Neurally adjusted ventilatory assist vs pressure support ventilation: short-term effects on shunt and dead space...

Postoperative pulmonary complications, particularly atelectasis, are common after cardiac surgery and may contribute to impaired gas exchange or acute lung injury (ALI). Neurally Adjusted Ventilatory Assist (NAVA) delivers ventilatory support proportional to the patient's respiratory drive, offering theoretical advantages over Pressure Support Ventilation (PSV), including improved synchrony, enhanced diaphragmatic efficiency, and reduced risk of ventilator-induced lung injury. However, comparative data on gas exchange, dead space, and regional ventilation during weaning after cardiac surgery remain limited. This prospective crossover study evaluated 12 mechanically ventilated patients with mild ALI following cardiac surgery across three ventilation phases: two PSV phases (PSV1 and PSV2) separated by a phase of NAVA. Intrapulmonary shunt fraction was calculated from measurements obtained via a Swan-Ganz catheter. Physiological dead space fraction (V_D/V_T) was assessed using three methods: the Bohr-Enghoff equation, end-tidal CO₂-derived alveolar dead space fraction (AVDSf-ET), and a novel time-to-volume converted capnographic approach (V_CAP-CALC). Regional ventilation was assessed using electrical impedance tomography (EIT), and neuroventilatory efficiency (NVE) was calculated from diaphragmatic electrical activity (EAdi). Data were analyzed using linear mixed-effects models to account for repeated measures and within-subject variability. V_D/V_T was significantly lower during NAVA compared with PSV1 and PSV2 when assessed by V_CAP-CALC (58.5% vs. 63.8% and 61.3%, respectively; p < 0.001). The PaO₂/FiO₂ ratio and NVE were significantly higher during NAVA (p = 0.01 and p = 0.037, respectively). No significant difference in pulmonary shunt fraction was observed. EIT revealed a modest increase in dorsal end-expiratory lung volume during NAVA, without redistribution of tidal volume or Center of Ventilation. The V_CAP-CALC method showed strong agreement with established dead space measures (R2 = 0.77-0.82) and demonstrated high repeatability (mean coefficient of variation 3.5%). NAVA is a safe and feasible ventilatory mode following cardiac surgery, associated with reduced dead space fraction, improved oxygenation and enhanced neuroventilatory efficiency. Given that shunt fraction remained unchanged, the observed improvement in ventilation-perfusion (V/Q) matching reflects a reduction in V_D/V_T. The potential implications for postoperative recovery and long-term outcomes merit evaluation in larger clinical studies.ClinicalTrials.gov: NCT03217305. Initial Release 21/06/2017.