Background. Mechanical ventilation, although often a life-saving intervention, can, in some cases, lead to significant harmful consequences. Particularly, from a pulmonary perspective, ventilator-induced lung injury (VILI) and patient self-inflicted lung injury (P-SILI) represent relevant complications, associated with poor outcomes. These conditions are mediated by tidal volumes and elevated transpulmonary pressures, generated -in the case of VILI- by the ventilator, and, in assisted ventilation, also by the patient’s own respiratory effort. To decrease the incidence of these events, it is essential to monitor tidal volume and driving pressure (static and dynamic), in addition to respiratory drive and muscular effort, quantified by measuring, respectively, the electrical activity of the diaphragm (EAdi) and the airway pressure during an expiratory occlusion manoeuvre (ΔPocc). EAdi, specifically, is detected by eight electrodes placed along a nasogastric tube. In this context we can find NAVA ventilation mode (Neurally Adjusted Ventilatory Assist), where the pressure support given by the ventilator is proportional to EAdi, based on a conversion factor chosen by the physician, called NAVA level. Aim of the study. The aim of our study is to describe the relationship between NAVA level and tidal volume, based on the hypothesis that different patient phenotypes exist according to this relationship. Secondarily, our study aims to investigate the changes of respiratory drive and effort in these subgroups during the variation of the support given by NAVA. Materials and methods. In our monocentric observational study 26 patients were ventilated with incremental NAVA levels (from 0.5 to 3 cmH₂O/μV) and for each step we measured these parameters: tidal volume, respiratory rate, respiratory system static driving pressure (ΔP), peak EAdi, ΔPocc, dynamic transpulmonary driving pressure (ΔPL, dyn). Furthermore, blood gases were recorded. Results: for each patient, the relationship between NAVA level and tidal volume normalized to ideal body weight (VT/IBW) was described as a straight line. Based on its angular coefficient, a clustering analysis was then performed through a K-means algorithm. Two groups were identified, later named “safe” (n=16) and “danger” (n=8). The group with the lowest slope, referred to as Safe, constantly showed tidal volumes within the range of lung protection, regardless of the ventilator’s support (7.7±1.7 ml/kg of IBW at a maximal NAVA level of 3 cmH₂O/μV); the group with the highest slopes, referred to as Danger, instead, reached clearly non-protective values with the increasing NAVA level (15±7 ml/kg of IBW at a maximal NAVA level of 3 cmH₂O/μV, p<0.05). The difference between Safe and Danger phenotypes was shown to be present also with the trend of ΔP (Safe: 9±3; Danger: 17±9; p<0.05) and ΔPL, dyn (Safe: 16 [12-24]; Danger: 40 [20-51] cmH₂O; p<0.05): both parameters, in the Danger group, exceeded the target values for lung protection during the increase in NAVA level. At the same time, differences were also found in drive and effort indices: as support increased, EAdi and ΔPocc decreased in the safe group (respectively p<0.0001 and p<0.001), while showed no significant reduction in the danger group (p=0.29 and p=0.05). There were no changes in pH, PaCO2, PaO2/FiO2. Conclusions. In our study, based on the relationship between NAVA level and tidal volume, the existence of two distinct phenotypes of patients undergoing ventilation with NAVA was shown. In the Safe group, stable and protective values of the various ventilatory parameters studied were kept regardless of the level of support. In contrast, in the Danger group, despite initial parameters equivalent to those of the safe group, an increased, non-chemoregulated respiratory drive led to altered tidal volumes and transpulmonary pressures at higher NAVA levels, potentially increasing the risk of VILI and P-SILI.
