A 56-year-old man with SARS-COV2 infection was admitted to our hospital on November 1st, after 7 days of fever. Comorbidities comprised hypertension and obesity. On the third day after admission, he was transferred to our Respiratory Intensive Care Unit due to worsening of his respiratory condition, including notable shortness of breath with minimal exertion. On arrival, the patient was alert, haemodynamically stable. His respiratory rate was 25/min with a reservoir oxygen mask at 15L/min. Blood gas analysis (ABG) revealed a severe hypoxemia and alkalosis respiratory (pH 7.51, PaCO2 35mmHg, PaO2 118mmHg, HCO3− 27mmol/L, PaO2/FiO2 148). Due to patient’s gas exchange impairment and clinical progression, NHF therapy trial was attempted.

Initial temperature, inspired oxygen fraction (FiO2) and flow rate were 34°C, 60L/min and 60% respectively.

In order to reduce bio-aerosol dispersion, a surgical mask was placed over the patient’s mouth and nose.1Fig. 1 shows bilateral consolidations with ground-glass areas, mainly in the posterior dependent zones on CT-scan. Based on this radiological picture, we continued NHF in awake prone positioning. Fig. 2 illustrates oxygen pulse saturation/FiO2 (SpO2/FiO2), oxygen blood pressure/FiO2 (PaO2/FiO2) ratios and ROX index trends within the first four days of treatment. ROX index is defined as the ratio of pulse oximetry/fraction of inspired oxygen to respiratory rate (RR). This index, described by Roca et al.,2 identifies patients at risk of NHF failure. In particular, a ROX index of 4,88 is associated with a high risk of intubation. In conclusion, NHF setting was adjusted according to the maximum patient’s comfort and the best effects in terms of distending pressure and FiO2 stability. FiO2 was also targeted to achieve oxygen saturation between 94–96%. Prone positioning was implemented because we felt that the mild distension pressure generated by NHF could be enhanced by placing the ventilated areas in a dependent position, leading to increased V/Q mismatch.3

Figure 1.

CT-scan. Bilateral consolidations with ground-glass areas, mainly in the posterior dependent zones.

(0.53MB).

Figure 2.

SpO2/FiO2, PaO2/FiO2 and ROX index trends within the first four days of treatment. Patient’s SpO2/FiO2, PaO2/FiO2 and ROX index improve since NHF and prone position were implemented. (NHF: nasal high flow oxygen. SP: supine position, PP: prone position).

(0.14MB).

Thermo-humidified nasal high flow (NHF) oxygen therapy is increasingly used in the management of respiratory failure. This therapy has recently gained attention as an alternative non-invasive respiratory support in several clinical scenarios, acute and chronic,4 in order to enhance patient comfort and tolerance compared with standard oxygen by supplying heat and humidified mixture of air and oxygen at flows up to 60L/minutes.5

NHF can be delivered through different devices. Although few studies have compared the clinical effects of different NHF systems, the purpose of this paper is to describe the major benefits of NHF and to provide a quick guide on how to implement this therapy in daily practice. We have also included a brief description of the most frequently used NHF systems: BioRespira® (IBD), Airvo 2® and OptiFlow® (Fisher and Paykel), TNI softFlow 50® (Masimo) and Precision Flow® (Vapotherm).

High flows air/oxygen mixture can be beneficial for both chronic and acute patients, who need additional oxygen supply:

• -

  Matches patient’s inspiratory flow (stable FiO2). Conventional oxygen therapy devices such as Venturi masks or reservoir masks have holes to prevent carbon dioxide rebreathing. The patient inspires the delivered oxygen, but part of the air can be rebreathed. In addition, if the patient’s ventilatory demand exceeds the flows provided by the device, the patient will breathe part of atmospheric air. For this reason, because of the relationship with the patient-breathing pattern, FiO2 can be much lower than predicted. Pisani et al.4 observed that calculated FiO2 approached the prescribed FiO2 when the delivered gas flow rates were greater than the patient’s peak inspiratory flow rate. This means that, with low-flow systems, as the peak inspiratory demand increase, we cannot accurately estimate the FiO2 that the patient is breathing. NHF delivers a flow that is much higher than the patient’s spontaneous inspiratory flow in any condition, so actual FiO2 is close to the predicted FiO2.

• -

  Improves lung mucociliary clearance. In normal conditions, inhaled air is conditioned in the nasal airways. The air is warmed and humidified by the nasal mucosa because of its highly vascular sinusoid tissue. Humidification is always recommended for flow rates above 4L/min, because the humidification function of the nasal mucosa could be insufficient. In addition, unwarmed and dry gas leads to a poor tolerance of oxygen therapy.

