory grunting and cyanosis. This disease is quite common in preterm infants as the risk of developing RDS in infants born before the 30 week is about 50% and in the future it may contribute to the development of chronic lung disease or bronchopulmonary dysplasia. The most common therapy for IRDS in neonatal intensive care units (NICUs) is positive pressure mechanical ventilation (such as Continuous Positive Airways Pressure, CPAP, e Bilevel Positive Airways Pressure, BiPAP), often delivered in couple with surfactant administration. However, mechanical ventilation may cause major injuries to the lung, called Ventilator-Induced Lung Injury (VILI), which increase the morbidity of this pathology. In order to achieve enough gas exchange surface, promote lung fluid reabsorption, reduce the need of mechanical ventilation and prevent VILI, recruitment maneuvers (RMs) have been proposed. They aim to promote alveolar recruitment by applying a high-level pressure at airways opening for a given period of time. However, RMs effects and use is still controversial, in fact both in animal model experimentations and on infants clinical studies the hypothesis and the result obtained differs. Moreover, there is not accordance on the parameter used in recruitment maneuver, and they do not take into account the actual status of the lung that, for sure, it varies from infant to infant, since parameters are defined a priori. An incorrect use of RMs it cannot only have no effect on the health status of the preterm infant, but it may even overdistend the lung parenchyma causing additional adverse effects. All these issues limits the application of RMs in clinical practice, although if they were properly adapted to each infants they could improve the clinical outcome. Therefore there is necessity for a system that helps in the evaluation of current status of the lung so that RMs could be tuned (in pressure and time) for each different situation. At present, an ongoing study [26] proposes to drive RMs on lung volume during the maneuver. Although the results were promising, lung volume does not provide knowledge about the lung stress and, besides, this parameter is evaluated with a technique hardly applicable in clinical practice (Electrical Impedance Tomography, EIT). The estimation of mechanical properties of the respiratory system can overcome current limitations and may be a good indicator for optimizing the ventilation strategy and RMs in mechanically ventilated infants. Having mechanical knowledge about the lungs in fact could be used to assess the alveolar recruitment and the distention of the parenchyma. This could really help avoiding secondary injuries dealt by an incorrect ventilation modality. Unfortunately none of the traditional techniques used for the evaluation of lung function is suitable for this application, because they either require the cooperation of the subject or they are affected by respiratory muscles activity or by non-linearity of the system. A possible solution to overcome this technological problem is the use of Forced Oscillation Technique (FOT). The most attractive features of FOT are its non-invasiveness that permits it to be applied even in very preterm newborns and its independency from spontaneous breathing and from RMs and ventilation as external driving signal are superimposed on normal breathing. This kind of method permits to compute the impedance of the respiratory system, which its imaginary part evaluated at 5 Hz oscillations (respiratory reactance, Xrs) it has been proved to provide information on lung recruitment/distension. [13, 14] Currently there is no commercial device that allows measurements during recruitment maneuvers and permits to adapt RMs on lung mechanics bases. Therefore, the goal of this thesis project is the development of an adaptive system based on mechanical information of the respiratory system that can be used for recruiting the lung and to automatically adapt pressure and duration of the recruitment maneuver according to the chosen strategy in order to minimize lung stress, avoid ventilation induced lung injury and, hopefully, reduce the need for mechanical ventilation in preterm infants. To reach this goal a system composed of two major parts was designed and developed: an Electro-Pneumatic Unit and a Supervisor Unit. The Electro-Pneumatic unit is a low-level system constituted of: • A Control Unit, which is the core of the Electro-Pneumatic Unit and directs all the function of the low-level system. • A Pneumatic Unit, composed of a blower and a voice coil valve, that generates the flow and the pressure in the respiratory circuit. • A Signal condition and data acquisition system composed of two sensors, which measure the pressure and the flow inside the respiratory circuit and a conditioning board for their signal