Presupposti. La ventilazione meccanica invasiva si associa in conseguenze lesive significative. In particolare, a livello polmonare, il danno indotto da ventilazione (VILI, Ventilator-Induced Lung Injury) e quello autoindotto dal paziente (P-SILI, Patient Self-Inflicted Lung Injury) rappresentano complicanze rilevanti, associate a prognosi sfavorevoli. Si tratta di condizioni determinate da volumi correnti e pressioni transpolmonari elevate, generati, nel caso del VILI, dal ventilatore e, nella ventilazione assistita, anche dagli sforzi del paziente. Per limitare l’incidenza di questi eventi è necessario monitorare volume corrente e driving pressure (statica e dinamica), oltre che drive respiratorio ed effort muscolare, valutati tramite la misurazione rispettivamente dell’attività elettrica del diaframma (EAdi) e della pressione delle vie aeree durante una manovra di occlusione espiratoria (ΔPocc). In questo contesto si inserisce la modalità di ventilazione NAVA (Neurally Adjusted Ventilatory Assist), in cui il supporto pressorio erogato dal ventilatore è proporzionale all’EAdi, in base a un coefficiente impostato dal medico, il NAVA level. Scopi. Lo scopo dello studio è descrivere la relazione tra NAVA level e volume corrente, nell’ipotesi che, in base a questa, esistano diversi fenotipi di pazienti. Secondariamente, lo studio mira a studiare i cambiamenti di drive ed effort in questi eventuali sottogruppi al variare del supporto NAVA. Materiali e metodi. I pazienti arruolati in questo studio sono stati ventilati con livelli di NAVA level incrementali (da 0.5 a 3 cmH₂O/μV) e per ogni step sono stati rilevati i seguenti parametri: volume corrente, frequenza respiratoria, driving pressure statica del sistema respiratorio (ΔP), EAdi picco, ΔPocc, driving pressure transpolmonare dinamica (ΔPL, dyn). Sono stati inoltre registrati valori dall’emogasanalisi arteriosa. Risultati. Sono stati arruolati 26 pazienti. Per ogni paziente la relazione tra NAVA level e volume corrente normalizzato sul peso corporeo ideale (VT/IBW) è stata descritta come una retta. In base al valore del coefficiente angolare è stata effettuata un’analisi di clustering attraverso algoritmo K-means, ottenendo due gruppi, successivamente soprannominati safe (n=18) e danger (n=8). Il gruppo con pendenza minore, safe, ha mostrato volumi correnti costantemente entro i valori target di lung protection, a prescindere dal supporto (7.7±1.7 ml/kg di peso corporeo ideale a NAVA level massimale di 3 cmH₂O/μV); il gruppo con pendenze maggiori, danger, invece, con la crescita del NAVA level ha raggiunto volumi francamente non protettivi (15±7 ml/kg di peso corporeo ideale a NAVA level massimale di 3 cmH₂O/μV, p<0.05). La differenza tra fenotipi safe e danger è stata dimostrata anche nell’andamento di ΔP (safe: 9±3 cmH₂O; danger: 17±9 cmH₂O; p<0.05) e ΔPL, dyn (safe: 16 [12-24] cmH₂O; danger: 40 [20-51] cmH₂O; p<0.05): entrambe, nel gruppo danger, sono fuoriuscite dai valori target di lung protection durante l’incremento del NAVA level. Sono state trovate, contestualmente, differenze anche negli indici di drive ed effort: all’aumentare del supporto EAdi e ΔPocc calano significativamente nel gruppo safe (rispettivamente p<0.0001 e p<0.001), mentre rimangono costanti nel gruppo danger (p=0.29 e p=0.05). Non ci sono stati cambiamenti in pH, PaCO2, PaO2/FiO2. Conclusioni. Nel nostro studio, sulla base della relazione tra NAVA level e volume corrente, è stata documentata l’esistenza di due fenotipi di pazienti ventilati tramite NAVA, safe e danger. Nel gruppo safe sono mantenuti valori stabili e protettivi dei vari parametri ventilatori studiati, a prescindere dal supporto. Nel gruppo danger, invece, nonostante parametri iniziali simili al safe, un aumentato drive non chemoregolato porta a volumi correnti e pressioni transpolmonari alterati a NAVA level elevati, potenzialmente aumentando il rischio di VILI e P-SILI.
