Design and simulation-based evaluation of a novel tracheostomy high-flow therapy device interface: a preclinical study

Introduction

Tracheostomy is performed in 10–15% of critically ill patients requiring prolonged mechanical ventilation to facilitate the weaning process (Abe et al., 2018; Aquino Esperanza, Pelosi & Blanch, 2019), with increasing adoption over the past decade (Esteban et al., 2013). Unlike conventional oxygen therapy, high-flow tracheal oxygen (HFTO) delivers inhaled gas with stable temperature, humidity, and oxygen concentration. This maintains mucociliary clearance, ensures airway hydration, and promotes normothermia. Case reports document successful weaning with HFTO (Mitaka et al., 2018; Lionello et al., 2023; Lytra et al., 2024).

However, as an upper airway-bypassing system, HFTO generates minimal positive end-expiratory pressure (PEEP). Limited studies (Moorhouse et al., 2015; Mitaka et al., 2018; Chen et al., 2019; Natalini et al., 2019) have quantified the PEEP delivered by HFTO, which is no more than 2 cmH₂O—insufficient to improve respiratory and cardiac function. This evidence-practice gap may limit the clinical adoption of HFTO in patients with tracheostomy.

Chen et al. (2019) modified the HFTO system by adding a 5-cmH₂O/L/s resistor to the expiratory port and found it induced an elevation in airway pressure. We hypothesized that adding an exhalation valve to the exhalation port of a traditional interface would significantly increase PEEP levels beyond baseline, and accordingly designed a novel HFTO interface (Chinese Utility Model Patent ZL2024 2 0280842.6). We conducted a randomized crossover study to evaluate the effects of the new interface and the traditional interface on PEEP and other respiratory physiological indicators across different flow rates.

Materials and Methods

We conducted an in vitro laboratory experiment to quantify and characterize the PEEP generated by high-flow therapy (HFT). This methodology was implemented to specifically assess HFT’s isolated mechanical contribution to PEEP without physiological interference. The experiment was undertaken in February 2024. Ethics approval was not required for this study, as it involved no human participants, animals, or biological materials, and was classified as non-biomedical device research under institutional guidelines.

The apparatus involved in the benchtop experiment (Fig. 1)

HFT device:

The Micomme HFT Device (OH-70C, Hunan, China) and its matching heated delivery tube were selected. Gas flow rates ranged from 10 to 80 L/min, with oxygen concentration adjustable from 21% to 100% and temperature from 29 °C to 37 °C. Its advantage lies in the built-in accurate oxygen concentration monitoring module, which can accurately monitor the oxygen concentration without loading oxygen batteries and other consumables.

High-flow tracheal interface (HFTI):

The traditional HFTI (tHFTI, HFC-09-P-T, Hunan, China, Fig. 1A) adopts the tracheal interface configured by Micomme HFT Device (OH-70C, Hunan, China).

A novel HFTI (nHFTI) has been developed and patented. The core innovation of nHFTI is a 7.5-mm diameter exhalation valve integrated at the exhalation port to increase exhalation resistance. The nHFTI features an 80-mm-long four-way tubular structure with dimensions comparable to the tHFTI. Its ports comprise:

• Exhalation port (ID: 13 mm) with integrated valve

• Inhalation port (ID: 13 mm)

• Patient port (ID: 15 mm)

• Sealed spare port (ID: 13 mm; used as an exhalation end during bronchoscopy or suctioning procedures)

The exhalation port of the HFTI has an internal diameter of 7.5 mm, comparable to that of a standard adult endotracheal tube. The pathway is designed to be short and straight to minimize secretion retention. The interface is intended for use with a commercial high-flow device equipped with real-time pressure monitoring and high-pressure alarms, which detect and alert clinicians to any obstruction in the circuit.

The 3D-printed nHFTI is shown in Fig. 1B and comparison of the two is shown in Fig. 1C.

Tracheostomy tube and simulated trachea:

A size 8 cuffed Shiley tracheostomy tube (313-80, Covidien LLC, Downey, CA, USA) was used. A 100-mm-long ventilator tube with a 22-mm inner diameter was employed to simulate the trachea of an average adult.