pre-processing. The Supervisor Unit is based on an A13-Olinuxino board which mounts Android. An Android application (BabyFlow) has been developed in order to: • Interface with the low-level system • Compute impedance in real time • Visualize real time graph and save values of sensor’s data and respiratory impedance • Provide a user-friendly UI to control and tune (online and offline) the implemented strategies and their parameters. On the Supervisor Unit, in addition to standard ventilation modalities (CPAP, BiPAP), following recruitment maneuvers were implemented: • All Fixed (AF): represent a standard SLI maneuver. The user is asked to manually tune a low-level pressure, a high-level one and a fixed duration. • Xrs Stability (XS): similar to AF, but the pressure is maintained high until Xrs reaches a stable condition or when the maneuver reached the maximum time allowed. A stable Xrs means that there are no more alveoli opening during the maneuver, so maintaining a high airway pressure cannot provide further significant improvement in lung recruitment and, additionally, it can be potentially harmful. • Xrs Target Stability (XTS): the algorithm aims to reach a user-defined Xrs target by automatically set the pressure to apply to the respiratory system. This maneuver stops when delivered pressure is too high, the target is stable or maximum time is reached. XTS tries to keep a constant lung stress adapting airway pressure automatically taking account on lung status changes during the maneuver. • Xrs AutoTarget Stability (XATS): similar to XTS, but the algorithm automatically select an Xrs target as the minimum Xrs reached after the Plow/Phigh transition. The minimum Xrs reached in the transition might represent a measure of the “tolerable” stress of the lung, so the system aims to maintain that target avoiding unnecessary overstressing of alveoli. The system was then tested both in vitro and in ex-vivo. In vitro studies shown that the developed system had an accuracy of the estimation of impedance comparable to a previous validated system, whereas on ex-vivo studied we analyzed the mechanical response of both excise lung and total respiratory system, also in presence of saline, to the implemented recruitment maneuvers.
Questo lavoro di tesi è indirizzato alla gestione dei trattamenti respiratori dei neonati prematuri affetti da sindrome da distress respiratorio (Respiratory Distress Syndrome, RDS). Un neonato è definito prematuro quando la sua età gestazionale è inferiore alle 37 settimane complete. La prematurità può essere ulteriormente classificata come tarda (33-37 settimane), moderata (28-32 settimane) e grave (20-27 settimane). I neonati prematuri, particolarmente quelli gravemente prematuri, soffrono di mancanza di surfattante, la loro parete toracica è scarsamente ossificata e i loro muscoli intercostali non sono pienamente funzionali. Queste condizioni, e in particolar modo la carenza di surfattante, possono portare a insufficienza respiratoria e, nelle ore prossime alla nascita, ad una particolare malattia chiamata sindrome da distress respiratorio infantile (Infant Respiratory Distress Syndrome, IRDS), caratterizzata da un’elevata frequenza respiratoria, difficoltà di respirazione, gemiti di fine espirazione e cianosi. Questa malattia è abbastanza comune nei neonati prematuri, infatti il rischio di sviluppare RDS in bambini nati prima di 30 settimane è circa pari al 50% e può portare allo sviluppo di malattie polmonari croniche tra cui broncopneumo displasia. La terapia più comune per la IRDS nelle unità di terapia intensiva neonatale (Neonatal Intensive Care Unit, NICU) è la ventilazione meccanica a pressione positiva (tra cui Continuous Positive Airways Pressure, CPAP, e Bilevel Positive Airways Pressure, BiPAP), spesso associata alla somministrazione di surfattante. Tuttavia, la ventilazione può causare gravi lesioni al polmone, definite come danni polmonari indotti dalla ventilazione (Ventilator-Induced Lung Injury, VILI), che aumentano la probabilità di insorgenza di malattie croniche respiratorie. Per ottenere una adeguata superficie di scambio dei gas, promuovere il riassorbimento dei liquidi polmonari, ridurre la necessità di ventilazione meccanica e prevenire i VILI, vengono eseguite delle manovre di reclutamento (Recruitment Maneuvers, RMs) che consistono nell’aumento di pressione alle vie aeree fino ad un livello specifico nel mantenerla per un tempo predeterminato. Tuttavia gli effetti e l'uso delle RMs sono ancora controversi, infatti sia in studi su modelli animali sia studi clinici su bambini le ipotesi e i risultati ottenuti sono diversi. Inoltre, non c’è conformità nei parametri utilizzati nelle RMs e non viene tenuto conto dello stato attuale del polmone che può variare notevolmente da bambino a bambino in quanto i parametri vengono