Indici di drive ed effort respiratori nel paziente in ventilazione invasiva assistita a trigger neurale
DISSEGNA, PIERO
2024/2025
Abstract
Background. Mechanical ventilation, although often a life-saving intervention, can, in some cases, lead to significant harmful consequences. Particularly, from a pulmonary perspective, ventilator-induced lung injury (VILI) and patient self-inflicted lung injury (P-SILI) represent relevant complications, associated with poor outcomes. These conditions are mediated by tidal volumes and elevated transpulmonary pressures, generated -in the case of VILI- by the ventilator, and, in assisted ventilation, also by the patient’s own respiratory effort. To decrease the incidence of these events, it is essential to monitor tidal volume and driving pressure (static and dynamic), in addition to respiratory drive and muscular effort, quantified by measuring, respectively, the electrical activity of the diaphragm (EAdi) and the airway pressure during an expiratory occlusion manoeuvre (ΔPocc). EAdi, specifically, is detected by eight electrodes placed along a nasogastric tube. In this context we can find NAVA ventilation mode (Neurally Adjusted Ventilatory Assist), where the pressure support given by the ventilator is proportional to EAdi, based on a conversion factor chosen by the physician, called NAVA level. Aim of the study. The aim of our study is to describe the relationship between NAVA level and tidal volume, based on the hypothesis that different patient phenotypes exist according to this relationship. Secondarily, our study aims to investigate the changes of respiratory drive and effort in these subgroups during the variation of the support given by NAVA. Materials and methods. In our monocentric observational study 26 patients were ventilated with incremental NAVA levels (from 0.5 to 3 cmH₂O/μV) and for each step we measured these parameters: tidal volume, respiratory rate, respiratory system static driving pressure (ΔP), peak EAdi, ΔPocc, dynamic transpulmonary driving pressure (ΔPL, dyn). Furthermore, blood gases were recorded. Results: for each patient, the relationship between NAVA level and tidal volume normalized to ideal body weight (VT/IBW) was described as a straight line. Based on its angular coefficient, a clustering analysis was then performed through a K-means algorithm. Two groups were identified, later named “safe” (n=16) and “danger” (n=8). The group with the lowest slope, referred to as Safe, constantly showed tidal volumes within the range of lung protection, regardless of the ventilator’s support (7.7±1.7 ml/kg of IBW at a maximal NAVA level of 3 cmH₂O/μV); the group with the highest slopes, referred to as Danger, instead, reached clearly non-protective values with the increasing NAVA level (15±7 ml/kg of IBW at a maximal NAVA level of 3 cmH₂O/μV, p<0.05). The difference between Safe and Danger phenotypes was shown to be present also with the trend of ΔP (Safe: 9±3; Danger: 17±9; p<0.05) and ΔPL, dyn (Safe: 16 [12-24]; Danger: 40 [20-51] cmH₂O; p<0.05): both parameters, in the Danger group, exceeded the target values for lung protection during the increase in NAVA level. At the same time, differences were also found in drive and effort indices: as support increased, EAdi and ΔPocc decreased in the safe group (respectively p<0.0001 and p<0.001), while showed no significant reduction in the danger group (p=0.29 and p=0.05). There were no changes in pH, PaCO2, PaO2/FiO2. Conclusions. In our study, based on the relationship between NAVA level and tidal volume, the existence of two distinct phenotypes of patients undergoing ventilation with NAVA was shown. In the Safe group, stable and protective values of the various ventilatory parameters studied were kept regardless of the level of support. In contrast, in the Danger group, despite initial parameters equivalent to those of the safe group, an increased, non-chemoregulated respiratory drive led to altered tidal volumes and transpulmonary pressures at higher NAVA levels, potentially increasing the risk of VILI and P-SILI.| File | Dimensione | Formato | |
|---|---|---|---|
|
Dissegna_Piero.pdf
Accesso riservato
Dimensione
2.35 MB
Formato
Adobe PDF
|
2.35 MB | Adobe PDF |
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
https://hdl.handle.net/20.500.12608/86969