• -

  Washes out anatomical dead space and carbon dioxide (CO2), allowing a higher fraction of minute ventilation to participate in gas exchange. Additionally, the nasal cannula, used for NHF, is the only respiratory interface that does not increase the instrumental dead space. With an oxygen mask, especially at low flow, carbon dioxide is rebreathed. NIV interfaces also increase the dead space. As a consequence, to maintain PaCO2 (and alveolar ventilation), the minute volume has to increase, either through respiratory rate or through tidal volume (VT), or even both. For these reasons, although NHF does not provide an active inspiratory support, it may improve the alveolar ventilation and decrease respiratory rate (RR).6 Fraser et al.7 found that NHF leads to a reduction in respiratory rate of 10%, and in some patients, this reduction was associated with a reduction in hypercapnia.

• -

  Provides a mild distending pressure. Although NHF is an open system, high flow from the nasal cannula increases the airway pressure usually around 3cmH2O,8 depending on the flow level and the nasal prong size in relation to the nostrils. Richie et al.6 conclude that NHF could deliver a clinically relevant mean positive airway pressure directly proportional to delivered gas flow rates. In addition, high flows could act as a resistance to exhalation when the mouth is closed.9,10 However, it remains unclear whether NHF is sufficient to increase lung volume or recruit the collapsed alveoli. Evaluating end-expiratory lung volume (EELV), Corley et al.11 reported increased end-expiratory lung impedance, suggesting increasing volumes and functional residual capacity with NHF compared with low-flow oxygen therapy. Also using electrical lung impedance tomography on healthy subjects in supine and prone position, Riera et al.12 found that NHF increased global EELV, suggesting an increase in functional residual capacity, regardless of the body position. This effect could be important in patients with a higher body mass index.

• -

  Attenuates inspiratory resistance and increases expiratory resistance. The attenuation of inspiratory resistance takes place essentially by providing an adequate warmed and humidified flow. It is known that in normal subjects,13 inhalation of cold air causes the activation of specific receptors or osmoreceptors in the nasal mucosa, which may induce bronchoconstriction through a mechanism mediated by the activation of muscarinic receptors.14 Besides, NHF is able to increase the expiratory resistance depending on the size of the nasal cannula and the patient’s expiratory flow.15

A typical NHF system consists of a flow generator, active heated humidifier, single-limb heated circuit, and nasal cannula (or tracheostomy interface). There are three types of flow generators: air-oxygen blender devices, built-in flow generators (turbines), and entrainment systems (Venturi Systems). The humidifier can be a traditional “pass-over heated humidifying system” or a “filter-type humidifying system” in which humidification is generated by passing gas through a bundle of narrow tubes, or which requires an evaporative surface in contact with the gas. A heated-wire inspiratory circuit is preferred to ensure delivery of adequately humidified medical gas as well as to avoid circuit vapour loss and condensation. The lower the ambient temperature, the more likely there is to be condensation. One major reason for the patient’s discomfort or intolerance is the interface. Beyond the circuit, condensation may also accumulate in the nasal prongs, which results in water droplet spray into the nostrils. NHF interfaces are specifically for this purpose. They can vary from a slender nasal cannula similar in appearance to a regular nasal oxygen cannula in which both the internal diameter and nasal prong bore are narrow, to a large bore nasal prong. Using slender nasal cannulas results in high flow out of the nasal prongs. The volumetric flow is the volume of gas that passes in a certain unit of time. The speed at which the volume moves is the velocity. The speed, at a constant volumetric flow, varies inversely proportional to the cross-sectional area of the tube. This is the reason why the smaller the cross-sectional area of the nasal cannula, the greater the speed (Fig. 3).16

Figure 3.

Fluid velocity, at a constant flow, varies inversely with the cross-sectional area of the tube.

(0.06MB).

• a)

  General considerations (regardless of the NHF device used)

• -

  Gas connection to the device (O2, and compressed air if required).

• -

  Place distilled water in the humidifier.

• -

  Connect the circuit, appropriate to the system used.

• -

  Switch on and heat up the water until the temperature reaches 31−37°C.

• -

  Select the size of the nasal cannula properly. When using a large bore nasal prong, it must not occlude more than 50% of the nostril.

• -

  Position the cannula correctly in the patient’s nostrils, to avoid lateral leakage.

• -

  Program the parameters, according to the equipment used (temperature, flow, and FiO2 if the setting is available).

• -

  Switch on the device (some devices allow parameters to be programmed once the machine is on).

• b)

  General monitoring

• -

  Patients must be spontaneously breathing, alert and must be able to protect the airway.

• -

  Check that there is no accumulation of condensed water in the tube, ensuring its inclination towards the heater.

• -

  Aspiration of patient’s secretions, if needed.

• -

  Check that the cannula is not occluded by a mucus plug or against the patient’s nose.

• -

  Make sure the patient is in a semi-seated position at 45°. NHF can be implemented in “awake prone position”, if needed.

• -

  Check that the device is not disconnected from the electrical source when using high-flow devices, because not many have an internal battery.

• c)

  General set up

• -

  Temperature: it must be targeted according to the patient’s comfort. The temperature needs to be set at higher values at cold weathers or high flows.