High-fidelity lung simulator:

A high-fidelity lung simulator (HFLS, TestChest^®, ORGANIS GmbH, Switzerland, Software version 5.07) was used as a reference tool to simulate normal human lungs in the benchtop study. The TestChest^® is a full physiologic artificial lung that can simulate gas exchange and hemodynamic responses of the healthy and pathological adult lungs. The device was calibrated according to the manufacturer’s description.

Non-invasive ventilator (NIV):

The Yi’an VG55 NIV (Hebei Yi’an Aomei Medical Equipment Co., Ltd, China) was used to provide a constant airflow, simulating HFT output through a heating and humidification device to maintain normal HFT operation and prevent humidified gases from entering and damaging the simulated lung.

Establish an HFLS-based trachea-tracheostomy tube-HFTI-HFT device model (Figs. 1D–1F)

The simulated trachea was connected to the airway interface of the HFLS. The tracheostomy tube was inserted into the simulated trachea, and its cuff was inflated with 10 mL of air. The seal was checked to confirm the absence of air leakage. The air supply tube of the HFT device was directly connected to the tracheostomy tube via the HFTI, bypassing the heating and humidification device. This configuration delivered dry gas output from the HFT device directly to the HFLS through the HFTI, preventing humidified gas from damaging the simulated lung and affecting test results.

However, bypassing the heated humidifier could trigger alarms or cause malfunction in the HFT device. To maintain normal operation, we connected the air supply tube of the NIV to the inlet of the heated humidifier, and the outlet was connected to a simplified test lung—equipped with a platform outlet valve on the adjacent tubing—via a bypass circuit. The NIV was then activated in CPAP mode at 6 cmH₂O to generate a constant airflow through the heated humidifier. This simulated the typical output of the HFT system and ensured stable operation of the device during testing. The gas flow pathways are shown in Fig. 2.

Results

Changes in the primary endpoint (PEEP, the positive end-expiratory pressure)

As the HFT flow rate increased, PEEP significantly increased with both tHFTI and nHFTI (P < 0.010). Compared with tHFTI, the nHFTI generated significantly higher PEEP at each flow rate (P < 0.010, Table 1, Fig. 3A).

Table 1:

Primary and secondary endpoint parameters during high-flow tracheal oxygen in the bench experiment.