definiti a priori. Un non corretto utilizzo delle RMs non solo potrebbe non portare un effettivo miglioramento della condizione del prematuro, ma addirittura potrebbe sovradistendere il tessuto polmonare causando ulteriori danni collaterali. Questi problemi limitano l’applicazione delle RMs nella pratica clinica, che se correttamente adattate ad ogni bambino potrebbe migliorarne l’esito clinico. Pertanto vi è la necessità di un sistema che aiuti nella valutazione dello stato del polmone in modo che le RMs possano essere regolate in tempo reale (in pressione e tempo) per ogni differente situazione. Attualmente è in corso uno studio [26] che propone di pilotare le RMs basandosi sul volume polmonare durante la manovra. Nonostante i risultati ottenuti siano incoraggianti, il volume polmonare non fornisce informazioni circa lo stato di stress del polmone e, oltretutto, la tecnica con cui questo parametro viene stimato (tomografia ad impedenza elettrica, Electrical Impedance Tomography, EIT) è difficilmente applicabile nella pratica clinica soprattutto alla nascita. La stima delle proprietà meccaniche del sistema respiratorio può superare i limiti odierni e potrebbe essere un buon indicatore per l’ottimizzazione delle strategie di ventilazione e delle manovre di reclutamento. La conoscenza della meccanica potrebbe essere infatti usata per stimare il reclutamento polmonare e la distensione del parenchima. Ciò potrebbe essere d’aiuto per evitare danni secondari dovuti ad una scorretta modalità di ventilazione. Purtroppo nessuna delle tecniche tradizionali per la valutazione della funzionalità polmonare è adatta per questa applicazione, perché richiedono la cooperazione del soggetto e sono influenzate dall’attività dei muscoli o dalle non-linearità del sistema. Una possibile soluzione per superare queste limitazioni tecnologiche è ricorrere alla tecnica delle oscillazioni forzate (Forced Oscillation Technique, FOT). Le caratteristiche più interessanti della FOT sono la sua non- invasività, che le permette di essere applicata anche in neonati prematuri, e la sua indipendenza dalla respirazione spontanea, dalle RMs e dalla ventilazione. Questo tipo di metodo permette di calcolare l'impedenza del sistema respiratorio di cui è stata dimostrata la specificità al reclutamento/dereclutamento polmonare della sua componente immaginaria (reattanza respiratoria, Xrs) valutata a 5 Hz. [13, 14] Oggigiorno non esistono dispositivi commerciali che permettano di misurare la meccanica polmonare durante una RM e adattarla di conseguenza. L'obiettivo di questo progetto di tesi è lo sviluppo di un sistema adattativo basato su informazioni meccaniche del sistema respiratorio per adattare automaticamente la pressione e la durata delle RMs, in modo da minimizzare lo stress polmonare, evitare i VILI e ridurre la necessità di ventilazione meccanica nei neonati prematuri. Per raggiungere questo obiettivo è stato progettato e sviluppato un sistema composto da due parti principali: una unità Elettro-Pneumatica e una unità Supervisor. L’unità Elettro-Pneumatica è un sistema di basso livello costituito da: • Un’unità di controllo, che rappresenta il nucleo dell’unità Elettro-Pneumatica e dirige tutte le funzioni del sistema di basso livello. • Un gruppo pneumatico, composto da un soffiatore e da una valvola voice-coil, che genera il flusso e la pressione nel circuito respiratorio. • Un sistema di condizionamento del segnale e di acquisizione dati composto da due sensori che misurano la pressione e il flusso all'interno del circuito respiratorio e da una scheda di condizionamento per la preelaborazione dei segnali. L'unità Supervisor si basa su una scheda A13-Olinuxino su cui è installato il sistema operativo Android. Su questa unità stata sviluppata un'applicazione (BabyFlow) al fine di: • Interfacciarsi con il sistema di basso livello • Calcolare l’impedenza del sistema respiratorio in tempo reale • Visualizzare i grafici in tempo reale e salvare i valori dei dati provenienti dai sensore e dell’impedenza respiratoria. • Fornire una interfaccia user-friendly per controllare e regolare (online e offline), le strategie messe in atto e i loro parametri. Sull'unità Supervisor, oltre alle modalità di ventilazione standard (CPAP, BiPAP), sono state implementate le seguenti manovre di reclutamento: • All Fixed (AF): rappresenta una manovra SLI standard. All' utente è richiesto di impostare manualmente un livello di pressione basso, uno alto e una durata della fase di alta pressione. • Xrs Stability (XS): simile ad AF, ma la pressione è mantenuta alta fino a che Xrs raggiunge una condizione stabile o quando la manovra ha raggiunto il tempo massimo consentito. Un Xrs stabile durante la manovra significa che non vengono reclutati nuovi alveoli, quindi il mantenimento di una elevata pressione alle vie aeree non