• -

  Flow: the maximum benefit of this treatment is achieved at higher flows, although the flow should also be titrated according to the patient's comfort. It is suggested starting with flows close to 50L/m, according to the patient's tolerance, in order to increase or decrease the flow.

• -

  FiO2: depending on the NHF device used, FiO2 can be a parameter to be programmed or a parameter that depends on the flow. By using an air-oxygen blender device, oxygen concentrations are stable. On the other hand, by using turbine devices, oxygen is supplied via a low-pressure system, while the device just monitors the oxygen concentration (FiO2). Any rise in the flow, will decrease the FiO2, and vice versa. Venturi high flow systems use a flow generator to create a high flow by Venturi. It is composed of a flow meter and an oxygen concentration monitor. Regardless of the system used, in all cases, FiO2 titration should depend on the SpO2 target, recommending an SpO2 range between 94% and 96%, or between 88%–92% in hypercapnic COPD patients.

Table 1 summarizes the main characteristics associated with the currently available devices.

Table 1.

Main characteristics of most frequent used NHF devices.

	BioRespira® IBD	Airvo2 (F&P®)	Optiflow (F&P®)	TNI softFlow 50 Masimo	Precision Flow Vapotherm®
FLOW (L/M)	5−120 (NHF AND CPAP USE)	10−60 (5L/M STEPS)	30- 60	1060 (0,5 STEPS)	5−40 (1L/M STEPS)
FLOW GENERATION	BLOWER	BLOWER	BLENDER	BLOWER	ELECTRONIC BLENDER
FIO2 INDEPENDENT OF SETUP FLOW	YES	NO	YES	NO	YES
POSSIBILITY OF FIO2 100%	YES	NO	YES	NO	YES
TYPE OF CONNECTION O2	HIGH PRESSURE SYSTEM O2 LOW PRESSURE SYSTEM O2	LOW PRESSURE O2 (UP TO 30L/M)	HIGH PRESSURE SYSTEM AIR AND O2	LOW PRESSURE SYSTEM O2 (UP TO 60L/M)	HIGH PRESSURE SYSTEM AIR AND O2
NEED FOR COMPRESSED AIR FROM THE WALL	NO	NO	YES	NO	YES
BUILT-IN HUMMIDIFIER	EXTERNAL GENERIC	INCORPOATED F&P	EXTERNAL F&P	INCORPOATED INTERSURGICAL®	INCORPORATED
	PASS OVER	PASS OVER	PASS OVER	PASS OVER	FILTER-TYPE
	SYSTEM	SYSTEM	SYSTEM	SYSTEM	HUMIDIFYING SYSTEM
TUBES (INSPIRATORY CIRCUIT)	GENERIC (HEATED WIRED SYSTEM RECCOMENDED)	SPECIFIC (HEATED WIRED SYSTEM)	GENERIC	SPECIFIC (HEATED WIRED SYSTEM)	SPECIFIC TRIPLE LUMEN PATIENT CIRCUIT HEATED BY WATER RESERVOIR
INTERFASE	GENERIC	OPTIFLOW INTERFASE (SILICONE)	OPTIFLOW INTERFASE. (SILICONE)	SPECIFIC: (SLENDER SILICONE HEATED WIRED NASAL CANNULA)	Hi-VNI, Vapotherm: (SLENDER SILICONE CANNULA)
		NASAL: 3 SIZES, REGARDLES THE FLOW SET	NASAL: 3 SIZES, REGARDLES THE FLOW SET	Standard-Plus: Flows up to 35L/m	TRACH MASK ADAPTER
				Large: Flows up to 60L/m
		TRAQUEAL INTERFASE	TRAQUEAL INTERFASE	Tracheal: Flows up to 50L/m
ON-SCREEN MONITORING	SET FIO2	MEASURED FIO2	MEASURED FIO2	MEASURED FIO2	SET FIO2
	SET FLOW	SET FLOW	SET FLOW	SET FLOW	SET FLOW
		SET GAS TEMPERATURE	SET GAS TEMPERATURE	SET GAS TEMPERATURE	SET GAS TEMPERATURE
	CONTINUOUS MONITORING OF PATIENT’S PARAMETERS: AIRWAY PRESSURE	CONTINUOUS MONITORING OF PATIENT’S PARAMETERS: NO	CONTINUOUS MONITORING OF PATIENT’’S PARAMETERS: NO	CONTINUOUS MONITORING OF PATIENT’S PARAMETERS: NO	CONTINUOUS MONITORING OF PATIENT’S PARAMETERS: NO
	SPO2
	BPM
	RR
	ROX INDEX

NHF: nasal high flow. FiO2: graction of inspired oxygen. O2: oxygen. SpO2: pulse oxygen saturation. BPM: beats per minute. RR: respiratory rate. ROX index: ratio of pulse oximetry/fraction of inspired oxygen to respiratory rate (SpO2/FiO2/RR).

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.