Effects of an asymmetrical high flow nasal cannula interface in hypoxemic patients

BACKGROUND: Optimal noninvasive respiratory support for patients with hypoxemic respiratory failure should minimize work of breathing without increasing the transpulmonary pressure. Recently, an asymmetrical high flow nasal cannula (HFNC) interface (Duet, Fisher & Paykel Healthcare Ltd), in which the caliber of each nasal prong is different, was approved for clinical use. This system might reduce work of breathing by lowering minute ventilation and improving respiratory mechanics. METHODS: We enrolled 10 patients ≥ 18 years of age who were admitted to the Ospedale Maggiore Policlinico ICU in Milan, Italy, and had a PaO(2)/FiO(2) < 300 mmHg during HFNC support with a conventional cannula. We investigated whether the asymmetrical interface, compared to a conventional high flow nasal cannula, reduces minute ventilation and work of breathing. Each patient underwent support with the asymmetrical interface and the conventional interface, applied in a randomized sequence. Each interface was provided at a flow rate of 40 l/min followed by 60 l/min. Patients were continuously monitored with esophageal manometry and electrical impedance tomography. RESULTS: Application of the asymmetrical interface resulted in a −13.5 [−19.4 to (−4.5)] % change in minute ventilation at a flow rate of 40 l/min, p = 0.006 and a −19.6 [−28.0 to (−7.5)] % change at 60 l/min, p = 0.002, that occurred despite no change in PaCO(2) (35 [33–42] versus 35 [33–43] mmHg at 40 l/min and 35 [32–41] versus 36 [32–43] mmHg at 60 l/min). Correspondingly, the asymmetrical interface lowered the inspiratory esophageal pressure–time product from 163 [118–210] to 140 [84–159] (cmH(2)O*s)/min at a flow rate of 40 l/min, p = 0.02 and from 142 [123–178] to 117 [90–137] (cmH(2)O*s)/min at a flow rate of 60 l/min, p = 0.04. The asymmetrical cannula did not have any impact on oxygenation, the dorsal fraction of ventilation, dynamic lung compliance, or end-expiratory lung impedance, suggesting no major effect on PEEP, lung mechanics, or alveolar recruitment. CONCLUSIONS: An asymmetrical HFNC interface reduces minute ventilation and work of breathing in patients with mild-to-moderate hypoxemic respiratory failure supported with a conventional interface. This appears to be primarily driven by increased ventilatory efficiency due to enhanced CO(2) clearance from the upper airway. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s13054-023-04441-6.