High Flow Tracheal Oxygen for Weaning of Tracheostomized Patients (HFTO-WEAN)

Background High Flow Nasal Cannula (HFNC) use has increased over the past decade in critically ill patients with acute hypoxic respiratory failure. HFNC involves the delivery of oxygen at a higher flow reaching up to 60 L/min, which is heated and humidified to enhance tolerability of the device. HFNC improves oxygenation and decrease work of breathing through several mechanisms including: (1) improving high oxygen delivery through decrease in the inspired fraction of inspired oxygen (FiO2) dilution, (2) decreasing the patient's work of breathing through matching of the patient's needed flow and the minimal amount of positive end-expiratory pressure (PEEP) exerted on the upper airway and alveoli caused by the constant flow, which could help in alveoli recruitment and improvement in oxygenation (up to 5 cm of end-expiratory pressure), (3) anatomical dead space washout leading to washout of carbon dioxide (CO2) and thus decreases the work of breathing by the continuous flow to the pharynx and upper part of the trachea.

Evidence on HFNC in critically ill patients Several randomized controlled trials (RCT) have assessed the role of HFNC in critically ill patients for different indications. In a systematic review of 9 RCTs, HFNC compared to conventional oxygen therapy, reduced the need for intubation (relative risk [RR] 0.85, 95% confidence interval [CI] 0.74-0.99, low certainty) or escalation of oxygen therapy to non-invasive positive pressure ventilation (NIPPV) (RR 0.71, 95% CI 0.51-0.98, low certainty) for patients with acute hypoxic respiratory failure without an increase in mortality. In addition, compared to NIPPV in acute hypoxic respiratory failure, HFNC showed similar risk for intubation (RR, 0.93; 95% CI 0.69-1.27, low certainty) without survival benefit. These results prompted a strong recommendation (moderate certainty) for the use of HFNC over conventional oxygen therapy in acute hypoxic respiratory failure from the European Society of Intensive Care Medicine. The use of HFNC was expanded to the peri-extubation period for patients at high-risk of extubation failure [5]. The high-risk population was defined as: age > 65, congestive heart failure, moderate-severe chronic obstructive pulmonary disease (COPD), APACHE II score > 12, body mass index > 30 Kg/m2, airway patency or secretion problems, difficulty weaning, ≥2 comorbidities, or duration of invasive mechanical ventilation >7 days. Evidence from four RCTs showed that HFNC use post-extubation reduced the need for intubation when compared to conventional oxygen therapy (RR 0.46, 95% CI 0.30-0.70), and it resulted in little to no difference in intubation rates when compared to NIPPV (RR 1.16, 95% CI 0.86-1.57). The utilization of HFNC for hypoxic respiratory failure has been extensively studied, but not for hypercapnic respiratory failure. Few observational studies and RCTs assessed the safety of HFNC in hypercapnic respiratory failure. When comparing HFNC to NIPPV for patients with hypercapnic respiratory failure, a systematic review of five RCTs showed no difference in intubation rates (odds ratio [OR]= 0.92, 95% CI 0.45-1.88) and mortality (OR 1.33, 95% CI 0.68-2.60). Similar results were obtained for patients with COPD. However, the evidence is limited by imprecision and heterogeneity and a conclusion about the safety of such an approach needs to be studied further.

Evidence of high-flow tracheal oxygen (HFTO) in tracheostomized patients Recently, HFTO has been used for weaning mechanical ventilation in tracheostomy patients at high risk of weaning failure. In a single-arm cross-over study, 14 mechanically ventilated patients through tracheostomy with prolonged weaning after more than 7 days from the first separation attempt according to Weaning according to a New Definition (WIND) study criteria were enrolled [16]. After a successful spontaneous breathing trial (SBT), the patients were placed on HFTO for 2 hours interrupted by 1-hour of conventional oxygen therapy through T-Piece to assess the physiological effect of HFTO. The study showed a lack of HFTO effect on the neuro-respiratory drive using electrical diaphragmatic activity measurement, work of breathing, and the ratio of peripheral arterial oxygen saturation to the inspired fraction of oxygen (PaO2:FiO2) ratio in comparison to conventional oxygen therapy. This study might suggest that HFTO works differently from HFNC. In contrast, another study of 26 tracheostomized patients who were weaned from mechanical ventilation and spontaneously breathing on conventional oxygen therapy for at least 24 hours, were randomly placed on different gas flow rates through HFTO for 30 minutes (10L /min, 30 L/min, or 50 L/min) with no washout period [17]. Compared to conventional oxygen therapy, HFTO improved the PaO2:FiO2 ratio but the CO2 remained stable. In addition, HFTO slightly reduced the respiratory rate and the negative swings in airway pressures during inspiration, and increase the mean and peak expiratory pressures at flows of 50 L/min. At a lower flow of 30 L/min, conventional oxygen therapy was similar to HFTO with regard to oxygenation and tracheal pressure. Of interest, although there was an increment in the measured tracheal expiratory pressure, it was lower than the pressures measured during HFNC. This could be explained by the limited resistance in tracheostomized patients in comparison to HFNC where the upper airway creates resistance and a higher expiratory pressure. A crossover RCT (n=20), in which tracheostomized patients were randomized to either HFTO at 50 L/min or T-piece and data was collected at 5 minutes and 15 minutes of each intervention. At 15 minutes, the patients on HFTO showed higher ratio of peripheral arterial oxygen saturation to the inspired fraction of oxygen (SPO2: FiO2) ratio and mean airway pressures with lower FiO2 requirements. However, respiratory rate and CO2 remained the same. Recently, a few case reports were also published about the utilization of HFTO for successful weaning, which showed that higher positive mean airway and tracheal pressures, improved PaO2: FiO2 ratio, and reduced inspiratory effort. Furthermore, a recent RCT (n=330) of tracheostomized critically ill patients who were weaned from mechanical ventilation and deemed ready for decannulation were randomized to either continuous HFTO with tracheostomy capping or capping with intermittent HFTO for 24 hours. Continuous HFTO utilization during the capping procedure was associated with shorter time to decannulation (median 6 [interquartile range (IQR) 5-7] days vs. 13 [IQR, 11-14] days) without decannulation failure.

Knowledge gap and aims of this study The evidence on the efficacy of using HFTO in patients with tracheostomy is limited. Such an approach might be costly due to the use of HFTO set and the need for a special interface to fit the HFTO through tracheostomy. If the use of HFTO proves to a safe and advantageous strategy, then the timing, population, dose, and average duration of weaning need to be addressed in controlled trials. In this pilot RCT, the investigators aim to explore the feasibility of conducting a powered RCT that examines the efficacy and safety of HFTO in weaning critically ill tracheostomy patients from mechanical ventilation.