Neural Control of Pressure Support Ventilation Improved Patient-Ventilator Synchrony in Patients with Different Respiratory System Mechanical Properties: A Prospective, Crossover Trial
Pressure support ventilation (PSV) is one of the most widely used modes of mechanical ventilation for patients with respiratory failure. It aims to reduce the patient’s work of breathing by providing partial ventilatory support. However, conventional pressure support ventilation (PSP), which relies on pneumatic signals such as flow or pressure to trigger and cycle off the ventilator, often results in patient-ventilator asynchrony. This asynchrony can lead to increased inspiratory effort, prolonged mechanical ventilation, and worsened clinical outcomes. To address these limitations, neurally controlled pressure support (PSN) has been proposed as an alternative. PSN uses the electrical activity of the diaphragm (EAdi) to trigger and cycle off the ventilator, theoretically improving patient-ventilator synchrony and reducing inspiratory effort.
This study aimed to compare the effects of PSN and PSP on patient-ventilator synchrony, inspiratory effort, and breathing patterns in patients with different respiratory system mechanical properties. The study included three groups of patients: post-operative patients without respiratory comorbidities, patients with acute respiratory distress syndrome (ARDS) and restrictive acute respiratory failure (ARF), and patients with chronic obstructive pulmonary disease (COPD) and mixed restrictive and obstructive ARF. The primary objective was to evaluate the extent of patient-ventilator asynchrony, as measured by the total asynchrony index (AI), and to assess the impact of PSN and PSP on inspiratory effort and breathing patterns.
The study enrolled 24 patients, divided into three groups of eight patients each. The post-operative group consisted of patients who had undergone recent surgery and had no respiratory comorbidities. The ARDS group included patients with ARDS and restrictive ARF, characterized by low static compliance of the respiratory system (CRS). The COPD group comprised patients with COPD and mixed restrictive and obstructive ARF, characterized by high respiratory system resistance (RRS). Patients were excluded if they had conditions that could interfere with the study, such as neuromuscular diseases, hemodynamic instability, or contraindications for nasogastric tube insertion.
The study used a crossover design, with each patient receiving both PSP and PSN at two levels of support: 100% and 150% of the initial pressure support level. The initial pressure support level was set to achieve a tidal volume (VT) of 6 mL/kg of predicted body weight (PBW). The 150% support level was achieved by increasing the pressure support to 1.5 times the initial level. Each condition was maintained for 20 minutes, and the order of the modes was randomized.
PSP was delivered using a flow trigger of 1.4 L/min and a cycling-off criterion of 30% of peak inspiratory flow. PSN was delivered using the neurally adjusted ventilatory assist (NAVA) mode, with the NAVA level set to the maximum (15 cmH2O/mV) and upper pressure limits adjusted to achieve the targeted pressure support above positive end-expiratory pressure (PEEP). PSN was triggered by EAdi (0.5 mV) and cycled off when EAdi fell to 70% of its peak value.
Patient-ventilator interactions were analyzed using macro and micro asynchronies. Macro asynchronies included ineffective triggering, auto-triggering, and double triggering. Micro asynchronies included inspiratory trigger delay, premature cycling, and late cycling. The total AI was calculated as the sum of macro and micro asynchronies divided by the total neural respiratory rate. Inspiratory effort was assessed using the pretrigger Pes-time product (PTPes-trig) and the total inspiratory Pes-time product (PTPes).
The results showed that PSN consistently reduced the total AI compared to PSP in all patient groups. In the COPD group, the total AI was significantly lower during PSN at both 100% (3% vs. 93%, P = 0.012) and 150% (8% vs. 104%, P = 0.012) support levels. Similarly, in the ARDS group, the total AI was lower during PSN at 100% (8% vs. 29%, P = 0.012) and 150% (16% vs. 41%, P = 0.017) support levels. In post-operative patients, the total AI was also lower during PSN at 100% (21% vs. 35%, P = 0.012) and 150% (15% vs. 50%, P = 0.017) support levels.
The reduction in total AI during PSN was primarily due to a decrease in micro asynchronies, particularly inspiratory trigger delay and cycling-off errors. In patients with COPD, PSN significantly reduced late cycling-off errors, which were prevalent during PSP. In ARDS and post-operative patients, PSN improved cycling-off errors by bringing the values closer to zero, indicating better synchronization between the ventilator and the patient’s neural inspiratory effort.
Trigger delays were also significantly reduced during PSN. In patients with COPD, the median trigger delay decreased from 197 ms during PSP to 81 ms during PSN at the 100% support level. In ARDS patients, the trigger delay decreased from 126 ms during PSP to 81 ms during PSN. In post-operative patients, the trigger delay decreased from 144 ms during PSP to 81 ms during PSN. These reductions in trigger delay were consistent across both support levels.
The inspiratory effort for triggering (PTPes-trig) and total inspiratory effort (PTPes) were significantly lower during PSN compared to PSP in patients with COPD and ARDS. In COPD patients, PTPes-trig decreased from 0.6 cmH2O·s/min during PSP to 0.3 cmH2O·s/min during PSN at the 100% support level. In ARDS patients, PTPes-trig decreased from 0.4 cmH2O·s/min during PSP to 0.2 cmH2O·s/min during PSN. In post-operative patients, the reduction in inspiratory effort was less pronounced but still significant at the 100% support level.
The breathing patterns, including tidal volume (VT), respiratory rate (RR), and neural inspiratory time (TiN), were similar between PSN and PSP in all patient groups. However, increasing the support level from 100% to 150% led to significant increases in VT and peak airway pressure (Ppeak) and a decrease in peak EAdi in all groups. In COPD patients, the increase in support level also led to a decrease in RR during PSP but not during PSN.
The study concluded that PSN improves patient-ventilator synchrony and generates a respiratory pattern similar to PSP, regardless of the level of support or the patient’s respiratory system mechanical properties. PSN reduces the trigger and total inspiratory effort in patients with COPD or ARDS, making it a viable alternative to PSP for these patients. The study also highlighted that increasing the support level during PSP worsens patient-ventilator synchrony, while PSN maintains synchrony even at higher support levels.
In summary, PSN offers significant advantages over PSP in terms of patient-ventilator synchrony and inspiratory effort reduction, particularly in patients with COPD or ARDS. The use of EAdi to control ventilation ensures that the ventilator’s assistance is more closely aligned with the patient’s neural respiratory drive, reducing asynchronies and improving patient comfort. These findings suggest that PSN should be considered as an alternative mode of ventilation for patients with respiratory failure, especially those with complex respiratory mechanics.
doi.org/10.1097/CM9.0000000000001357
Was this helpful?
0 / 0