può fornire un ulteriore significativo miglioramento nel reclutamento polmonare e, oltretutto, può essere potenzialmente dannoso. • Xrs Target Stability (XTS): l'algoritmo mira a raggiungere un target Xrs definito dall'utente regolando automaticamente la pressione da applicare al sistema respiratorio. Questa manovra si arresta quando la pressione erogata è troppo alta, il target è stabile o viene raggiunto il tempo massimo. XTS cerca di mantenere uno stress polmonare costante adattando la pressione alle vie aeree automaticamente in base ai cambiamenti di stato nel polmone durante la manovra. • Xrs AutoTarget Stability (XATS): simile alla XTS, ma l'algoritmo seleziona automaticamente un valore target di Xrs definito come la minima Xrs raggiunta dopo la transizione in pressione da un valore basso ad uno più alto scelto dall’utilizzatore. La Xrs minima raggiunta nella transizione potrebbe rappresentare una misura dello stress "tollerabile" dal polmone, quindi il sistema mira a mantenere quella condizione evitando inutili sollecitazioni sugli alveoli. Il sistema è stato poi testato sia in vitro che in ex-vivo. Gli studi in vitro hanno dimostrato che il sistema sviluppato fornisce una stima dell'impedenza paragonabile ad un sistema precedentemente validato, mentre in ex-vivo sono state analizzate le risposte del polmone isolato sia del sistema respiratorio totale, anche in presenza di soluzione fisiologica, alle RMs implementate.
An adaptive system for optimizing assisted ventilation and Recruitment Maneuvers
EDANTIPPE, PIETRO;CARDANI, MATTEO
2012/2013
Abstract
This thesis work addresses the respiratory care of preterm infants suffering from Respiratory Distress Syndrome. Preterm birth is defined as a delivery at less than 37 completed weeks, and can be further divided in late preterm (33-37 weeks), moderate preterm (28-32 weeks) and severe preterm (20-27 weeks). Preterm infants, especially severe preterm, suffer from lack of surfactant, their chest wall is poorly ossified and their intercostal muscle are not fully functional. All this causes, but especially the deficiency of surfactant, may leads to respiratory failure and, in the next hours to the birth, to a particular disease called Infant Respiratory Distress Syndrome (IRDS), characterized by rapid respiratory rate, respiration difficulties, expiratory grunting and cyanosis. This disease is quite common in preterm infants as the risk of developing RDS in infants born before the 30 week is about 50% and in the future it may contribute to the development of chronic lung disease or bronchopulmonary dysplasia. The most common therapy for IRDS in neonatal intensive care units (NICUs) is positive pressure mechanical ventilation (such as Continuous Positive Airways Pressure, CPAP, e Bilevel Positive Airways Pressure, BiPAP), often delivered in couple with surfactant administration. However, mechanical ventilation may cause major injuries to the lung, called Ventilator-Induced Lung Injury (VILI), which increase the morbidity of this pathology. In order to achieve enough gas exchange surface, promote lung fluid reabsorption, reduce the need of mechanical ventilation and prevent VILI, recruitment maneuvers (RMs) have been proposed. They aim to promote alveolar recruitment by applying a high-level pressure at airways opening for a given period of time. However, RMs effects and use is still controversial, in fact both in animal model experimentations and on infants clinical studies the hypothesis and the result obtained differs. Moreover, there is not accordance on the parameter used in recruitment maneuver, and they do not take into account the actual status of the lung that, for sure, it varies from infant to infant, since parameters are defined a priori. An incorrect use of RMs it cannot only have no effect on the health status of the preterm infant, but it may even overdistend the lung parenchyma causing additional adverse effects. All these issues limits the application of RMs in clinical practice, although if they were properly adapted to each infants they could improve the clinical outcome. Therefore there is necessity for a system that helps in the evaluation of current status of the lung so that RMs could be tuned (in pressure and time) for each different situation. At present, an ongoing study [26] proposes to drive RMs on lung volume during the maneuver. Although the results were promising, lung volume does not provide knowledge about the lung stress and, besides, this parameter is evaluated with a technique hardly applicable in clinical practice (Electrical Impedance Tomography, EIT). The estimation of mechanical properties of the respiratory system can overcome current limitations and may be a good indicator for optimizing the ventilation strategy and RMs in mechanically ventilated infants. Having mechanical