Parameter	HFTI	Flow rate (L/min)
		10	20	30	40	50	60	70	80	Pa
PEEP (cmH2O)	tHFTI	0.1 ± 0.0	0.3 ± 0.1	0.6 ± 0.1	1.0 ± 0.0	1.9 ± 0.1	3.1 ± 0.2	4.4 ± 0.1	5.7 ± 0.1	<0.001
	nHFTI	0.2 ± 0.1	0.9 ± 0.1	1.7 ± 0.1	3.4 ± 0.1	5.4 ± 0.1	7.3 ± 0.2	9.6 ± 0.2	12.7 ± 0.2	<0.001
	P*	0.004	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
FRC (mL)	tHFTI	2,031.9 ± 1.3	2,039.4 ± 0.5	2,053.3 ± 1.1	2,069.2 ± 1.3	2,097.2 ± 2.3	2,145.6 ± 4.0	2,278.0 ± 2.5	2,465.0 ± 3.3	<0.001
	nHFTI	2,031.8 ± 0.7	2,054.3 ± 0.8	2,086.3 ± 1.1	2,155.0 ± 3.3	2,375.2 ± 2.3	2,610.0 ± 6.2	2,877.9 ± 4.0	3,282.8 ± 6.7	<0.001
	P*	0.759	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
Vt (mL)	tHFTI	504.3 ± 8.2	502.3 ± 12.7	494.1 ± 2.3	489.9 ± 8.7	481.0 ± 11.7	475.9 ± 12.3	469.4 ± 12.8	459.8 ± 18.3	<0.001
	nHFTI	505.6 ± 5.4	496.2 ± 7.9	476.1 ± 7.8	454.2 ± 11.3	426.2 ± 11.3	403.1 ± 3.7	379.1 ± 6.1	343.7 ± 5.7	<0.001
	P*	0.679	0.211	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
Ptp-EI (cmH2O)	tHFTI	6.8 ± 0.2	6.9 ± 0.1	6.9 ± 0.1	7.0 ± 0.2	7.3 ± 0.2	7.6 ± 0.1	8.0 ± 0.0	8.6 ± 0.1	<0.001
	nHFTI	7.1 ± 0.2	7.4 ± 0.2	7.5 ± 0.1	7.9 ± 0.2	8.3 ± 0.1	8.9 ± 0.1	9.6 ± 0.2	11.4 ± 0.1	<0.001
	P*	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
Ptp-EE (cmH2O)	tHFTI	0.1 ± 0.1	0.3 ± 0.1	0.5 ± 0.1	1.1 ± 0.1	1.7 ± 0.1	2.9 ± 0.1	3.4 ± 0.1	4.0 ± 0.1	<0.001
	nHFTI	0.1 ± 0.1	0.6 ± 0.0	1.5 ± 0.1	2.9 ± 0.1	3.8 ± 0.1	4.6 ± 0.1	5.6 ± 0.1	6.7 ± 0.1	<0.001
	P*	0.500	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
PIP (cmH2O)	tHFTI	−1.7 ± 0.1	−1.7 ± 0.2	−1.6 ± 0.2	−1.4 ± 0.1	−1.1 ± 0.2	−0.6 ± 0.1	0.6 ± 0.2	1.9 ± 0.1	<0.001
	nHFTI	−1.8 ± 0.2	−1.7 ± 0.1	−1.2 ± 0.3	−1.0 ± 0.2	0.7 ± 0.2	2.7 ± 0.2	4.6 ± 0.3	7.0 ± 0.2	<0.001
	P*	0.089	0.472	0.003	<0.001	<0.001	<0.001	<0.001	<0.001
PEP (cmH2O)	tHFTI	1.5 ± 0.1	1.9 ± 0.2	2.7 ± 0.1	3.3 ± 0.1	4.2 ± 0.1	5.5 ± 0.1	6.7 ± 0.3	8.4 ± 0.2	<0.001
	nHFTI	2.1 ± 0.2	3.2 ± 0.1	3.8 ± 0.4	5.4 ± 0.2	7.3 ± 0.4	9.0 ± 0.1	11.4 ± 0.2	14.9 ± 0.3	<0.001
	P*	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
PEIC (cmH2O)	tHFTI	−6.9 ± 0.1	−6.9 ± 0.1	−6.8 ± 0.1	−6.7 ± 0.1	−6.5 ± 0.1	−6.2 ± 0.1	−5.6 ± 0.1	−4.9 ± 0.1	<0.001
	nHFTI	−6.8 ± 0.1	−6.8 ± 0.1	−6.8 ± 0.1	−6.5 ± 0.1	−5.8 ± 0.2	−4.4 ± 0.1	−3.1 ± 0.1	−2.1 ± 0.1	<0.001
	P*	0.400	0.230	0.492	0.001	<0.001	<0.001	<0.001	<0.001
PEEC (cmH2O)	tHFTI	0.16 ± 0.01	0.20 ± 0.01	0.27 ± 0.01	0.37 ± 0.01	0.56 ± 0.02	1.21 ± 0.06	1.96 ± 0.08	2.79 ± 0.07	<0.001
	nHFTI	0.16 ± 0.01	0.28 ± 0.01	0.47 ± 0.03	1.17 ± 0.09	2.19 ± 0.11	3.35 ± 0.08	4.70 ± 0.11	6.64 ± 0.13	<0.001
	P*	0.310	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001
FiO2 (%)	tHFTI	39.2 ± 0.7	43.6 ± 0.4	47.4 ± 0.5	48.7 ± 1.2	49.6 ± 1.1	50.4 ± 0.9	50.9 ± 1.1	51.1 ± 0.8	<0.001
	nHFTI	38.8 ± 0.7	43.6 ± 1.6	47.7 ± 0.8	49.0 ± 0.4	49.8 ± 0.4	50.6 ± 1.1	51.1 ± 0.2	51.2 ± 0.4	<0.001
	P*	0.132	0.939	0.274	0.478	0.489	0.532	0.561	0.741

DOI: 10.7717/peerj.20445/table-1

Notes:

FiO₂, fraction of inspired oxygen; FRC, functional residual capacity; HFTI, high-flow tracheal interface; nHFTI, novel high-flow tracheal interface; PEEC, end-expiratory cardiac pressure; PEEP, positive end-expiratory pressure; PEIC, end-inspiratory cardiac pressure; PEP, peak expiratory pressure; PIP, peak inspiratory pressure; Ptp, transpulmonary pressure; Ptp-EE, end-expiratory transpulmonary pressure; Ptp-EI, end-inspiratory transpulmonary pressure; tHFTI, traditional high-flow tracheal interface; Vt, the tidal volume.