knowledge about the lungs in fact could be used to assess the alveolar recruitment and the distention of the parenchyma. This could really help avoiding secondary injuries dealt by an incorrect ventilation modality. Unfortunately none of the traditional techniques used for the evaluation of lung function is suitable for this application, because they either require the cooperation of the subject or they are affected by respiratory muscles activity or by non-linearity of the system. A possible solution to overcome this technological problem is the use of Forced Oscillation Technique (FOT). The most attractive features of FOT are its non-invasiveness that permits it to be applied even in very preterm newborns and its independency from spontaneous breathing and from RMs and ventilation as external driving signal are superimposed on normal breathing. This kind of method permits to compute the impedance of the respiratory system, which its imaginary part evaluated at 5 Hz oscillations (respiratory reactance, Xrs) it has been proved to provide information on lung recruitment/distension. [13, 14] Currently there is no commercial device that allows measurements during recruitment maneuvers and permits to adapt RMs on lung mechanics bases. Therefore, the goal of this thesis project is the development of an adaptive system based on mechanical information of the respiratory system that can be used for recruiting the lung and to automatically adapt pressure and duration of the recruitment maneuver according to the chosen strategy in order to minimize lung stress, avoid ventilation induced lung injury and, hopefully, reduce the need for mechanical ventilation in preterm infants. To reach this goal a system composed of two major parts was designed and developed: an Electro-Pneumatic Unit and a Supervisor Unit. The Electro-Pneumatic unit is a low-level system constituted of: • A Control Unit, which is the core of the Electro-Pneumatic Unit and directs all the function of the low-level system. • A Pneumatic Unit, composed of a blower and a voice coil valve, that generates the flow and the pressure in the respiratory circuit. • A Signal condition and data acquisition system composed of two sensors, which measure the pressure and the flow inside the respiratory circuit and a conditioning board for their signal pre-processing. The Supervisor Unit is based on an A13-Olinuxino board which mounts Android. An Android application (BabyFlow) has been developed in order to: • Interface with the low-level system • Compute impedance in real time • Visualize real time graph and save values of sensor’s data and respiratory impedance • Provide a user-friendly UI to control and tune (online and offline) the implemented strategies and their parameters. On the Supervisor Unit, in addition to standard ventilation modalities (CPAP, BiPAP), following recruitment maneuvers were implemented: • All Fixed (AF): represent a standard SLI maneuver. The user is asked to manually tune a low-level pressure, a high-level one and a fixed duration. • Xrs Stability (XS): similar to AF, but the pressure is maintained high until Xrs reaches a stable condition or when the maneuver reached the maximum time allowed. A stable Xrs means that there are no more alveoli opening during the maneuver, so maintaining a high airway pressure cannot provide further significant improvement in lung recruitment and, additionally, it can be potentially harmful. • Xrs Target Stability (XTS): the algorithm aims to reach a user-defined Xrs target by automatically set the pressure to apply to the respiratory system. This maneuver stops when delivered pressure is too high, the target is stable or maximum time is reached. XTS tries to keep a constant lung stress adapting airway pressure automatically taking account on lung status changes during the maneuver. • Xrs AutoTarget Stability (XATS): similar to XTS, but the algorithm automatically select an Xrs target as the minimum Xrs reached after the Plow/Phigh transition. The minimum Xrs reached in the transition might represent a measure of the “tolerable” stress of the lung, so the system aims to maintain that target avoiding unnecessary overstressing of alveoli. The system was then tested both in vitro and in ex-vivo. In vitro studies shown that the developed system had an accuracy of the estimation of impedance comparable to a previous validated system, whereas on ex-vivo studied we analyzed the mechanical response of both excise lung and total respiratory system, also in presence of saline, to the implemented recruitment maneuvers.| File | Dimensione | Formato | |
|---|---|---|---|
|
2014_04_Cardani_Edantippe.pdf
solo utenti autorizzati dal 09/04/2017
Descrizione: Testo della tesi
Dimensione
6.03 MB
Formato
Adobe PDF
|
6.03 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/92533