*Between tHFTI and nHFTI.

aAmong flow rates.

Discussion

At the same flow rate, the novel high-flow tracheal interface (nHFTI) generated higher PEEP, peak inspiratory pressure (PIP), peak expiratory pressure (PEP), end-inspiratory transpulmonary pressure (Ptp-EI), end-expiratory transpulmonary pressure (Ptp-EE), end-inspiratory cardiac pressure (PEIC) and end-expiratory cardiac pressure (PEEC) compared to the traditional high-flow tracheal interface (tHFTI). Additionally, it also led to an increase in functional residual capacity (FRC). In contrast, fraction of inspired oxygen (FiO₂) remained relatively stable with both interfaces, while tidal volume (Vt) was slightly lower with nHFTI.

PEEP is the positive pressure that remains in the airways at the end of the respiratory cycle (end of exhalation) and is greater than the atmospheric pressure in mechanically ventilated patients. The application of PEEP was first described in the 1930s (Poulton & Oxon, 1936; Barach, Martin & Eckman, 1938) and came into common use for treating acute respiratory distress syndrome (ARDS) in the 1960s (Ashbaugh et al., 1967).

Few studies have explored the airway pressures delivered by HFTO and how they vary with different gas flow rates. In a benchtop experiment, Thomas et al. (Thomas, Joshi & Cave, 2021) found that HFT can generate measurable and variable PEEP despite the open system design. At flow rates of 40, 50, and 60 L/min, PEEP levels were approximately 0.3, 0.5 and 0.9 cmH₂O, respectively. In a single-center randomized crossover study, Natalini et al. (2019) compared standard oxygen therapy with HFTO in tracheostomized patients. Peak and mean expiratory pressures with HFTO at 50 L/min were 1.8 (1.4–2.2) cmH₂O and 1.2 (1–1.5) cmH₂O, respectively.

The study also compared HFTO and high-flow nasal cannula oxygen (HFNO) in a subgroup of five patients who underwent tracheostomy decannulation, and found that compared with HFNO, changes in expiratory pressure were significantly smaller with HFTO. At 50 L/min flow, the median peak expiratory pressure was 5.1 cmH₂O with HFNO and 1.8 cmH₂O with HFTO; the mean expiratory pressure was 3.9 cmH₂O with HFNO and 1.5 cmH₂O with HFTO. This within-patient comparison of HFTO and HFNO provided a unique opportunity to highlight the contribution of upper airway resistance to positive-pressure generation during HFNO.

We hypothesized that therapeutic PEEP could be generated by adding an exhalation valve to the exhalation port of the traditional interface to simulate nasal resistance during exhalation. In our benchtop study, the nHFTI was modified by incorporating an exhalation valve, thereby mimicking the nasal resistance during expiration and resulting in the generation of significant PEEP. In contrast to HFNO, which generate minimal intrathoracic PEEP (Rubin et al., 2014), the nHFTI is designed to intentionally modulate expiratory resistance to achieve therapeutic PEEP levels. At 80 L/min, maximum PEEP reached 12.7 ± 0.2 cmH₂O, while at 50 L/min, PEEP was 5.4 ± 0.1 cmH₂O, which was comparable to the maximum PEEP (5.7 ± 0.1 cmH₂O) generated by tHFTI at 80 L/min.

PEEP generated by nHFTI in our bench study was approximately 3.4 cmH₂O at a flow rate of 40 L/min, which was comparable to the pharyngeal pressure reported in adult patients receiving HFNO (Groves & Tobin, 2007; Parke, McGuinness & Eccleston, 2009; Corley et al., 2011; Parke, Eccleston & McGuinness, 2011). Previous HFNO studies showed that the airway pressure was determined by flow, with a linear (Groves & Tobin, 2007; Parke, Eccleston & McGuinness, 2011) or quadratic (Kumar, Spence & Tawhai, 2015; Luo et al., 2017) relationship between pressure and flow rate. In our bench model with nHFTI, PEEP and flow rate fit well with a power function: PEEP = 0.002 × Flow^1.986 (R² = 0.998, see Fig. 4), suggesting a much steeper increase in PEEP at higher flow rates.

PEEP increases FRC, thereby improving oxygenation through recruitment of collapsed alveoli and reduction of intrapulmonary shunt. In heart failure, PEEP may enhance hemodynamics by reducing venous return, decreasing left-ventricular afterload, and lowering transmural pressure. Excessive PEEP, however, can impair cardiac output and cause alveolar overdistension, potentially leading to volutrauma or barotrauma.

Our research confirmed that nHFTI produced higher PEEP than tHFTI. Additionally, the nHFTI slightly increased external cardiac pressure at both end-inspiration and end-expiration (PEIC and PEEC, particularly PEEC), primarily due to the higher effective PEEP. This addresses limitations of traditional interfaces for heart failure management. Based on our findings, the nHFTI can generate a PEEP of approximately 3–7 cmH₂O at gas flow rates between 40 and 60 L/min. The increase in PEEP and external cardiac pressure observed with nHFTI was mild under benchtop conditions. These findings suggest that the device may produce hemodynamic effects that are less likely to impair cardiac output or induce hypotension, which are potential concerns with higher levels of PEEP in clinical settings.

Pulmonary complications such as alveolar collapse and atelectasis are common in critically ill patients and are associated with adverse outcomes, including prolonged mechanical ventilation, increased risk of ventilator-associated pneumonia, and extended intensive care unit and hospital lengths of stay. The main physiological consequence of alveolar collapse and atelectasis is decreased FRC. FRC is the volume of gas remaining in the lungs at the end of expiration during spontaneous breathing at atmospheric pressure (Leith & Brown, 1999). It represents the equilibrium point for the spring out forces of the chest wall and the collapsing tendency of the lung (Rahn et al., 1946) and is a major determinant of pulmonary oxygenation capacity. The term end-expiratory lung volume (EELV) is used to indicate FRC, if PEEP is applied.

A previous observational study indicated that HFNO increases EELV and alveolar ventilation compared with low-flow oxygen therapy (Corley et al., 2011). In this study, the nHFTI significantly increased FRC compared with tHFTI, and additionally, increased FRC was highly correlated (Spearman’s correlation r_s = 1.0, P < 0.001) with increased PEEP.

We also observed that an increase in FRC was accompanied by a decrease in Vt, which can be attributed to increased expiratory resistance. While this study was conducted under simulated conditions, the observed increase in FRC suggests potential benefits in maintaining alveolar patency and improving ventilation-perfusion (V/Q) matching, which may support oxygenation in clinical settings. Although Vt decreased slightly—possibly affecting alveolar ventilation—the magnitude of this change was considerably smaller than the increase in FRC. For example, at a flow rate of 50 L/min, the nHFTI increased FRC by 278 mL compared to tHFTI, while decreasing Vt by only 54.8 mL on average. These results suggest that the new device may offer a favorable balance between lung volume recruitment and tidal volume preservation, which could support better oxygenation compared to the traditional device.

The difference between the alveolar pressure and the intrapleural pressure is known as transpulmonary pressure (Ptp). Ptp is the force that determines the expansion and contraction of the alveoli and is also key to maintaining alveolar patency at the end of expiration.

Our study demonstrated that increasing gas flow rates led to higher Ptp-EI and Ptp-EE, with a more pronounced increase observed with nHFTI compared to tHFTI. Elevated Ptp-EI reduces inspiratory work, while the concomitant rise in Ptp-EE increases expiratory work. These counteracting effects largely offset each other, resulting in only modest changes to the overall work of breathing.

In this study, Ptp-EI reached 11.4 ± 0.1 cmH₂O at a flow rate of 80 L/min, a level unlikely to cause alveolar overdistension and subsequent alveolar injury. Maintaining Ptp-EE ≥0 cmH₂O helps preserve alveolar patency at end-expiration. Clinically, Ptp-EE should be maintained at 0–5 cmH₂O to prevent end-expiratory alveolar collapse while minimizing the risk of barotrauma and shear stress-related injury. In this model, Ptp-EE with nHFTI at flow rates of 40–60 L/min fell within this recommended range (0–5 cmH₂O).

Increasing flow rate leads to a concomitant rise in peak airway pressure. Compared to tHFTI, the nHFTI increased both PIP and PEP. These elevated pressures, however, remains well below thresholds for pressure-induced lung injury. Notably, in ARDS management, it is the plateau pressure—rather than peak pressure—that is used to assess the risk of ventilator-induced lung injury, with current guidelines recommending maintenance below 30 cmH₂O.

In this crossover study, we found that the nHFTI, compared with tHFTI, mitigates the negative swing in airway pressure during inspiration (allowing for a limitation of the negative swing in inspiratory airway pressure). Importantly, reducing excessive negative inspiratory pressure swings in airway (and pleural) pressure helps prevent negative pressure pulmonary edema, which can cause lung injury and impair oxygenation (Yoshida et al., 2017). Concurrently, the increase in expiratory pressure with nHFTI may contribute to higher EELV, reduced shunt fraction, improved lung mechanics, and better oxygenation (Corley et al., 2011; Parke, Bloch & McGuinness, 2015; Mauri et al., 2017a; Mauri et al., 2017b).

Theoretically, HFT delivers warmed and humidified gas with a relatively constant oxygen concentration. However, at low flow rates, significant room air entrainment during inspiration results in a delivered FiO₂ substantially lower than the set value, compromising the accuracy of oxygen delivery and clinical monitoring. As flow increases, the delivered oxygen concentration rises due to reduced entrainment of room air. This study demonstrated that at flow rates of 40 L/min and above, the delivered FiO₂ closely matched the preset concentration, with no statistically significant difference between the two.

In recent years, HFTO has been increasingly utilized as a supportive therapy in tracheostomized patients challenging to wean from invasive mechanical ventilation. Clinical studies on HFTO, however, remains limited to case reports (Mitaka et al., 2018; Ramachandran, Jha & Sircar, 2021; Vadi, Phadtare & Shetty, 2021) and studies investigating HFTO surrounding the process of decannulation (Hernández Martínez et al., 2020; Lionello et al., 2023).

Our findings contribute to the limited evidence on HFTO use in tracheostomized patients, a practice that appears to be increasing (Corley et al., 2017). Conventional HFTO delivery resembles HFNO administration with an open mouth, during which the extra expiratory resistance vanished, and the PEEP effect disappeared (Parke, McGuinness & Eccleston, 2009; Parke, Eccleston & McGuinness, 2011; Luo et al., 2017). In the bench study of nHFTI incorporating the exhalation valve, we observed a PEEP effect of 3.5–7.5 cmH₂O at flow rates of 40–60 L/min, with a significant increase in FRC and a slight decrease in Vt. These findings highlight the efficacy of the new interface.

We must acknowledge several study limitations. As a benchtop investigation, we evaluated PEEP effects within a simulated respiratory system; consequently, the findings may not directly translate to clinical practice. A further limitation is the absence of gas humidification, which deviates from standard clinical care and may alter airway surface conditions—such as mucus viscosity and epithelial function—thereby reducing the physiological relevance of the results. Additionally, although the controlled environment of this model eliminates many clinical confounders and enables precise measurement of respiratory parameters, it inherently lacks patient-level variability. This means critical physiological responses—such as cardiovascular effects and safety risks related to auto-PEEP or dynamic hyperinflation—could not be assessed. Therefore, the speculative nature of clinical translation warrants caution in interpreting these results.

The study demonstrated that nHFTI increased PEEP while simultaneously reducing tidal volume, highlighting a potential trade-off between alveolar recruitment and ventilation efficiency. While this underscores the importance of selecting an appropriate inspiratory flow rate tailored to individual patient condition, such clinical decisions must ultimately be informed by future in vivo studies that rigorously evaluate both therapeutic efficacy and patient safety.

Supplemental Information